gasification and pyrolysis characterization and heat transfer …757954/... · 2014-10-23 · msw...
TRANSCRIPT
Gasification and Pyrolysis Characterization and
Heat Transfer Phenomena During Thermal
Conversion of Municipal Solid Waste
Chunguang Zhou
Doctoral Dissertation
2014
KTH-Royal Institute of Technology School of Industrial Engineering and Management Department of Materials Science and Engineering
Division of Energy and Furnace Technology SE-100 44 Stockholm, Sweden
___________________________________________________________________________Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan I Stockholm framlägges för offentlig granskning för avläggande av teknologie doktorsexamen, måndagen den 10 November 2014, kl. 10:00 i Sal F3, Lindstedtvägen 26, Kungliga Tekniska Högskolan, Stockholm.
ISBN 978-91-7595-284-0
Chunguang Zhou. Gasification and Pyrolysis Characterization and Heat Transfer Phenomena During Thermal Conversion of Municipal Solid Waste KTH-Royal Institute of Technology School of Industrial Engineering and Management Department of Materials Science and Engineering Division of Energy and Furnace Technology SE-100 44 Stockholm, Sweden ISBN 978-91-7595-284-0
Copyright Chunguang Zhou (周春光)
Dedicated to my beloved parents and grandparents
谨以此文献给我挚爱的父母和祖父母
i
Abstract
The significant generation of municipal solid waste (MSW) has become a controversial global
issue. Pyrolysis and gasification technologies for treating rejects from solid waste disposal sites
(SWDSs), for which over 50 % of MSW is attributed to combustible species, have attracted
considerable attention. MSW is an alternative energy source that can partly replace fossil
resources; there is an increasing awareness that global warming caused by the utilization of fossil
resources is occurring.
The goal of this thesis is to realize the efficient and rational utilization of MSW and decrease the
harmful impact of pollutants, such as dioxin, HCl, and CO2, on the environment. To achieve this
goal, some fundamental studies have been experimentally and numerically conducted to enhance
the understanding of the properties of municipal solid waste thermal conversion.
In this thesis, the pyrolysis behaviors of single pelletized recovered fuel were tested. A detailed
comparison of the pyrolysis behaviors of typical recovered solid waste and biomass particles was
conducted. A swelling phenomenon with a swelling ratio of approximately 1.6 was observed on
the surface of pelletized recovered fuels. Subsequently, a particle model was constructed to
describe the thermal conversion process for large recovered fuel particles that are composed of a
high fraction of polyethylene (PE) and a comparable low fraction of cardboard. The results
indicate that an understanding of the heat transfer mechanism in highly porous and molten
structures and the selection of a heat transfer model are crucial for accurate prediction of the
conversion process.
MSW pyrolysis is a promising method for producing liquid products. With the exception of
lignocellulosic materials, such as printing paper and cardboard, PE, polystyrene (PS),
polypropylene (PP), polyethylene terephthalate (PET), and polyvinyl chloride (PVC) are the six
main polymers in domestic waste in Europe. Characterization studies of the products obtained
from these individual components, such as PE, PET, PVC, printing paper, and cardboard, have
been conducted on a pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) system
and a fixed-bed reactor. The possible pathways for the formation of the main primary/secondary
products in rapid and conventional pyrolysis were also discussed.
MSW steam gasification with CaO was performed in a batch-type fixed-bed gasifier to examine
the effect of CaO addition on the heat transfer properties, pollutant removal, and devolatilization
and char gasification behaviors in the presence of steam.
A new carbon capture and recycle (CCR) system combined with an integrated municipal solid
waste system was proposed. The foundation of the system is the development of a novel method
to remediate CO2 using a high-temperature process of reforming CH4 and/or O2 and/or H2O
without catalysts. Thermodynamic and experimental studies were performed. High temperatures
significantly promoted the multi-reforming process while preventing the problem of catalyst
deactivation. Potential improvements in the efficiency of the novel technology can be achieved
ii
by optimizing the reforming reactants. Landfill gas (LFG) and fuel gas from bio-waste treatment
contain a considerable fraction of CH4, which may be a source of CH4 for this process.
Keywords: municipal solid waste; pyrolysis; gasification; heat transfer; CaO; CO2; reforming;
numerical model.
iii
Acknowledgments
I am sincerely grateful to my supervisors Docent Weihong Yang and Professor Wlodzimierz
Blasiak for their constant encouragement and guidance during my studies at KTH.
I am also deeply indebted to my supervisors at Zhejiang University, Prof. Shurong Wang and
Prof. Mengxiang Fang, for helping me venture into an interesting and promising research area. I
appreciate Madam Yingfang He for her help in my studies abroad.
I am grateful to the European Commission for funding a scholarship under the Erasmus Mundus
Programme for my PhD research. I am also thankful to JERNKONTORET for funding my 6-
month scholarship, which enabled me to complete this thesis.
I am grateful to Dr. Efthymios Kantarelis, who assisted with my work in the lab. I sincerely
appreciate Dr. Qinglin Zhang and Mr. Liran Dor for their discussions regarding municipal solid
waste gasification and Pelle Mellin for his help with the application for the JERNKONTORET
scholarship.
I thank Ståhlgren Mathias and Per-Håkan Nilsson for their assistance in supplying plastic
granules.
I am grateful to my colleagues at the Division of Energy and Furnace Technology, Dr. Xiaolei
Zhang, Dr. Jun Li, Dr. Yueshi Wu, Mohsen Pour, Duleeka Gunarathne, Mersedeh Ghadamgahi,
Condo Adolfo, and Carlos Cuvilas, for their help and pleasant discussions.
I thank Dr. Thorsak Kittikorn, Dr. Baolin Guo, Xianfeng Hu, Dr. Zuogang Guo, Bo Wei, Dr.
Yingying Zhu, Rui Li, Xin Du, Shen Li, Cong Wang, and my badminton partners at the MSE
Department for their endless assistance.
I am grateful to my beloved parents and grandparents, my brother Chiguang, and my sisters
Zhaoguang and Fanguang, for their continuous support during my research. I extend a special
thank you to my beloved wife Yuehui Chen. Without her endless love and support, I would not
have been able to complete my PhD research.
Chunguang Zhou
2014-09-30 Stockholm, Sweden
iv
List of Papers in the Thesis
I. Chunguang Zhou, Qinglin Zhang, Leonie Arnold, Weihong Yang, Wlodzimierz Blasiak.
A study of the pyrolysis behaviors of pelletized recovered municipal solid waste fuels,
Applied Energy. 107:pp.173-182.2013.
II. Chunguang Zhou, Weihong Yang. Effect of heat transfer model on the prediction of
municipal solid waste (MSW) pyrolysis process. Revised version, Submitted to Fuel,
second round of review, June, 2014.
III. Chunguang Zhou, Weihong Yang, Wlodzimierz Blasiak. Characteristics of waste printing
paper and cardboard in a reactor pyrolyzed by preheated agents. Fuel Processing
Technology. 116:pp.63-71.2013.
IV. Chunguang Zhou, Thomas Stuermer, Rathnayaka Gunarathnea, Weihong Yang,
Wlodzimierz Blasiak. Effect of calcium oxide on high-temperature steam gasification of
municipal solid waste. Fuel.122: pp.36-46.2014.
V. Chunguang Zhou, Lan Zhang, Artur Swiderski, Weihong Yang, Wlodzimierz Blasiak.
Study and development of a high temperature process of multi-reformation of CH4 with
CO2 for remediation of greenhouse gas. Energy. 36:pp.5450-5459.2011.
Contribution Statement:
In Paper V, I performed the simulations, results analyses, and writing.
In the remaining papers, I performed all simulations, experimental tests, data analyses, and
writing.
v
List of Papers Omitted From the Thesis
I. Chunguang Zhou, Weihong Yang. Characterization of the products from spruce and pine
sawdust pyrolysis at various temperatures. 21st European Biomass Conference and
Exhibition, 2013, Copenhagen.
II. Chunguang Zhou, Weihong Yang. Study of the heat transfer properties and gasification
behaviors of a single solid waste particle for Plasma Gasification Melting. The 27th
international conference on solid waste technology and management, Philadelphia,2012,
USA.
III. Pelle Mellin, Efthymios Kantarelis, Chunguang Zhou, and Weihong Yang. Simulation of
bed dynamics and primary products from fast pyrolysis of biomass: steam compared to
nitrogen as a fluidizing agent. Industrial & Engineering Chemistry Research.
IV. Pelle Mellin, Qinglin Zhang, E. Kantarelis, Chunguang Zhou, W. Yang. Accuracy and
potential use of a developed CFD-pyrolysis model used for simulating lab-scale bio oil
production. 20sthEuropean Biomass Conference and Exhibition, 2012, Milan, Italy.
V. Qinglin Zhang, Leonie Arnold, Chunguang Zhou, Yueshi Wu, Weihong Yang, et al. A
thermodynamic analysis of waste to energy systems using plasma gasification melting
technology. The 31st international conference on thermal treatment technologies &
hazardous waste combustions, October, 2012, USA.
vi
Contents Chapter 1 .................................................................................................................................................. 1
Introduction ............................................................................................................................................. 1
1.1 Background..................................................................................................................................... 1
1.2 Municipal solid waste thermal conversion process ........................................................................... 2
1.2.1 Incineration ................................................................................................................................. 2
1.2.2 Pyrolysis ..................................................................................................................................... 3
1.2.3 Gasification................................................................................................................................. 4
1.2.4 MSW to RDF or SRF ................................................................................................................ 5
1.3 Problem statement .......................................................................................................................... 5
1.3.1 Integrity and volume changes during thermal conversion......................................................................... 6
1.3.2 Heat transfer properties in the porous structure ................................................................................... 6
1.3.3 Liquid product characterizations ...................................................................................................... 8
1.3.4 MSW gasification ........................................................................................................................ 9
1.3.5 Greenhouse gas remediation .......................................................................................................... 10
1.4 Objective and framework of the thesis .......................................................................................... 11
1.5 Outline of the study and objectives of this thesis ........................................................................... 12
Chapter 2 ................................................................................................................................................ 14
2 Literature review .................................................................................................................................. 14
2.1 Kinetic modeling ........................................................................................................................... 14
2.2 Modeling of chemical and physical processes ................................................................................ 16
2.3 Heat transfer models ..................................................................................................................... 18
2.4 Experimental studies on the products of MSW pyrolysis and gasification ...................................... 19
2.4.1 Pyrolysis ................................................................................................................................... 19
2.4.2 MSW gasification ...................................................................................................................... 21
Chapter 3 ................................................................................................................................................ 24
3 Methodology ....................................................................................................................................... 24
3.1 Experimental methodology ........................................................................................................... 24
3.1.1 Materials .................................................................................................................................. 24
3.1.2 Experimental facility................................................................................................................... 25
3.1.3 Characterization of the products .................................................................................................... 28
3.2 Modeling study for MSW particles ................................................................................................ 29
3.2.1 Melting behavior of polyethylene ..................................................................................................... 30
3.2.2 Kinetics model ............................................................................................................................ 31
vii
3.2.3 Mass and heat balance equations ................................................................................................... 32
3.2.4 Effective thermal conductivity due to thermal radiation ....................................................................... 33
3.3 Thermodynamic analysis ............................................................................................................... 40
Chapter 4 ................................................................................................................................................ 42
4. Pyrolysis behaviors of pelletized recovered municipal solid waste fuels ............................................... 42
4.1 Results and discussion ................................................................................................................... 42
4.1.1 Heat transfer properties ............................................................................................................... 42
4.1.2 Devolatilization behavior ............................................................................................................. 43
4.1.2 Swelling/shrinkage phenomena during the pyrolysis process ................................................................. 45
4.1.2 Char properties .......................................................................................................................... 48
4.2 Summary ....................................................................................................................................... 51
Chapter 5 ................................................................................................................................................ 53
5. Heat transfer mode in porous MSW particles ...................................................................................... 53
5.1 Results and Discussion .................................................................................................................. 53
5.1.1 TGA-DSC analysis ................................................................................................................... 53
5.2.2 Prediction of MSW pyrolysis ........................................................................................................ 56
5.2 Summary ....................................................................................................................................... 62
Chapter 6 ................................................................................................................................................ 64
6. Pyrolysis of the main components of MSW ........................................................................................ 64
6.1 Results and Discussion .................................................................................................................. 64
6.1.1 Product yields distribution ............................................................................................................ 64
6.1.2 Syngas characterization ................................................................................................................ 66
6.1.3 Liquid product characterizations .................................................................................................... 72
6.1.4 Elemental analysis of chars from paper board pyrolysis ....................................................................... 89
6.2 Summary ....................................................................................................................................... 91
Chapter 7 ................................................................................................................................................ 93
7. Steam gasification of MSW with CaO ................................................................................................. 93
7.1 Results and Discussion .................................................................................................................. 94
7.1.1 Heat transfer properties ............................................................................................................... 94
7.1.2 Evolutionary behavior of syngas composition ..................................................................................... 95
7.1.3 Syngas flow rate ......................................................................................................................... 96
7.1.4 Solid analysis ............................................................................................................................ 98
7.1.5 Mechanistic discussion of catalytic effect ......................................................................................... 102
7.2 Summary ..................................................................................................................................... 104
viii
Chapter 8 .............................................................................................................................................. 106
8. Multi-reformation of CH4 with CO2 for the remediation of greenhouse gas ...................................... 106
8.1 Results and Discussion ................................................................................................................ 107
8.2 Summary ..................................................................................................................................... 112
Chapter 9 .............................................................................................................................................. 113
9 Conclusion and future studies ............................................................................................................ 113
9.1 Conclusion .................................................................................................................................. 113
9.2 Future studies ............................................................................................................................. 115
References ............................................................................................................................................ 116
ix
Abbreviations
BET Brunauer-Emmett-Teller
CCR Carbon capture and recycle
DSC Differential scanning calorimetry
DTA Differential thermal analysis
EDS Energy-dispersive X-ray spectroscopy
FF Furfural
GC-MS Gas chromatography-mass spectrometry
GHG Greenhouse gas
GWF Global warming factor
GWP Global warming potential
HA Hydroxyacetone
HAA Hydroxyacetaldehyde
HiTAC High-temperature air combustion
HMF Hydroxymethyl-furfural
HTAG High-temperature steam/air gasification
KTH Kungliga Tekniska högskolan
LFMR Landfill mining and reclamation
LFG Landfill fuel gas
MSW Municipal solid waste
PAHs Polycyclic aromatic hydrocarbons
PCDD Polychorinated dibenzo-p-dioxin
PCDF Polychlorinated dibenzofuran
PMMA Polymethyl methacrylate
PE Polyethylene
PET Polyethylene terephthalate
x
PP Polypropylene
PVC Polyvinyl chloride polymer
Py-GC/MS Pyrolysis gas chromatography mass spectrometry
RDF Refuse-derived fuel
SEM Scanning electron microscopy
SRF Solid recovered fuel
SWDS Solid waste disposal site
TGA Thermogravimetry
VOCs Volatile organic compounds
Symbols
𝐴𝑒 Pre-exponential factor s-1
𝐴𝑖 Pre-exponential factor in the 𝑖th reaction rate s-1
𝑎1 Stoichiometric parameter for cardboard degradation
𝑎2 Stoichiometric parameter for cardboard degradation
𝑎𝑠𝑘 Stoichiometric parameters for yield of gases
𝐷 Heating rate K·min-1
𝑑𝑝 Voids center distance or distance between neighboring particle surfaces
m
𝒅𝒑,𝑐𝑎𝑟𝑑𝑏𝑜𝑎𝑟𝑑 Particle diameter m
𝐸 Activation energy kJ mol-1
𝐸𝑖 Activation energy in the 𝑖th reaction rate kJ mol-1
𝐹 Radiation exchange factor
xi
𝑓𝑐𝑜𝑛𝑡𝑎𝑐𝑡 Contact surface
(𝐻/𝐶)𝑒𝑓𝑓 Effective hydrogen index
𝐻𝑚 Heat of melting J·g-1· K-1
ℎ External heat transfer coefficient W·m-2·K-1
ℎ𝑟 Radiation heat transfer coefficient W·m-2·K-1
ℎ𝑠𝑘 Enthalpy of solid J·kg-1
ℎ𝑔𝑗 Enthalpy of gases J·kg-1
𝑘 Reaction rate kg·m-3·s-1
𝑘𝑠𝑟 Thermal conductivity accounting for void-to-void radiation
W·m-1·s-1
𝑘𝑒𝑓𝑓 Effective thermal conductivity W·m-1·s-1
𝑘𝑒𝑓𝑓𝑠𝑐 Thermal conductivity through the contact surface W·m-1·s-1
𝑘𝑒𝑓𝑓𝑠𝑟 Thermal conductivity accounting for radiation W·m-1·s-1
𝑘𝑒𝑓𝑓𝑠𝑓
Thermal conductivity within the voids filled by the gas W·m-1·s-1
𝑘𝑔 Gas thermal conductivity W·m-1·s-1
𝑘𝑠 Solid thermal conductivity W·m-1·s-1
𝑚 Initial mass of pellet kg
𝑚0 Transient mass of pellet kg
𝑚𝑐 Concentration of cardboard kg·m-3
𝑚𝑐ℎ𝑎𝑟1 Concentration of intermediate kg·m-3
𝑚𝑐ℎ𝑎𝑟2 Concentration of final char kg·m-3
xii
𝑚𝑔𝑎𝑠 Concentration of gases kg·m-3
𝑚𝑃𝐸 Concentration of PE kg·m-3
𝑚𝑢𝑛𝑖𝑡 Rate of mass generation of the solid phase per unit volume
kg·m-3
𝑛 Avrami exponent
𝑄𝐶 Heat source term
𝑞𝑟 Radiation heat flux J·kg-1·m-2
𝑅 Constant
𝑅𝑉 Volumetric swelling and shrinkage ratio
𝑅𝑏 Particle radius of the MSW cylinder m
𝑟 Radius coordinate m
𝑇 Temperature K
𝑇0 Onset temperature of the melting stage K
𝑇∞ Completion temperature of the melting stage K
𝑇𝑓 Furnace temperature K
𝑡 Time s
𝑡0 Onset time of the melting stage s
𝑡∞ Completion time of the melting stage s
𝑉𝑐𝑎𝑟𝑑𝑏𝑜𝑎𝑟𝑑 Initial volume fraction of cardboard m
𝑉0 Initial volume of pelletized pellet m3
𝑉𝑝𝑒 Initial volume fraction of PE m3
xiii
𝑉𝑡 Transient volume m3
𝑉𝑢𝑛𝑖𝑡 Unit volume m3
𝑋 Volatile conversion of pelletized material
𝑋𝑀 Degree of de-crystallinity during heating
𝑋𝑝𝑒 Conversion rate of PE
𝑍𝑡 Crystallization rate constant
Greek symbols
∅𝑟 Volatile flux toward bed surface at radius kg·m-2
α Conversion degree
β Heating rate K·min-1
𝛼𝑟𝑠 Radiation coefficient for the solid
𝛼𝑟𝑣 Radiation coefficient for the voids
𝜌𝑠𝑘 Density kg·m-3
𝜎 Boltzmann radiation constant W·m-2·K-4
𝜀 Porosity
𝜖 Emissivity
Subscripts
xiv
𝑓 Furnace
𝑀 Melting
𝑝𝑒 Polyethylene
𝑟𝑠 Radiation from particle surface to particle surface
𝑟𝑣 Radiation from void to void
1
Chapter 1
Introduction
1.1 Background
Municipal solid waste (MSW) refers to byproducts that are generated during the
urbanization of human society, such as product packaging, furniture, clothing, bottles,
food waste, newspapers, and batteries [1]. Typical waste compositions of MSW in Europe
are listed in Table 1.1. According to estimates from the Global Waste Management
Market Report (2007), the maximum total amount of municipal solid waste that was
globally generated in 2006 was 2.02 billion tons [2]. Reliable forecasts indicate that the
generation of global MSW is expected to increase to 4.9 billion tons per year by 2054 [3].
Table 1.1 Waste composition of municipal solid waste in the EU.
Fraction Sweden (1991) [4] France (1991) [4] Finland (1991) [4] Europe (2009) [5]
Organic 30 24 28 35
Paper 40 32 29 22 Plastic 9 7 7 10
Glass 7 10 4 6 Metal 3 4 4 4
Textile 3 3 5 3
Other 8 20 23 20
Sum 100 100 100 100
Landfilling is a conventional disposal method that requires a large amount of land for
dumping MSW. These landfill sites are limited by local geology and the natural stability of
the underground soil because poorly maintained landfill sites are prone to contaminating
groundwater. Landfill fuel gas (LFG) that consists of approximately 50-60 vol. %
methane and 30-40 vol. % CO2 from the anaerobic decomposition of solid waste and
wastewater is a significant contributor to greenhouse gas (GHG) emissions, which is
estimated in the range of 19-40 Tg/year [6] and [7]. Methane is a particularly potent GHG
because it has a global warming potential (GWP) that is 25 times the GWP of CO2 for a
100-year time-horizon GHG. Thus, landfilling is not an environmentally friendly disposal
method for MSW.
In the majority of the member states in Europe, a shift from landfilling toward preferred
waste management approaches has begun [8]. All collected MSW is sent to solid waste
2
disposal sites (SWDSs) for separation and sorting via a combined process of manual
inspection and automatic selection. Recycling is considered to be a preferred waste
management method because recycled materials replace an equal or almost equal quantity
of virgin materials in a closed-loop recycling system. All countries in Europe significantly
increased their material recycling rate (as a percentage of municipal waste generation)
during the period 2001-2010. Sweden increased this rate from 29 % to 36 % [8]. These
recyclable materials primarily include metals, plastics, glass, and paper products.
Alternative methods for treating recyclable plastic waste include mechanical recycling and
chemical recycling processes. Mechanical recycling aims to reuse the material in new
applications by melting and remolding the recyclable plastic waste. Thus, a complex
separation system is required to reduce the impurities that may be mixed with recyclable
plastics. Post-consumer recycled plastics are not permitted in the food sector. Raw plastic
materials with high purity are always mixed with some additives to produce different
products according to market demand. Generally, the simultaneous disposal of recyclable
materials and impurities is simple and economical. Chemical recycling involves a series of
processes in which polymers are broken down by heat or chemical processes to produce a
large variety of chemical products.
Bio-waste, which constitutes a relatively small proportion of total municipal waste, is also
separated and recycled at a low recycling rate in European countries. Bio-waste includes
food and garden waste but does not include wood, paper, cardboard, and textile waste. In
2010, the recycling rate (as a percentage of municipal waste generation) in Sweden was
approximately 13 % [8]. Some biochemical conversion technologies, including anaerobic
digestion, composting, and bioreactor landfilling, are used to treat bio-waste with
products or by-products of methane, carbon dioxide, compost, and digestate.
Improved waste management is an essential element in efforts to enhance the resource
utilization efficiency. After recycling and the separation of the bio-waste at the solid waste
disposal sites (SWDSs), more than 50 % of the composition of the rejects involves
combustible species. Thus, this fraction of MSW can be used as an energy source.
Moreover, the solid wastes that have previously been landfilled may also be excavated and
processed. This process is called as landfill mining and reclamation (LFMR).
1.2 Municipal solid waste thermal conversion process
Recently, thermochemical treatment processes for energy recovery from MSW have
become an essential component of an integrated municipal solid waste management
system. Some primary waste-to-energy technologies for MSW treatment are shown in
Figure 1.1.
1.2.1 Incineration
3
Approximately 130 million tons of waste is currently incinerated across 35 countries [9].
The main advantages of incineration include a significant reduction of waste in terms of
mass (70-80 %) and volume (90 %), a significant conservation of land, and the reduction
of GHG emissions from anaerobic decomposition. Among the thermochemical
treatment processes, grate combustion is the oldest method; however, it is extensively
used industrially for power generation or heating systems. Grate combustion is also the
most advanced method in the waste disposal hierarchy compared with dumping,
landfilling, and composting. However, some environmental problems have also emerged
because incineration does not completely eliminate MSW and numerous toxic pollutants
are produced in the contaminated residues, such as bottom ash, fly ash, and dioxin [10]. A
detectable risk of some cancers in the populations living near the sites was reported based
on a study of the health effects related to active incinerators from 1969-1996 [11]. Thus,
the need to develop alternative new technologies has gained significant attention from an
environmental standpoint and an efficiency energy utilization standpoint.
Heat,
Electricity
Materials
Temperature
CH4, CO2,
etc
CO,H2, etc
DME/ME,
F-T,etc
Pyrolysis
oil
Direct gasification
Synthetic reactions
Non-charring
materials
Pyrolysis (plastics)
Charring materials
Pyrolysis
(cellulosic
materials)
Energy
Pre-treatment
(MSW-
RDF,separati
on and
sorting)
LFG collection,
Anaerobic
digestion
(Bio-waste)
Solid
Dry reforming of
methane
Indirect gasification
EthanolFermentation
(Bio-waste)
Combustion(MSW incinerator,
gas/steam turbine)
1500-2000 °CRoom
Liquid
Gas
Energy flow/Steam flow
Feedstock/solid residue flow
Gaseous fuel flow
Scope of this study area
Ma
in p
rod
uc
ts
Figure 1.1 Waste-to-energy technologies for municipal solid waste
1.2.2 Pyrolysis
In addition to the traditional thermal conversion method of incineration, the energy
efficient, environmentally friendly, and economically effective technologies of pyrolysis
and gasification have attracted significant attention. Pyrolysis is an endothermic
thermochemical process involving carbonaceous material decomposition in an oxygen-
4
deficient atmosphere to produce pyrolytic oil, gas, and solid char residues. The process
can be conducted using various types of equipment, such as an entrained-flow reactor, a
fixed-bed gasifier, a cyclone gasifier, fluidized beds, and a plasma furnace. The yield and
composition of the products is dependent on the type of materials, the temperature of the
processes, the heating rate of the samples, the gas residence time, the presence of catalysts,
and the presence of hydrogen donors, such as hydrogen and steam vapor. Organic carbon
materials in MSW can be divided into non-charring materials and charring materials,
which correspond to plastic mixtures and lignocellulosic materials, respectively. Two
approaches can be employed according to the target products: the first approach is
referred to as fast or flash pyrolysis, which aims to maximize the yield of liquid product
by reducing the residence time and increasing the heating rate, and the second approach is
referred to as traditional or conventional pyrolysis, which aims to maximum the fuel gas
yield in conditions involving a high temperature, a long gas residence time and a low
heating rate. Pyrolysis, which is regarded as the first stage of thermal conversion, is
conducted at a low temperature and begins prior to gasification or combustion, which
occurs at a high temperature. Due to the lower temperature and reducing conditions, the
pyrolysis of MSW has the advantage of reducing and preventing corrosion and emissions
by retaining alkali, heavy metals, chlorine and sulfur within the process residues,
preventing dioxin and PCDD/PCDF formation, and reducing thermal NOx formation
[12].
1.2.3 Gasification
Gasification is a thermal conversion process with a high energy utilization efficiency for
producing combustible gases that are abundant in CO and H2 via partial combustion or a
heat supply from external sources with temperatures above 700 °C. Gasification enables
the use of solid fuel in gas applications, including chemical synthesis, small-scale electricity
production in motors or turbines, combustion in industrial furnaces, and reducing gas in
steel plants. From a global warming and economic standpoint and compared with power
generation in a nuclear-dominated country such as Sweden, a more significant global
warming factor (GWF) benefit can be expected if the same amount of syngas is used to
replace reducing gas from fossil fuel coal in steel plants or to synthesize liquid chemical
fuels from crude oil refineries.
Gasification is an endothermic process. In terms of the heat sources that are required to
gasify a solid fuel into gaseous products, two types of gasification processes exist: the first
process is referred to as direct gasification or conventional gasification, in which reaction
heat from the partial combustion of feedstock is employed, and the second process is
referred to as indirect gasification, in which sensible heat from preheated gasification
agents, hot sand or plasma is employed. In some technologies discussed in the literature,
preheated agents, such as high-temperature steam, have been used to improve the
efficiency of the gasification process [13], [14], and [15]. In a high-temperature steam/air
5
gasification system (HTAG), steam agents are preheated in a combustion chamber
(HiTAC) and subsequently added to a traditional fixed-bed gasifier [14]. In an MSW
plasma gasification system, the heat source for preheating steam agents originates from a
series of plasma torches [15]. The adoption of an indirect gasification system can produce
the following benefits: first, the total calorific value of syngas increases; second, the
concentrations of the two main combustible gases—CO and H2—increase because the
dilution of N2 from air can be prevented.
1.2.4 MSW to RDF or SRF
Compared with biomass resources, the quality of municipal solid waste is more regionally
dependent and can vary over a more extensive range. Nearly 45–50 % by mass of
household waste is combustible, and certain sources can generate 85–90 % combustible
waste by mass [ 16 ] and [ 17 ]. The particle sizes also significantly vary. To provide
feedstock with relatively stable physical and chemical characteristics, the combustible
fraction in the municipal solid waste (known as refuse-derived fuel (RDF) or solid
recovered fuel (SRF)) is mechanically sorted and subsequently pelletized at the production
site to reduce the cost of transportation. Because the RDF is dried and pelletized during
the pre-treatment process, the maximum heating value of the RDF ranges from 4000–
6000 kcal/kg, and the quality can be comparably stabilized throughout the year [18] and
[19]. Because the introduction of pelletized fuel enables a variety of fuels to be used in the
same gasifier, pelletized feedstock is extensively used in large-scale commercial plants [20].
Other benefits include lower pollutant emissions and a lower ash concentration. Thus, the
mechanical separation of the incombustible fraction from MSW can enhance the energy
density per unit weight of feedstock for pyrolysis and gasification reactors and improve
the energy utilization efficiency. This process can be conducted at solid waste disposal
sites (SWDSs). The separated materials include HDPE, film (LDPE), PET, food/drink
cartons, and mixed plastic. Each obtained material is compacted and sent to the
corresponding recycler. Both mechanical and chemical recycling methods can be used to
treat plastics based on their purity and quality. The rejected fraction can be used for
gasification and incineration. Some rejected materials from a waste separation and
classification industrial plant have a plastic concentration that varies from 97.03 % to
43.46 %, whereas the concentration of paper materials ranges from 0.35 % to 33.25 %
[21]. Pyrolysis and gasification are the preferred method for producing RDF with a high
concentration of plastic and paper materials.
1.3 Problem statement
An understanding of the pyrolysis characteristics of MSW particles and the main
components of MSW is crucial for the design and operation of MSW thermal treatment
facilities. Considerable effort regarding experimental studies and the development of
6
reliable models of coal and biomass pyrolysis has been made over the past several decades.
However, relatively few related studies of municipal solid waste have been performed.
Although lignocellulose material is one of the main components in RDF, some large
differences in the physical and chemical properties between RDF and biomass thermal
degradation remain. A typical recovered solid waste particle contains a considerable
fraction of plastics, which exhibit different degradation properties compared with the
degradation properties of lignocellulose fuels. High-density polyethylene (HDPE), low-
density polyethylene (LDPE), polystyrene (PS), polypropylene (PP), and polyethylene
terephthalate (PET) are six polymers that are most commonly found in domestic waste;
they represent more than two-thirds of the polymer production in western Europe [22].
Of these six polymers, polyethylene in the form of disposable bags represents more than
60 % of MSW [23 ]. Thus, new phenomena that have emerged during the thermal
conversion of MSW were investigated in this thesis.
1.3.1 Integrity and volume changes during thermal conversion
The physical integrity of fuel is a critical factor in the formation of a bed with proper void
spacing, which facilitates the evolution of volatile components and gas-solid reactions in a
fixed-bed gasifier. Recovered solid waste is known to melt and become sticky at high
temperatures due to its high plastic content. The test conducted by Marsh et al. [24]
showed that 50 mm RDF pellets possess minimal compressive strength once the core
temperature of the pellets reached 150-200 °C. Therefore, it is unlikely that these RDF
pellets would have enough integrity to maintain the bed height in the gasifier.
The swelling behavior and the generation of a sticky material may trigger bridging in the
pyrolysis zone, which is a phenomenon in which solid fuel forms a stable structure across
an opening and causes interruptions in the material flow in the reactors [25]. In industry,
the free swelling index is used as a measure of the amount of swelling experienced by coal.
Coal with free swelling indices greater than 2.7 is deemed to be impractical in certain
types of fixed-bed gasifiers [26] and [27]. For some feedstock with a high swelling index,
such as certain types of coal, an automatic poking system or stirrer in the top portion of
the reactor is required to break up any bridging that occurs during devolatilization [26] [28]
and [29] because the gases that cause the swelling can be released from the coal structure
via agitation or poking. However, this additional process consumes a large amount of
energy and increases the complexity of the gasifier. An alternative gasifier design is
required for feedstocks with high swelling properties and a tendency to generate sticky
and melted materials during devolatilization. Thus, the pyrolysis behaviors of feedstock,
which are related to volume change, are crucial for the design and operation of a gasifier.
1.3.2 Heat transfer properties in the porous structure
7
Typical wooden particles can shrink and lose approximately 40 % of their volume after
decomposition, as shown in Figure 1.2 (a) and (b). The solid char residue remains a
compact structure. The main heat conduction mechanism in the unreacted wood and char
layer can be attributed to heat conduction. However, RDF particles consist of different
components, which are primarily classified into the lignocellulose group and the plastic
group and present a heterogeneous structure, as shown in Figure 1.2 (c). The char layer in
an RDF particle exhibits a porous structure after non-charring materials, such as
polyethylene, lose the majority of their volume during pyrolysis. RDF materials that are
not pelletized exhibit considerable porosity, on the order of 65.1 % for large bricks
composed of cardboard and polyethylene, according to Salvador et al. [30]. The porosity
of the packing structure increases with the continuous consumption of plastic during
RDF pyrolysis if the entire structure does not collapse. Porosity serves an important role
in the accurate estimation of effective thermal conductivity, which is a crucial transport
property for modeling in a porous medium [31]. However, a suitable model that describes
the pyrolysis process of MSW particles with high plastic concentrations has not been
identified in the literature. The only empirical model proposed and used in the MSW
combustion process, as developed by Salvador [32], was considered to be inadequate
when the PE concentration exceeds 30 %. Significant differences in the physical and
chemical properties between RDF and biomass thermal degradation may contribute to
this inadequacy. One goal of this thesis is to construct a proper heat transfer model for a
numerical particle model that correctly describes the mass- and heat-transfer processes
during municipal solid waste decomposition.
Drying front
Residue charPyrolysis layer
Molten layer
Macro-porous layer Pyrolysis layer of
component #2
Pyrolysis layer of
component #1
Molten layer
Drying layer
Component
#1
Component
#2
(b) Layers fronts and control volume inside a large wooden
particle during devolatilization
(c) Layers fronts and control volume inside a large hetergeneous
RDF particle during devolatilization
Homogeneous
structure
Heterogeneous
structure consisting
component #1,#2,etc
Time (s)
Co
nv
ers
ion
ra
te o
f s
olid
pa
rtic
le (
%)
100
(a) A typical wooden particle conversion curve and particle
volume changing during devotilization process
Figure 1.2 Volume change of wooden and RDF particles during decomposition
8
1.3.3 Liquid product characterizations
The majority of popular transportation fuels, such as gasoline, kerosene, and diesel, are
primarily produced from crude oil. Gasoline is a mixture of paraffins, olefins, naphthenes
and aromatics in the C5-C10 range. The preferred structure for diesel fuels is linear and
consists of saturated hydrocarbons characterized by cetane number. The concerns of
climate change and the depletion of crude oil are driving the conversion of alternative
feedstock to crude oil. A promising method for realizing the targets is to produce liquid
products from MSW pyrolysis and upgrade the liquid into hydrocarbon fuels using
hydrotreating or hydroprocessing in conventional oil refineries because the use of existing
standard refinery units reduces the capital investment.
MSW exhibits a significant potential to satisfy the challenges of finding energy sources
with a renewable nature and a sustainable supply due to its large global quantities. It can
be converted to desired products using catalytic or non-catalytic pyrolysis that is either
fast or conventional. Many of the oils derived from MSW contain paraffins, olefins and
aromatics. However, the products are highly dependent on the compositions of MSW. In
the non-catalytic pyrolysis of polyethylene, polypropylene is adequate for the industrial
application of liquid products in oil production. PET and PS are not suitable for oil
production because they lack long-chain aliphatic fragments. The hydrocarbons derived
from lignocellulosic materials exhibit a high oxygen concentration, which creates barriers
in the utilization of bio-oil in a conventional petroleum refinery [33]. These oxygenate
components can advance the formation of coke and light organics via polymerization and
cracking.
The effective hydrogen index defined by Chen et al. [34] can be used to evaluate the
potential amount of hydrogen in the fuel that is available for energy production. It is
calculated by Equation (1-1)
(𝐻/𝐶)𝑒𝑓𝑓 =𝐻 − 2𝑂 − 3𝑁 − 2𝑆 − 𝐶𝑙
𝐶 (1-1)
As shown in Figure 1.3, the index for highly oxygenated lignocellulosic materials is less
than 1, and the index for polyethylene is approximately 2.25. Polyethylene can be a good
H donor during the co-pyrolysis of polyethylene and lignocellulosic materials.
9
Figure 1.3 Effective Hydrogen Index
Generally, solid fuels in high-temperature surroundings undergo primary pyrolysis via a
series of processes, including the de-polymerization of the polymers or fragmentation of
macromolecules, the decomposition of each monomer or molecular unit sequentially or in
parallel, and the production of carbonized char, tar, light hydrocarbons, and light gases.
Subsequently, the tar and light hydrocarbons with a high reactivity exhibit a high risk of
cracking. Two major product classes of volatiles can be detected: (1) primary tar and (2)
secondary/tertiary tar, including phenolics, benzenes, and polycyclic aromatic
hydrocarbons (PAHs). Both temperature and residence time have a significant effect on
this process.
1.3.4 MSW gasification
The goal of gasification technology is to realize the chemical recycling of waste plastics to
generate useful incondensable gases (H2, CO et al.) and hydrocarbons. Compared with
biomass, plastics have a higher concentration of hydrogen and a lower char yield, which
ensures a higher hydrogen production potential in syngas from the gasification of plastics.
RDF steam gasification is the preferred process to produce syngas that is abundant in H2
and CO. Calcium oxide (CaO) has been extensively used in coal and biomass gasification
processes because it can be used to absorb CO2 and improve H2 and CO concentrations.
CaO also serves the role of catalyst during CaO/biomass devolatilization. Angel et al.
reported that gasification rates rectilinearly increased with an increase in calcium loading
for lignite char gasification in air, CO2, steam, and H2 [35]. The opposite effect of calcium
in the air gasification of cokes was also reported [36] and [37]. Arash et al. reported that
PET Cardboard PVC RDF PE0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
2.50
Eff
ec
tiv
e h
yd
rog
en
in
de
x (
H/C
eff)
10
CaCO3, CaO, and MgO did not show any catalytic activity during bitumen coke steam
gasification due to their poor reactivity to insufficient contact with the coke and steam
[38]. As previously discussed, plastics exhibit a molten process, which favors contact
between CaO and a solid. Thus, MSW and CaO steam gasification is proposed here.
Some studies have tested the roles of CaO during MSW and CaO blending gasification.
1.3.5 Greenhouse gas remediation
The heavy exploitation and utilization of coal and other fossil fuels since the industrial
revolution have produced a series of global climate issues since the 1990s. Global
warming caused by the large amount of greenhouse gas (GHG) emissions (CO2, CH4,
N2O, et al.) is the primary factor in abnormal climate change. The awareness of global
warming has resulted in the consensus that action must be taken to stabilize climate
changes. The European Commission established a target for limiting the average increase
in the Earth’s surface temperature to 2 °C and proposed that the EU reduce GHG
emissions relative to 1990 levels by 20 % in 2020 and by 60 % in 2050.
Thus, a new carbon capture and recycle (CCR) system, which is combined with the
integrated municipal solid waste system and based on the multi-reforming of CH4 with
CO2 (as shown in Figure 1.4), is proposed in this chapter. The aim was to develop a novel
method to remediate greenhouse gases (CO2) using a high-temperature (greater than 1173
K) process of reforming CH4 and/or O2, and/or H2O without catalysts. LFG and the
fuel gas produced from bio-waste treatment contain a considerable fraction of methane,
which is a potential source for this process. The entire system requires a powerful energy
source for its highly endothermic reactions. A high-temperature gas reactor-type nuclear
reactor can be a suitable energy source for a CCR system because maximum high-
temperature outputs of 950 °C can be achieved with non-carbon emissions and because a
sufficient amount of nuclear fuel is available to satisfy a country’s demand [39]. According
to current data, Sweden operates 10 nuclear power reactors that provide more than 40 %
of its electricity [40]. Thus, the nuclear reactors in Sweden can satisfy the energy demand
for a CCR system. The energy may also be derived from wind, hydro, waves or an MSW
combustion system.
11
Figure 1.4 Principle of a CCR system combined with MSW treatment
1.4 Objective and framework of the thesis
The general objectives of this thesis are focused on two main topics:
1. Experimental investigation and mathematical modeling to enhance the
understanding of the fundamentals of municipal solid waste thermal conversion.
2. Development of a novel routine to remediate CO2 via the multi-reformation of
CH4.
According to the goal of this thesis, the specific objectives of this study are as follows:
1. To analyze the mechanism of the swelling phenomena during the pyrolysis of
pelletized recovered municipal solid waste.
2. To investigate the heat transfer and devolatilization properties of recovered
municipal solid waste.
3. To develop a mathematical model for the pyrolysis of MSW.
4. To determine the characteristics of the liquid products from the main cellulosic
materials and plastics contained in MSW.
5. To provide possible pathways for the mechanism of tar formation and cracking.
6. To identify the catalytic effects of CaO on the syngas during MSW steam
gasification.
7. To reduce CO2 emissions in the presence of CaO during MSW steam gasification.
12
8. To identify the influence of plastics on catalyst mobility on the char surface during
MSW steam gasification.
9. To present a new routine for CO2 remediation via the multi-reformation of CH4
1.5 Outline of the study and objectives of this thesis
Physical process
Ch
em
ica
l p
roc
es
s
Tempera
ture
risin
g
Room
Temp.
Initial.
MSW
samples
Getting
molten
Swelling
Cat
alys
t mobili
ty o
n the
char
surfac
e in
the
solid
ble
nding
Heat conduction mode
in porous beds
Tar formation and
crackingChar
gasification
Pap
er b
oard,P
last
ics
pyroly
sis
Pollu
tant c
ontrol
(such
as
CO
2, H
Cl,
etc)
Work 1
Work 3
Work 4
Work 2
Multi-reformation of CH4 with CO2 Work 5
Figure 1.5 Overview of the research
13
Table 1.2 Overview of supplements and their objectives
Supplements Events: Objectives:
Ⅰ
Pyrolysis behaviors of pelletized recovered municipal solid waste fuels
*Heat transfer and devolatilization properties
*Volume changes during swelling *The mechanism of the swelling phenomena
Ⅱ A mathematical model of a large MSW particle
*Heat transfer model modification *The mechanisms of heat transfer inside porous MSW particles
Ⅲ
Characterization of the products from the pyrolysis of paper-board and plastics
*Influence of pyrolysis temperature
*Influence of residence time
*Tar formation and cracking *Pyrolysis behaviors of printing paper, cardboard, PE, PET, PVC, and RDF
Ⅳ
Steam gasification of MSW & CaO blending
*Catalytic effects on the syngas
*Catalyst mobility on the char surface
*CO2 absorption
Ⅴ
A new routine for multi-reformation of CH4 with CO2
*Validation of the thermodynamic analysis
*Influence of temperature
*Influence of other oxidizers including oxygen and steam
14
Chapter 2
2 Literature review
As previously described, pyrolysis and gasification are promising disposal methods that
have attracted significant attention. One of the main challenges is that MSW presents
different heat and pyrolysis behaviors for wooden materials due to its complex
heterogeneous structures. In the first part of this literature review, the progress in
previous experimental and modeling studies related to mass and heat transfer inside RDF
particles is introduced. The review focuses on relevant literature on the characterizations
of products obtained from MSW pyrolysis and gasification.
2.1 Kinetic modeling
Some studies of kinetic mechanisms have focused on the pyrolysis of biomass materials
and different types of plastics. Biomass materials consist of three main components:
cellulose, hemi-cellulose, and lignin. The kinetic models used in the literature that are
related to large-particle solid thermal conversion can be classified into three main groups
[41].
(1) One-step global models are used to interpret the solid fuel degradation in terms of
a one-step reaction, in which the reaction rate is proportional to weight loss, and
can be described using an Arrhenius law dependence on temperature. The
parameters of the Arrhenius equation can be obtained via the experimental
measurement of rates of weight loss in TGA [42], fluidized bed reactors [43], or
tube furnaces [44]. The Arrhenius equation has been employed to describe solid
fuel degradation for chemical regime conditions (intrinsic kinetics) and ablation
regime conditions (apparent kinetics).
Solid Volatiles Char+k
(Scheme 2-1)
(2) One-stage multi-competitive reactions[45] and [46] are used to interpret solid fuel
degradation in terms of two or three competitive reactions, which represent the
generation of each group i of products from the direct decomposition of virgin
solid fuel; for instance, char, gas, and tar. Two competitive reactions for the
formation of volatiles and char are shown as follows:
15
Solid
Volatiles
Char
+
k1
k2
(Scheme 2-2)
The kinetics of each independent reaction can be modeled through a unimolecular
first-order reaction rate.
(3) Two-stage or multi-stage. If the tar cracking is considered, the reaction sequence
can be expressed as follows:
Solid
Volatiles 1
Char
+
k1
k2
k3 Volatiles 2 + Gas
(Scheme 2-3)
Koufopanos et al. [47] proposed a two-step mechanism for describing the kinetics
of biomass. The volatiles from the biomass primary decomposition may react with
char to produce different types of volatiles, gases and char.
Solid
Volatiles 1 Char 1+
k1 k2
k3Volatiles 2 + Char 2
(Scheme 2-4)
Alves and Figueiredo [48] proposed the use of consecutive reactions to describe
cellulose decomposition. Chao-Hsiung Wu et al. [49] proposed a two-reaction
model for coated printing and writing paper, as shown in Scheme 2-5.
RDF is a heterogeneous substrate compared with biomass and coal. Thus, methods for
characterizing the kinetic schemes are needed. Three different approaches to modeling
the thermal degradation kinetics of complex solid fuels have been discussed in the
literature. The first approach considers the fuel as a single homogeneous species. The
second approach considers the pyrolysis rate of RDF as the weighted sum of the rates of
cellulose, lignin, hemicellulose, polyethylene and other plastics. The third approach, which
is similar to the second approach, considers the pyrolysis rate of RDF as the weighted
sum of the rates of paper, cardboard, wood, polyethylene and other plastics. The second
and third approaches assume that the possible interactions among RDF components have
a negligible effect on pyrolysis. In this paper, the third model, which has been thoroughly
Printing and
writing paper
Volatiles 1
Intermediates
k1 k2
Volatiles 2
Char
(Scheme 2-5)
16
discussed in the literature, was employed to predict the experimental data in the TGA and
the isothermal runs [50] and [51].
2.2 Modeling of chemical and physical processes
Charring materials
In practice, solid fuel thermal conversion occurs due to a strong interaction between
chemical and physical processes. Considerable effort has been expended constructing
models that describe cellulosic and coal material thermal degradation based on a series of
assumptions that consider chemical and transport phenomena. Transport properties are
dependent on the pore distribution. For example, permeability and effective diffusivity are
dependent on the third- and second-order powers of porosity. The pore distribution is
dependent on the reaction conditions. The increase in medium porosity produces shorter
transport times and longer heat transfer times, unless radiation is important. Thus, the
pore diameter, through which volatile products escape, serves an important role in the
description of these processes.
The viewpoint that assumes that the porosity is fine and uniformly distributed has been
employed in the Refs for single wood particles or logs [52]. The materials are considered
to act as a homogeneous medium with a specified porosity and permeability. The solid
and gas phases occur at the same temperature because gases and solids achieve superior
thermal contact.
The simplest model only considered a heat conduction equation with a heat source term
due to chemical processes [53] and [54]. The characteristic time of convective heat
transfer is at the micro level, which is very short. Considering that the solid volumetric
heat capacity is approximately 600 times larger than the solid volumetric heat capacity of
the gas phase, the volatile products are rapidly heated to the temperature of the char,
which signifies that a local thermal equilibrium is attained. Heat flows from the solid-
phase char to the volatiles. Kuug [55] and Villermaux et al. [56] improved the model by
considering that convective heat transfer due to the flow of volatiles produced the
assumption of a quasi-steady-state gas phase (the accumulation of mass and energy of the
gaseous species within the solid is neglected). Similar models considered conductive heat
transfer, a pseudo-steady-state gas phase, and no pressure gradients within a single solid
particle [57]. However, these assumptions were removed from the model of Kansa et al
[58]. The pressure variations inside the porous medium were considered and described
using the Darcy law [58]. Some studies considered the convective heat transfer within the
particle due to the flow of volatiles [59] and [60]. The assumption that the generated
volatiles are in thermal equilibrium with the solid phase was suggested to cause
17
overestimation of the heat loss [60]. A particle model that neglects the heat convection
terms was determined to correspond with the experimental data [60]. In some models, the
Darcy law or a heat transfer coefficient was adopted to account for the non-steady-state
convective flow of volatiles toward a solid surface [61], [62], and [32]. The effect of intra-
particle convection on mass evolution and temperature profiles during pyrolysis is
considered to be significant [61] and [63]. Thus, the effect of volatile flux convection
should be considered.
Non-charring materials
Few studies describe the thermal conversion process of thermal plastic degradation.
Vovelle et al. [ 64 ] proposed a model to describe the mass loss rate of polymethyl
methacrylate (PMMA) that is subjected to a radiant heat flux. Landau transformation was
introduced and combined in the model to predict surface regression. The surface
temperature is predicted to rapidly increase. Wichman [65] examines the effect of bubble
formation inside the molten layer on the steady-state transport of volatiles during
thermoplastic polymer degradation. The translational velocity, growth rate and
distribution of bubbles were determined from a series of mathematical equations.
Staggs [66] developed a mathematical model to describe the change in volume during the
pyrolysis of polymers with different loadings of inert filler. The model assumed that the
produced volatile species escape as soon as they are generated. The results showed that
the change in volume during pyrolysis has a minimal effect on the initial mass loss rate
but has a considerable subsequent effect.
Although non-charring materials exhibit a different degradation behavior than charring
materials, a heat transfer model that did not consider the heat of pyrolysis of PE remains
adequate for estimating the conversion time of pure PE particles [66] and [67].
Municipal solid waste
Valerio et al. [68] and [69] proposed a model for RDF conventional pyrolysis in a fixed-
bed reactor. The RDF samples consist of 53 % cellulose, 14 % lignin, 1 % hemicellulose,
20 % polyethylene, and inerts. The model assumed that the interactions between these
species are negligible, and the RDF pyrolysis behaviors were considered to be the sum of
the behaviors of these species. Other assumptions were as follows: no pressure gradients
were considered; the contribution of radiation in the bed was disregarded; a pseudo-
steady-state was assumed for the flow of volatiles; and local thermal equilibrium between
volatiles and the solid matrix was assumed. Satisfactory agreement of the model results,
including temperature transients, gas generation rate, and product yields, with the
experimental data was achieved.
Salvador [32] proposed a 1D model to describe the combustion of a porous medium
composed of cardboard and polyethylene. Fuel in the shape of a brick is subjected to a
18
radiative heat flux at its surface; thus, no forced convection of air flows through the
medium. The medium was considered to be a continuum, in which local thermal
equilibrium and chemical equilibrium were assumed. The majority of the required
parameters were obtained from specific experimental tests. An acceptable prediction of
mass loss rate and temperature levels inside the porous bed was obtained for fuels with a
low PE concentration from 0 % PE to 30 % PE. For a maximum PE concentration of
70 %, the model prediction of solid mass evolution significantly deviated from the
experiment.
2.3 Heat transfer models
Bluhm-Drenhaus et al. investigated the devolatilization of large thermoplastic (PE)
particles with different geometries—“slab” (2 × 16 × 16 mm3) and “cube” (8 × 8 × 8
mm3)—in co-firing conditions [67]. The conversion time for the slab at 900 °C is
approximately 0.5 s, which is substantially less than the conversion time for a cubic
particle, which is approximately 3 s. The slab decomposes in the thermally thick regime of
the entire volume of the particle, whereas the brick and cubic particle remain in the
ablation regime in the beginning and subsequently exhibit a changeover from the ablation
regime to the thermally thick regime. Because no charring layer remains after the
decomposition, the radiation heat flux is the only contributor that supports heat for PE
decomposition at the surface, if convection is disregarded. Because heat from radiation is
quite high at high temperatures, the area of the particle surface available to absorb the
radiation heat flux from the furnace wall is the main factor that affects the conversion
time of a particle. A thin slab is easier to spread after becoming molten, and it absorbs
more heat within a short time compared with thick particles. A similar phenomenon is
demonstrated by Salvador’s experimental results, in which the time evolution of a
normalized mass of bricks with a PE concentration that ranges from pure cardboard to
pure PE was presented [30]. The fastest evolution time was obtained from pure PE due
to a lack of char layer at the surface to inhibit the heat flux via radiation from the furnace
wall, whereas the other particles exhibited a relatively similar mass evolution rate.
A proper understanding of the mechanisms of heat transfer in a gas-solid system is critical
to the construction of a reliable model of the effective thermal conductivity in a packed
composite. In the absence of flow, the heat conduction occurs by two modes of heat
transfer: conduction and radiation in the gas and solid phases at typical temperatures for
gasifiers and chambers. The effective thermal conductivity due to the first mode is
defined by heat conduction through both the solid and fluid phases with stagnant flow
and point contact conduction [70]. According to a study by James, the series and parallel
layers of solid and fluid phases serve as the lower bounds and upper bounds, respectively,
for the effective conductivity of heterogeneous materials when the radiation term is
19
neglected [71]. Kunii and Smith developed a correlation, in which an equation to calculate
an empirical constant related to the number of solid particle contacts is only valid in the
range of porosity from 0.26 to 0.476, for a porous structure by discretizing the solid and
fluid phase into separate modes that act in series and parallel [72]. Zehner and Schlünder
drew an empirical curve to calculate the effective thermal conduction for porosity, which
varies from 0 to 1, in a cylindrical unit cell that contains both solid and fluid phases [73].
However, this curve was obtained from mass transfer experiments that were performed in
the bulk region of packed beds with specific porosities. The contact area, which varies
substantially with changes in the external loads acting onto the spherical particles or from
contact in a packed bed, may also play a role in the calculation of the effective thermal
conductivity [74]. The effective contact area is also significantly determined by the surface
roughness. For irregular particles, such scraps of cardboard, the contact area should be
larger and much more difficult to estimate than that for spherical particles with the same
volume.
2.4 Experimental studies on the products of MSW pyrolysis and
gasification
2.4.1 Pyrolysis
2.4.1.1 MSW
Several studies have assessed the suitability of pyrolysis for MSW. A.M Li et al. [75]
conducted experimental studies of municipal solid waste pyrolysis in a laboratory-scale
rotary kiln. The effects of heating methods, moisture content and size of waste on
pyrolysis gas yields and compositions were evaluated. A rapid heating rate enhances gas
production compared with a slow heating rate. Wood chips and PE plastics present
similar trends of heating values for the pyrolysis gases. The pyrolysis gases have high
HHV levels. Mohammad et al. [76] investigated the pyrolysis of municipal solid waste in a
fixed-bed reactor to analyze pyrolytic oil. The results show that the liquid yield increased
with an increase in temperature below 450 °C. The collected liquid is highly oxygenated
when the oxygen content is 52.91 %, compared with 53.15 % for raw feedstock. The
carbon and hydrogen concentrations are similar to the carbon and hydrogen
concentrations of biomass-derived oil. W.K. Buah et al. [77] analyzed the oil/wax from
municipal solid waste and discovered that the oil contains carboxylic acids and their
derivatives: alkanes, alkenes, and mono and polycyclic and substituted aromatic groups.
High temperatures enhance the yield of aromatic groups and lower the yield of aliphatic
groups. Velghe et al. [ 78 ] examined the pyrolysis of municipal solid waste for the
20
production of valuable products in a semi-batch reactor. The presence of long aliphatic
HCs reveals incomplete breakdown in the conditions of high heating rates and short
residence times. The liquid with the lowest water content, lowest O/C, and maximum
yield is obtained by fast pyrolysis at 510 °C. The oil fraction is abundant in 63.5 % of
aliphatic compounds and 23.5 % of aromatic compounds. No waxy material is obtained
in the liquid fractions produced by slow pyrolysis up to 550 °C. The liquid product
consists of a water phase and an oil phase. The oil has a low yield, low water content, low
O/C value, and 70.2 % aliphatic compounds. The analysis of pyrolysis liquid from the
pyrolysis of real waste and simulated waste composed of PE, PP, and PS revealed
significant amounts of styrene, toluene, and ethyl-benzene [79]. Thus, MSW is highly
suitable as a chemical feedstock.
2.4.1.2 Plastic materials
The properties of pyrolytic products from different types of plastics vary significantly. For
example, PS and poly(methymethacrylate) have high monomer yields that approach 100 %
from pyrolysis, whereas PP yields only approximately 2 % of its weight as its monomer
[80]. PE and PP are the main sources of alkane and alkene, respectively, in pyrolytic oil
from MSW pyrolysis, whereas PET contributes significantly to aromatic compounds [81],
[ 82 ], and [ 83 ]. To obtain the desired liquid product, numerous studies have been
performed to assess the effect of the plastic mixture on product yield and composition
[79], [84], [85], and [86].
The catalytic cracking of plastic waste has also attracted attention in recent years [87].
Solid acids such as zeolites are the main catalysts because their acidity favors cracking
reactions. However, other catalysts, such as red mud, have the advantage of lower cost
compared with zeolites. Red mud primarily consists of Al2O3 and Fe2O3, which promotes
cracking and aromatization reactions that yield more gases and lighter, aromatic and fluid
liquids [79]. Polyethylene samples were also catalytically pyrolyzed using different catalysts,
such as ZnO, MgO, CaC2, and SiO2 [88]. Maoyun He et al. [89] explored the MSW and
calcined dolomite pyrolysis in a bench-scale downstream fixed-bed reactor in the
temperature range of 750-900 °C. The results showed that the presence of calcined
dolomite significantly influenced the product yields and gas composition. In the presence
of calcined dolomite, the molar concentration of hydrogen is 66.30 %, which is
substantially higher than 36.69 % from non-catalytic pyrolysis processes, whereas CO2,
CH4, C2H4, and C2H6 concentrations exhibited opposite tendencies.
One problem with MSW pyrolysis is associated with the production of HCl and chloro-
organic compounds from PVC thermal degradation. Two methods for reducing the
influence of chlorine on the processes and utilization of products have been investigated
21
[90], [91], and [92]. In the first method, thermal degradation of plastic that contains PVC
is conducted in two steps: the sample was heated to 320 °C for dehydrochlorination and
subsequently heated to a higher temperature for additional degradation. In the second
method, HCl is fixed by adding absorbent, which is intimately mixed with PVC. These
absorbents primarily include MaO, BaO, CaO, TiO2, and Fe2O3. The Al-Zn composite
catalyst (AZCC) was also determined to act as both a dechlorination sorbent and a
catalyst in both the liquid phase and the vapor phase [93]. Metals (aluminum, iron and
zinc) and metal oxides (aluminum, titanium, copper and iron) depress HCl formation and
hinder benzene formation [94].
2.4.1.3 Lignocellulosic materials
Printing paper and cardboard predominantly consist of hemicellulose, cellulose, and ash.
Numerical studies on the thermal conversion technologies of pure cellulose and
hemicellulose, including pyrolysis kinetics, the effects of reaction conditions, and the
mechanisms of reactions of pure cellulose or hemicellulose have been performed [95],
[96], [97], [98], [99], [100], and [101]. An inorganic salt concentration as low as 0.005
mmoles/g of cellulose was determined to be sufficient for significantly affecting the
resulting pyrolysis speciation [ 102 ]. Some studies evaluate the thermal pyrolysis
characteristics of printing paper by Py-GC/MS and TGA [103], [104], [105], and [106].
However, relatively few studies have investigated the effect of reaction conditions on the
yields of products and the detailed characterization of the products from paper board
pyrolysis. Many primary products act either as a product of cellulose pyrolysis during
primary pyrolysis or as an intermediate for the formation of other products during
secondary reactions. The effect of residence time on secondary reactions is also
insufficient.
2.4.2 MSW gasification
Experimental studies on MSW gasification have already been investigated [107] and [108].
Compared with pyrolysis, gasification was shown to increase the yields of combustible
gases due to char gasification [109]. Tar formation is a primary concern during gasification
because it can cause problems in the downstream processing equipment. Considerable
efforts have been directed to tar removal from fuel gas and hydrogen-rich fuel gas to
improve the quality of fuel gas. The injection of steam has been shown to have a positive
effect on the syngas calorific value and can reduce the total amount of detected tar [110].
22
The cracking of some model organic compounds, such as toluene and volatile organic
compounds (VOCs), and their effect on downstream equipment in the presence of
catalytic CaO have also been shown to be very effective in some cases [111], [112], [113],
and [ 114 ]. The additional catalytic cracking of crude syngas from the pyrolysis/
gasification process using catalyst has been proven to be an effective method for tar
removal, the adjustment of the H2/CO molar ratio, and the increase of the gas yield and
the hydrogen yield [ 115 ] and [ 116 ]. In this concept, the catalytic reactor in the
downstream system is directly connected to the gasification reactor because of the high
sensible heat carried by the crude syngas. Siyi Luo et al. [116] performed a study of MSW
catalytic steam gasification in this type of fixed-bed reactor system and noted that the
concentrations of H2 and CO in syngas with calcined dolomite or NiO/γ-Al2O3 as a
catalyst were 61.5 % and 76.9 %, respectively, compared with MSW steam gasification at
900 °C in the absence of catalyst. Compared with catalytic pyrolysis, the injection of
steam during MSW gasification causes a rapid increase in the dry gas yield and carbon
conversion efficiency. An optimum value of 2.41 for steam/catalyst was obtained from
the study. The increase in temperature has a significant influence on the catalytic steam
gasification performance and contributes to a higher dry gas yield, H2 and CO
concentrations, and carbon conversion efficiency. Jianfen Li et al. [115] also conducted a
similar study at 800 °C in a similar reactor system. The optimum mass ratio of steam to
MSW is approximately 1.5. The effect of the catalyst to MSW mass ratio on the gas yield
and gas composition was investigated in their study. They discovered that the gas yield
and H2 concentration rapidly increased when the catalyst/MSW is 0.5; a smaller increase
was observed with an increase in the catalyst/MSW up to 2.0.
Catalysts were also shown to play a significant role in the devolatilization of a blended
CaO/feedstock material during gasification or pyrolysis. Metals in mineral form in coal
and biomass have a negative effect on pyrolysis, which results in lower tar and higher char
and gas yields [117] and [118]. The catalytic activity ranking of four catalysts that are used
in the steam gasification of a sludge-oil-coal agglomerate (SOCA) was K2CO3 ≫ CaO >
NiO > Fe2O3 [119]. Dolomite with high concentrations of CaO and MgO as a bed
material is suggested to enhance the dehydrogenation and isomerization reactions of
fragments produced by thermal cracking; however, it does not enhance the
dehydrogenation/carbonization reactions in a fluidized bed [120]. During the thermal
degradation of a PE/PP/PS/PVC polymer mixture, the addition of a calcium catalyst and
CaCO3 increased the liquid product yield by 4–6 % and reduced the yield of wax residue
by 2 % [121]. The pyrolysis of PET with Ca(OH)2 in a steam atmosphere produced a high
yield of liquid product (mainly benzene) and a low yield of solid [122].
Angel et al. noted that gasification rates rectilinearly increased with an increase in calcium
loading for lignite char gasification in air, CO2, steam, and H2 [35]. Ljubisa et al. attributed
the relatively high gasification reactivity of char obtained from lignite and CaO co-
pyrolysis to the catalytic effect of highly dispersed CaO on the char surface [123]. The
23
opposite effect of calcium in the air gasification of cokes was also reported [36] and [37].
Arash et al. [38] noted that CaCO3, CaO, and MgO did not show any catalytic activity
during bitumen coke steam gasification and attributed the reason for their poor reactivity
to insufficient contact with the coke and steam. Zuo-Liang et al. [124] suggested that the
effect of the methods and conditions employed for CaO loading is notable for the degree
of calcium dispersion on the char surface. Impregnation at a higher temperature and
longer time favors the distribution of calcium on the char surface and increases the
catalytic effectiveness of CaO.
24
Chapter 3
3 Methodology
3.1 Experimental methodology
3.1.1 Materials
Table 3.1 lists the tested materials with the proximate and ultimate analysis of their
physical and chemical properties. RDF pellets and wheat straw pellets with a diameter of
8 mm were obtained from external providers. Raw granules of PE, PVC (grade Norvinyl
S6770), and PET were provided by Borealis, INEOS ChlorVinyls, and Petainer,
respectively. For the gasification test, MSW and CaO (provided by Ecoloop) were ground
into small pieces using a mill and then sieved to a particle size of less than 2 mm. Both
particle samples were subsequently mixed into different proportions and packed into a
basket with a height of 70 mm and a diameter of 30 mm. Different experimental
conditions were achieved by changing the steam flow rate, gasification temperature, and
mass ratio of CaO to MSW. The input conditions for the MSW and CaO gasification
experiments are presented in Table 3.2.
Table 3.1 Proximate and ultimate analysis of the tested materials
Proximate analysis
MSW Printing paper
Cardboard
PET
PVC
PE
Wheat straw pellets
RDF pellets
Moisture content at 105 °C (%)
1.5 3.8 5.4 - - - 10.1 2.9
Ash content (550 °C) (% .dry)
8.3 29.2 9.6 <0.3 0.3 <0.3 3.8 6.0
Volatile matter 86.595 79.4 83.1 90.0 93.0 >99.0 77.7 84.4
Ultimate analysis (dry composition)
Sulfur (%) 540
mg/kg 0.015 0.094 <0.012 0.04
<0.012
0.04 0.09
Carbon (%) 69.7 34.9 43.0 60.2 38.4 80.5 46.5 63.3
Hydrogen (%) 10.1 4.4 5.4 4.4 5.0 15.5 5.81 8.9
Nitrogen (%) 0.57 0.14 0.16 <0.10 <0.1
0 <0.10 0.48 0.3
Oxygen (%) 10.0 31.3 41.7 35.4 4.2 3.9 43.39 20.95
Chlorine (%) 0.75 - - - 52.0 - 0.084 0.52
25
Table 3.2 Experimental conditions for MSW and CaO steam gasification
Case No. MSW (g) CaO (g) Temp. (°C) Steam flow
(g/min)
Elapsed time
(minute)
#1 15 0 800 5.6 17.1
#2 15 3 800 5.6 20.0
#3 15 7.5 800 5.6 20.6
#4 15 15 800 5.6 24.3
#5 15 0 900 19.3 17.6
#6 15 15 900 19.3 17.9
#7 15 0 800 19.3 19.8
#8 15 15 800 19.3 18.8
#9 15 0 700 19.3 19.5
#10 15 15 700 19.3 19.2
3.1.2 Experimental facility
3.1.2.1 Batch-type HiTAC fixed-bed gasifier
The experiments were conducted in a lab-scale facility, which was built in Kungliga
Tekniska Högskolan (KTH) in Sweden. Figure 3.1 shows a schematic of this facility. The
facility was preheated to a desired temperature via methane combustion in the first
chamber. One ceramic honeycomb heat exchanger is installed in the chamber; it is used to
accumulate heat from the methane combustion process and provide energy to the
pyrolysis reaction via convection and radiation. A thermocouple is used to record the
furnace temperature; once the temperature attains the desired level, the gas burner is
switched off and different types of gases, including N2, helium, oxygen, air, methane from
a gas cylinder and steam vapor from a steam boiler, are controlled by flow meters and fee
into the chamber. After passing though the honeycomb, the gas or steam vapor can be
preheated to attain a high temperature. Prior to the test, the sample was hung in a zone
above the hot zone by adjusting the length of the wire. Nitrogen was fed and passed
through this zone into the hot chamber to keep the zone cool. When the appropriate
temperature of the chamber was attained, the samples were placed in the furnace. The
inlet hole for the wire that moves upward and downward was connected with the mass
balance, and the nitrogen flow was also switched off. Nitrogen fed at the sample inlet was
also used to cool the char residue prior to removing it from the furnace. Two exhaust
outlets were provided. The large exhaust outlet was used to vent flue gas during the
26
preheating process. The small exhaust outlet was designed for gas sampling during the
test. During the pyrolysis experiment, the valve installed at the large exhaust hole was
closed, and a pump was used to suck all the gas produced from the chamber. A cooling
system with ice water and cooling liquid (-20 °C) outside the washing bottle and acetone
or tetrahydrofuran solvent inside the washing bottle was used to immediately cool and
trap the high-temperature organic vapors. Helium controlled by a mass flow controller
can be used as a tracer gas during the experiment. Two additional thermocouples were
used to measure the surface and center temperatures of the pellet during the reaction. An
observation hole composed of glass existed in the second combustion chamber; the hole
enabled the reaction to be recorded by a digital camera. A CCD camera was installed
opposite the glass window to record the entire process. ImageJ software was employed to
determine the change in the size of the pellet particles via edge detection of the pictures
obtained from the CCD camera.
FLOW
Cooling N2
Nit
org
en
Hel
ium
CH4
Exh
aust
Gas
es
Air
GC
Cooling system
Sample
Ceramic Honeycomb
Data Recording
Valve
NutScrew thread
Thermocouple
Stea
m
Ф 89mm
Thermocouple
0.01g
Electronic Scale
Acetone/Tetrahydrofuran
-20 °CIce
Figure 3.1 Lab-scale fixed-bed reactor
Selected images of the samples during the pyrolysis are analyzed in ImageJ software. The
swelling and shrinkage are analyzed and calculated based on the transient diameters and
heights of the images. The transient volumetric swelling and shrinkage ratio is expressed
as
𝑅𝑉 = (𝑉𝑡 − 𝑉0)/𝑉0 (3-1)
where 𝑅𝑉 is the volumetric swelling and shrinkage ratio, 𝑉𝑡 is the transient volume, and 𝑉0
is the initial volume.
27
The mass change in the sample particles during the pyrolysis was investigated by varying
the temperature from 800 °C to 450 °C. The volatile conversion was calculated according
to Equation (3-2)
𝑋 = 𝑚/𝑚0 (3-2)
where 𝑚0 and 𝑚 are the initial mass and the transient mass, respectively, of the pellet.
3.1.2.2 TGA/DSC
A Mettler Toledo TGA/DSC was employed to study the thermal degradation
characteristics of the printing paper, cardboard, and PE. A 6 mg sample was placed in an
alumina crucible. The heating program is as follows: 30 °C to 100 °C at
10 °C/min→100 °C (30 min)→100 °C to 700 °C at 10 °C/min→700 °C (5 min). The
change in weight was continuously registered. To prevent the secondary pyrolysis of
volatile vapors, which can affect the heat flux and promote char formation, all tests were
run without lids [125]. The Coats and Redfern method, which is based on the following
equations, was used to calculate the kinetic parameters [126].
𝐹(𝛼) = ∫𝑑𝛼
𝑓(𝛼)
𝛼
0
=𝐴𝑒𝛽∫ exp(
−𝐸
𝑅𝑇)𝑑𝑇
𝑇
0
≈𝐴𝑒𝑅𝑇
2
𝛽𝐸(1 −
2𝑅𝑇
𝐸)exp(−
𝐸
𝑅𝑇) (3-3)
ln (𝐹(𝛼)
𝑇2) = 𝑙𝑛
𝐴𝑒𝑅
𝛽𝐸(1 −
2𝑅𝑇
𝐸) −
𝐸
𝑅𝑇 (3-4)
where α is the conversion degree, β is the heating rate, 𝐴 is the pre-exponential constant,
and f(α) is a function that is dependent on the reaction mechanism. Because the activation
energy from lignocellulose biomass pyrolysis is in the range of 100-250 kJ/mol,
2𝑅𝑇
𝐸≈ 0 (3-5)
Then, Equation (3-4) becomes Equation (3-6),
ln (𝐹(𝛼)
𝑇2) = 𝑙𝑛
𝐴𝑒𝑅
𝛽𝐸−𝐸
𝑅𝑇 (3-6)
When a first-order reaction mechanism is applied, the pre-exponential constant and
activation energy can be determined by plotting the thermogravimetric data by applying
Equation (3-6).
28
3.1.2.3 Py-GC/MS
A pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) system was employed
to identify the volatiles from the primary pyrolysis of solid wastes. The pyrolyzer (Pyrola
2000) consists of a process unit with a Pt filament, a control unit, an optic cable and
software with the Agilent 6890A gas chromatograph (GC) and the Agilent 5973C MSD
mass spectrometer (MS). The chromatographic conditions were as follows: DB-1701 ms
(length: 60 m, diameter: 0.25 mm); injector temperature: 280 °C; column temperature:
40 °C; 40 °C –2 °C/min→180 °C; and 180 °C –5 °C/min—>300 °C. The samples were
pyrolyzed under isothermal conditions at temperatures that ranged from 400 °C to 700 °C.
3.1.3 Characterization of the products
3.1.3.1 Syngas analysis with an online micro gas chromatograph
Syngas was analyzed using the online Agilent 490 Micro gas chromatograph (GC). The
Agilent 490 is equipped with three columns; each column represents one gas
chromatograph. Column 1 in channel 1 is Molsieve 5 Å and can detect hydrogen, oxygen,
nitrogen, methane, carbon monoxide and noble gases. Channel 2 is equipped with a CP-
PoraPlotU column and is used for the analysis of CO2 and C2 hydrocarbons. Channel 3 is
equipped with an Al2O3 column and is used to analyze the C3 and C4 hydrocarbons.
Argon (Ar) was used as a carrier gas in channels 1 and 2, whereas Helium (He) was used
in channel 3. Helium was detected in Column 1. The samples of produced gases were
analyzed every 112 s. Because the syngas flow rate is expected to change significantly with
time, the addition of a tracer gas (carrier gas, He) with a known flow rate was required to
quantify this variation. The inert tracer gas (He) was used and controlled by a digital flow
controller with high accuracy. The instantaneous syngas flow rate can be calculated from
the known He flow rate and the molar ratios of He to incondensable gases (H2, CO, CO2,
CH4) and other low-molecular-weight hydrocarbon gases. Syngas was analyzed by the
online Agilent 490 Micro GC. The syngas flow rate is expected to significantly change
with time. To quantify the variation, the addition of a tracer gas (carrier gas, He) with a
known flow rate is necessary. The inert tracer gas—helium—was controlled by a digital
flow controller with high accuracy. The instant flow rate of the syngas was calculated by
the helium flow rate and the molar ratio of helium to an incondensable gas, such as H2,
CO, CO2, and CH4, and other low-molecular-weight hydrocarbon gases.
29
3.1.3.2 Solid sample tests using SEM/EDS and TGA
Scanning electron microscopy (Hitachi S-3700n) and energy-dispersive X-ray
spectroscopy (SEM/EDS) were employed to observe the microscopic morphologies and
to identify the relative elemental compositions and proportions of the solid residues.
The TGA experiments were performed in the SETARAM TG-24 double-furnace unit
with a 0.001 mg detection limit to examine the concentrations of calcium hydroxide,
calcium carbonate, and unreacted carbon in the solid residues. Approximately 100 mg of
solid residue was loaded in a crucible with an inner diameter of 8 mm and a height of 8
mm. The crucible was hung at one end of the balance beam, and a crucible containing
alumina pieces was hung at the other end of the balance beam as a reference. The heating
program was as follows: 20 °C to 1200 °C with 15 °C/min→1200 °C
(15 °C/min)→1200 °C to 20 °C with 50 °C/min. Ar served as the carrier gas at
temperatures below 1200 °C. Once a temperature of 1200 °C was attained, the carrier gas
was switched to oxygen to burn the unreacted carbon. The change in weight was
continuously registered. The calculation of the concentrations of calcium hydroxide and
calcium carbonate was based on the release of H2O and CO2.
3.1.3.3 Liquid product analysis using GC/MS
The GC/MS (gas chromatography-mass spectrometry) analysis of the tetrahydrofuran
soluble fraction of the liquid products from plastic materials such as
PE/PET/PVC/RDF was conducted using the Agilent 63440A/Waters Auto Spec with
1-methyl-7-(1-methylethyl)-naphthalene (case number 490-65-3) as an internal standard.
The chromatographic conditions were as follows: BPX5 (SGE, length: 30 m, diameter:
0.25 mm); injector temperature: 280 °C; column temperature: 50 °C (1 min); 50 °C –
5/min—>300 °C; 300 °C (5 min); model: electron impact (EI); electron energy: 70 Ev;
and mass range: 40-400 m/z. The liquid products from printing paper and cardboard
degradation were trapped using acetone in the washing bottle.
3.2 Modeling study for MSW particles
In this study, a model was constructed to describe the mass- and heat-transfer processes
inside a porous MSW particle composed of polyethylene and cardboard. The heat of PE
melting was also considered and combined into the particle model. A one-step global
model and a scheme of two consecutive first-order reactions were adopted to describe
polyethylene degradation and cardboard degradation, respectively. The particle model, in
which the contribution of the volatiles’ convective mass flows to the thermal balance was
30
considered, considers the conductive and convective heat transfer within the porous
particle. The mechanisms of heat transfer in the absence of gas flow were considered,
including a conduction term and a radiation term. Several typical heat transfer models,
which are based on these mechanisms and are used to estimate the effective thermal
conductivity, were selected and modified to fit with the experimental results. The effect of
the heat transfer model on the prediction of the municipal solid waste (MSW) pyrolysis
process has been investigated in this paper.
3.2.1 Melting behavior of polyethylene
The melting of a collection of crystalline long-chain molecules has been demonstrated to
be a first-order phase transition, which is accompanied by an endothermic process [127].
The heat of fusion of crystalline polyethylene also significantly contributes to a decrease
in the heating rate in the temperature zone approximately 100 °C. In this paper, a model
that is extensively applied to describe the de-crystallization process of polyethylene during
heating was combined with a particle model to describe the endothermic heat [128].
During the heating process, a classical equation can be used to describe the isothermal
crystallization kinetics of polymers, as shown by Equation (3-7)
𝑋𝑀 = exp(−𝑍𝑡𝑡𝑛) (3-7)
where Zt is the crystallization rate constant; n represents the Avrami exponent, which
represents the mechanism and dimensional geometry of crystal growth; and 𝑋𝑀 denotes
the degree of de-crystallinity during heating, which is a function of time t according to
Equation (3-8)
𝑋𝑀 =∫ {
𝑑𝐻𝑚𝑑𝑡
}𝑑𝑡𝑡𝑡0
∫ {𝑑𝐻𝑚𝑑𝑡
}𝑑𝑡𝑡∞𝑡0
∙ 100% (3-8)
where t0 and t∞ denote the onset and completion time, respectively, of the melting stage.
For the non-isothermal process, t can be transformed into T, as shown in Equation (3-9)
𝑡 =|𝑇0 − 𝑇|
𝐷 (3-9)
where 𝐷 is the heating rate. Combining with Equation (3-9), Equation (3-8) and Equation
(3-7) can be rewritten as
𝑋𝑀 = exp(−𝑍𝑡(𝑇 − 𝑇0𝐷
)𝑛) (3-10)
𝑋𝑀 =∫ {
𝑑𝐻𝑚𝑑𝑡
}𝑑𝑇𝑇𝑇0
∫ {𝑑𝐻𝑚𝑑𝑡
}𝑑𝑇𝑇∞𝑇0
∙ 100% (3-11)
31
Thus, the heat required during melting can be calculated according to Equation (3-12)
∫ {𝑑𝐻𝑚𝑑𝑡} 𝑑𝑇
𝑇
𝑇0
= 𝑋𝑀 ∙ ∫ {𝑑𝐻𝑚𝑑𝑡} 𝑑𝑇
𝑇∞
𝑇0
= 𝑋𝑀 ∙ 𝐻𝑚 (3-12)
The kinetic parameters were obtained from [128], and the heating rate is assumed to be 20
min/°C.
3.2.2 Kinetics model
Numerous studies focus on the determination of kinetic mechanisms and kinetic
constants for cellulosic material degradation and polyethylene. The actual reaction
schemes for the pyrolysis of biomass and plastics are extremely complex due to the
formation of more than one hundred intermediate products. The reaction schemes for
paper board degradation have been discussed in a previous paper [129]. Two consecutive
first-order reactions were proposed, as shown in Scheme 3-1. The first and second
reactions cover a mass loss of 70 % and 30 %, respectively. The values for the kinetic
parameters are obtained from [130]. The values of stoichiometric parameters a1 and a2
were determined based on the experimental results.
Paper-board a1 char1 (1-a1) Volatiles
a2 char2 (1-a2) Volatiles
+
+
K1, 220-375°C
K2, 380-450°C
Scheme 3-1
Based on this scheme of two consecutive first-order reactions, the evolution of the total
mass of cardboard can be expressed as the sum of the mass of cardboard for the
intermediate species of char. The derivatives are given as follows:
𝑑𝑚𝑐𝑑𝑡
= −𝑘1𝑚𝑐 (3-13)
𝑑𝑚𝑐ℎ𝑎𝑟1
𝑑𝑡= 𝑎1𝑘1𝑚𝑐 − 𝑎1𝑘2𝑚𝑐ℎ𝑎𝑟1 (3-14)
𝑑𝑚𝑐ℎ𝑎𝑟2𝑑𝑡
= 𝑎2𝑘2𝑚𝑐ℎ𝑎𝑟1 (3-15)
A one-step reaction can satisfactorily describe the thermal degradation of PE. The mass
source term can be expressed as
32
𝑑𝑚𝑃𝐸
𝑑𝑡= −𝑘3𝑚𝑃𝐸 (3-16)
where 𝑘 is expressed as
𝑘𝑖 = 𝐴𝑖exp(−𝐸𝑖/𝑅𝑇) (3-17) (For the degradation of cardboard, intermediate species of char, and polyethylene:
𝑖 = 1, 2, 3, respectively.)
The reaction heats for the devolatilization of PE and the two reactions during cardboard
degradation were determined during the TG/DSC (thermogravimetry/differential
scanning calorimeter) tests. These values are grouped in Table 3.3.
Table 3.3 Kinetic parameters
Drying [131]
Cardboard devolatilization
PE devolatilization Reaction1 [130] Reaction2 [130]
A (s-1) 5.13 𝖷 106 3.98 𝖷 108 2.04 𝖷 104 2.09 𝖷 1011
E (J·mol-1) 8.79 𝖷 104 1.074 𝖷 105 7.44 𝖷 104 2.02 𝖷 105
H (J·g-1) 2257 𝖷 103 93.41 𝖷 103 84.82 𝖷 103 1.57 𝖷 106 a1 - 0.42 - - a2 - - 0.29 -
3.2.3 Mass and heat balance equations
The rate of mass generation of the solid phase per unit volume can be written as follows:
𝑚𝑢𝑛𝑖𝑡 = −𝑉𝑢𝑛𝑖𝑡 ∙∑𝜕𝜌𝑠𝑘𝜕𝑡
𝑘
(3-18)
Based on the assumption, the mass balance equation incorporates the following equation
for gas generation per unit
𝑚𝑔𝑎𝑠 = −𝑉𝑢𝑛𝑖𝑡 ∙∑𝑎𝑠𝑘𝜕𝜌𝑠𝑘𝜕𝑡
𝑘
(3-19)
where ask represents the stoichiometric parameters of the four reactions that represent
the yield of the volatile.
Thus, the volatile flux toward the bed surface at radius 𝑟 is expressed as follows:
∅𝑟 = −1
𝑟∑∫ 𝑎𝑠𝑘
𝜕𝜌𝑠𝑘𝜕𝑡
𝑟𝑑𝑟𝑟
0𝑘
(3-20)
The energy balance equation is written with the assumption of local thermal equilibrium
as
33
𝜕
𝜕𝑡(∑𝜌𝑠𝑘ℎ𝑠𝑘𝑘
)
=1
𝑟
𝜕
𝜕𝑟(𝑟𝜕𝑇
𝜕𝑟) + (
1
𝑟∑∫ 𝑎𝑠𝑘
𝜕𝜌𝑠𝑘𝜕𝑡
𝑟𝑑𝑟𝑟
0𝑘
)1
𝑟
𝜕
𝜕𝑟(𝑟𝑇)
+∑𝑎𝑠𝑘𝜕𝜌𝑠𝑘𝜕𝑡
ℎ𝑔𝑗𝑘
+ 𝑄𝐶
(3-21)
The heat transfer from the surrounding gas and furnace walls to the basket is assumed to
be caused by a combination of conduction, convection, and radiation. The temperature of
the gas surrounding the basket is also assumed to be constant and equal to the
temperature of the gas surrounding the furnace wall. Thus, the boundary conditions can
be obtained as follows:
𝑘𝑒𝑓𝑓𝜕𝑇
𝜕𝑟│𝑟=𝑅𝑏 = ℎ(𝑇𝑓 − 𝑇) + 𝜎𝜀(𝑇𝑓
4 − 𝑇4) (3-22)
𝜕𝑇
𝜕𝑟│𝑟=0 = 0 (3-23)
The initial conditions are as expressed as follows:
𝑇(0, 𝑟) = 𝑇00 ≪ 𝑟 ≪ 𝑅𝑏 (3-24)
3.2.4 Effective thermal conductivity due to thermal radiation
Due to the low velocity of the volatiles that pass through the packed bed inside the basket,
only a static effective radial conductivity was considered in this study. At the temperature
of the thermal conversion of solid fuels in a reactor, radiation becomes significant,
especially for a conductive solid material such as wood. Thus, the two modes of heat
transfer, conduction and radiation, were included in the developed heat transfer model to
estimate the effective thermal conductivity.
Heat transfer models that are employed have been developed to estimate the effective
thermal conductivity. The predictions from the Kunii and Smith model and the Breitbach
and Barthels model were presented and compared with the experimental results. The
possible mechanisms of heat transfer in a porous medium were comprehensively
discussed. A coefficient as a function of porosity was applied to the radiation exchange
factor in the Kunii and Smith model, which was constructed based on a simplified model
of heat transfer in a packed bed. The effect of the contact surface area between solid
particles on the heat transfer via conduction was also considered in the B & B model.
Both modified models were validated with the experimental results obtained at different
temperatures, with different PE concentrations and initial porosities.
34
3.2.4.1 Kunii & Smith model
According to Yagi and Kunii [132], the following mechanisms contribute to the effective
conduction:
1. Transport through the gas phase in the voids
a) Conduction;
b) Radiation between neighboring voids.
2. Transport through the solid phase
a) Conduction through the contact surface between particles;
b) Conduction through the stagnating film in the vicinity of the contact surface;
c) Radiation from particle to particle;
d) Conduction through the particles.
Thus, a model for heat transfer in a packed bed, which was written in a form that is based
on the assumption that the gas phases and solid phase were organized as parallel layers
perpendicular to the heat flux direction, was proposed by Yagi and Kunii, as shown in
Figure 3.2 (a). Kunii and Smith [72] neglected the heat transfer through the contact
surface between particles and proposed the model shown in Figure 3.2 (b). The following
equation (Equation (3-25)) is obtained by combining the contributions of these
mechanisms depending on whether they operate in series or parallel.
𝑘𝑒𝑟,𝑠𝑒𝑟𝑖𝑒𝑠0
𝑘𝑔= 𝜀 (1 + 𝛽𝑐
𝑑𝑝𝛼𝑟𝑣
𝑘𝑔) +
𝛽𝑐(1 − 𝜖)
11∅+𝑑𝑝𝛼𝑟𝑠𝑘𝑔
+ 𝛾𝑘𝑔𝑘𝑠
(3-25)
with
𝛼𝑟𝑣 =4𝜎
1 +𝜀
2(1 − 𝜀𝜖)∙1 − 𝜖𝜖
𝑇3 (3-26)
𝛼𝑟𝑠 = 4𝜎𝜖
2 − 𝜖𝑇3 (3-27)
where kg and ks represent the conductivities of a fluid and a solid, respectively; ε is the
void fraction; αrv is the radiation coefficient from void to void; ϵ is the emissivity of the
solid; and αrs is the radiation coefficient for the solid.
35
Contact area
1-ε
ε1a
1b
2b
2c
2d
(a) (b)
Figure 3.2 Simplified model: (a) Simplified model by Yagi and Kunii [132] and (b) Electric
network of the thermal resistance.
3.2.4.2 Modification to the Kunii and Smith model
The real status of the porous structure in an MSW particle is shown in Figure 3.3(a). All
scraps of paper board and plastics are stacked together layer by layer, and some voids are
randomly distributed. Four radiation mechanisms exist in the gas-solid phase, including
void to void, particle surface to surface, particle surface to void, and void to particle
surface. To simplify the radiation heat transfer mode in MSW particles, the heat absorbed
by voids from particle surface emission is assumed to be equivalent to the heat absorbed
by the particle surface from voids emission. In the Kunii and Smith model, the radiation
coefficient for the void-to-void mechanism is a function of porosity and decreases with
an increase in porosity. However, a brief review in the introduction of this paper showed
that the radiation coefficient can increase with an increase in porosity in a highly porous
medium. A general form for the conductivity of radiation between voids can be defined
as
𝑘𝑠𝑟 = 𝛼𝑟𝑣 = 4𝜎𝜖𝑇3𝑑𝑝 ∙ 𝐹 (3-28)
where 𝐹 is defined as the radiation exchange factor.
To obtain a proper exchange factor, a satisfactory understanding of the porous structure
of MSW is needed, as shown in Figure 3.3 (a). The conductivity in Equation (3-28) is a
function of the radiation exchange factor and the distance between two neighboring voids.
A detailed derivation of dp is provided in this thesis. In a packed bed with regularly
shaped particles, a suitable formula that relates the particle diameter or sizes to the
distance between neighboring voids exists and can be mathematically derived. Because
MSW material is a structure that is packed with irregular scraps of paper board, the
calculation of the distance between neighboring voids is highly dependent on the shape of
the particles. Thus, the coefficient “𝑎” is adopted to modify the distance. As shown in
Figure 3.3 (a), the value of this coefficient for paper board scraps should exceed 1. The
radiation shape factor between neighboring voids is also highly affected by the porosity.
36
For a cylindrical unit cell in the Zehner-Bauer-Schlünder model [73], this value is
expressed as [1-(1-ε)1/2], as shown in Figure 3.4 (a). Note that this value is always less
than 1. A different expression was applied here to modify the Kunii and Smith model.
Because the radiant heat conductivity coefficient is considered to be a function of 𝜀
1−𝜀, a
new expression (cε
1-ε)B was proposed here. The expression is as follows:
𝛼𝑟𝑣 = 4𝜎𝜖𝑇3 ∙ 𝑎(
𝑐𝜀
1 − 𝜀)𝐵 (3-29)
By rearranging the expression, a new expression with only two unknowns can be obtained
as
𝛼𝑟𝑣 = 4𝜎𝜖𝑇3 ∙ 𝐴(
𝜀
1 − 𝜀)𝐵 (3-30)
where
𝐴 = 𝑎(𝑐𝐵) (3-31)
Figure 3.3 Porous structure of MSW composed of cardboard and PE: (a) Porous
structure prior to decomposition and (b) Porous structure after decomposition.
3.2.4.3 A model based on the Zehner-Bauer-Schlünder model
The detailed correlation of the Zehner-Bauer-Schlünder model [74] is shown in Figure
3.4(a). The following modes of the heat transfer mechanism are included:
Conduction through the solid phase and across the contact interface between particles
(keffsc ).
Conduction through the solid phase and radiation between particles (keffsr ).
Conduction through the solid phase and across the stationary fluid phase, which fills the
interstitial voids between the particles (keffsf ).
37
The total effective conductivity of the porous packed particles can be expressed as
follows:
𝑘𝑒𝑓𝑓𝐻 = 𝑘𝑒𝑓𝑓
𝑠𝑟 + 𝑘𝑒𝑓𝑓𝑠𝑓+ 𝑘𝑒𝑓𝑓
𝑠𝑐 (3-32)
The effective thermal conductivity that considers void-to-void radiation is given by the
Zehner-Schlünder equation as modified by Breitbach and Barthels [133]
𝑘𝑒𝑓𝑓𝑠𝑟 =
(
[1 − (1 − 𝜀)1/2]𝜀 +
(1 − 𝜀)1/2
2𝜖 − 1
𝛽𝑓 + 1
𝛽𝑓
1
1 +1
(2𝜖 − 1)𝜑)
4𝜎𝑇3𝑑𝑝 (3-33)
where
𝛽𝑓 = 1.25(1 − 𝜀
𝜀)10/9 (3-34)
and
𝜑 =𝑘𝑠
4𝜎𝑇3𝑑𝑝 (3-35)
The equation that describes conduction through the pebbles across the contact between
the pebbles is given by
𝑘𝑒𝑓𝑓𝑠𝑐 = √1 − 𝜀(𝑓𝑐𝑜𝑛𝑡𝑎𝑐𝑡𝑘𝑠 ) (3-36)
where 𝑓𝑐𝑜𝑛𝑡𝑎𝑐𝑡 is the flattening factor that represents the contact area.
The effective thermal conductivity that considers the thermal conduction within the voids
filled by the gas and conduction of the solid particles is expressed as
𝑘𝑒𝑓𝑓𝑠𝑓
𝑘𝑔𝑔 = 1 − √1 − 𝜀 + 𝒇𝒄𝒐𝒏𝒕𝒂𝒄𝒕
2√1 − 𝜀
1 − 𝑘𝛽𝑓[(1 − 𝑘)𝛽𝑓
(1 − 𝑘𝛽𝑓)2ln (
1
𝑘𝛽𝑓) −
𝛽𝑓 + 1
2−𝛽𝑓 − 1
1 − 𝑘𝛽𝑓] (3-37)
where
𝑘 = 𝑘𝑔𝑔/𝑘𝑠
𝑠 (3-38)
38
(a)
(b)
(c)
Gas
1-
Contact area
Contact area
PE
Cardbboard
q1 q2 q3
Contact area
1-
1-
Figure 3.4 Unit cell model: (a) unit cell model from Zehner and Schlünder [73], (b) unit
cell without PE, and (c) unit cell with PE.
3.2.4.4 Determination of dp
The derivation of the radiation coefficient from void to void in a porous bed has been
given by Argo and Smith and William Schotte [134] and [135]; an improvement and a new
explanation have been applied to dp in this paper. Radiation is transferred from void1 to
void2 across the void space and past the particles, as shown in Figure 3.5 (a). Considering
the void space, the total area for the heat transfer for a particle can be approximated by
𝜋
4𝑑𝑝2(
1
1 − 𝜀) (3-39)
and a void area of
𝜋
4𝑑𝑝2(
𝜀
1 − 𝜀) (3-40)
Thus, the radiation heat transfer across the void area is
𝑞𝑟 = −ℎ𝑟 (𝜋
4𝑑𝑝2 𝜀
1−𝜀) ∙ ∆𝑇 (3-41)
where ∆T represents the temperature difference between the two voids, which can be
rewritten as the temperature gradient multiplied by the distance between the central
points of two voids.
∆𝑇 = 𝑑𝑝 ∙𝑑𝑇
𝑑𝑥 (3-42)
39
For the porous bed of municipal solid waste, as shown in Figure 3.5 (b), part of these
voids is occupied by the melting polyethylene. Assuming the pressure from the bed height
is neglected, the void space surrounding the cardboard particles will increase with the
conversion of non-charring materials, such as polyethylene. Thus, a modified dp is given
for the blending of cardboard and polyethylene as
𝒅𝒑 = 𝒅𝒑,𝑐𝑎𝑟𝑑𝑏𝑜𝑎𝑟𝑑 +𝑉𝑝𝑒
𝑉𝑐𝑎𝑟𝑑𝑏𝑜𝑎𝑟𝑑𝒅𝒑,𝑐𝑎𝑟𝑑𝑏𝑜𝑎𝑟𝑑 ∗ (1 − 𝑋𝑝𝑒) (3-43)
where dp,cardboard is the effective diameter of the cardboard particle with the same
volume, which can be calculated; Vpe and Vcardboard are the initial volume fractions of
polyethylene and cardboard, respectively; and Xpe is the conversion rate of polyethylene,
which varies from 1 to 0.
Figure 3.5 Determination of 𝒅𝒑: (a) MSW beds with cardboard scraps and (b) MSW beds
with cardboard and polyethylene composites.
Table 3.4 Input parameters used in the model
Input parameters Value
Diameter of cardboard particles, D (mm) 5× 3× 0.1 mm3
Intrinsic density of cardboard particle, (kg/m3) 1765 [32]
Initial density of polyethylene, (kg/m3) 922
Specific heat of cardboard, (J/(kg K)) 1120+4.85·(T-273) [68]
Specific heat of polyethylene, (J/(kg K)) 2300+2·(T-273) [136],[137],[138]
Specific heat of char, (J/(kg K)) 1003+2.09·(T-273) [68]
Specific heat of volatiles, (J/(kg K)) 770+0.629·T+0.000191·T2 [68]
Thermal conductivity of cardboard, (W/(m K)) 0.13+0.0003·(T-273) [68]
Thermal conductivity of char, (W/(m K)) 0.08-0.0001·(T-273) [68]
Thermal conductivity of polyethylene, (W/(m K)) 0.34 [137]
Thermal conductivity of melting polyethylene, (W/(m K)) 0.28 [139]
40
3.3 Thermodynamic analysis
The minimization of the total Gibbs free energy is a commonly employed and suitable
method for calculating the equilibrium compositions of any reacting system [140]. The
method of thermodynamic analysis by the minimization of Gibbs free energy has been
discussed in many studies [141] and [142]. The minimization function of the total Gibbs
free energy for a reaction system of N compounds with iN moles of species i can be
expressed as
1
0 0
, ,01
ln ( ) 0N
i ii i T j ij c c T
ji
y PN G RT a N G
P
(3-44)
where 𝐺𝑖,𝑇0 is the standard Gibbs free energy, 𝑁𝑖 is the number of moles of species 𝑖, 𝑅 is
the molar gas constant, 𝑇 is the temperature of the system, 𝑦𝑖 is the gas-phase mole
fraction, Φ𝑖 is the fugacity coefficient of species 𝑖, 𝜋𝑖 is the Lagrange multiplier, and 𝑎𝑖,𝑗
is the number of atoms of the𝑗th element in each molecule of species 𝑖.
A thermodynamic equilibrium calculation that employs Gibbs energy minimization for
each type of reforming process was modeled in the Aspen Plus program, which is capable
of single-phase or multi-phase multi-component equilibrium modeling.
Each reforming process was modeled separately. The processes have different reactants;
therefore, different products were expected in the total reactions. In the models, the
stoichiometric ratio of the reactants was investigated. The settings for each reforming
process are listed in Table 3.5. The reforming temperature ranged from 300 K to 2000 K,
and the pressure ranged from 0.5 to 3 atm. The calculations were computed using the
Soave-Redlich-Kwong method. The proportions of the reactants and the products and
the reforming temperatures were clarified and compared.
Several parameters, such as CH4 conversion, CO2 conversion, H2 yield, and H2/CO, were
introduced to evaluate the performance of multi-reforming. The higher the conversion of
reactants, the better the results; therefore, the conversion ratios of CH4 and CO2 were
defined as
4 4 44 , , ,% 100 ( )CH in CH out CH inCH conversion F F F
(3-45)
2 2 22 , , ,% 100 ( )CO in CO out CO inCO conversion F F F (3-46)
2 4 22 , , ,% 100 (2 )H out CH in H O inH yield F F F
(3-47)
22 , ,/ H out CO outH CO F F
(3-48)
41
where 𝐹𝑖𝑛 is the molar flow rate input and 𝐹𝑜𝑢𝑡 is the molar flow rate output. The
H2/CO ratio is another important parameter. According to the Fischer-Tropsch synthesis,
a ratio less than 3 is required, but a value of approximately 2 is preferred for methanol.
Therefore, the H2/CO ratio is a useful parameter for evaluating the optimal reforming
conditions; it is frequently related to the input reactants.
Table 3.5 Reactants and theoretical products of each reforming type in the multi-
reforming of CH4 with CO2.
Reforming type
Input reactants Products
Dry reforming
CH4:CO2=1:1 C,CO,CO2,CH4,H2,H2O,C2H2,C2H4,C2H6
Bi-reforming CH4:CO2:H2O=2:1:1 C,CO,CO2,CH4,H2,H2O,C2H2,C2H4,C2H6
Auto-thermal reforming
CH4:CO2:O2=2:1:0.5 C,CO,CO2,CH4,H2,H2O,O2,C2H2,C2H4,C2H6
Tri-reforming CH4:CO2:H2O:O2=3:1:1:0.5 C,CO,CO2,CH4,H2,H2O,O2,C2H2,C2H4,C2H6
Table 3.6 Conditions of experiment
Case Number
CH4 input, Nm3/h CO2 Input, Nm3/h Pressure, atm
Temperature, K
1 0.5 0.5 1 1153
2 0.5 0.5 1 1169
3 0.5 0.5 1 1213
42
Chapter 4
4. Pyrolysis behaviors of pelletized recovered
municipal solid waste fuels
Pelletized recovered solid waste fuel is always applied in a pyrolysis/gasification system to
obtain feedstock with a stabilized quality and a high heating value and to prevent the
bridging behavior caused by a high moisture content, a low particle density, and an
irregular particle size. The pyrolysis behaviors of pelletized recovered fuels were tested in
this chapter. A detailed comparison of the pyrolysis behaviors between typical recovered
solid waste and biomass particles was performed.
4.1 Results and discussion
4.1.1 Heat transfer properties
The center temperature profiles of the pyrolysis stage for different temperatures varied
from 900 °C to 500 °C, as shown in Figure 4.1. Two phases were observed during heating
to 900 °C; they were more visible at lower temperatures. The primary phase involves the
removal of volatile organic matter from the pellets. This phase can be clearly identified
from the change in the shape of the curve between 350 °C and 400 °C. The primary
decomposition of the majority of the components in the center begins in the temperature
zone from 350 °C to 400 °C. The heating rate slows due to the endothermic reactions of
primary pyrolysis in the center and the decrease in the conductivity caused by the porous
structure of the char, which seems to be lower compared with the former phase. The
second phase heats the remaining material, which primarily consists of carbonaceous char
after the de-volatilization of the volatile components.
The liquid product from the pyrolysis of the plastic groups contained in the recovered
solid waste serves a crucial role in triggering the risk of bridging in the retorting zone. The
lower the temperature, the longer the duration of the primary phase. The risk of bridging
can increase at lower temperatures due to the higher viscosity of the sticky material in the
low-temperature zone. Figure 4.1 shows no significant increase of the rate with an
increase in the center temperature when the atmosphere temperature is varied from
43
700 °C to 900 °C. From 500 °C to 700 °C, however, a significant change is apparent,
including a longer duration of the primary phase and a lower rate of temperature increase.
These results demonstrate the risk of bridging for recovered solid waste in the lower
temperature zone, especially between 500 °C and 700 °C.
Figure 4.1 Center temperature profiles for different temperatures
4.1.2 Devolatilization behavior
Compared with the straw pellets, the effect of temperature on the conversion time of the
RDF is much more significant, as shown in Figure 4.2. The RDF particles have a higher
thermal resistance and require a longer conversion time. A total of 83 % of the RDF
sample was converted in the first 100 seconds at 800 °C, whereas the residues for 700 °C
and 600 °C were 26 % and 72 %, respectively, of the initial sample mass. At 700 °C,
approximately 150 s is required to convert 83 % of the initial sample weight compared
with 100 s at 800 °C. When the temperature is reduced to 500 °C or a lower temperature,
the high conversion proportions obtained at 700 °C cannot be attained, even if a much
longer running time is provided. Only 54 % of the sample weight was converted in the
first 500 s at 450 °C. The final char yield at different temperatures for both the recovered
solid waste and the straw pellets are shown in Figure 4.3. The pyrolysis temperature
should be maintained at approximately 500 °C to increase the recovered solid waste
conversion. The char yield of the recovered solid waste rapidly decreases with increasing
temperature compared with the char yield of the straw pellets. This result is attributed to
the high plastic group concentration in the recovered solid waste, which requires a high
0 20 40 60 80 100 120 140 160 180 200 220 240
0
50
100
150
200
250
300
350
400
450
500
550
600
650
700
750
800
850
900°C
800°C
700°C
600°C
500°CCen
ter
Tem
pera
ture
(°C
)
Time (S)
44
decomposition temperature and produces a lower char yield compared with the cellulosic
groups.
Figure 4.2 Mass loss curves of RDF pellets for different conditions.
Figure 4.3 Char yields from the pyrolysis of RDF pellets and straw pellets
The areas under the curves in Figure 4.4(a, b) correspond to the conversion of the
recovered solid waste particles and straw pellets during pyrolysis at different temperatures,
and the heights of the peaks indicate the reaction rates. The recovered solid waste
particles display a much slower maximum reaction rate compared with the straw pellets at
the same temperature. Two stages that correspond to the moisture evaporation and
0 50 100 150 200 250 300 350 400 450 500
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
RDF,800°C
RDF,700°C
RDF,600°C
RDF,500°C
RDF,450°C
Straw,700°C
Straw,600°C
Straw,500°C
Straw,450°C
m/m
0
Time (S)
400 500 600 700 800 900
5
10
15
20
25
30
35
40
45
50
RDF pellet
Straw pellet
Ch
ar
yie
ld (
%)
Temperature (°C)
45
devolatilization processes for the straw pellets at low temperature are observed. With an
increase in temperature, the overlap of the two stages also increases. In Figure 4.4(a, b),
the first stage is represented by one shoulder of the entire peak at 500 °C. For the
recovered solid waste particles, the three reaction rate curves at 600 °C, 500 °C and
450 °C show similar trends with time; the main difference is that the corresponding peaks
are shifted forward and the overlap is increased at high temperatures. The existence of the
peaks and shoulders is attributed to the complex components in these materials. When
the temperature exceeds 700 °C, all small peaks and shoulders overlap to form one large
peak. A high temperature can increase the conversion. With the rapid heat transfer at high
temperatures, all components can be emitted within a short time, despite their different
decomposition temperatures. Therefore, a lower temperature presents a more distinct risk
of bridging because the variety of plastics with different decomposition temperatures and
liquid tars with higher viscosities at lower temperatures produce a much longer
thermoplastic transition zone. In real cases in a fixed-bed gasifier, the feedstock passes
through the retorting zone at a low temperature—from 500 °C to 700 °C—before
transferring to the char gasification zone. Thus, the temperature zone from 500 °C to
700 °C is crucial for studying the effect of swelling on the risk of bridging.
Figure 4.4 Reaction rate profiles of straw and RDF pellets.
4.1.2 Swelling/shrinkage phenomena during the pyrolysis process
The swelling and shrinkage curves for the pyrolysis of the RDF pellets and straw pellets
are presented in Figure 4.5(a, b). The majority of the biomass materials display significant
shrinkage phenomena during the pyrolysis stages. The sample initially undergoes a slight
swelling process and subsequently experiences a significant contraction. As shown in
Figure 4.5(a, b), the volume of the final char is only 56 % of the initial volume. The
swelling phenomenon that occurs during the pyrolysis of RDF pellets is significant
compared with the swelling phenomenon of the straw pellets. The swelling ratio climbs to
1.18 during the first 10 s at 660 °C but produces a value of only 1.01 for straw at the same
temperature. The highest maximum swelling ratio can reach 1.54 and 1.58 at 550 °C and
0 50 100 150 200 250 300 350 400 450 500
0.000
0.002
0.004
0.006
0.008
0.010
0.012
0.014
0.016
Mass lo
ss r
ate
(1/S
)
Time (S)
RDF, 600°C
Straw,600°C
RDF, 500°C
Straw,500°C
RDF, 450°C
Straw,450°C
(a)
0 50 100 150 200
0.000
0.002
0.004
0.006
0.008
0.010
0.012
0.014
0.016
0.018
0.020
dm
/dt
(1/S
)
Time (S)
RDF,900°C
RDF,800°C
RDF,700°C
Straw,700°C
(b)
46
660 °C, respectively. At lower temperatures, the process requires additional time to attain
the highest swelling ratio: 70 s at 550 °C or 40 s at 660 °C. After attaining the maximum
swelling ratio, this stage lasts approximately 30 seconds and is followed by a long
contraction stage. Both the final swelling/shrinkage ratios remain above 1, which
indicates that no shrinkage was detected compared with the initial volume.
Figure 4.5 Volume change during pyrolysis: (a) Straw pellets and (b) RDF pellets
The volume change for both RDF samples indicates a rapid response time. The particles
begin swelling approximately 10 s after being dropping into the reactor at 550 °C. The
surface temperature, center temperature, and mass loss rate are shown in Figure 4.6 (a, b).
As shown in Figure 4.6 (a, b), the swelling ratio increases significantly when the surface
and center temperatures reach 180 °C, which is a temperature at which many plastics
begin to melt and enter the thermoplastic transition. The evaporation of the contained
moisture and volatiles from the pyrolysis of the cellulosic groups and part of the plastic
groups simultaneously burst through the particle. This process suggests that the
formation of unstable liquid from the melting of plastic significantly contributes to the
swelling ratio. At higher temperatures, additional transient liquid forms within the particle,
and bubbles of volatiles from lignin group decomposition join the eruption and cause
swelling in the softened RDF mass, as evidenced by the significant increase in the particle
diameters compared with the diameters of the original particles. The same phenomenon is
observed in Figure 4.6 (a, b). Both the material structure and the mass transfer serve a
significant role in the process. Certain coal, i.e., bituminous coal, exhibit plasticity and
swelling phenomena during heating. The study showed that the transport of volatiles via
the growth of a large population of bubbles that are uniformly dispersed in the plastic
coal melt is highly controlled by the pyrolysis behavior of the softening coal [143]. The
RDF particles exhibit a mass transfer phenomenon similar to the mass transfer
phenomenon of the softening coal. However, the majority of the plastic materials in the
RDF have a lower thermoplastic temperature compared with the softening coal, which
explains the rapid response time of the swelling during the pyrolysis process. The typical
temperature at which coal softening begins is approximately 500 °C, and the maximum
swelling ratio of the RDF particles is attained when the center temperature is only 300 °C.
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Swelling/Shrinkage ratio, under 660°C
Mass loss rate,under 660°C
Time (S)
Sw
ell
ing
/sh
rin
ka
ge
ra
tio
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
0.040
0.045
0.050
Ma
ss
lo
ss
ra
te (
1/s
)
(a)
0 20 40 60 80 100 120 140 160 180 200
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
Swelling under 660°C
Swelling under 550°C
Sw
ell
ing
/sh
rin
kag
e r
ati
o
Time (S)
(b)
47
The RDF particles remain in the second stage after reaching the maximum ratio and prior
to entering the contraction process. When a temperature of 400 °C is attained, the
particles begin to contract, which is a process that is attributed to the frequent bubble
ruptures caused by the significant decrease in the viscosity of the melting material with an
increase in temperature. The increase in the internal volatile pressure competes with
changes in the plastic liquid viscosity and surface tension, which are highly dependent on
the temperature. The mechanism of volatile transport through the RDF particles must be
substantially different from the process that occurs in the straw particles. Additional
explanations should be investigated based on the test results from the char materials.
(a) (b)
Figure 4.6 Surface and center temperatures and reaction rate profiles during pyrolysis: (a)
RDF, 550 °C and (b) RDF, 660 °C.
The apparent change in density of the RDF pellets during pyrolysis at both temperatures
is shown in Figure 4.7. The deviation of the densities is dependent on the
swelling/shrinkage ratios during the test. The most distinct deviation of both apparent
densities is observed in Figure 4.7. The real density that considers the effect of swelling
on the volume change is much lower than the real density that does not consider swelling.
Low-density feedstock in a fixed bed may also trigger the risk of bridging. Thus, the
significant decrease in the real density of the RDF pellets caused by the swelling may
increase the risk of bridging in the retorting zone.
48
Figure 4.7 Change in RDF pellet density during the test.
4.1.2 Char properties
The final char results obtained from the pyrolysis of the RDF and straw at 500 °C are
shown in Figure 4.8. A porous structure can be detected on the surface of the RDF char,
whereas the straw pellet char exhibits a more compact structure. The formation of pores
results from the escape of volatiles from the non-char materials when the plastic groups at
the outer layer are melted during pyrolysis. The diameters of the final RDF and straw char
are shown in Figure 4.9, in which the diameter of the RDF char is slightly larger than the
diameter of the original RDF.
Figure 4.8 RDF and straw char
49
Figure 4.9 Diameter of the char produced from the pyrolysis of straw and RDF
Scanning electron microscopy was performed on the char obtained from the pyrolysis to
examine the difference in the swelling/shrinkage behaviors of the recovered solid waste
and the straw pellets. The SEM images of the char are shown in Figure 4.10(a-f). The
straw char pellets exhibit a compact structure with a regular shape. In Figure 4.10(c, d),
the fibers are stacked layer-by-layer due to the compression during the pelletized process
and remain as a multilayered structure in the longitudinal direction. The surface is distinct.
Abundant pores can also be detected from the cross-area of the fibers. In Figure 4.10(a, b,
c), the RDF char consists of filamentous structures and irregular particles. The remaining
entangled filamentous structures originate from the cellulosic fibers in the recovered solid
waste, and the irregular particles dispersed over the surface of the filamentous fiber are
attributed to the char and mineral matter obtained from the plastic and wood
components [144]. As shown in Figure 4.10(a), the surface of the char is covered by a thin
layer of smooth material, which is generated and retained during the plastic melting and
pyrolysis processes and can be compared with the surface of the char produced from
higher-temperature decomposition. Circles from the rupture of bubbles randomly cover
the surface in Figure 4.10(a). This appearance suggests that the recovered particles
exhibited plasticity during the pyrolysis. With an increase of the temperature, the fluffy
structure of the char becomes more visible. The fluffy filamentous fibers, which can
increase the integrity of the recovered solid waste particles, serve an important role in the
pyrolysis process. Compared with a typical coal particle or biomass particle, the recovered
solid waste particle exhibits a more heterogeneous structure. Because the softening
temperatures of the majority of the plastics fall within the temperature zone of 110 °C to
260 °C, the plastic groups contained in the RDF particles exhibit plasticity during the
heating phase. The volatiles from the pyrolysis of the cellulosic groups with lower
decomposition temperatures begin to burst and escape from the particle. Thus, many
bubbles form throughout the RDF particle mass and cause the swelling phenomenon.
500 550 600
6.0
6.5
7.0
7.5
8.0
8.5
9.0
Straw pellet
Straw char
RDF pellet
RDF char
Dia
mete
r (
mm
)
Temperature (°C)
50
Figure 4.10 SEM images of char: (a) RDF, 550 °C, 300 µm; (b) RDF, 660 °C, 300 µm; (c)
RDF, 780 °C, 300 µm; (d) Straw, 660 °C, 200 µm; (e) RDF, 780 °C, 100 µm; and (f) Straw,
780 °C, 100 µm
To explain the differences in the surface area of the chars obtained at 550 °C, the BET
surface area, the total pore volume and the average pore diameter are listed and compared
in Table 4.1. The pellet particles were ground to a particle size of less than 2 mm. The
RDF char exhibits a lower BET surface area and total pore volume compared with the
straw char. This result suggests that the thin layer of smooth material generated from the
plastic melting and pyrolysis has a significant influence on the BET surface area and the
total pore volume of RDF char at 550 °C, which may be attributed to the finding that the
thin layer of smooth material covered the surface of the RDF char and blocked the pores.
(c) (d)
(b)(a)
(e) (f)
51
Table 4.1 BET analysis of RDF char and straw char obtained at 550 °C
Sample BET surface area (m2/g) Total pore volume
(cm3/g)
Average pore diameter
(Å)
Straw char 0.882 0.0019 87
RDF char 0.13 0.00027 77
This analysis concludes that the swelling behaviors of the RDF particles is significant and
may be highly dependent on their thermoplastic properties and heterogeneous structures.
In addition to these factors, the pressure and particle size can also affect the swelling and
shrinkage behavior. Numerical studies of the effect of pressure on the swelling behaviors
of softening coal have been performed and published. The swelling process is actually a
mass transfer process of the volatiles through the solid and liquid masses, and the
increased pressure provides a larger resistance to the transport of liquids to the surface of
the particle. Due to the change in the external pressure, the vapor-liquid equilibrium
becomes more important by forcing the higher-molecular-weight material to remain as a
liquid rather than evaporating at a higher pressure [145]. However, the phenomenon by
which a coal particle exhibited almost no swelling at high pressures but swelled
significantly at elevated pressures was also observed [143]. These results indicate the
existence of a complex variation in the materials during the devolatilization and swelling
behavior with applied pressure. The density and size of the particle, which significantly
affect the volatile mass and transfer process, may also affect the mass transfer process.
A single RDF pellet and straw pellet were considered to demonstrate the different
swelling and shrinkage behaviors, and the melting phenomenon can be observed during
pyrolysis due to the considerably high concentration of plastic groups in the pellets. A
fragile and porous structure from the residue obtained from RDF pyrolysis was also
observed. A solid material bed with a proper height is necessary to maintain stable and
steady operation of a fixed-bed gasifier. Thus, the exposure of single particles to the bed
pressure is unavoidable during decomposition. Given that the RDF generally has a lower
compressive strength than coal and biomass materials at elevated temperatures and that
the majority of plastic constituents will soften at 150 °C, these particles most likely will
merge during the devolatilization process. Subsequently, the pellet-shaped particles will
break, and part of the void spacing in the bed will be occupied by the melted plastic
constituents.
4.2 Summary
Pelletized recovered solid waste fuel has a higher thermal resistance and a longer
conversion time compared with straw pellets. Due to the high concentration of plastic
52
groups with a high decomposition temperature, the char yield of pelletized recovered
solid waste significantly decreases with increasing temperatures, especially between 450 °C
and 700 °C.
The volumetric change in pelletized recovered solid waste fuels displayed a rapid response
time. A significant transient volumetric swelling behavior was also observed during the
pyrolysis of pelletized recovered solid waste fuels. The swelling phenomenon begins when
a surface and center temperature of 180 °C is attained, which is lower than the
decomposition temperature of the majority of the components in the recovered solid
waste. This observation can be explained by the evaporation of the moisture in the plastic
liquid. With an additional increase in temperature, additional volatiles from the pyrolysis
of the cellulosic groups and part of the plastic groups burst through the particle. The
maximum swelling ratio is attained when the center temperature is approximately 300 °C.
The material structure and mass transfer play significant roles in the entire process. Once
a center temperature of 400 °C is attained, a significant contraction process begins, which
is most likely due to the frequent bubble ruptures caused by a decrease in the viscosity
and surface tension of the plastic liquid with increasing temperature.
The SEM results confirmed that the RDF particles exhibit plasticity during heating. The
increased temperature improves the visibility of the fluffy structure of the char. The
remaining filamentous structure, which originates from the entangling of cellulosic fibers,
is important in the swelling process. Its original material has a lower decomposition
temperature and may provide most of the volatiles that form bubbles when the plastic
groups melt into a liquid.
A significant reduction in the apparent density of the RDF particles also occurs during the
swelling process and may increase the risk of bridging in a fixed-bed gasifier.
53
Chapter 5
5. Heat transfer mode in porous MSW particles
In the previous chapter, the pyrolysis behaviors of pelletized RDF particles were
identified. In this study, a model was constructed to describe the mass and heat transfer
processes inside a porous MSW particle that is composed of polyethylene and cardboard.
The heat of PE melting was also considered and incorporated into the particle model. A
one-step global model and a scheme that involves two consecutive first-order reactions
were adopted to describe polyethylene degradation and cardboard degradation,
respectively. The particle model, in which the contribution of the convective mass flow
of the volatiles on the thermal balance was considered, simulates conductive and
convective heat transfer within a porous particle. The mechanisms of heat transfer in the
absence of gas flow were considered, including the conduction term and the radiation
term. Several typical heat transfer models, which are based on these mechanisms and are
used to estimate the effective thermal conductivity, were selected and modified to fit with
the experimental results. The effect of the heat transfer model on the prediction of the
pyrolysis process of municipal solid waste (MSW) has been investigated in this paper.
5.1 Results and Discussion
5.1.1 TGA-DSC analysis
TGA, DTG and DSC curves from the pyrolysis of printing paper and cardboard are
displayed in Figure 5.1 (a) and Figure 5.1 (b). The TGA curves reveal that the release of
moisture causes a minor loss in weight. At this stage, printing paper and cardboard
experience a loss of 3.2 % and 3.6 %, respectively. The volatile release stage is the most
important stage of the thermogram. As shown in Figure 5.1 (a), a large mass loss in the
temperature range between 250 °C and 375 °C is observed, which is primarily attributed
to the cellulose pyrolysis. Cardboard exhibits a loss of 64.5 %, whereas printing paper
only exhibits a loss of 55.5 % in the same temperature zone—between 105 °C and 370 °C.
The rate of mass loss derived from the derivative TGA is an indication of reactivity of the
sample feedstock. Cardboard has a higher reactivity than printing paper. In the
temperature zone between 250 °C and 280 °C, a shoulder can be observed on the DTG
54
curve of the printing paper. Similar phenomena are not observed during the cardboard
pyrolysis, which indicates that the printing paper contains considerable hemicellulose or
extractives.
As shown in Figure 5.1 (b), the results from the DSC curves correspond with the findings
of the DTG curves and provide more information about char carbonization. No lids were
used in the experiment. The emissivity of chars is significantly higher than the emissivity
of an alumina crucible, which is used as a reference. A rising baseline, which is attributed
to the increasing heat flux caused by the radiation at elevated temperatures, can be
detected from the curves at high temperatures. However, several peaks of heat flow are
observed in the rising baseline. Two peaks, which are located in the DSC curves of
printing paper at temperatures between 250 °C and 300 °C, correspond to the
decomposition of hemicellulose or extractives in printing paper. Similar peaks are not
observed in the cardboard DSC curves. With increasing temperature, the highest
endothermic heat flow, which is contributed by the cellulose pyrolysis, is detected at
350 °C for both printing paper and cardboard. With an increase in cellulose consumption,
the transient heat flux decreases until approximately 370 °C. Subsequently, a new heat
flow peak, which corresponds to char carbonization, is detected. Note that this process is
an endothermic process; however, char carbonization has exhibited an exothermic
reaction at temperatures above 350 °C [146]. This finding indicates that an intermediate is
produced from the primary pyrolysis, which can decompose and form new char. This
intermediate is completely different from anhydrocellulose and active cellulose, which act
as an intermediate prior to primary pyrolysis because it begins at approximately 380 °C.
The process exhibits a minor mass loss but significantly contributes to the heat of
pyrolysis.
Researchers have suggested that the heat of pyrolysis is dependent on the extent of char
formation [147]. A nearly linear relationship between mass loss and heat absorption below
360 °C is observed in Figure 5.1 (c), which indicates that a large portion of char formation
is distributed over the entire period of pyrolysis. Competing processes of volatiles and
char formation completely overlap. Due to an increase in temperature, another linear
relationship derived from the endothermic reactions of char (intermediate) decomposition
is observed. However, a deviation in the curves is apparent at high conversion, which
suggests that the exothermic reaction from char formation begins to dominate at high
temperatures.
55
Figure 5.1 TGA-DTG-DSC curves of cardboard and printing paper at a heating rate of
10 ︒C/min: (a) TGA-DTG, (b) DSC (all crucibles without lids, 225 °C-450 °C), and (c)
Enthalpy of pyrolysis as a function of mass loss.
Many possible reaction schemes have been proposed. These pyrolysis models can be
divided into three principal categories: single-step global reaction models, multiple-step
models, and semi-global models [148]. Based on the previous discussion about the DSC
test, DSC curves cannot be created from a single reaction of any order. Alves and
Figueiredo suggested that the cellulose pyrolysis can be modeled using three consecutive
first-order reactions [48]. The first reaction and the third reaction release approximately
30 % and 70 %, respectively, of the volatile matter, whereas the second reaction releases
no volatile matter. A suitable linear relationship that is based on the Coats and Redfern
method was obtained in this paper, as shown in Table 5.1. It encompasses more than 70 %
of the devolatilization and indicates that a model of first-order kinetics is adequate for
temperatures between 220 °C and 375 °C. A two-reaction model based on the previous
discussion is proposed for the pyrolysis of cardboard and printing paper, as shown in
Scheme 3-1. The first reaction includes the mass loss of 70 % at temperatures from
200 °C to 375 °C, and the second reaction encompasses the mass loss of 30 %, which is
primarily attributed to char carbonization at temperatures between 375 °C and 500 °C.
0 50 100 150 200 250 300 350 400 450 500 550 600 650
0
10
20
30
40
50
60
70
80
90
100
Cardboard,TG
Paper,TG
Paper, TGA
Cardboard,TGA
Temperature (°C)
TG
(%)
-0.20
-0.15
-0.10
-0.05
0.00
DT
G (
%/S
)
(a)
200 250 300 350 400 450
-2
0
2
4
6
8
10
Printing paper (5.90mg)
Cardboard (5.95mg)
He
at
flu
x (
mW
)
Temperature (°C)
(b)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8-250
-200
-150
-100
-50
0
Cardboard
Printing paper
En
thalp
y o
f p
yro
lysis
(J/g
)
(M0-M)/M
0 (100%)
(c)
56
Table 5.1 Kinetic parameters of the thermal degradation of cardboard and printing paper
calculated by Coats and Redfern, which approach 10 °C/min.
Samples Temperature E (kJ/mol) n Pre-exponential Factor, A (1/S)
Correlation coefficients
Cardboard
200-375 °C
101.108
1
1335509
0.9966
Printing paper
200-375 °C
94.426
1
327278
0.9931
The TGA, DTG and DSC curves from the pyrolysis of polyethylene are displayed in
Figure 5.2 (a) and Figure 5.2 (b). The comparison with paper board materials suggests that
polyethylene degradation is an endothermic process that requires a significantly higher
heat flux.
Figure 5.2 TGA-DTG-DSC curves of PE at a heating rate of 10 °C /min: (a) TGA-DTG
and (b) DSC (all crucibles without lids, 200 °C-600 °C).
5.2.2 Prediction of MSW pyrolysis
The results from the Kunii and Smith model and the modified Kunii and Smith model
were compared with the experimental results and are shown in Figure 5.3. According to
Figure 5.3 (a), the prediction from the Kunii and Smith model does not correspond with
the experimental mass evolution. For the Kunii and Smith model, the mass evolution
exhibits a good fit with the experimental results in the first 100 s. The reason for this
finding is that the contribution of radiation in the heat transfer remains insignificant at
low temperatures. A slight increase in porosity is caused by the low conversion degree of
PE. With an increase in elapsed time, the deviation between the model prediction and the
experimental measurement in mass conversion becomes distinct. The mean error of the
absolute differences between the modeled and experimental results is shown in Table 5.2.
It indicates that the Kunii and Smith model does not properly describe the heat transfer
200 250 300 350 400 450 500 550 6000
20
40
60
80
100
TG
DTG
Temperature (°C)
TG
(%
)
(a)
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
DT
G (
%/S
)
200 250 300 350 400 450 500 550 600
0
2
4
6
8
10
12
14
16e
nd
oth
erm
ic
(b)
He
at
flu
x (
mW
)
Temperature (°C)
PE (6.62mg)
57
process in a municipal solid waste particle with a high plastic concentration. Center
temperature profiles are also illustrated in Figure 5.3 (b). Due to the low contribution of
radiation in the Kunii and Smith model, the lowest heating rate of the center temperature
can be observed after 200 s compared with the other models. The model that was
modified based on the Kunii and Smith model corresponds with the experimental results,
with the exception of the beginning and the end of the entire conversion process. This
finding implies that the radiation exchange factor of the radiation term in the heat transfer
model directly varies with porosity, instead of inversely with porosity in the Kunii and
Smith model.
To obtain proper coefficients for “A” and “B”, a series of simulations was performed; the
corresponding curves are shown in Figure 5.3. In Figure 5.3 (a, b), “A” has a significant
effect on the mass evolution and center temperature profiles. With a decrease in “A”, a
longer mass evolution time and lower heating rate are observed from the curves. The
proper values for “A” and “B” were determined to be 0.4 and 0.33, respectively.
Figure 5.3 Experimental results and modeling (K and S, modified K and S) predictions
during the pyrolysis of an MSW cylinder with a diameter of 28 mm at 973 K (initial
porosity, 0.526; PE mass fraction, 75 %; cardboard mass fraction, 25 %): (a) mass
evolution and (b) center temperature profile.
Table 5.2 Error analysis of the simulated results
Setting in the models Mean error* (%)
Modified K&S, A=0.5; B=0.33 2.6 Modified K&S, A=0.4; B=0.33 2.4 Modified K&S, A=0.3; B=0.33 4.1 Modified K&S, A=0.4; B=0.34 2.3 Modified K&S, A=0.4; B=0.32 2.5 K&S model 14.2 *calculated as the mean of the absolute differences between the modeled results and the experimental
results of mass loss: 𝑒𝑟𝑟𝑜𝑟 = ∑│𝑋𝑚𝑜𝑑−𝑋𝑒𝑥𝑝│
𝑛∙ 100%𝑛
0 100 200 300 400 500 6000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
A
Re
sid
ua
l m
as
s f
rac
tio
n (
10
0%
)
Time (s)
Experimental
Modified K&S, A=0.5;B=0.33
Modified K&S, A=0.4;B=0.33
Modified K&S, A=0.3;B=0.33
Modified K&S, A=0.4;B=0.34
Modified K&S, A=0.4;B=0.32
K&S model
(a)
0 100 200 300 400 500 600300
400
500
600
700
800
900
1000
(b)
B
Te
mp
era
ture
(K
)
Time (s)
Experimental
Modified K&S, A=0.5;B=0.33
Modified K&S, A=0.4;B=0.33
Modified K&S, A=0.3;B=0.33
Modified K&S, A=0.4;B=0.34
Modified K&S, A=0.4;B=0.32
K&S model
A
58
The results from the modified Breitbach and Barthels model, which are shown in Figure
5.4, corresponded with the experimental mass evolution. This agreement suggests that the
model adequately predicts the mass evolution during the MSW pyrolysis process.
The effect of the contact surface area on the mass evolution and temperature profile was
also considered in this study. When a larger contact surface area was utilized in the
modified Breitbach and Barthels model, a more rapid increase in the temperature was also
observed. The modeling prediction with a high contact surface area corresponds with the
experimental mass evolution. At temperatures below 450 K, the contribution of the
radiation in the porous beds is smaller than the contribution of the radiation in
conduction mode. Because the shape of all cardboard particles resemble thin chips, the
contact area should be larger than the contact area of a spherical particle with the same
volume, when these chips are stacked layer by layer. This finding implies the importance
of accurately estimating the contact surface area during numerical modeling at low
temperatures. With an increase in the conversion degree of plastics, a loose-packed bed,
in which the contact area will decrease significantly, is expected. In the temperature range
of 500 K to 800 K, the heating rate from the modified Breitbach and Barthels model is
higher that the heating rate of the experimental study. One of the reasons for this
difference is that this model did not consider the reduction in the contact area during
MSW pyrolysis.
Figure 5.4 Experimental results and modeling (modified B and B) predictions during the
pyrolysis of MSW particles with a diameter of 28 mm at 973 K (initial porosity, 0.526; PE
mass fraction, 75 %; cardboard mass fraction, 25 %): (a) mass evolution and (b) center
temperature profile.
The predictions from the two modified heat transfer models—the modified Kunii and
Smith model and the modified Breitbach and Barthels model—at a temperature of 670 °C
were compared with the experimental results, as shown in Figure 5.5. Both predictions
correspond with the experimental data.
0 100 200 300 400 500 6000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
fcontact
Experimental
fcontact=1.0
fcontact=0.5
fcontact=0.2
fcontact=0.0
Resid
ual
mas
s f
rac
tio
n (
100
%)
Time (s)
fcontact
(a)
0 100 200 300 400 500 600300
400
500
600
700
800
900
1000
(b)
fcontact
Cen
ter
tem
pera
ture
(K
)
Time (s)
Experimental
fcontact=1.0
fcontact=0.5
fcontact=0.2
fcontact=0.0
59
Figure 5.5 Experimental results and predictions during the pyrolysis of MSW particles
with a diameter of 28 mm at 943 K (initial porosity, 0.526; PE mass fraction, 75 %;
cardboard mass fraction, 25 %): (a) mass evolution and (b) center temperature profile.
To validate the adoption of the dp determination in the models, an experimental test with
a lower PE concentration and different porosity was conducted at a temperature less than
700 °C. Figure 5.6 shows the experimental and predicted mass evolution of MSW
particles with the same cylinder diameter. The predictions from both modified models
correspond with the experimental data, which suggests that both models are adequate for
the calculation of the thermal conductivity of feedstock with a significant variation in PE
concentration. These results also imply that the distance between two neighboring-void
centers has a significant effect on the radiation heat transfer inside a porous medium,
which is a crucial factor in the calculation of the radiation terms in effective thermal
conductivity.
Figure 5.6 Experimental results and predicted mass evolution during the pyrolysis of
MSW particles with a diameter of 28 mm at 973 K (initial porosity: 0.692; PE mass
fraction: 55 %; cardboard mass fraction: 45 %).
0 100 200 300 400 500 600 7000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Resid
ual
mas
s f
rac
tio
n (
100
%)
Time (s)
Experimental
Modified B&B: fcontact=0.6
Modified K&S:A=0.4;B=0.33
(a)
0 100 200 300 400 500 600 700300
400
500
600
700
800
900
1000
(b)
Tem
pera
ture
(K
)
Time (s)
Experimental
Modified B&B: fcontact=0.6
Modified K&S:A=0.4;B=0.33
60
The case with a high PE concentration was also considered for these heat transfer models.
As shown in Figure 5.7, the maximum PE concentration of the MSW samples in weight
that were used in the experiment is 89 %. A significant deviation in the mass evolution
between the modeling predictions and the experimental results was detected. Note that a
significant collapse of the structure was observed after the experimental test. The MSW
cylinder with a diameter of 28 mm in the basket lost almost 50 % of its height after the
test. The reduction in the porosity affects the size of the voids. In this paper, the
modeling predictions with and without consideration of the collapse of the porous
structure were compared in Figure 5.7. The distance between neighboring voids formed
due to PE conversion was assumed to decrease with the same degree of height of the
solid particles in the vertical direction. Despite the consideration of these factors, more
distinct deviations were observed in Figure 5.7. According to the discussion in the
previous chapter, a large population of bubbles from the particle surface is observed for
MSW particles with high PE concentrations or densities. Thus, a large amount of molten
plastic liquid can be carried with the gaseous volatiles toward the particle surface. No char
layer existed to inhibit the heat transfer; this molten liquid on the surface exhibits a much
faster degradation process compared with the inner layer inside the MSW particle. These
modified models failed to estimate the phenomenon of plastic liquid mobility and are not
adequate to accurately estimate the mass evolution process for maximum plastic
concentrations of 90 %. The comparison of the prediction of total conversion time
between the modified models and experimental results indicates that the modified K and
S model can predict the same conversion time as the experimental results; however,
additional studies are needed to support this conclusion. Although non-charring materials
exhibit a different degradation behavior than charring materials, a heat transfer model that
does not consider the heat of pyrolysis of PE remains adequate to estimate the
conversion time of pure PE particles [66] and [67]. For the mixture of plastic and
cardboard, the particle surface would be covered by molten plastic in the early stage,
whereas the molten plastic on the surface would vanish and exhibit a similar heat transfer
mode with normal porous wooden materials. These situations hamper the construction of
a proper model for the mixture of plastic and cardboard materials.
The heat transfer inside the MSW particles is complicated by the convection of released
volatile matter flux and the heat transfer mode in porous particles. To clarify the effects
of these two factors on heating and mass evolution, four different cases were represented
in the simulation. In case 1, the effects of heat radiation and convection flow were
considered; in case 2, only the heat radiation was considered; in case 3, neither heat
radiation nor convection flow was considered; and in case 4, only convection flow was
considered. The temperatures and mass evolutions of the particles with a diameter of 28
mm are illustrated in Figure 5.8. A significantly lower heating rate and slower mass
evolution rate at high temperatures are observed for case 4, in which the contribution of
heat radiation is disregarded, compared with the curves for case 1 in Figure 5.8. Due to a
higher pyrolysis heat for PE compared with cellulosic materials, the endothermic effect of
61
PE pyrolysis significantly affects the temperature profiles inside the particle, especially
when the effective thermal conductivity inside the porous medium is relatively low. The
effect of convection is significant based on the comparisons of the temperature profiles
and the mass evolution curves between case 1 and case 2 and between case 3 and case 4.
Because the volatile evolution is rapid at high temperatures, the deviation is more
significant at high temperatures.
Figure 5.7 Experimental and predicted mass evolution during the pyrolysis of MSW
particles with a diameter of 28 mm at 973 K (initial porosity: 0.41; PE mass fraction: 89 %;
cardboard mass fraction: 11 %).
Figure 5.8 (a) Center temperature and (b) mass evolution for different cases from the
predictions of the Kunii and Smith model: case 1—heat radiation with volatile convection;
case 2—heat radiation without volatile convection; case 3—without heat radiation and
volatile convection; and case 4—without heat radiation but with volatile convection.
0 100 200 300 400 500 600 700 800 900 10000.0
0.2
0.4
0.6
0.8
1.0
ma
ss
ev
olu
tio
n, m
/m0 (
10
0%
)
Time (s)
Case 1
Case 2
Case 3
Case 4
(a)
0 100 200 300 400 500 600 700 800 900 1000
300
400
500
600
700
800
900
1000
(b)
Ce
nte
r te
mp
era
ture
(K
)
Time (s)
Case 1
Case 2
Case 3
Case 4
62
5.2 Summary
In this study, a model was constructed to describe the mass and heat transfer inside a
porous MSW particle with a high PE concentration. Two extensively used heat transfer
models—the Kunii and Smith model and the Breitbach and Barthels model—were
selected and modified to fit with the experimental results. The effect of the heat transfer
model on the prediction of the municipal solid waste (MSW) pyrolysis process has been
investigated in this paper.
(1) A model used to describe the de-crystallization and endothermic process of PE
was incorporated in the particle model in this paper. This endothermic process
during PE melting slows the heating rate of particles at the temperature zone at
approximately 100 °C.
(2) The prediction from the Kunii and Smith model does not correspond with the
experimental results for the MSW particles, which suggests that the K and S
simplified model does not adequately estimate the heat transfer process inside a
highly porous medium. A coefficient as a function of porosity was proposed to
modify the radiation term in the Kunii and Smith model. Agreement with the
experimental results was obtained. A modified Breitbach and Barthels model
corresponded with the experimental data.
(3) A method to estimate the void size and center distance in MSW particles, which
was incorporated in the heat transfer models, was proposed and validated. A
strong relationship between the accurate estimation of heat transferred by the
radiation mode and the void center distance implies that the void size has a
significant effect on the radiation heat transfer inside a porous medium and is a
crucial factor in the calculation of the radiation terms of effective thermal
conductivity.
(4) All heat transfer models failed to accurately predict the mass evolution of MSW
particles when the PE concentration is 89 %. The mass evolution rate is rapid in
the early stage but subsequently decreases because a large amount of molten plastic
liquid can be carried with the gaseous volatiles toward the particle surface. No char
layer exists to inhibit the heat transfer; this molten liquid on the surface would
exhibit a much faster degradation process compared with the inner layer inside the
MSW particle. Thus, a new heat transfer model that considers the liquid mobility is
needed to predict the MSW degradation process for maximum plastic
concentrations of 90 %.
(5) The convection flow of released volatile matter has a significant effect on MSW
conversion, which implies that the convection term in the heat equation cannot be
disregarded. The radiation mode is the dominant mechanism at high temperatures
63
inside an MSW particle with a high PE concentration because an MSW particle can
form a highly porous structure with the increased porosity caused by PE
degradation.
64
Chapter 6
6. Pyrolysis of the main components of MSW
In this chapter, the products from the rapid and conventional pyrolysis of municipal solid
waste were characterized. To relate the products of MSW pyrolysis with their origins, the
main individual components in MSW, including PE/PET/PVC/printing
paper/cardboard, were characterized. The mechanisms for forming aromatic
hydrocarbons during pyrolysis were also discussed.
6.1 Results and Discussion
6.1.1 Product yields distribution
Tar, syngas, and char yields from printing paper and cardboard pyrolysis at three
temperatures of 400 °C, 500 °C, and 600 °C are shown in Figure 6.1 (a) and Figure 6.1 (b).
For both feedstocks, the yield of syngas increases with an increase in temperature,
whereas char decreases with an increase in temperature. At a flow rate of 1.5 l/min, paper
pyrolysis at 400 °C yields 45.7 % char and 30 % syngas, whereas pyrolysis at 600 °C yields
41.1 % char and 41.6 % syngas. For lignocellulose biomass pyrolysis at low temperatures
between 400 °C and 600 °C, higher pyrolysis temperatures generally result in higher
heating rates, which may result in lower char production and higher tar production. The
tar yield decreases with a continual increase in temperature. For the paper pyrolysis in
Figure 6.1 (a), tar yield continuously decreases with increasing temperature. The tar yield
is 33.19 %, 26.2 %, and 18.3 % at the temperatures of 400 °C, 500 °C, and 600 °C,
respectively, which suggests that the temperature for the maximum tar yield from printing
paper pyrolysis is approximately 400 °C.
Compared with printing paper pyrolysis, cardboard pyrolysis presents a lower char yield, a
lower syngas yield, and a higher tar yield. Cardboard pyrolysis at 400 °C yields 33.4 %
char and 17.6 % syngas, whereas pyrolysis at 600 °C yields 26.5 % and 34.5 %. The tar
yield decreases from 33.76 % at 400 °C to 31.4 % at 500 °C. The differences in the yields
between cardboard and printing paper are consistent with the thermal degradation
characteristics of the printing paper and cardboard obtained from the TGA/DTG.
Compared with the decreased tar yields of printing paper pyrolysis, a more significant
65
decrease in tar yield from cardboard pyrolysis is observed when the temperature increases
from 500 °C to 600 °C. The temperature for maximum tar yield falls within the
temperature zone between 400 °C and 500 °C.
The effect of the flow rate of carrier gas on the product yield is shown in Figure 6.1(b).
The flow rate of carrier gas and the pyrolysis temperature serve important roles. The flow
rate of carrier gas directly determines the residence times of the volatile vapors generated
in the reactor. As the residence time increases at a given temperature, the tar yield
decreases, and the gas yield increases. In Figure 6.1(b), the tar yield increases by more
than 15 % when the flow rate increased from 1.5 l/min to 8 l/min. The gas yield
decreases by the same extent, whereas the char yield decreases by less than 2 %. The
significant change in the gas yield and the tar yield can be attributed to the intensive
vapor-phase cracking in the chamber because the relative mass change for gas
corresponds to the relative mass change for tar.
Figure 6.1 Product yields of char, gas and tar: (a) Printing paper and (b) Cardboard.
Tar, syngas and residue yields from PE/PET/PVC pyrolysis at the temperature of 500 °C
and 600 °C are shown in Figure 6.2. The yield of gas increased with an increase in
pyrolysis temperature, whereas the yields of wax and tar decreased with an increase in
pyrolysis temperature. PE pyrolysis at 500 °C yielded 42.6 % wax and 8.0 % syngas,
whereas PE pyrolysis at 600 °C yielded 15.0 % wax and 42.8 % syngas. The tar yield
decreased from 49.4 % at 500 °C to 42.3 % at 600 °C. PET pyrolysis at 500 °C yielded
18.4 % char and 44.8 % syngas, whereas PET pyrolysis at 600 °C yielded 9.4 % char and
61.7 % syngas. The tar yield decreased from 36.8 % at 500 °C to 28.9 % at 600 °C.
Tar, syngas and char yields from RDF pyrolysis at the temperatures of 400 °C, 500 °C
and 600 °C are shown in Figure 6.2. The optimum temperature for the high yield of liquid
products is approximately 500 °C. Due to the variation in temperature increase from
400 °C to 600 °C, the yield of solid decreased, and the gas yield significantly increased.
400 450 500 550 6000
10
20
30
40
50
60
70
(a)
Yie
ld (
wt.
init
ial s
am
ple
%)
Temperature (°C)
Char, 1.5l/min
Gas, 1.5l/min
Tar+H2O (caculated), 1.5l/min
400 500 6000
10
20
30
40
50
60
70
(b)
Yie
ld (
wt.
init
ial
sa
mp
le %
)
Tempearture (°C)
Char, 1.5l/min
Gas, 1.5l/min
Tar+H2O (caculated), 1.5l/min
Char, 8l/min
Gas, 8l/min
Tar+H2O (caculated), 8l/min
66
Figure 6.2 Product yields of the char, gas and tar (Helium flow rate: 1.5 l/min).
6.1.2 Syngas characterization
6.1.2.1 Paper board
Syngas composition varies with a change in pyrolysis temperature. As shown in Figure 6.3
(a), the molar concentrations of the H2, CH4, and C2-C4 light hydrocarbons in syngas
increase with increasing temperature, whereas the molar concentrations of CO and CO2
in syngas decrease with increasing temperature. Similar trends for syngas from cardboard
pyrolysis are shown in Figure 6.3 (b). The decrease in the concentrations of CO and CO2
is attributed to the significant increase of the yields of H2 and the light hydrocarbon
species, which were derived from the secondary cracking of volatile vapors. CO2 occupies
the largest fraction in syngas from printing paper pyrolysis. The molar concentrations of
CO and CO2 at 400 °C is 44.0 % and 50.1 %, respectively, whereas pyrolysis at 600 °C
yields 33.7 % CO and 35.9 % CO2. These results suggest that the secondary pyrolysis of
volatile vapors intensifies with an increase in temperature and yields more CO than CO2.
The CO2 molar concentration increases with an increase in the flow rate of carrier gas,
whereas H2 and CH4 decreased with an increase in the flow rate of carrier gas. These
results suggest that a low flow rate favors CO, H2, and light hydrocarbons.
0
10
20
30
40
50
60
70
80
90
100
PE
Temperature (°C)
600500
Yie
ld (
wt.
init
ial sam
ple
%)
Liquid
Gas
Wax
0
10
20
30
40
50
60
70
80
90
100
PET
Temperature (°C)
600500
Yie
ld (
wt.
init
ial sam
ple
%)
Liquid
Gas
Char
0
10
20
30
40
50
60
70
80
90
100
PVC
600500
Temperature (°C)
Yie
ld (
wt.
init
ial sam
ple
%)
Others (Liquid +HCl)
Gas without HCl
Char
400 450 500 550 6000
10
20
30
40
50
60
70
80
90
100
Yie
ld (
wt.
init
ial sam
ple
%)
Temperature (°C)
RDF
Char
Gas
Liquid
67
The CO/CO2 ratio exhibits a complicated relationship with evolution time and
temperature, as shown in Figure 6.3 (c). The CO/CO2 ratio initially increases with
evolution time. After attaining its highest point, the ratio begins to slowly decrease. Three
stages are observed. Due to the low heating rate of the particles, the detected product is
primarily derived from the primary pyrolysis of printing paper and cardboard during the
first tens of seconds. As shown in Scheme 6-1, CO2 and CO are the main byproducts
from the generation of levoglucosan and formaldehyde, respectively, during primary
pyrolysis [149]. CO2 is released from the cracking and reforming of functional groups of
carboxyl (C=O) and COOH, whereas the release of CO is primarily caused by carbonyl
(C-O-C) and carboxyl (C=O) groups [150]. This finding indicates that levoglucosan
formation dominates during the process at low temperatures. At high temperatures,
CO/CO2 increases, especially for printing paper, which is consistent with the results that
indicate that formaldehyde formation requires a higher activation energy compared with
levoglucosan. With an increase in time, a greater concentration of volatiles is produced to
participate in the secondary decomposition. The CO/CO2 ratio significantly increases,
which suggests that a large number of oxygenated CO products are produced via the
decarbonylation reactions. After 600 s, the CO/CO2 ratio exhibits a decreasing trend with
time, which indicates that char carbonization favors the CO2 yield.
Figure 6.3 Syngas distribution: (a) Printing paper pyrolysis; (b) Cardboard pyrolysis; (c)
and CO/CO2 ratio (Helium flow rate: 1.5 l/min).
400 500 6000
10
20
30
40
50
60
70
Mo
le c
on
ce
ntr
ati
on
(%
)
Temperature (°C)
H2, 1.5l/min
CH4, 1.5l/min
CO, 1.5l/min
CO2, 1.5l/min
C2-C4, 1.5l/min
(a)
400 500 6000
10
20
30
40
50
(b)
Mo
le c
on
ce
ntr
ati
on
(%
)
Temperature (°C)
H2,1.5l/min
CH4,1.5l/min
CO,1.5l/min
CO2,1.5l/min
C2-C4,1.5l/min
H2,8l/min
CH4,8l/min
CO,8l/min
CO2,8l/min
C2-C4,8l/min
0 200 400 600 800 1000 1200 1400 1600 18000
1
2
3
CO
/CO
2 r
ati
o
Time (S)
600°C, printing paper
500°C, printing paper
400°C, printing paper
600°C, cardboard
500°C, cardboard
400°C, cardboard
CO/CO2<1
(C)
68
Cellulose Depolymerizing Cellulose
Tar/Levoglucosan+Char+CO2
Hydroxyacetaldehyde+Formaldehyde+CO
Scheme 6-1
CO and CO2 yields by mass are shown in Figure 6.4 (a). With an increase in temperature,
both CO yields from printing paper and cardboard pyrolysis increase, whereas CO2 yields
change slightly between 500 °C and 600 °C. This finding indicates that the secondary
pyrolysis of volatiles does not contribute to the CO2 formation in this temperature zone.
Figure 6.4 (b) and Figure 6.4 (c) show the yields by mass of C2-C4 hydrocarbons from
printing paper and cardboard pyrolysis at the temperatures of 400 °C, 500 °C, and 600 °C.
Ethylene, ethane, and propylene are the three main species detected in light hydrocarbon
gases. The yields of ethane and propane by mass obtained from cardboard pyrolysis at
different temperatures and flow rates of carrier gas are shown in Figure 6.4 (c). Higher
temperatures produce a higher yield of hydrocarbons. At temperatures between 500 °C
and 600 °C, a higher flow rate of carrier gas yields a greater concentration of alkane gases.
Alkane pyrolysis can yield olefins via a series of reactions with H radicals [151]:
H•+ C3H8→H2+2-C3H7
• Reaction 6-1
H•+ C3H8→H2+1-C3H7
• Reaction 6-2
2- C3H7•→ H
•+C3H6 Reaction 6-3
1-C3H7•→ CH3
•+C2H4 Reaction 6-4
69
Figure 6.4 Syngas yields by mass: (a) CO and CO2 (flow rate: 1.5 l/min); (b) C2-C4
hydrocarbons (cardboard, Helium flow rate: 1.5 l/min); and (c) Ethane and propane
(cardboard).
6.1.2.2 Plastic
As shown in Figure 6.5, H2, CH4, and C2H4 are the main products in the gas from PE
pyrolysis. In the gaseous product, alkenes exhibit a significantly higher yield than alkanes
with the same C number. Jude A. Onwudili et al. [83] noted that alkanes have a
significantly higher yield in the gaseous products from the tests with experimental
pressures between 0.8 and 4.3 MPa. This comparison suggests that the test with a
standard pressure favors the formation of alkenes because the saturation of olefins with
H2 is not predominant. Two peaks are observed in Figure 6.5: the first peak can be
attributed to the primary degradation of PE, and the second peak can be attributed to the
secondary cracking of wax produced from PE degradation. This result is consistent with
the TGA-DSC test in the previous chapter.
Due to the high oxygen concentration, the compositions of gas from PET degradation
primarily consist of CO and CO2. A small fraction of H2 is also detected. For PVC, H2 is
400 500 6000
10
20
30
40
CO
2, C
O y
ield
(w
t.in
itia
l s
am
ple
%)
Temperature (°C)
CO, printing paper
CO2, printing paper
CO, cardboard
CO2, cardboard
(a)
0.00
0.25
0.50
0.75
1.00
(b)
cis-2-C4H
8iC
4H
81-C
4H
8tr-2-C
4H
8C
3H
6C
2H
2
(wt.
init
ial sam
ple
%)
600°C
500°C
400°C
C2H
4
Ehane,1.5l/min Propane,1.5l/min Ethane,8l/min Propane,8l/min
0.0
0.5
1.0
1.5
2.0
2.5
3.0
(c)
(wt.
init
ial
sa
mp
le %
)
600°C
500°C
400°C
70
the main product in gaseous products. Because a considerable amount of H is released in
the form of HCl during PVC pyrolysis, the formation of alkanes and alkenes is highly
suppressed.
Figure 6.5 Molar concentration of syngas yields at 600 °C (Helium flow rate: 1.5 l/min):
(a), PE; (b), PET; and (c), PVC.
6.1.2.3 RDF
The compositions of syngas are shown in Figure 6.6. At low temperatures, the major
collected gases, including CO2 and CO, are primarily derived from the pyrolysis of
lignocellulosic materials, as shown in Figure 6.6 (b). At high temperatures, the yields of
gases, including H2, CH4, C2H6, C2H4, C3H8, and C3H6, significantly increased. H2, CH4,
C2H4, are the main products. C2H4, which exists in the form of a monomer of the PE
structure, is predominantly derived from PE degradation.
0 250 500 750 1000 1250 1500 1750 20000
1
2
3
4
5
6
7
8
Mo
le C
on
ce
ntr
ati
on
(%
)
Time (s)
H2
CH4
CO
CO2
C2H4
C2H6
C2H2
C3H8
C3H6
(a)
0 250 500 750 1000 1250 1500 1750 20000
1
2
3
4
5
6
7
8
9
10
(b)
Mo
le C
on
ce
ntr
ati
on
(%
)
Time (s)
H2
CH4
CO
CO2
C2H4
C2H6
C2H2
C3H8
C3H6
0 250 500 750 1000 1250 15000
1
2
3
4
5
6
(c)
Mo
le C
on
ce
ntr
ati
on
(%
)
Time (s)
H2
CH4
CO
CO2
C2H4
C2H6
C2H2
C3H8
C3H6
71
Figure 6.6 Molar concentration of syngas yields (Helium flow rate: 1.5 l/min): (a) 600 °C,
RDF and (b) 500 °C, RDF.
6.1.2.4 Carbon yield in gas versus (𝑯/𝑪)𝒆𝒇𝒇
Carbon yield is calculated as follows:
𝐶𝑎𝑟𝑏𝑜𝑛𝑦𝑖𝑒𝑙𝑑 =𝑀𝑜𝑙𝑒𝑠𝑜𝑓𝑐𝑎𝑟𝑏𝑜𝑛𝑖𝑛𝑎𝑝𝑟𝑜𝑑𝑢𝑐𝑡
𝑀𝑜𝑙𝑒𝑠𝑜𝑓𝑐𝑎𝑟𝑏𝑜𝑛𝑓𝑒𝑑𝑖𝑛𝘟100% (6-1)
Figure 6.7 shows the carbon yield in CO, CO2, and lower olefin (C2H4 & C3H6) as a
function of the effective hydrogen ratios from the main components in MSW. With an
increase in the effective hydrogen ratio, the yields of CO and CO2 rapidly decrease,
whereas the carbon yield in lower olefins increases. Thus, gaseous products with the
desired qualities can be obtained via adjustment to the effective hydrogen ratio of
municipal solid waste.
Figure 6.7 Carbon yield in gas versus (𝐻/𝐶)𝑒𝑓𝑓
0 250 500 750 1000 1250 15000
1
2
3
4
5
6
(a)M
ole
Co
nce
ntr
ati
on
(%
)
Time (s)
H2
CH4
CO
CO2
C2H4
C2H6
C2H2
C3H8
C3H6
0 250 500 750 1000 1250 15000
1
2
3
4
(b)
Mo
le C
on
ce
ntr
ati
on
(%
)
Time (s)
H2
CH4
CO
CO2
C2H4
C2H6
C2H2
C3H8
C3H6
0.0 0.5 1.0 1.5 2.0 2.50
5
10
15
20
25 CO
CO2
Ca
rbo
n y
ield
(%
)
H/Ceff ratio (mol/mol)
0.0 0.5 1.0 1.5 2.0 2.50
5
10
15
20
25
30 C
2H
4+C
3H
6
Ca
rbo
n y
ield
(%
)
H/Ceff ratio (mol/mol)
72
6.1.3 Liquid product characterizations
6.1.3.1 Paper board
The main primary products of volatiles from the fast pyrolysis of cellulosic materials
consist of three types of compounds: pyrans, such as levoglucosan (LG) and
levoglucosenone; furans, such as 5-hydroxymethyl furfural and furfural; and linear small
molecular chemicals, such as 1-hydroxy-2-propanone, acetaldehyde, acetic acid, and
propanoic acid [99]. Levoglucosan (LG) and hydroxyacetaldehyde (HAA) were
considered to be the two main products from two competing mechanisms, as shown in
Scheme 6-2. The production distribution from levoglucosan pyrolysis is similar to the
production distribution of cellulose pyrolysis [152] and [153].
Cellulose
Levoglucosan
Hydroxyacetaldehyde
Hydroxyacetaldehyde
Scheme 6-2
The results of the products from the fast pyrolysis using Py-GC/MS are shown in Table
6.1. The main products consist of HAA, hydroxyacetone (HA), acetic acid, and propanoic
acid, 2-oxo-, methyl ester. No LG was detected from this test due to the high ash
concentration in cardboard. In the absence of minerals, levoglucosan is the main product
from cellulose pyrolysis, whereas the presence of minerals favors the formation of low-
molecular-weight compounds, especially hydroxyacetaldehyde (HAA) [154] and [155].
However, HAA is a potential product of levoglucosan pyrolysis. Thus, two mechanisms
can act on the minerals: one is to accelerate the formation of HAA from the primary
competing reactions, and the second mechanism is to act as a catalyst for the secondary
degradation of levoglucosan. Piskorz et al. [155] observed the formation of HAA was
increased at the expense of levoglucosan yield, when metal salts were added. Pushkaraj et
al. [102] observed that the yield of HAA and formic acid reached the maximum with
inorganic salt concentrations of 0.005 mmoles/g. However, with the further increase of
NaCl loading, the formation of HAA and formic acid were reduced and accompanied
with an exponential decline in the levoglucosan yield. From the proximate analysis of
cardboard in Table 3.1, there is a content of 10 wt. % of alkaline earth metals, which are
associated with the organic macromolecules. Yang et al. [156] proposed that the ‘ionic
forces’ associated with the metal ions and pyranose ring interaction exists and leads to
hemolytic scission of various bonds. Based on the test on Py-GC-MS, possible pathways
for hydrocarbons derived from paper-board primary pyrolysis are shown in Figure 6.8.
The formation of formic acid and HAA is from the primary reactions catalyzed by
73
inorganic salts/ash. The detection of propanoic acid, 2-oxo-, methyl ester could be a
product from secondary pyrolysis of LG, which was produced from heterocyclic cleavage
of glycosidic linkages and further pyrolyzed and converted to propanoic acid, et.al. But
based on the result of this study, LG is not the main primary products during fast
pyrolysis. All these suggest that the direct conversion to HAA is the main pathway for the
cellulosic materials during fast pyrolysis of cardboard at temperatures above 500 °C. This
result is also consistent with the analysis of the yields of CO and CO2 in gases, which
shows levoglucosan formation dominate during the process at low temperature
formaldehyde formation needs higher activation energy compared with levoglucosan.
These primary hydrocarbon products serve either as a product from cellulose primary
pyrolysis or as an intermediate for the formation of other products during secondary
pyrolysis. According to the discussion about syngas characterization, the gas is the main
product during the conventional pyrolysis of paper board compared with fast pyrolysis.
The main pathways are shown in Figure 6.8. Both HAA and HA are unstable and apt to
crack to form light gases, including CO, CO2, and CH4. Formaldehyde can be a source of
H2 production. Tar recovered from cardboard conventional pyrolysis in a fixed-bed
reactor at a flow rate of 1.5 l/min was analyzed via GC/MS. The tar composition
primarily consists of phenolics, benzenes, naphthalenes, benzofurans and cyclopentenes,
as shown in Figure 6.9 (a). Aliphatic compounds occupy a small fraction, which suggests
that an intensive secondary cracking occurred in the reactor due to a long residence time.
All generated aromatic compounds with 1-2 rings, which are also referred to as polycyclic
aromatic hydrocarbons (PAHs), can be the precursors of soot and atmospheric aerosols
[157].
Table 6.1 Organic compounds determined by Py-GC-MS
RT Compounds Area (%), 500 °C Area%, 600 °C
4.221 Carbon dioxide 22.9 14.1
4.291 Nitrous oxide
7.43
4.545 Acetaldehyde
3
5.24 Acetone
0.52
5.367 Pentane/Propanal, 2-methyl 4.97 4.97
6.444 3,4-Methylenedioxy-N-ethylamphetamine
0.45
6.687 2-Butanone, 3-methyl-
2.3
7.127 Butanoic acid, 2-oxo- 1.99 0.63
7.764 Acetaldehyde, hydroxy- (HAA) 23.77 16.44
8.818 Acetic acid 4.94 3.79
10.532 2-Propanone, 1-hydroxy- (HA) 10.87 8.01
13.728 Furan, 2-methyl-
0.48
16.171 Acetic acid, methyl ester
1.63
16.854 2-Butenal, 2-methyl- 2.58 2.32
18.892 Propanoic acid, 2-oxo-, methyl ester 5.31 5.26
74
19.471 Styrene
2.44
19.783 2-Cyclopenten-1-one
0.64
19.969 Furfural 2.31 2.65
22.771 2-Furanmethanol 1.47 0.94
27.599 1,2-Cyclopentanedione 2.86 3.43
29.846 Benzoic acid 1.18 0.84
31.362 2(5H)-Furanone 1.68 1.51
33.308 Oxazolidine, 2,2-diethyl-3-methyl- 2.15 1.76
34.686 1,2-Cyclopentanedione, 3-methyl- 2.02 2.59
38.46 Phenol, 2-methoxy- 1.19 1.54
45.72 Creosol
2.21
51.371 Phenol, 4-ethyl-2-methoxy-
1.14
55.111 2-Methoxy-4-vinylphenol
1.67
56.709 3-Allyl-6-methoxyphenol
0.44
63.552 trans-Isoeugenol
1.18
76.312 D-Allose
1.41
Scission at C2-
C3 bond
position
O CH2OHO CHO
O CHOHOH2C
CH
CH
CHO
COH
COH
CH2OH
CHOH
CH
CHO
CHOH
COH
CH2OH
COCH2O
O
O
O
OH
or(13) (14)
(15)
(16)
O
O
OH
OHHO
O
O
OH
OH
CH3
O
O
H3C
O
CH3
O
OH
OH
OHHO
OH
C
CH3
CH2OH
OC O
CHOH
CHOH
CH2OH
CHO
CHOH
CH3
CHO
OO
(22) (23)
(19)
(1)
(17) (18) (20)
(21)
CHO
CH2OH
C
CH2
CHO
CH2OH
O
+CO CH3OH+C
CH2
OCOOH
CH3
C
CH3
CH2OH
O
CHO
CH3+ CH2O
(8)
(12)(11)
(10)
(9)
CHO
CH2OH
CHO
CHOH
CHO
CHOH +
AcetolO
O
OH
OH
OH
1
6
54
32O
O
OH
OHOH
O
O
OH
OHOH
O
Scission at C4-
C5 bond
position
Scission at C1-O
bond position
Hemolytic scission
(Presence of inorganic salts)
Heterocyclic cleavage
(2)
(5)
(6)(7)
(4) Scission at
C1-C2 bond
position
(3)Scission at C3-
C4 bond
position
CO2, formic acid
Figure 6.8 Pathways for hydrocarbons derived from paper-board primary pyrolysis on Py-
GC/MS.
75
Relatively few studies have investigated the formation of PAHs from cellulose pyrolysis at
temperatures less than 650 °C. The O-containing species generated from the breakdown
of cellulose/hemicellulose are highly reactive and may dissociate to form radicals at
relatively low temperatures, e.g., 350 °C or less [158]. As shown in Figure Scheme 6-3,
ethane can be formed in the presence of O radicals [159]. The yields of several non-
aromatic compounds detected at different temperatures are shown in Figure 6.9 (b). 2-
Hydroxycyclopent-2-en-1-one and 5-methyl furfural exhibit an increase when the
temperature increases from 400 °C to 500 °C and exhibit a subsequent decrease for
additional increases in temperature. A similar trend for 2-cyclopenten-1-one, 2-hydroxy-3-
methyl is also observed. The formation of cyclopentenes is attributed to the formation of
cyclization organics, which are produced by the activation and combination of reactive
radicals with other structures. A decrease in 2-cyclopenten-1-one and 2-hydroxy-3-methyl
at 500 °C and 600 °C can be attributed to the dehydroxylation process, which is proven
by the increase in the 2-cyclopenten-1-one, 3-methyl yield. Compared with the lighter
olefins, such as ethylene and propylene in syngas, 2-hexene shows a different relationship
with the pyrolysis temperature. At temperatures of 400 °C and 500 °C, the 2-hexene yield
remains stable but significantly decreases at 600 °C. This finding suggests that the
combination reactions dominate in the temperature range of 400 °C to 500 °C, whereas
the decomposition reactions dominate at higher temperatures.
CHO
CH3
CO·
CH3
CO
CH3
CO
CH3
CO+C2H6
Scheme 6-3
Phenolics and benzenes are the two main aromatic groups. The mechanism to yield
phenols and benzenes at low temperatures is complicated. The secondary pyrolysis of
HMF at 600 °C favors the decomposition of HMF into FF, which exhibits superior
stability and produces a small amount of benzene at 600 °C, some acetic acid and 5-
methyl furfural [99]. However, another pathway for aryl compounds, such as benzene and
phenol, can be formed from the rearrangement reaction, which has also been proposed
for the secondary reaction of 5-HMF [160].
According to the composition analysis of the gases, the concentration of light olefins and
alkynes increases with temperature. 2-Hexene and 1, 2-dimethyl-cyclohexene have been
detected in tar at 600 °C. The mechanisms of benzene, toluene, and xylene formation by
the Diels-Alder reaction with the light olefins produced by the thermal degradation of the
alkanes are shown in Scheme 6-4 [161].
76
Figure 6.9 Tar yield from cardboard pyrolysis at a flow rate of 1.5 l/min.
CH3H3C C3H6C2H4
Diels-Alder reactions
CH3
CH3
CH3
CH3
CH3
CH3
+ O·
·O
+ H·
Scheme 6-4
400 500 6000
200
400
600
800
(µg
/g)
Temperature (°C)
Phenolics
Benzenes
Naphthalenes
Indenes
Benofurans
Pyridines
Cyclopentens
(a)
400 500 6000
50
100
150
200
(µg
/g)
Temperature (°C)
2-hexene
2-Hydroxycyclopent-2-en-1-one
2-Cyclopenten-1-one, 3-methyl-
2-Cyclopenten-1-one, 2-hydroxy-3-methyl-
5-methyl furfural
(b)
400 500 6000
50
100
150
200
250
300
350
400
(µg
/g)
Temperature (°C)
Phenol
Phenol,2-methyl&Phenol,3-methyl
Pheno, 2-methoxy-
Phenol,2,3-dimethyl
Phenol,3-(1-methylethyl)-
Phenol,5-methyl-2-(1-methylethyl)
(c)
400 500 6000
2
4
6
8
10
12
14
16
(d)
(µg
/g)
Temperature (°C)
Biphenyl
Phenanthrene
77
Benzene can also be formed from acetylene decomposition at approximately 900 K [162]:
2C2H2 → C4H4→ C4H3•+H Reaction 6-5
C4H3•+C2H2 → C6H5
•+H→ C6H6 Reaction 6-6
The reaction of benzene with O atoms to produce phenolic and H is an important
pathway [163]:
C6H6+O ↔ C6H5O+H Reaction 6-7
Via the reaction with acetylene or other radicals, benzenes can act as a precursor to form
heavier PAHs, such as naphthalenes. As shown in Figure 6.9 (d), heavy compounds with
two rings, such as naphthalenes, have a quite low yield. An increase in temperature causes
a significant increase in the yield of polycyclic aromatic hydrocarbons (PAHs). Thomas E.
[164] proposed that aromatics directly form and evolve from the carbonization process of
the solid char residue at temperatures between 400 °C and 600 °C. The yield of
phenanthrene from cellulose char solid at 600 °C is approximately 7 µg/g at a residence
time of 18 s. The yield is significantly lower than the results shown in Figure 6.9 (d),
which indicates that phenanthrene can also be produced from the homogeneous
secondary pyrolysis of gases at 600 °C. When the pyrolysis temperature is less than
600 °C, the total yield of phenanthrene from the homogeneous secondary pyrolysis and
carbonization of solid char significantly decreases.
6.1.3.2 PE
Four different mechanisms are proposed for PE degradation: end-chain scission, chain-
stripping of the side chain, random-chain scission, and cross-linking [165]. The main
liquid products from conventional pyrolysis are shown in Table 6.2. The main products
from the fast pyrolysis of polyethylene are long-chain alkanes and long-chain alkenes. 1-
Olefin is the dominant type of olefin. The product distribution shows that the random
degradation route of the polymers is the main mechanism for this process. Free radicals
serve an important role in the polyethylene thermal pyrolysis of polyethylene. They are
randomly formed at high temperatures and undergo either a process of chain
rearrangement or fission of the C-C bond. The alkenes with the compounds C11-C18
obtained in this test are also referred to as alpha-olefin compounds and are highly desired
in the petrochemical industry for plastic and detergent manufacture.
78
No significant levels of aromatic compounds were detected, as shown in Table 6.2. Three
schemes are proposed in the literature to describe the formation mechanism of aromatic
compounds. Westerhout et al. [166] proposed that the aromatics can be produced from
wax degradation of the cyclization reactions of low alkanes and alkenes in the gas-phase
for a fluidized-bed reactor. Elordi et al. [167] noted that the aromatics are produced from
wax degradation in a conical sprouted-bed reactor. In addition to these two mechanisms,
Jude A. Onwudili et al. [83] reported that aromatics can be produced from liquid pyrolysis
in a batch-type pressurized reactor. Two possible mechanisms for the formation of
aromatics in this study are shown in Figure 6.10. As shown in Table 6.2, alkyl-benzene,
alkyl-naphthalene, and alkyl-indene are the main aromatics detected in Table 6.2. Scheme
1 dominates aromatics formation during PE conventional pyrolysis. With an increase in
temperature and pressure, alkyaromatics will form benzene and toluene, indanes, and
naphthalenes via demethylation and cyclization. This conclusion is also consistent with
the tests performed at pressures between 1.6 MPa and 4.31 MPa by Jude A. Onwudili [83],
which showed that a concentration of 3.1 wt.% of toluene at 450 °C increased to 24.3 %
at 500 °C at the cost of lower yields of alkanes and alkenes.
Olefins, alkenes AlkylaromaticCyclization Cyclization, dehydrogenation
Demethylation
Indanes
NaphthalenesCyclization, dehydrogenation
Benzene,TolueneLower alkanes,
lower alkenes Diels-Alder Benzene, or toluene Naphthalene
PE
Wax
Scheme 1 for
aromatics formation
Scheme 2 for
aromatics formation
Figure 6.10 Possible mechanisms for aromatics formation in PE conventional pyrolysis
Table 6.2 Chromatographic peak identification of the main products from PE
conventional pyrolysis at 600℃
Suggested compounds Area.%
Decane 2.62 1-Undecene 2.62 Undecene 3.85 5-Isopropylidene-3,3-dimethyl-dihydrofuran-2-one 0.13 Benzene,1,2,4,5-tetramethyl- 0.53
1-Dodecene 3.34 Dodecene 3.64 1H-Indene,1-ethyl-2,3-dihydro-1-methyl- 0.42 1H-Indene,1-methylene- 0.85 8-Dodecen-1-ol, acetate,(Z)- 0.30 1-Tridecene 3.61
79
Tridecene 2.95 2-Undecene,2,5-dimethyl- 0.22 Naphthalene,2-methyl- 0.79 Naphthalene,1-methyl- 0.71 Heptylcyclohexane 0.26 1-Dodecene 4.18 Undecane 2.72 Naphthalene,1-ethyl- 0.25 Naphthalene.1,4-dimethyl- 0.39 trans-2-Dodecen-1-ol, trifluoroacetate 0.73 1-Pentadecene 4.10 Pentadecene 2.76 1-Undecene,7-methyl- 0.25 11-Hexadecen-1-ol,(Z)- 0.64 1-Hexadecene 3.56 Hexadecane 2.27 Unknown 0.54 1-Heptadecene 2.89 Heptadecane 3.56 1-Octadecene 3.43 Octadecane 2.66 1-Nonadecene 3.20 Nonadecane 2.77 1-Eicosene 2.43 Eicosane 2.45 1-Heneicosene 2.06 Heneicosane 2.35 1-Docosene 1.37 Docosane 2.69 Unknown 0.14 1-Tricosene 1,50 Tricosane 1.88 Tetracosane 1.82 Pentacosane 1.66 Hexacosane 1.72 Heptacosane 1.95 Octacosane 1.41 Nonacosane 1.98 Triacontane 1.81 Hentriacontane 2.02 Dotriacontane 1.04
6.1.3.3 PVC
80
The main products from the fast pyrolysis of PVC by Py-GC-MS are shown in Table 6.3.
A considerable fraction of HCl was detected because dehydrochlorination is the first step
of PVC thermal decomposition. In addition to HCl, benzene and naphthalene are the two
main components in the products. The yields of toluene and 1-methyl or 2-methyl
naphthalene are substantially lower than those of benzene and naphthalene. Thus, the
main mechanisms for forming benzene and naphthalene are proposed in Figure 6.11. The
cyclisation of a triene radical was considered as the main mechanisms for benzene
formation. As the low yield of alkyl aromatics, these triene radicals should be situated at
the end of chains, rather than the inner part of the macromolecules.
In the test on a fixed bed, the identification of the main products from PVC conventional
pyrolysis at 600 °C in Table 6.4 reveals that o-xylene, methylnaphthalene, and 1-
methylene-1H-indene are the main products. Naphthalene was not detected. The analysis
of the gas composition reveals a low yield of alkanes and alkenes in gases. This result
implies that the demethylation is not intense during conventional pyrolysis and tar
cracking in the conditions for this case. Thus, these three products primarily originated
from PVC primary decomposition. Some dominant mechanisms during conventional
pyrolysis are proposed in Figure 6.11. A crosslinked network containing cyclohexene and
cyclohexadine rings was first formed via intramolecular Diel-Alder condensation. And
then the alkyl aromatics can be formed followed by scission of aliphatic bonds and
dehydration of thecyclohexene and cyclohexadine rings [168 ]. Thus, a slow heating rate
can enhance the degree of crosslinked structure compared to fast pyrolysis on Py-GC-MS.
Table 6.3 Chromatographic peak identification of the main products in PVC pyrolysis by
Py-GC-MS at 500 °C
RT (min) Area (%) Components
4.348 30.12 Hydrogen chloride
5.159 0.85 1,3-Cyclopentadiene
7.185 0.34 4-Methylenecyclopentene
7.417 27.96 Benzene
11.238 2.03 Toluene
16.425 0.4 Ethylbenzene
19.471 0.75 Styrene
22.643 0.41 Benzene, cyclopropyl-
29.382 0.32 Indane
29.521 0.5 Benzene, 1-ethenyl-2-methyl- -
31.362 1 Indene
39.062 0.28 Naphthalene, 1,2,3,4-tetrahyd
40.139 2.15 1,4-Dihydronaphthalene
43.115 7.13 Naphthalene
50.711 0.58 Naphthalene, 2-methyl-
81
51.788 0.6 Naphthalene, 1-methyl-
56.037 0.27 1,1'-Biphenyl, 2-methyl-
56.153 1.73 Biphenyl
69.515 0.44 Fluorene
69.677 0.79 Fluorene
73.684 0.44 Benzene, 1-methyl-3-(2-phenyl
74.992 0.82 Stilbene
75.351 0.51 9H-Fluorene, 2-methyl-
75.478 0.96 4a,9a-Methano-9H-fluorene
75.861 0.63 9H-Fluorene, 2-methyl-
76.081 0.92 9H-Fluorene, 9-methyl-
76.382 2.64 (E)-Stilbene
76.799 0.77 Naphthalene, 2-(1-cyclopenten)-
77.192 0.54 4a,9a-Methano-9H-fluorene
77.354 0.55 Tricyclo[4.4.1.0(2,5)]undeca-
77.505 1.05 Phenanthrene, 1,2,3,4-tetrahy
77.968 0.62 Anthracene, 1-methyl-
78.142 1.36 cis-Stilbene
79.137 0.45 5,8,11,14-Eicosatetraynoic ac
79.612 4.35 Anthracene
79.948 1.02 Phenanthrene
80.967 1.08 Naphthalene, 1-phenyl-
81.789 0.61 5H-Dibenzo[a,d]cyclohepten-5-
82.947 0.41 Anthracene, 1-methyl-
83.086 0.39 Phenanthrene, 1-methyl-
83.572 0.41 Anthracene, 2-methyl-
83.734 0.29 Phenanthrene, 2-methyl-
84.776 0.29 1,1,4a-Trimethyl-5,6-dimethyl
Table 6.4 Chromatographic peak identification of main products from PVC conventional
pyrolysis at 600 °C
Suggested compounds Area.% o-xylene 12.40 Benzene,1-ethyl-2-methyl- 2.60
Indane 2.21
Indene 3.97
Benzaldehyde,4-[(2-methylphenyl)methoxy]- 0.33
Indan,1-methyl- 1.09
Benzene,4-ethenyl-1,2-dimethyl- 0.67
1H-Indene,1-methylene- 15.99
Cyclohexasiloxane, dodecamethyl- 0.43
Tridecane 0.51
1H-Indene,2,3-dihydro-2,2-dimethyl- 0.25
Naphthalene,2-methyl- 17.14
82
Tetradecane 0.52
Naphthalene,2-ethenyl- 2.54
Benzaldehyde,4-(1-naphthalenylmethoxy)- 1.55
Hydroxylamine,o-(2-naphthalenylmethyl)- 1.17
4,4-Bipyridine 0.92
Pentadecane 0.44
Naphthalene,2-ethenyl- 1.54
Hexadecane 0.51
Fluorene 3.95
2,2-Diphenylethanol 1.93
4a,9a-Methano-9H-fluorene 1.96
Anthracene 7.85
Unknown 0.66
1H-Indene,1-(phenylmethylene)- 0.67
Phenanthrene,2-methyl 1.74
Anthracene,2-methyl- 5.09
8,9-Dihydrocyclopenta[def]phenanthrene 0.81
Phenanthrene,2-methyl- 1.64
Naphthalene,2-phenyl- 0.88
Pyrene,4,5,9,10-tetrahydro- 1.17
Fluoranthene 1.68
Benzanthrene 0.94
11H-Benzo[b]fluorene 0.93
Chrysene 1.31
83
CH CH2CH CH2
Cl
CH CH2
Cl
CH CH2
Cl
CH CH2
Cl
CH CH2
Cl
CH CH2
Cl Cl
CH CH2CH CH2
Cl Cl
CH CH2CH CH2
Cl Cl
CH CH2
Cl
CH CH CH CH CH CHCH CH2
Cl
CH CH CHCH
CH CH2
Cl
CH CH CH CH CH CHCH CH2
Cl
CH2 CH CH2CH
CH2 CH CH2
ClCl
CH2
CH CH2CH CH2
Cl Cl
CH CH2CH CH2
Cl Cl
CH CH2CH CH2
Cl Cl
+
+
PVC
CHCH CH2
Cl Cl
+CH3
Cl Cl
CH CH2CH CH2
Cl Cl
CH CH2CH CH2
Cl Cl
CHCH CH2
Cl Cl
+
CHCl
CH3
CH3
H3C
CH CH2CH CH2
Cl Cl
HC
Cl
CH2 CHCl
CH CH2CH CH2
Cl Cl
ClCH2
CHCl
CH CH2CH CH2
Cl Cl
+
HC
Possible pathways for alkyl aromatics
Pathways for benzene and naphthalene (Dominated)(1)
(12)
(10)
(15)
(14)
(13)
(9)
(8)
(7)
(6)(5)
(4)
(2)
(3)
CH CH2
Cl
C CH CH CH2CH CH2
Cl
CH CH2
Cl
C CH CH CH2CH CH2
Cl
Cl
Cl
CH CH2
Cl
C CH CH CH2CH CH2
Cl
CH CH2
Cl
C CH CH CH2CH CH2
Cl
CH3
H3C
CH CH2
Cl
CH CH CH CH CH CH2CH CH2
Cl
CH
CH CH2
Cl
CH CH CH CH CH CH2CH CH2
Cl
CH
Cl
Cl
Alkyl naphthalenes
CH CH2
Cl
CH CH CH CH2 CH CH2CH CH2
Cl
CH
CH CH2
Cl
CH CH CH CH2 CH CH2CH CH2
Cl
CH
Cl
Cl
Cl
Cl
Reaction scheme for PVC conventional pyrolysis
Reaction scheme for PVC fast pyrolysis
Figure 6.11 Pathways for fast PVC pyrolysis and conventional PVC pyrolysis.
6.2.3.4 PET
The main products from virgin PET pyrolysis consist of vinyl benzoate, benzoic acid, and
4-(vinyloxycarbonyl) benzoic acid [ 169 ]. The maximum yield of 4-(vinyloxycarbonyl)
benzoic acid is 27.08 wt.%, whereas the maximum yield of benzoic acid is approximately
10 wt.%. The pathways (1-5) to form these products are shown in Figure 6.12.
Identification of the products from PET pyrolysis in the fixed bed at 600 ℃ indicates that
the biphenyl yield is significantly higher compared with fast pyrolysis. A possible
mechanism explaining the formation of biphenyl involves the formation of its
precursor—benzene—which is predominantly produced via the decarboxylation of
benzoic acid. This finding can also be confirmed by the high yield of CO2 in gases.
84
Another pathway for producing benzoic acid in the presence of Ca(OH)2 has also been
discussed by Toshiak [170]. The mechanism is that Ca(OH)2 can contribute the amount
of water required for the pathway (2). No Ca(OH)2 is produced in this case, whereas water
from the dehydration reaction can serve as a catalyst and promote this reaction. Low
yields of CH4, C2H4, and H2 in gases were attained. Their formation can be attributed to
the cracking of 1,4-butanediol.
O
O
O
OO
O
O
O
O
O
O
O
O
O
O
O
H2C
O
O
O
OO
O
O
O
H2C
O
O
O
O
O
O
O
O
CH2
O
O
O
O
O
O
O
O
O
O
O
O
H2C CH2
CH2
O
O
O
O
H2C
O
H3CCO
OH
O
CO2
+
+
+
+
+
+
CO2+
Pathways for primary pyrolysis of PET Pathways for secondary pryolysis
HO
O
OH
O
+
HOOH
CO + CO2 + H2
2H2O + CH4, C2H4
O
+ H2O
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(12)
(8)
(9)
(10)
(14)
(13)(15)
•
•
• •
••
•
•••
•
•
• • •
•
•
+ H2O
H2C•CO2+
H2C
(11)
Figure 6.12 Pathways for the primary pyrolysis of PET and the secondary pyrolysis of
their derived hydrocarbons.
Table 6.5 Chromatographic peak identification of the main products from the
conventional pyrolysis of PET at 600 °C
Suggested compounds Area.%
Styrene 6.45
a-Methylstyrene 0.85
Benzene,1,3,5-trimethytl- 0.29
1H-Indene,1-chloro-2,3-dihydro- 1.40
Acetophenone 5.77
Naphthalen,1,2-dihydro- 0.72
1H-Indene,1-methylene- 0.63
Naphthalene 3.33
Benzoic acid 1.64
1H-Indene,1-3-dimethyl- 0.31
Naphthalene,2-methyl-
2.90
85
1-Methylindan-2-one 0.32
Tetradecane 7.48
Biphenyl 28.35
Naphthalene,1-ethyl- 0.36
Naphthalene,1,4-dimethyl- 0.73
Diphenylmethane 2.26
Naphthalene,1,4-dimethyl-
0.68
1,1-Biphenyl,4-methyl- 1.14
1,1-Biphenyl,3-methyl- 2.35
Ethylene,1,1-diphenyl- 0.16
2,2-Dimethylbiphenyl 0.23
Naphthalene,2,3,6-trimethyl- 0.15
Fluorene 5.10
9H-Fluorene,2-methyl- 1.21
Benzophenone 1.39
9H-Fluorene,4-methyl- 0.36
1,2-Diphenylethylene 2.57
Diphenylethyne 1.25
9H-Fluoren-9-one 2.84
1H-Indene,1-phenyl- 0.19
Phenanthrene 2.70
Ethanone,1-[1,1-bipheny]-4-yl- 0.69
1H-Indene,1-(phenylmethylene)- 0.60
o-Terphenyl 1.27
Phenanthrene,3-methyl- 0.47
Phenanthrene,1-methyl- 0.48
Naphthalene,2-phenyl- 1.10
Fluoranthene 0.34
9H-Fluorene,9-phenyl- 0.38
p-Terphenyl 2.35
4-Benzylbiphenyl 0.11
p-Terphenyl 3.02
1H-ben[f]indene,2-phenyl- 0.10
11H-Benzo[b]fluorene 0.21
7H-Benzo[c]fluorene 0.20
4-Benzylbiphenyl 0.20
9H-Fluorene,9-phenyl- 0.60
Tetracosamethyl-cyclododecasiloxane 0.14
Chrysene 0.44
9,10[1,2]-Benzenoanthracene,9,10-dihydro- 0.41
Tetracosamethyl-cyclododecasiloxane 0.16
86
1,1:2,1-Terphenyl,4-phenyl- 0.08
7,8-Dihydro-9H-cyclopenta[a]pyren-9-one 0.15
Phenanthrene,2-phenyl- 0.13
1,1:2,1-Terphenyl,4-phenyl- 0.22
6.1.3.5 RDF
The main products from the conventional pyrolysis of RDF at 600 °C and 400 °C are
shown in Table 6.6 and Table 6.7, respectively. Different organic compounds, including
aliphatic alkanes, aliphatic alkenes, benzenes, naphthalenes, aliphatic acids, indenes, and
furans, were detected. Benzenes and naphthalenes are the two main groups in aromatic
hydrocarbons. The high yield of aliphatic alkanes and aliphatic alkenes is attributed to the
pyrolysis of PE, which contributes a high fraction to MSW.
Table 6.6 Chromatographic peak identification of main products from the conventional
pyrolysis of RDF at 600 °C
Suggested compounds Area.%
Dodecanoic acid,2-phenylethyl ester 1.18
Deltacyclene 1.80
Benzene,1,3,5-trimethyl- 0.83
Indane 0.96
Benzene,1-propynyl- 3.19
Benzene,2-ethyl-1,4-dimethyl- 0.80
1-Undecene 1.50
Undecane 1.49
Benzene,1,2,3,5-tetramethyl- 0.68
Benzene,4-ethynyl-1,2-dimethyl- 0.38
1H-Indene,1-methyl- 1.66
Naphthalene,1,2-dihydro- 1.20
2-Dodecene,(Z)- 1.95
Dodecane 1.97
1H-Indene,1-methyl- 4.33
1,11-Dodecadiene 0.17
1-Tridecene 2.61
Tridecane 1.69
Naphthalene,1-methyl- 3.55
Naphthalene,2-methyl- 2.69
Benzenebutanenitrile 0.52
87
1-Tetradecene 2.17
Tetradecane 1.38
Biphenyl 2.13
Naphthalene,1-ethyl- 1.23
Naphthalene,2,6-dimethyl- 0.82
Naphthalene,1,4-dimethyl- 1.41
Naphthalene,2,6-dimethyl- 0.53
Naphthalene,2-ethenyl- 0.43
Naphthalene,1,4-dimethyl- 0.60
1,15-Pentadecanediol 0.46
1-Pentadecene 2.10
Pentadecane 1.46
1,1-Biphenyl,4-methyl- 0.63
Naphthalene,2-(1-methylethyl)- 0.55
Naphthalene,1,6,7-trimethyl-
1.15
Carbonic acid, ethyl tetradecyl ester 0.54
1-Hexadecene 1.23
Hexadecane 1.42
1H-Phenalene 1.03
Benzene.1,1-(1,3-propanediyl)bis- 0.62
E-14-Hexadecenal 1.86
Heptadecane 1.40
9H-Fluorene,4-methyl- 0.41
Cyclododecene,1-methyl- 0.48
1-Octadecene 1.97
Octadecane 1.58
Anthracene 1.30
1,19-Eicosadiene 0.50
1-Nonadecene 3.10
Nonadecane 2.15
Naphthalene,1-phenyl- 0.24
Naphthalene,1-methyl- 0.38
Naphthalene,2-methyl- 1.30
Naphthalene,4-methyl- 0.28
Naphthalene,2-methyl- 0.79
Cycloeicosane 1.42
Eicosane 1.98
Naphtalene,2-phenyl- 0.64
1-Heneicosyl formate 1.44
Heneicosane 1.16
Fluoranthene 0.18
88
1-Docosene 1.41
Docosane 1.30
2-Isopropyl-10-methylphenanthrene 0.24
1-Nonadecene 1.17
Tricosane 1.09
n-Tetracosanol-1 0.94
Tetracosane 1.22
Pentacosane 1.64
Hexacosane 1.09
Heptacosane 1.53
Tetratetracontane 1.62
Octacosane 0.91
Nonacosane 1.59
Tetratetracontane 1.52
Pentatriacontane 1.15
Table 6.7 Chromatographic peak identification of main products from the typical
pyrolysis of RDF at 400 °C
Suggested compounds Area.%400
Styrene 7.41
Benzene,1-methyl-3-(1-methylethyl)- 6.02
Undecane 1.92
1-Dodecene 2.28
Dodecane 2.48
Naphthalene 5.57
1-Tridecene 2.78
Tridecene 3.63
Naphthalene,2-methyl- 5.97
3-Tetradecene,(2)- 4.26
Tetradecane 3.97
Naphthalene,2-methyl- 2.76
1-Pentadecene 3.59
Pentadecane 4.01
1-Hexadecene 2.37
Hexadecane 3.40
1-Heptadecene 1.76
Heptadecane 2.33
Octadecane 1.86
1,2-Benzenedicarboxylic acid,bis(2-
methylpropyl)ester 0.47
89
Heneicosane 1.22
Eicosane 1.06
Heneicosane 0.93
Dodecane 1.08
Tricosane 1.11
Unknown 2.70
Pentacosane 1.42
Unknown 2.05
1,2-Benzenedicarboxylic acid,mono (2-
methylpropyl) ester 2.28
Hexacosane 1.46
Unknown 2.81
Heptacosane 1.22
Unknown 3.35
Octacosane 1.16
Unknown 3.22
Nonacosane 1.08
Unknown 3.01
6.1.4 Elemental analysis of chars from paper board pyrolysis
The elemental composition of chars is presented by a comparison of hydrogen/carbon
(H/C) and oxygen/carbon (O/C) ratios in the form of the Van Krevelan diagram in
Figure 6.13. The corresponding temperatures for the chars are given in the figure. Both
O/C and H/C continuously decrease with an increase in temperature. This result
indicates a process of loss of hydrogen and oxygen and an enrichment of carbon in the
solid char. The direction of the curve from cardboard char suggests that the reaction
primarily involves dehydration at temperatures below 400 °C. As dehydration is an
important step for the formation of oxo-groups [171], either aldehyde grups at C-6, or
keto-groups at C2 and C3, it is likely the effect of ionic forces associated with inorganic
earth metals and pyranose ring interaction is not significant enough to lead hemolytic
cleavage at low temperature. And levoglucosenone that is mainly produced via
dehydration might be the main products compared to light hydrocarbons like HAA at low
temperature. At the temperature between 400 °C and 500 °C, the decomposition, which is
dominated by decarbonylation and decarboxylation, suggests the effect of inorganic earth
metals is becoming significant at high temperature and high heating rate. While printing
paper exhibits a different process: decarbonylation and decarboxylation also play
important roles below 400 °C. The reason may be attributed to that printing paper has a
much higher ash content compared to cardboard. It’s interesting that the hydrogen loss
appears to be more rapid relative to the oxygen above 500 °C. It seems dehydrogenation,
90
demethanation occur and dominate at temperature above 500 °C. According to the
discussion in CO and CO2 yield of printing paper, CO2 yield almost doesn’t change when
temperature increases from 500 °C to 600 °C. Oxygen content in printing paper char that
obtained at 500 °C and 600 °C respectively is quite low. All these results indicate that the
CO2 is presumably produced in early stage of cellulose pyrolysis below 380 °C when ash
content is high in paper-board and ash content has a significant influence on the char
carbonization at temperatures below 400 °C. The effect of temperature, heating rate, and
ash concentration in solid samples, on the paper-board decomposition was proposed in
Figure 6.14.
Figure 6.13 Van Krevelan diagram for chars and original material.
Hemolytic scission
Heterocyclic cleavagePaper-board LG
Solid residue
with Oxo-
groups
HAA,HA, etc
Te
mp
era
ture
Ash
He
atin
g r
ate
Dehydration
At C6,C2,or C3 before hydrolysis or
pyrolytic scission begins
High
Low Low
High
Slow
Rapid
Char
Figure 6.14 Effect of temperature, heating rate, and ash concentration in paper-board, on
the pathway of paper-board decomposition
91
6.2 Summary
With increasing temperature in the temperature range between 400 °C and 600 °C, the gas
yield increases, and the char yield decreases. The temperature for maximum tar yield from
printing paper is approximately 400 °C. Compared with printing paper, cardboard
pyrolysis produces a higher tar yield and a lower char yield. The temperature for the
maximum tar yield falls in the temperature range between 400 °C and 500 °C.
Temperature has a significant effect on the yield of wax for PE pyrolysis at low
temperatures, which varied from 42.6 % at 500 °C to 15 % at 600 °C. Gas occupies a
significant fraction of the products for PET and PVC due to the amount of produced
CO2 and HCl, respectively. The optimum temperature for a high yield of liquid products
from RDF pyrolysis is approximately 500 °C.
The effect of residence time on lignocellulosic materials is significant. For paper board,
CO2 is primarily produced in the early stage of cellulose pyrolysis and char carbonization.
High temperatures and long residence times favor the formation of H2, CH4, C2-C4 and
CO. The abundance of C2-C4 is formed by the decarbonylation of oxygenated products.
Alkanes are apt to decompose and form alkenes via H-transfer, radical formation and
decomposition. H2, CH4, and C2H4 are the main products in the gas formed by PE
pyrolysis. CO and CO2 are the main gases in syngas from PET pyrolysis, and a small
fraction of H2 is also detected. For PVC, H2 is the main product in gaseous products.
Because a considerable amount of H is released in the form of HCl during PVC pyrolysis,
the formation of alkanes and alkenes is highly suppressed.
With an increase in the effective hydrogen ratio, the CO and CO2 yields rapidly decrease,
whereas the carbon yield in lower olefins increases. The effective hydrogen ratio is
adequate for evaluating the potential yield of lower olefins, CO and CO2 in syngas from
RDF degradation.
Paper-board in fast pyrolysis is mainly converted to HAA directly at temperatures above
500 °C, as the hemolytic cleavage of several bonds in the pyranose ring caused by the
alkaline earth metals which are associated with the organic macromolecules. It is also
consistent with the analysis of the yields of CO and CO2 in gases, which shows
levoglucosan formation dominate during the process at low temperature. While for
conventional lignocellulose materials pyrolysis, phenolics and benzenes formed by the
reactions among the abundance of olefins and alkynes are the two main groups in tar,
suggesting that an intensive secondary cracking of primary volatiles occurred in the
reactor due to a low flow rate of carrier gas. H radical and O radicals play key roles in the
formation of olefins, alkynes, benzenes, and phenols. Benzenes can act as precursor to
form heavier PAHs. At 600 °C, the yield of naphathanes and phenanthrene increased
significantly. With temperature above 500 °C, the process of demethoxylation and
deethylation of phenols and benzenes exhibits. The yields of several non-aromatic
92
compounds formed from cyclization reactions, are presented and exhibit a peak at 500 °C.
It suggests that the combination reactions dominate at the temperature zone of 400 °C to
500 °C, while decomposition reactions dominate at higher temperature.
A random degradation route is the main mechanism for PE degradation. The liquid
product, which primarily consists of C11-C18 alkanes and 1-olefines, is highly desired in
the petrochemical industry for plastic and detergent manufacture. No significant level of
aromatic compounds were obtained from PE conventional degradation. Alkyl-benzenes,
alkylnaphthalenes, and alkyl-indene are the main aromatics detected. Compared with the
cyclization reactions from lower olefins, cyclization reactions from long-chain olefins
comprise the dominant mechanism for aromatic formation during PE conventional
pyrolysis at temperatures below 600 °C.
PVC conventional pyrolysis exhibits different pyrolysis behaviors with fast pyrolysis.
Dehydrochlorination is the first step in PVC thermal decomposition. Benzene and
naphthalene are the main two products from the fast pyrolysis of PVC. The yields of
toluene and 1-methyl or 2-methyl naphthalene are significantly lower than those of
benzene and naphthalene. For conventional pyrolysis of PVC in a fixed bed, the yields of
o-xylene, methylnaphthalene, and 1-methylene-1H-indene are relatively high, and no
naphthalene was detected. According to the analysis of the gas composition, the yields of
low alkanes and low alkenes in gases are quite low, which implies that the demethylation
is not intense during conventional pyrolysis and tar cracking in the conditions of this case.
Thus, these alkyl aromatics are mainly produced by scission of aliphatic bonds and
dehydrogenation of the cyclohexene and cyclohexadine rings contained in a crosslinked
network, which was formed via intramolecular Diel-Alder condensation during PVC
conventional pyrolysis. A slow heating rate can enhance the degree of crosslinked
structure compared to fast pyrolysis on Py-GC-MS.
The high yield of CO2 is primarily attributed to the decarboxylation reactions of benzoic
acid, which are the main products from PET pyrolysis. The biphenyl yield is the main
product detected in liquid from the conventional pyrolysis of PE in a fixed-bed reactor.
The possible mechanism explaining the formation of biphenyl involves the formation of
its precursor—benzene—which is the other main product from benzoic acid
decarboxylation.
93
Chapter 7
7. Steam gasification of MSW with CaO
In the previous chapter, CO2 is one of the main components in the gases obtained from
RDF pyrolysis. To improve the hydrogen yield and the heating value of the gases, CaO
can be used as a sorbent in CO2 adsorption. In the presence of steam vapor, CaO reacts
with H2O to form Ca(OH)2, which subsequently reacts with CO2 to form CaCO3. These
two reactions occur at relatively low temperatures and are not affected by the partial
pressure of CO2. Due to a decrease in CO2 partial pressure caused by CO2 isolation from
syngas by CO2 sorption, the water-gas shift reaction is driven forward, giving high H2
yields [172] and [173].
The detailed reaction sequence for steam gasification of MSW and CaO blending in a
batch-type fixed bed is shown in Figure 7.1. These reactions occurred in heterogeneous
and homogeneous zones. The study examines the effect of the addition of CaO on the
heat transfer properties, the CO2 absorption, and the catalytic effect on the
devolatilization and char gasification behaviors in the presence of steam.
Figure 7.1 Reaction sequence in a small batch-type gasifier.
94
7.1 Results and Discussion
7.1.1 Heat transfer properties
The temperature profiles in the center of the basket for different conditions are shown in
Figure 7.2. The curves for MSW and CaO at 900 °C exhibit a temperature profile similar
to the curves of MSW at 900 °C in the first three minutes. However, the comparison of
both curves after 3 minutes indicates that the addition of CaO has a significant effect on
the heat transfer properties. The plastic in MSW as a non-charring material significantly
decreases during its decomposition at high temperatures. Consequently, the thermocouple
inserted into the basket was laid bare to measure the temperature of the gas agent instead
of the temperature of the solid residue. In the presence of CaO, the integrity of the
mixture of CaO and MSW can be significantly improved to support the entire bed in a
fixed-bed gasifier. The rate of temperature increase of the particle was also slowed
compared with the case without CaO. The maximum temperatures of both curves at
900 °C were approximately 780 °C, which is significantly lower than 900 °C. The reason
for this result is that the temperature measurement inside the mesh basket screened the
effect of radiation from the furnace wall. The final temperatures of the solid residue at
800 °C and 700 °C were approximately 700 °C and 600 °C, respectively.
Figure 7.2 Center temperature profiles at different temperatures of 900 °C, 800 °C, and
700 °C.
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.00
100
200
300
400
500
600
700
800
Part
icle
Tem
pera
ture
(°C
)
(min)
MSW without CaO, 900°C (case #9)
MSW&CaO, 900°C(case #10)
MSW&CaO, 800°C(case #8)
MSW&CaO, 700°C(case #6)
95
7.1.2 Evolutionary behavior of syngas composition
The evolutionary behaviors of the molar concentrations of major gases at 800 °C are
shown in Figure 7.3 (a) and (b) without the addition of CaO and with a CaO:MSW mass
ratio of 1:1. Due to a certain degree of overlap between MSW pyrolysis and char
gasification, the entire process can be divided into two stages—devolatilization and char
gasification—according to the product distribution. The devolatilization stage can be
determined by calculating the molar evolution of CH4, which is primarily produced from
MSW pyrolysis. During the char gasification stage, H2, CO, and CO2 were the main gases
in syngas. In Figure 7.3 (a), CO2 and CO gases had higher molar concentrations at the
beginning of gasification compared with H2 and CH4 because lignocellulosic components
in MSW decompose at a much lower temperature and release relatively higher amounts of
CO and CO2 compared with some plastic components. With increasing temperature, the
decomposition of plastic groups produces an increase in both H2 and CH4 molar
concentrations. CH4 had a higher yield than H2 in the first half of the pyrolysis stage,
which can be attributed to the use of a long sampling pipe. The close proximity of the
sampling pipe to the basket did not provide sufficient time for CH4 steam reforming,
which produces H2. The CH4 molar concentration decreased with the consumption of
MSW materials, whereas the H2 and CO2 molar concentrations significantly increased,
which indicates that the carbonization of the solid residue and char steam gasification
serve an important role in the syngas evolution. However, the CO molar fraction
continued to decrease until the end of the second half of the devolatilization stage. This
result suggests that the carbonization of solid residue favors the release of CO2 compared
with CO in the second half of the pyrolysis stage. The devolatilization stage ended at
approximately 600 s, at which time the H2 mole fraction began to decrease, and the CO
mole fraction slowly increased. The comparison of the results shown in Figure 7.3(a) and
(b) indicated that CaO had a significant effect on the gas mole composition. For a
CaO:MSW mass ratio of 1:1, the H2 concentrations significant increased to 10 % at the
highest point compared with MSW gasification without the addition of CaO. The effect
of CaO was also observed for CO2 evolution, which decreased by 10 % at the lowest
point in the presence of CaO, which indicates that CaO has a distinct adsorption capacity
for CO2 in the devolatilization stage. Lower CO molar concentrations were also observed
in the devolatilization and char gasification stages when CaO was added.
96
Figure 7.3 Evolutionary behavior of syngas composition at 800 °C at a steam flow rate of
19.3 g/min: (a) case #7, without CaO and (b) case #8, CaO:MSW=1:1.
7.1.3 Syngas flow rate
Figure 7.4 shows the flow rate of the main gases in syngas for different CaO:MSW mass
ratios at two steam flow rates (5.6 and 19.3 g/min). The CO2 flow rate during the
devolatilization stage increased with an increase in the steam flow rate from 5.6 g/min to
19.3 g/min in the absence of CaO, whereas the hydrogen and CH4 flow rates decreased.
This finding is attributed to the notion that the residence time of volatiles inside the
chamber, which is related to the distance between the basket and the sampling line, serves
an important role during tar cracking. During devolatilization, syngas primarily consists of
two parts, including syngas from MSW devolatilization (primary pyrolysis) and syngas
from tar homogeneous reactions. Because the sampling line was quite close to the sample,
the residence time of volatiles from pyrolysis was decreased when the steam flow rate
increased from 5.6 g/min to 19.3 g/min. Thus, syngas from tar homogeneous reactions
contributed less to the syngas amount at 19.3 g/min compared with 5.6 g/min.
As shown in Figure 7.4, MSW gasification with CaO addition yielded a greater amount of
H2 and reduced CH4, CO, and CO2 amounts compared with raw MSW steam gasification
at a steam flow rate of 5.6 g/min. At a high steam flow rate (19.3 g/min), the CaO
addition caused an opposite effect on the evolution of CH4, CO, and H2, with the
exception of CO2, which is due to CO2 adsorption. CH4 exhibited a significant decrease
in the presence of CaO, as shown in Figure 7.4 (b). CH4 is a product from the primary
pyrolysis of feedstock and tar cracking and can react with steam to produce more
hydrogen and CO. As shown in Figure 7.4 (a), the H2 flow rate in the devolatilization
stage decreased in the presence of CaO at a steam flow rate of 19.3 g/min compared with
the absence of CaO, which indicates that the severity of tar cracking to form
incondensable small molecular gases was significantly reduced in the presence of CaO.
0 2 4 6 8 10 12 14 16 18 20
0
10
20
30
40
50
60
70
80G
as
co
mp
os
itio
n (
%)
Time (min)
H2, CH4, CO
CO2, C2H6, C3H8
(a)
0 2 4 6 8 10 12 14 16 18 20
0
10
20
30
40
50
60
70
80
(b)
Gas
co
mp
os
itio
n (
%)
Time (min)
H2, CH4, CO
CO2, C2H6, C3H8
97
Figure 7.4 Flow rates (liter per minute) of main gases in syngas at 800 °C, cases #1, #2,
#3, #4, #7, and #8: (a) H2; (b) CH4; (c) CO; and (d) CO2.
The carbon conversion in syngas is shown in Figure 7.5. The calculation of carbon
conversion is based on the local syngas flow rate and only considers the main gases in
syngas, including CO2, CO, and CH4. Because a large amount of removed tar did not
completely participate in homogeneous reactions, the carbon in syngas only occupied a
small fraction. Carbon conversion at different steam flow rates in the absence of CaO
produced similar curves, as shown in Figure 7.5 (a). During the devolatilization stage,
carbon conversion at a flow rate of 5.6 g/min showed a slightly faster increase, whereas
the char gasification stage was overtaken by carbon conversion at 19.3 g/min. The result
is consistent with the discussion regarding residence times at different steam flow rates.
The addition of CaO seemed to reduce the carbon conversion (Figure 7.5 (a)), which may
be attributed to CO2 adsorption. To eliminate the effect of CO2 adsorption on the
calculation, a carbon conversion value that did not consider CO2 in syngas was also
plotted (Figure 7.5 (b)). The actual mechanism of CaO on MSW devolatilization and char
gasification is unclear due to the homogeneous cracking and reforming of volatiles. Thus,
syngas that is produced in low-temperature and high-steam-flow-rate conditions is less
affected by the homogeneous reactions of volatiles and reliably reflects the real
devolatilization behavior compared with the case of high temperatures and low steam
flow rates. Based on the material balance, the lower yields of the main gases in syngas in
98
the devolatilization stage suggest that the presence of CaO most likely improved the yield
of condensable tar or the solid yield (reduced by the amount of adsorbed CO2). At a
higher temperature or lower steam flow rate, this tar may crack to yield more syngas,
which explains why CaO showed different effects on the carbon conversion at two steam
flow rates and completely opposite effects on the carbon conversion at 900 °C and
700 °C (Figure 7.5).
Figure 7.5 Carbon conversion in syngas: (a) 800 °C, two different flow rates, cases #1, #4,
#7, and #8; (b) same steam flow rate, 700 °C, 900 °C, cases #5, #6, #9, and #10.
7.1.4 Solid analysis
Ca(OH)2, CaCO3, and carbon contained in the solid residues were quantified by TGA.
With an increase in temperature, Ca(OH)2 and CaCO3 decomposed and lost weight, as
shown in Figure 7.6. Ca(OH)2 begins to decompose at approximately 450 °C and stops
decomposing at approximately 500 °C, whereas CaCO3 begins to decompose at
approximately 600 °C and stops decomposing at approximately 800 °C. Oxygen was fed
to burn the carbon instead of Ar when a temperature of 1200 °C was attained. The mass
99
fractions of these compounds were determined from the weight loss curves for H2O,
CO2, and carbon combustion. The calcium compounds in the residues are shown in
Table 7.1. According to the previous discussion, the final particle temperature at 700 °C,
which is approximately 600 °C, is significantly higher than the decomposition temperature
of Ca(OH)2 in Figure 7.6. However, the Ca(OH)2 concentration in calcium compounds is
67.4 %. A similar phenomenon for CaCO3 is observed according to the comparison in
Figure 7.2 and Figure 7.5. These results suggest that the decomposition temperatures for
Ca(OH)2 and CaCO3 significantly vary in steam atmosphere compared with inertia gas.
The CaCO3 decomposition temperature is crucial for the operations of CO2 absorption
and the calcination reactor in a system with two interactive fluidized beds. If the operating
temperature in the reactor can be promoted via lifting the CaCO3 decomposition
temperature by adjusting the operational conditions, the steam gasification efficiency can
also be promoted.
Figure 7.6 Results of the thermogravimetric analysis of solid residue sampled from case
#10.
To interpret these phenomena, the yields of solid residue are shown in Figure 7.7, in
which the elapsed experiment times were not identical. Although the elapsed times of
cases #2–4 were longer than the elapsed time of case #1, they presented higher residue
yields compared with case #1. The reasons for this observation can be attributed to CO2
absorption by CaO and higher char yields caused by CaO during the devolatilization stage.
The fraction of unreacted carbon and ash is shown in Figure 7.7 (a), which indicates that
CO2 adsorption has a significant effect on the fraction of solid residue. With regard to the
different elapsed times, whether the presence of CaO increased the yield of solid char
remains unclear.
0 1000 2000 3000 4000 50000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Time (s)
Mas
s lo
ss
(100%
)
0
200
400
600
800
1000
1200
Carbon combustionCaCO3 decomposition
Te
mp
era
ture
(°C
)
Ca(OH)2 decomposition
100
At a high steam flow rate of 19.3 g/min, a lower char yield (unreacted carbon and ash)
was observed for cases with CaO compared with cases without CaO, as shown in Figure
7.7 (b). This observation suggests that the increase in steam flow rate significantly
increases the char steam gasification rate in the presence and absence of CaO, which is
verified from the carbon conversion profile during the char gasification stage in Figure 7.7
(a).
Figure 7.7 Residue yield: (a) steam flow rate: 5.6 g/min, cases #1, #2, #3, and #4 and (b)
steam flow rate: 19.3 g/min, cases #5,#6, #7,,#8,,#9, and #10.
Table 7.1 Calcium compounds in residues
Calcium Species (wt. %) #6 #8 #10
Ca(OH)2 49.1 61.0 67.4
CaCO3 23.7 29.3 30.2
Others 27.2 9.7 2.4
101
Three different morphologies were observed for the solid residue, as noted from their
SEM images in Figure 7.8. A large number of white particles attributed to calcium species
were scattered in the solid mixture. These particles exhibited a fragile structure; Figure
7.8(a) shows the particle morphologies of a large CaCO3 particle that was easily crushed
during sample preparation for SEM imaging. The morphologies of a mixture of
filamentous fibers and white solid irregular particles are shown in Figure 7.8(b). The
entangled filamentous fibers originated from the cellulosic fibers contained in the
recovered solid waste, whereas the white irregular particles were primarily derived from
two materials: the mineral matter contained in MSW and fine powders from CaO. The
third morphology is shown in Figure 7.8(c, d). Circles dotting the particle surfaces were
observed, which indicates that bubbles were formed during the devolatilization process
and caused the swelling phenomenon. As the plastic components in MSW exhibited
plasticity properties, the liquid phase moved and surrounded the solid particles, which
improved the dispersion of calcium on the char surface and enabled calcium to make
sufficient contact with the solid char.
To examine the dispersion and mobility of calcium in char, the EDS spectra of the
samples, which exhibited all three morphologies, were compared. Several points were
selected for examination from the SEM images in Figure 7.8. As shown in Table 7.2,
chars obtained from both filamentous fibers and molten plastics had a significant content
of calcium atoms, which indicated that CaO has superior mobility in the steam
gasification of a blended CaO/MSW material.
Figure 7.8 SEM images of the solid residue: (a) CaCO3 from case #6, 900 °C; (b) Mixture
of char and particles of calcium species from case #10, 700 °C; (c) Mixture of char and
particles of calcium species from case #6, 900 °C; and (d) Mixture of char and particles of
calcium species from case #8, 800 °C.
(c) (d)
(b)(a)
102
Table 7.2 SEM/EDS analysis of the selected points on the char surface.
Element
Case #6, [atom.%] Case #8, [atom.%] Case #10, [atom.%]
P4’ P5’ P6’ P7’ P3’ P8’
Oxygen 50.79 69.09 51.31 45.54 63.92 56.66
Sodium 1.71 0.53 0.68 1.64 0.27 0.53
Magnesium 1.1 0.22 0.07 1.46 6.45 0.22
Aluminum 2.1 - 0.08 1.72 0.64 1.08
Silicon - - 0.61 1.9 6.6 0.54
Chlorine 19.88 8.16 12.16 9.9 8.41 13.17 Calcium 22.86 20.9 32.45 31.39 13.72 27.8
Potassium 1.55 1.11 - 1.72 - -
7.1.5 Mechanistic discussion of catalytic effect
The thermal decomposition reaction of polyolefins, which has been described as a
random scission, occurs in the molten state and proceeds via a free-radical mechanism.
PE decomposition is initiated by the preferential scission of weak bonds caused by
irregularities in the polymer chain and random scission of typical C–C bonds. Because the
symmetrical C–C bonding energy is higher than the symmetrical C–C bonding for other
C–C bonds in aliphatic compounds, greater amounts of macro-sized radicals are
predominantly produced via bond cleavage near the end of the macromolecular chain.
Various pyrolysis products are produced via intramolecular radical transfer processes. The
presence of a catalyst promotes carbon chain scissioning. According to the results of
syngas yields, the flow rates of H2, CH4, and CO from MSW steam gasification with CaO
were less than the flow rates from the MSW gasification without CaO. This finding
suggests a competing polyolefin degradation reaction that forms condensable
hydrocarbons. The presence of CaO is known to accelerate this competing reaction,
which results in a lower composite thermal degradation temperature [174]. Ca species can
diffuse into the polyolefin matrix when it is surrounded by molten liquid and cause
extensive random degradation along the polymer chain and subsequent backbiting
reactions [174]. Thus, the catalyst can increase the degradation rate and yield more liquid
products and smaller amounts of char [175] and [176]. At decomposition temperatures
below 300 °C, a lower gas yield from catalytic pyrolysis of LDPE was also observed
compared with thermal pyrolysis of polyethylene. At temperatures above 350 °C, gas
yields increased significantly [175]. This result indicates that catalytic degradation causes
the extensive fracturing of long chains and produces components with medium-length
chains in the radical initiation stage, which decreases the probability of branched chain
fracturing processes that favor the formation of incondensable gases. However, the
potential yield of gases for catalytic degradation is promising in terms of its high liquid
103
yield. For a high temperature and long residence time, a higher yield of gases via catalytic
degradation can be expected compared with the thermal pyrolysis of polyolefins.
A substantial amount of literature has been published regarding the mechanism of non-
catalytic and alkali-catalyzed char gasification [177] and [178]. C(O), which is produced via
an oxygen adsorption reaction on the carbon surface, is regarded as a key intermediate for
the non-catalytic steam gasification of carbon. The gasification, which is selective toward
the formation of H2, CO, and CO2, is governed by the competition between the
desorption of C(O) and the reaction of C(O) and H2O to form CO2. Once the reactive
C(O) intermediate desorbs as CO or releases as CO2, the next layer of previously
unreactive carbon atoms becomes uncovered and available for reaction. In this case, the
catalyst can act as an “oxygen reservoir” and supply oxygen to these recently exposed sites,
which results in the formation of additional C–O complexes. Wang et al. have proposed a
redox cycle mechanism to explain the selectivity toward CO, H2, and CO2, [179] and
[180]. Two more intermediates—K2O–C and K2O2–C, which refer to a reducing form
and oxidizing form, respectively, of the K–C–O intermediate—are involved in this
mechanism. Thus, the gasification selectivity toward CO and CO2 is governed by the
competition between the desorption of C(O) and the reaction of C(O) with K2O2–C. This
mechanism indicates that the gross molar ratio of H2 to (2CO2 + CO) is 1, regardless of
the effect of the water-shift reaction.
In Figure 7.9, the molar ratio of H2 to (2CO2 + CO) significantly decreased in the char
gasification stage. The ratio at the beginning of char gasification significantly exceeded 1
at a high CaO/MSW ratio, which indicates that a large amount of H2 is produced from
the catalytic steam gasification. Kim et al. revealed that hydrocarbon (C–H) was the major
carbon component at the surface of the spent catalyst used in the co-processing of coal
and waste plastics compared with other forms of carbon, such as C=O and C–O [181].
This finding suggests that a vast amount of C–H complexes are formed on the surface of
Ca species during the devolatilization stage. As previously indicated in Figure 7.9(b), the
effect of temperature on the molar ratio of H2 to (2CO2 + CO) is significant. The
significant decrease at 900 °C is attributed to the CO2 emission at the char gasification
stage caused by CaCO3 decomposition; conversely, only a slight change in the CO/CO2
ratio is observed with increased elapsed time during char gasification, as shown in Figure
7.9 (c) and (d). The effect of CaO addition on CO/CO2 is minimal, which suggests that
the high molar ratio of H2 to (2CO2 + CO) at the beginning of char gasification is likely
due to the remaining hydrogen as a C–H complex on the char surface. As C–H
complexes are consumed and C(O) species accumulate on the char, the molar ratio of H2
to (2CO2 + CO) decreases. After the majority of the C–H is consumed, the molar ratios
of H2 to (2CO2 + CO) and CO/CO2 become relatively stable with increases in elapsed
time.
104
Figure 7.9 The molar ratio of H2 to (2CO2+CO), and CO to CO2 versus elapsed time.
7.2 Summary
The plastic in MSW as a non-charring material shrinks significantly during its
decomposition at high temperatures, whereas CaO can significantly improve the integrity
of the blending of MSW and CaO, which is important for the operation of MSW
gasification in a large fixed bed. The effect of CaO on the temperature profile in the solid
phase begins to exhibit at the beginning of MSW degradation because a stable bed can be
formed in the presence of CaO.
The CaO used in the steam gasification of a blended MSW/CaO material showed a
superior capacity toward CO2 adsorption, which resulted in hydrogen-rich syngas. The H2
molar composition increased significantly with an increase in the mass ratio of CaO:MSW,
whereas both CO and CO2 decreased. The process temperature also has a significant
effect on the syngas composition.
Ca(OH)2 and CaCO3 decomposition temperatures, which are crucial for the operations of
CO2 absorption and the calcination reactor in a system with two interactive fluidized beds,
vary significantly in a steam atmosphere compared with in an inertial gas.
The catalytic effect of CaO on the degradation of plastics exists. A satisfactory dispersion
of calcium was obtained on the char surface during MSW/CaO gasification because the
plastic components in MSW are able to move and surround calcium particles in their
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00(c)
Char gasification stageDevolatilization stage
CO
/CO
2 m
ola
r ra
tio
Time (min)
#1
#2
#3
#4
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00(d)
Devolatilization stage Char gasification stage
CO
/CO
2 m
ola
r ra
tio
Time (min)
#5
#6
#9
#10
#8
#7
105
molten state. Calcium catalysts favored the extensive fracture of long chains and
produced components with medium-length chains at the radical initiation stage, which
acts to decrease the probability of branched-chain fracturing events that favor the
formation of incondensable gases. However, the potential gas yields via catalytic
degradation are promising in terms of high liquid yields; at a sufficiently high temperature
and prolonged residence time, a higher yield of gases can be expected via catalytic
degradation compared with thermal pyrolysis of polyolefins.
After co-pyrolysis of the blended MSW/CaO, a satisfactory calcium dispersion and
mobility in char was obtained. The molar ratio of H2 to (2CO2 + CO) at the char
gasification stage increased with the addition of CaO; this finding can be reasonably
explained by the remaining C–H complex on the surfaces of calcium species. Among the
three temperatures, the highest molar ratio was obtained at 800 °C.
106
Chapter 8
8. Multi-reformation of CH4 with CO2 for the
remediation of greenhouse gas
The process of CO2 reforming of CH4, which is usually referred to as dry reforming, has a
high selectivity to CO and offers a method for fixing CO2 for CO2-emitting processes.
However, it is a highly endothermic reaction that requires extra heat [182] and [183]. In
addition, the H2/CO ratio for dry reforming is 1:1, whereas the expected H2/CO ratio of
syngas for high-level hydrocarbons via Fischer-Tropsch synthesis is approximately 2:1
[184]. Three other types of reforming technologies (bi-reforming, auto-thermal reforming,
and tri-reforming), which are based on this basic type of reforming, were examined and
compared with dry reforming of CH4 via thermodynamic analysis, as shown in Figure 8.1.
Bi-reforming, which combines dry reforming and steam reforming, can achieve a higher
H2/CO ratio than dry reforming [185]. Auto-thermal reforming, which combines dry
reforming and partial oxidation, provides partial heat for endothermic reactions and
increases the H2/CO ratio [143]. Tri-reforming, in which both O2 and steam are added,
improves the H2/CO ratio and provides partial heat for endothermic reactions [186].
Tri-reforming
Auto-reformingBi-reforming Dry-reforming
H2O CH4 CO2 O2
Figure 8.1 Four types of reforming technologies
107
8.1 Results and Discussion
According to the thermodynamic equilibrium models of the multi-reforming of CH4 with
CO2 technology, the theoretical performances of four reforming processes at high
temperatures were investigated. The molar fractions of the main species are plotted versus
the reforming temperature in Figure 8.2.
Figure 8.2 The main species concentrations of four types of reforming on temperature,
from thermodynamic equilibrium modeling; the reactants are input using a stoichiometric
ratio, 1 atm pressure, temperatures from 300 K to 2100 K: a) dry reforming; b) bi-
reforming; c) autothermal reforming; d) tri-reforming
According to the Gibbs free energy change for the reaction, steam reforming is not
favorable below 900 K, and CO2 reforming will not proceed below 925 K. However, the
conversion can proceed at temperatures ranging from 500 K to 1100 K because of side
reactions, such as WGS and the methanation reaction, which serve a key role in the
reforming process. Although the reactant inputs in the four reforming processes differed,
the conversion phenomena were similar, with the exception of the intermediate H2O. In
400 600 800 1000 1200 1400 1600 1800 20000.0
0.1
0.2
0.3
0.4
0.5
Mo
le F
racti
on
Temperature (K)
CH4
CO2
CO
H2
H2O
(a)
400 600 800 1000 1200 1400 1600 1800 20000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mole
Fra
ctio
n
Temperature (K)
CH4
CO2
CO
H2
H2O
(b)
400 600 800 1000 1200 1400 1600 1800 20000.0
0.1
0.2
0.3
0.4
0.5
0.6
Mo
le F
racti
on
Temperature (K)
CH4
CO2
CO
H2
H2O
(c)
400 600 800 1000 1200 1400 1600 1800 20000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mo
le F
rac
tio
n
Temperature (K)
CH4
CO2
CO
H2
H2O
(d)
108
dry reforming, H2O significantly formed between 600 K and 1100 K and reached a
maximum at 860 K; methanation reactions reversed above 860 K and methane steam
reforming was favored above 900 K. In auto-thermal reforming, total CH4 oxidation was
performed at low temperatures. Therefore, H2O was detected from 300 K to 1100 K, at
which point almost all H2O was converted to H2 via steam reforming. H2O is one
reactant input in bi-reforming and tri-reforming,; it is consumed by steam reforming and
almost disappears at temperatures above 1100 K. The mole fraction plots indicate all four
reforming processes have high conversion ratios at temperatures above 1100 K. However,
different H2/CO ratios exist among the four reforming processes.
To compare the conversion of the reactants and the distribution of the products for the
four multi-reforming processes, CH4 conversion, CO2 conversion, the H2/CO ratio, and
the H2 yield are plotted versus temperature (Figure 8.3). As shown in Figure 8.3 (a) and
Figure 8.3 (b), the conversions increased with increasing temperature and the majority of
the reactants were consumed in the 700 K to 1100 K temperature range. When the
temperature exceeded 1100 K, 90 % of the CH4 was converted in all four types of
reforming. A higher conversion—99 %—was achieved above 1300 K. However, the four
types of reforming performed similarly in terms of CH4 conversion but performed very
differently for CO2 conversion at temperatures below 1300 K. Dry reforming maintained
the highest CO2 conversion ratio. For all other types of reforming, the phenomenon of
negative CO2 conversion was observed at temperatures below 825 K because the WGS
reaction proceeds while the methanation reaction is inhibited as a result of steam injection.
At 1100 K, CO2 conversion rates for dry reforming, bi-reforming, auto-thermal reforming,
and tri-reforming were 96 %, 94 %, 94 %, and 92 %, respectively. When the temperature
exceeded 1300 K, more than 99 % of the CO2 was converted for all four types of
reforming. Thus, high-temperature multi-reforming can theoretically proceed at a high
conversion ratio. For temperatures above 1300 K, no residual CH4 and CO2 are expected.
For all reactions, the H2 yield was related to H2O and to the hydrocarbon concentrations
in the product. In addition, the H2 yield was dependent on the CH4 conversion ratio. The
produced hydrocarbons only occupied a small mole fraction. The reactants were primarily
converted at temperatures above 1100 K, and the H2O concentration decreased to a low
value at 1100 K. As shown in the plot, all four types of reforming produced H2 yields
over 90 % at temperatures above 1100 K and yields of 99 % at temperatures above 1300
K. At a high temperature, most hydrogen atoms in the compounds were converted to H2.
The H2/CO ratio is an important parameter for syngas. As shown in Figure 8.3 (d), all
four types of reforming had a constant H2/CO ratio less than 2 at temperatures above
1000 K. Tri-reforming yielded the best H2/CO ratio of 1.75, whereas dry reforming only
yielded a H2/CO ratio of 0.99.
109
Figure 8.3 Plots of the main parameters for four types of reforming as a function of
temperature that were obtained from thermodynamic equilibrium modeling. The reactants
are added in stoichiometric ratios with 1 atm pressure and temperatures from 300 K to
2100 K: a) CH4 conversion; b) CO2 conversion; c) H2 yield; and d) H2/CO ratio
To investigate the feasibility of high-temperature multi-reforming technology, the
conversion ratio and product concentrations were discussed and compared with simulated
data in the same conditions.
Figure 8.4 indicates that the CH4 conversion, which performed slightly better than the
simulated results, was 1 in all three runs. Thus, almost all CH4 was consumed. The
average mole fraction of H2 was 42 %, which was less than the modeled value of 49 % at
the same temperature. These two phenomena are the result of the extra input of CO2
from the combustion flue gas. In the experiments, the reforming reactions were
performed after combustion of natural gas. Therefore, a small excess quantity of O2, CO2,
and steam remained inside the facility. The amount of CH4 was insufficient to convert all
of the CO2 in the tested products because CO2 remained and less CO and H2 were
produced. The high conversion of CH4 and the high H2 concentration support the high-
temperature multi-reforming technology.
400 600 800 1000 1200 1400 1600 1800 20000
20
40
60
80
100C
H4 C
on
ve
rsio
n %
Temperature (K)
Dry reforming
Bi-reforming
Autothermal reforming
Tri-reforming
(a)
400 600 800 1000 1200 1400 1600 1800 2000
-40
-20
0
20
40
60
80
100
CO
2 C
on
ve
rsio
n %
Temperature (K)
Dry reforming
Bi-reforming
Autothermal reforming
Tri-reforming
(b)
400 600 800 1000 1200 1400 1600 1800 20000
20
40
60
80
100
H2 y
ield
%
Temperture (K)
Dry reforming
Bi-reforming
Auto-reforming
Tri-reforming
(c)
700 800 900 1000 1100 1200 1300 1400 15000
1
2
3
4
5
6
7
8
9
H2/C
O
Temperature (K)
Dry reforming
Bi-reforming
Autothermal reforming
Tri-reforming
(d)
110
Figure 8.4 Comparison of CH4 conversion and H2 concentration between thermodynamic
equilibrium modeling and the high-temperature dry reforming experiment, in which the
temperatures range from 300 K to 2100 K; the reactants are input on a stoichiometric
ratio with 1 atm pressure
Regarding the H2 and CO flow rates, the H2 produced increased from 0.56 Nm3/h to
0.74 Nm3/h with increasing temperatures from 1153 K to 1213 K, as shown in Figure 8.5.
In the simulated results, the H2 flow rate was maintained above 0.94 Nm3/h. Less CO
was produced than suggested by the model. However, the CO flow rate significantly
increased as the temperature increased. At 1153 K and 1169 K, the CO flow rate was only
0.16 Nm3/h and 0.33 Nm3/h, respectively. At 1213 K, the CO flow rate reached 0.68
Nm3/h; at this temperature, the flow rate was expected to be 0.98 Nm3/h. Although the
H2 and CO flow rates were influenced by the extra CO2 input from the combustion flue
gas, the experiments at high temperatures, such as 1213 K, showed significant potential
for the use of this high-temperature multi-reforming technology.
Relatively higher values of hydrocarbons were produced than expected from the
simulated results. Although hydrocarbons only represent a small proportion of the total
products, they were detected at significantly higher levels in the experiment than
suggested by the models, as shown in Figure 8.6. C2H2, C2H4, and C2H6 were analyzed:
C2H2 decreased from 0.63 % to 0.24 %, C2H6 decreased from 1.53 % to 0.78 % as the
temperature increased and C2H4 was extremely high at 1153 K but was undetected from
1169 K and 1213 K. The formation of these hydrocarbons has been explained in the
literature [187]. A small amount of O2 remained inside the facility; therefore, after CH4
input, the environment contained an ultra-rich CH4-O2 mixture. Thus, hydrocarbons were
formed during the reaction of CH4 and O2. The hydrocarbons can be converted to useful
compounds that may generate new ideas for CH4 reforming with CO2.
111
According to the experimental results, a high temperature enhances the multi-reforming
of CH4 with CO2 in the absence of catalysts. This technology is feasible and can be
optimized by evaluating the temperature and ratio of input reactants.
Figure 8.5 Bar chart of the comparison of H2 and CO flow rates between a high-
temperature dry reforming experiment and thermodynamic equilibrium modeling; the
reactants are input using a stoichiometric ratio and 1 atm pressure
Figure 8.6 Bar chart of the comparison of hydrocarbon concentration between the
high-temperature dry reforming experiment and thermodynamic equilibrium modeling;
the reactants are input using a stoichiometric ratio and 1 atm pressure
112
8.2 Summary
The feasibility of a novel technology for high-temperature multi-reforming of CH4 with
CO2 was demonstrated using equilibrium modeling and experimental studies. The three
primary conclusions are as follows:
Thermodynamic equilibrium models demonstrated that at a high temperature (above 1300
K), more than 98 % reactant conversions and H2 yields can be achieved by all four types
of multi-reforming processes. Among all types of stoichiometric reforming, tri-reforming
yielded the best H2/CO ratio near 2 for methanol formation and near 1 for dry reforming.
The pressure also had a significant effect on reactant conversion and product yields. At
high temperatures, low pressure should be maintained to achieve better performance
results.
According to the experimental results, the multi-reforming process will be stable, achieve
high conversion, and yield satisfactory product outputs if the temperature is sufficiently
high. If a sufficiently high reaction temperature can be attained, a catalyst is unnecessary,
and coke formation is no longer a problem. Lower CO2 conversion and a greater number
of hydrocarbons were detected compared with thermodynamic equilibrium modeling.
This finding may be attributed to the small quantity of excess O2, produced CO2, and
remaining steam inside the facility after combustion.
113
Chapter 9
9 Conclusion and future studies
9.1 Conclusion
Thermal conversion behaviors and heat transfer phenomena of municipal solid waste
were investigated in this thesis. The main conclusions can be summarized as follows:
1. A significant transient volumetric swelling behavior was observed during the
pyrolysis of pelletized recovered solid waste fuels. A significant decrease in the
apparent densities of the RDF particles, which also occurs during the swelling
process, may increase the risk of bridging in a fixed-bed gasifier. A porous
structure with large pores was also observed.
2. A particle model was developed to describe the mass and heat transfers inside a
porous MSW particle with a high PE concentration. The void size has a significant
effect on radiation heat transfer inside a porous medium and is a crucial factor in
the calculation of the radiation terms in effective thermal conductivity. The
simplified Kunii and Smith model is inadequate for estimating the heat transfer
process inside a highly porous medium. A coefficient as a function of porosity was
proposed to modify the radiation term in the Kunii and Smith model. All heat
transfer models failed to accurately predict the mass evolution of MSW particles
for a PE concentration of 89 %. The reason for the failure is that a large amount
of molten plastic liquid can be carried with the gaseous volatiles toward the
particle surface. Because a char layer does not exist to inhibit the heat transfer, the
molten liquid on the surface exhibits a much faster degradation process than the
degradation process of the inner layer inside the MSW particle. Thus, a new heat
transfer model that considers the liquid mobility is needed to predict the MSW
degradation process for a maximum plastic concentration of 90 %.
3. Paper-board in fast pyrolysis is directly converted to HAA at temperatures above
500 °C via hemolytic cleavage in the presence of alkaline earth metals. For the
pyrolysis of conventional lignocellulosic materials, the effect of residence time on
these primary products is significant. H radical and O radicals serve key roles in
the formation of olefins, alkynes, benzenes, and phenols. Benzenes can act as a
precursor in the formation of heavier PAHs.
114
4. With an increase in the effective hydrogen ratio, the yield for CO and CO2 rapidly
decreases, whereas the carbon yield in lower olefins increases. The effective
hydrogen ratio is adequate for evaluating the potential yield of lower olefins, CO
and CO2 in syngas from RDF degradation.
5. A random degradation route is the predominant mechanism for PE degradation.
The liquid product, which primarily consists of C11-C18 alkanes and 1-olefines, is
highly desired in the petrochemical industry for plastic and detergent manufacture.
Low levels of the aromatics alkyl-benzene, alkylnaphthalene, and alkyl-indene were
detected. Cyclization reactions from long-chain olefins are the main mechanism
for aromatic formation during PE conventional pyrolysis below 600 °C.
6. Dehydrochorination is the first step for PVC thermal decomposition. Benzene and
naphthalene are the two main products of PVC fast pyrolysis. For PVC
conventional pyrolysis, the yields of o-xylene, methylnaphthalene, 1H-Indene, and
1-methylene- are relatively high, and no naphthalene was detected. No intense
demethylation reactions occurred due to the low yield of lower alkanes and low
alkenes. PVC conventional pyrolysis favors alkyl aromatics formation because of
the tendency of crosslinking between neighboring chains.
7. The high yield of CO2 is primarily attributed to the decarboxylation reactions of
benzoic acid, which is the main product of PET pyrolysis. Biphenyl yield is the
dominant product of PET conventional pyrolysis. The possible mechanism to
form biphenyl is attributed to the formation of its precursor—benzene—which is
the other main product from benzoic acid decarboxylation.
8. CaO used in the steam gasification of a blended MSW/CaO material showed a
superior capacity toward CO2 adsorption, resulting in hydrogen-rich syngas. The
catalytic effect of CaO on the degradation of plastics also exists. A superior
dispersion of calcium can be obtained on the char surface during MSW/CaO
gasification because the plastic components in MSW are able to move and
surround calcium particles in their molten state. The molar ratio of H2 to (2CO2 +
CO) in the char gasification stage increased when CaO was added; this result can
be reasonably explained by the C–H complex that remained on the surfaces of the
calcium species. The highest molar ratio was obtained at 800 °C.
9. The feasibility of a CCR system and an integrated municipal solid waste system
was demonstrated for CO2 remediation. In terms of the H2/CO ratio, tri-
reforming produced the best H2/CO ratio of all types of stoichiometric
reforming—approximately 2 for methanol formation and approximately 1 for dry
reforming. According to the experimental results, the multi-reforming process will
be stable if the temperature is sufficiently high and high conversion and acceptable
product outputs can be achieved. If a sufficiently high reaction temperature can be
reached, a catalyst is not necessary, and coke formation is no longer a problem.
115
9.2 Future studies
This thesis enhanced the understanding of the fundamentals of municipal solid waste
thermal conversion and formed some conceptions for the efficient and environmentally
friendly utilization of municipal solid waste. However, additional studies are needed in the
following areas:
1. A particle model should be developed for the steam gasification of MSW particles.
The pressure variations inside the porous medium were considered and described
using the Darcy law. Both heterogeneous and homogeneous reactions should be
considered for this process. A new heat transfer model should be developed to
describe heat transfer in a liquid phase with the passage of bubbling.
2. Detailed experiments for the rapid and conventional pyrolysis of MSW and the
main factors to characterize the liquid products are needed. An evaluation or
prediction model that is based on the effective hydrogen ratio can be developed
based on these detailed experiments.
3. The pyrolysis of MSW and CaO blendings (or PVC and CaO blendings) should
be conducted to develop an optimized approach to remove chlorine and increase
the effective hydrogen ratio of feedstock that contains PVC.
4. To understand the effect of ER, the mass ratio of MSW and CaO and other
operational conditions on the removal of pollutants and energy efficiency, steam
gasification of MSW and CaO blendings should be performed in an advanced
continuous gasification system such as an HTAG.
5. An environmental, economic and social sustainability assessment is a useful tool
for the evaluation of a CCR system that is combined with an integrated municipal
solid waste system.
116
References
[1] Bing Xue, Yong Geng, Wan-xia Ren, et al., "An overview of municipal solid waste
management in inner Mongolia autonomous region, " China J Mater Cycles Waste Manage,
vol.13, pp. 283–292, 2011.
[2] "UPEN year book: new science and developments in our changing environment, " United
Nations Environment Programme, 2010.
[3] Brian Bahor and Michael Van Brunt, "Integrated waste management as a climate change
stabilization wedge, " Waste management & Research, vol.27, p. 839-849, 2009.
[4] A.J.Chandler, T.T.Eighmy, J.Hartlen, et al, "Municipal solid waste incinerator residues. The
international Ash working group, " Elsevier Science B.V.Neterlands, p. 21-51, 1991.
[5] Christensen T.H., Simion F., Tonini D. and Moller J., "Global warming factors modelled
for 40 generic municipal waste management scenarios, "Waste Management & Research,
vol.27,p. 871-884, 2009.
[6] Khalil MAK, "Non-CO2 greenhouse gases in the atmosphere," Annual Review Energy and
Environment, vol.24, p. 645-61, 1999.
[7] Bonner J. and Matthews E, "Global methane emissions from landfills: New methodology
and annual estimates 1980-1996," Global Biogeochem Cycles, vol.17, p. 1-171065, 2003.
[8] "European Environment Agency (EEA) report: Managing municipal solid waste-a review
of achievements in 32 European countries," ISSN. 1725-9177, 2013.
[9] N.J. Themelis, "An overview of the global waste-to-energy industry," Waste Management
World, vol. 3, pp. 40-47, 2003.
[ 10 ] H. Vogg, M. Metzger and L. Stieglitz, "Recent findings on the formation and
decomposition of PCDD/PCDF in municipal solid waste incineration," Waste Management
& Research, vol. 5, pp. 285-294, 1987.
[11] Amalia Mattiello, Paolo Chiodini, Elvira Bianco, et al., "Health effects associated with the
disposal of solid waste in landfills and incinerators in populations living in surrounding
areas: a systematic review," International Journal of Public Health, vol. 58, pp. 725-735, 2013.
[12] Yangsheng Liu and Yushan Liu, "Novel incineration technology integrated with drying,
pyrolysis, gasification, and combustion of MSW and ashes vitrification," Environmental
Science & Technology, vol. 39, pp. 3855-3863, 2005.
[13] P.M Lv, Z.H. Xiong, J. Chang, C.Z. Wu, Y. Chen, J.X. Zhu, "An experimental study on
biomass air-steam gasification in a fluidized bed," Bioresource Technology, vol. 95, pp. 95-101,
2004.
[14] Weihong Yang, Anna Ponzio, and Carlos Lucas, "Wlodzimierz Blasiak. Performance
analysis of a fixed-bed biomass gasifier using high-temperature air," Fuel processing
Technology, vol. 87, pp. 235-245, 2006.
[ 15 ] Qinglin Zhang, Liran Dor, Amit Biswas, Weihong Yang, and Wlodzmierz Blasiak,
"Modeling of steam plasma gasification for municipal solid waste," Fuel processing Technology,
vol. 106, pp. 546-554, 2013.
[16] Muthmann R., "Waste generated and treated in Europe, Data 1990-2001," Office for
official publications of the European communities, ISBN. 92-894-6355-4, 2003.
117
[17] Eriksson O., "Environmental and economic assessment of Swedish municipal solid waste
management in a systems perspective, " Royal institute of technology, PhD Thesis 2003.
[18] Y. Kim, T. Matsuto, and N. Tanaka, "Effect of raw refuse composition on the fuel
properties of RDF from household waste, " Journal of the Japan Society of Waste Management
Experts, vol. 6, pp. 7–15, 1995.
[19] Nobusuke Kobyashia, Yoshinori Itayaa, and Guilin Piaob, "The behavior of flue gas from
RDF combustion in a fluidized bed," Powder Technology, vol. 151, pp. 87–95, 2005.
[20] Eduardo Ferrer a, Martti Aho, and Jaani Silvennoinen, "Fluidized bed combustion of
refuse-derived fuel in presence of protective coal ash," Fuel processing Technology, vol. 87, pp.
33–44, 2005.
[21] A. López, I. de Marco, B.M. Caballero, M.F. Laresgoiti, and A. Adrados, "Pyrolysis of
municipal plastic wastes: influence of raw material composition," Waste Management, vol.
30, pp. 620-627, 2010.
[22] Prospects for Plastics, Shell Briefing SC&c, Leaflet No.1, 1993.
[23] Elizabeth A. Willianms and Paul T. Williams, "The pyrolysis of individual plastics and a
plastic mixture in a fixed bed reactor," Journal of Chemical Technology and Biotechnology, vol. 70,
pp. 9-20, 1997.
[24] R. Marsh, A.J. Griffiths, and K.P. Williams, "Measurement of heat transfer and change in
compressive strength of waste derived solid fuels due to devolatisation," Fuel, vol. 87, pp.
1724–1733, 2008.
[25] Murat Dogru, Adnan Midilli, and Colin R. Howarth, "Gasification of sewage sludge using
a throating downdraft gasifier and uncertainty analysis," Fuel Processing Technology, vol. 75,
pp. 55–82, 2002.
[26] Kuerston Richard D, "Retort with anti-bridging mechanical agitator," US Patent 4563247,
1986.
[27] Visagie JP, "Generic gasifier modeling: evaluating model by gasifier type," University of
Pretoria, pp. 16–25, 2008.
[28] Harold H. Schobert, Robert C. Streeter, and Erle K. Diehl, "Flow properties of low-rank
coal ash slags: implications for slagging gasification,"Fuel, vol. 64, pp. 1611–1617, 1985.
[29] Bress Dellason F, and Murray Hill NJ, "Automated poking system for coal gasifier," US
4134738, 1979.
[30] S. Salvador, M. Quintard, and C. David, "Combustion of a substitution fuel made of
cardboard and polyethylene: Influence of the mix characteristics-experimental approach,
" Fuel ,vol. 32, pp. 451-462, 2004.
[31] W. Van Antwerpen, C.G. du Toit, and P.G. Rousseau, "A review of correlations to model
the packing structure and effective thermal conductivity in packed beds of mono-sized
spherical particles, " Nuclear Engineering and Design, vol. 240, pp.1803-1818, 2010.
[32] S. Salvador, M. Quintard, and C. David, "Combustion of a substitution fuel made of
cardboard and polyethylene: Influence of the mix characteristics-Modeling," Fire and
Materials, vol. 32, pp. 417-444, 2008.
118
[33] Michael S. Talmadge, Robert M. Baldwin, Mary J. Biddy, et al., "A perspective on
oxygenated species in the refinery integration of pyrolysis oil, " Green Chem., vol. 16, pp.
407-453, 2014.
[ 34 ] N.Y. Chen, J.T.F. Degnan and L.R. Koening, "Liquid fuel from carbohydrates,"
CHEMTECH, vol. 16, pp.506-511, 1986.
[35] Angel Linares-Solano, Edwin J. Hippo, and Philip L. Walker Jr., "Catalytic activity of
calcium for lignite char gasification in various atmosphere,"Fuel, vol. 65, pp.776–779,
1986.
[36] Philippe Barthe, Henri Charcosset, and Jean –Michel Guet, "Calcium efficiency in the low
temperature gasification of coke in air," Fuel, vol. 65, pp. 1330-1333, 1986.
[37] Radovic LR, Walker PL, R.G., "Proceedings of an international symposium fundamentals
of catalytic coal and carbon gasification," Amsterdam, 1983.
[38] Arash Karimi, Murray R. Gray, "Effectiveness and mobility of catalysts for gasification of
bitumen coke," Fuel, vol. 90, pp. 120-125, 2011.
[39] S. Fujikawa, H. Hayashi, T. Nakazawa, K. Kawasaki, T. Iyoku, S. Nakagawa, et al,
"Achievement of reactor-outlet coolant temperature of 950 °C," Journal of Nuclear Science
and Technology, vol. 41, pp. 1245-1254, 2004.
[40] http://www.world-nuclear.org/info/inf42.html.
[41] Colomba Di Blasi, "Modeling and simulation of combustion processes of charring and
non-charring solid fuels," Progress Energy Combustion Science, vol. 19, pp. 71-104, 1993.
[42] Ali Tabatabaie-Raissi,"Cellulose Pyrolysis Kinetics in a Simulated Solar Environment,"
Industrial & Engineering Chemistry Research, vol. 28, pp. 856-865, 1989.
[ 43 ] Jitendra N.Barooah and Victor D.Long, "Rates of thermal decomposition of some
carbonaceous materials in a fluidized bed," Fuel, vol. 55, pp. 116-120, 1976.
[44] Lewellen P.C., Peters W.A., and Howard J.B., "Cellulose pyrolysis kinetics and char
formation mechnism," Symp (Int) on Combust, 16th, MIT, Cambridge, MA, USA, 1976.
[45] A.G.W. Bradbury, Y. Sakai, and F. Shafizadeh, "A kinetic model for pyrolysis of cellulose,"
Journal of Applied Polymer Science, vol. 23, pp. 3271–3280, 1979.
[46] Vadim Mamleev, Serge Bourbigot, and Jacques Yvon, "Kinetic analysis of the thermal
decomposition of cellulose: the change of the rate limitation," Journal of Analytical and
Applied Pyrolysis, Vol. 80, pp. 141-150, 2007.
[47] Koufopanos CA, Papayannakos N, Maschio G, Lucchesi A., "Modelling of the pyrolysis
of biomass particles. Studies on kinetics, thermal and heat transfer effects," The Canadian
Journal of Chemical Engineering, vol. 69, pp. 907-915, 1991.
[48] Alves S.S and Figueiredo J.L., "Kinetics of cellulose pyrolysis modelled by three consecutive
first-order reactions," Journal of Analytical and Applied Pyrolysis, vol. 17, pp. 37-46, 1989.
[49] Chao-Hsiung Wu, Ching-Yuan Chang, Jrh-Ping Lin, and Jiann-Yuan Hwang, "Thermal
treatment of coated printing and writing paper in MSW: pyrolysis kinetics," Fuel, vol. 76,
pp. 1151-1157, 1997.
[ 50 ] Kuen-song Lin, H.Paul Wang, S.H.Liu, Ni-Bin Chang, Y.J.Huang, and H.-C.Wang,
"Pyrolysis kinetics of refuse-derived fuel," Fuel Processing Technology, vol. 60, pp. 103-110,
1999.
119
[51] P. Grammelis, P. Basinas, A. Malliopoulou, and G. Sakellaropoulos, "Pyrolysis kinetics and
combustion characteristics of waste recovered fuels, " Fuel, vol. 88, pp. 195-205, 2009.
[52] Colombia Di Blasi, "Modeling chemical and physical processes of wood and biomass
pyrolysis," Progree in Energy and Combustion Science, vol. 34, pp. 47-90, 2008.
[53] Miller, C.A and Ramohalli, N.R., "A Theoretical Heterogeneous Model of Wood Pyrolysis,"
Combustion Science Technology, vol. 46, pp. 249-265,1986.
[54] I. S. Wichman and A. Atreya, "A simplified model for the pyrolysis of charring materials,"
Combustion Flame, vol. 68, pp. 231-247, 1987.
[55] H. C. Kung. "A mathematical model of wood pyrolysis," Combustion Flame, vol. 18, pp.
185-195, 1972.
[56] J.Villermaux, B. Antoine, J. Lede, and F. Soulignac, "A new model for thermal volatilization
of solid particles undergoing fast pyrolysis," Chemical Engineering Science, vol.41, pp. 151-
157, 1986.
[57] L. J. Curtis and D. J. Miller, "Transport model with radiative heat transfer for rapid
cellulose pyrolysis, " Industrial & Engineering Chemistry Research, vol. 27, pp. 1775-1783,
1988.
[58] Kansa E.J., Perler H.E. and Chaiken R.F., "Mathematical model of wood pyrolysis
including internal forced convection, " Combustion and Flame, vol. 29, pp. 311-324,1977.
[59] W. R. Chan, M. Kelbon, B. Krieger-Brokett, "Modeling and experimental verification of
physical and chemical processes during pyrolysis of a large biomass particle," Fuel, vol. 64,
pp. 1505-1503, 1985.
[60] R. Bilbao, A. Millera, M. B. Murillo, "Temperature profiles and weight loss in the thermal
decomposition of large spherical wood particles," Industrial & Engineering Chemistry Research,
vol. 32, pp. 1811-1817, 1993.
[61] Anup Kumar Sadhukhan, Parthapratim Gupta, and Ranajit Kumar Saha, "Modeling of
pyrolysis of large wood particles," Bioresource Technology, vol. 100, pp. 3134-3139, 2009.
[62] C. Di Blasi, "Analysis of convection and secondary reaction effects within pourous solid
fuels undergoing pyrolysis," Combustion Science and Technology, vol. 90, pp. 315-340, 1993.
[63] Xunliang Liu, Gan Wang, Gang Pan, and Zhi Wen, "Numerical analysis of heat transfer
and volatile evolution of coal particle," Fuel, vol. 106, pp. 667-673, 2013.
[64] Voveller C., Delfau J.C., Reuillon M., Bransier J., and Laraqui N., "Experimental and
numerical study of the thermal degradation of PMMA," Combustion science Technolgoy, VOL.
53, pp. 187-201, 1987.
[ 65 ] Wichman, I.S., "A model describing the steady-state gasification of bubble-forming
thermoplastics in response to an incident heat flux," Combustion flame, vol. 63, pp. 217-229,
1986.
[66] J.E.J Staggs, "A simplified mathematical model for the pyrolysis of polymers with inert
additives," Fire Safety Journal, vol. 32, pp. 221-240, 1999.
[67] T. Bluhm-Drenhaus, A. Becker, S. Wirtz, and V. Scherer, "A model for the devolatilisation
of large thermoplastic particles under co-firing conditions," Fuel, vol. 101, pp. 161-170,
2012.
120
[68] Valerio Cozzani, Cristiano Nicolella, Mauro Rovatti, and Leonardo Tognotti, "Modeling
and experimental verification of physical and chemical process during pyrolysis of a
refused-derived fuel," Industrial & Engineering Chemistry Research, vol. 35, pp. 90-98, 1996.
[69] Valerio Cozzani, Cristiano NIcolella, luigi petarca, Mauro Rovatti, and Leonardo Tognotti,
"A fundamental study on conventional pyrolysis of a refuse-derived fuel," Industrial &
Engineering Chemistry Research, vol.34, pp. 2006-2020, 1995.
[70] W. Van Antwerpen, C.G. du Toit, and P.G. Rousseau, "A review of correlations to model
the packing structure and effective thermal conductivity in packed beds of mono-sized
spherical particles," Nuclear Engineering and Design, vol. 240, pp. 1803-1818, 2010.
[71] James K. Carson, Simon J. Lovatt, David J. Tanner, and Andrew C.Cleland, "Thermal
conductivity bounds for isotropic, porous materials," Heat and Mass Transfer, vol. 48, pp.
2150-2158, 2005.
[72] D. Kunii and J.M. Smith, "Heat transfer characteristics of porous rocks," AIChE Journal,
vol. 6, pp. 71-78, 1960.
[ 73 ] P. Zehner and E. U. Schlünder, "Wärmeleitfähigkeit von Schüttungen bei mäβigen
Temperaturen, " Chemie Ingenieur Technik , vol. 42, pp. 933–941, 1970.
[74] C.T. Hsu. "Modified Zehner–Schlünder models for stagnant thermal conductivity of
porous media," International Journal of Heat and Mass Transfer, vol. 37, pp. 2751–2759, 1994.
[75] A.M Li, X.D Li, S.Q Li, Y Ren, N Shang, Y Chi, J.H Yan, and K.F Cen, "Experimental
studies on municipal solid waste pyrolysis in a laboratory-scale rotary kiln," Energy, vol. 24,
pp. 209-218, 1999.
[76] Mohammad Nurul Islam, Mohammad Nurul Islam, Mohammad Rafiqul Alam Beg, and
Mohammad Rofiqul Islam, "Pyrolytic oil from fixed bed pyrolysis of municipal solid
waste and its characterization," Renewable energy, vol. 30, pp. 413-420, 2005.
[77] W.K. Buah, A.M. Cunliffe, and P.T. Williams, "Characterization of products from the
pyrolysis of municipal solid waste," Process Safety and Environmental Protection, vol. 85, pp.
450-457, 2007.
[78] I. Velghe , R. Carleer, , J. Yperman, and S. Schreurs, "Study of the pyrolysis of municipal
solid waste for the production of valuable products," Journal of Analytical and applied
pyrolysis, vol. 92, pp. 366-375, 2011.
[79] A.Adrado. I. de Marco, B.M. Caballero, A.Lopez, M.F Laresgoiti, and A.Torres, "Pyrolysis
of plastic packaging waste: A comparison of plastic residuals from material recovery
facilities with simulated plastic waste," Waste Management, vol. 32, pp. 826-862, 2012.
[80] Kaminsky W., "Plastics, recycling," Ullman’s Encyclopedia of industrial chemistry, vol. A21, pp.
57-73, 1992.
[81 ] Ilkay Cit, Ali Smag, Tugrul Yumak, Suat Ucar, Zarife Misirlioglu, Muammer Canel.
"Comparative pyrolysis of polyolefins (PP and LDPE) and PET," Polymer Bulletin, vol. 64,
pp. 817-834, 2010.
[82] Pinto F,Costa P, Gulyurtlu I, and Cabrita I, "Pyrolysis of plastic waste: 2. Effect of catalyst
on product yield," Journal of Analytical and Applied Pyrolysis, vol.51, pp. 57-71, 1999.
[83] Jude A. Onwudili, Nagi Insura, and Paul T. Williams, "Composition of products from the
pyrolysis of polyethylene and polystyrene in a closed batch reactor: Effects of
121
temperature and residence time," Journal of Analytical and Applied Pyrolysis, vol. 86, pp. 293-
303, 2009.
[84] F.Pinto, P.Costa, I.Gulyurtlu I, and Cabrita, "Pyrolysis of plastic wastes: 1. Effect of
plastic waste composition on product yield, " Journal of Analytical and Applied Pyrolysis, vol.
51, pp. 39-55, 1999.
[85] Nilgun Kiran CILIZ, Ekrem Ekinci, and Colin E.Snape, "Pyrolysis of virgin and waste
polypropylene and its mixtures with waste polyethylene and polystyrene," Waste
Management, vol. 24, pp. 173-181, 2004.
[86] Jingli Wu, Tianju Chen, Xitao Luo, Dezhi Han, Zhiqi Wang, and Jinhu Wu, "TG/FTIR
analysis on co-pyrolysis behavior of PE, PVC and PS," Waste Management, vol. 34, pp.
676-682, 2014.
[87] Y.-H. Lin and M.-H.Yang, "Tertiary recycling of polyethylene waste by fluidiseb-bed
reactions in the presence of various cracking catalysts," Journal of Analytical and paplied
pyrolysis, vol. 83, pp. 101-109, 2008.
[88] Jasmin Shah, M.Rasul Jan, Fazal Mabood, and Farah Jabeen, "Catalytic pyrolysis of LDPE
leads to valuable resource recovery and reduction," Energy Conversion and Management, vol.
51, pp. 2791-2801, 2010.
[89] Maoyun He, Bo Xiao, , Shiming Liu, Zhiquan Hu, Xianjun Guo, Siyi Luo, and Fan Yang,
"Syngas production from pyrolysis of municipal solid waste (MSW) with dolomite as
downstream catalysts," Journal of Analytical and Applied Pyrolysis, vol. 87, pp. 181-187, 2010.
[90] R.Miranda, H.Pakdel, C.Roy, and C.Vasile, "Vacuum pyrolysis of commingled plastics
containing PVC ii.Product analysis," Polymer Degradation and Stability, vol.73, pp.47-67,
2001.
[91] Laine Tiikma and Ille Johannes, "Hans Luik. Fixation of chlorine evolved in pyrolysis of
PVC waste by Estonian oil shales," Journal of Analytical and Applied Pyrolysis, vol. 75, pp.
205-210, 2006.
[92] Sanae Horikawa, Yoshifumi Takai, Hiroichi Ukei, Naoto Azuma, and Akifumi Ueno,
"Chlorine gas recovery from polyvinyl chloride," Journal of Analytical and Applied Pyrolysis,
vol. 51, pp. 167-179, 1999.
[93] Chao Tang, Yu-Zhong Wang, , Qian Zhou, and Li Zheng, "Catalytic effect of Al–Zn
composite catalyst on the degradation of PVC-containing polymer mixtures into pyrolysis
oil, " Polymer Degradation and Stability, vol. 81, pp. 89-94, 2003.
[94] M Blazso and E.Jakab, "Effect of metals, metal oxides, and carboxylates on the thermal
decomposition processes of poly(vinyl chloride)," Journal of Analytical and Applied Pyrolysis,
vol. 49, pp. 125-143, 1999.
[95] Juan A. Conesa, Jose A. Caballero, A. Marcilla, and R. Font, "Analysis of different kinetic
models in the dynamic pyrolysis of cellulose," Thermochimica Acta, vol. 254, pp. 175–192,
1995.
[96] Wu. Chunfei, Zichun Wang, Jun Huang, and Paul T. Williams, "Pyrolysis/gasification of
cellulose, hemicellulose and lignin for hydrogen production in the presence of various
nickel-based catalysts," Fuel, vol. 106pp. 697–706, 2013.
122
[97] R. Ragucci, P. Giudicianni, and A. Cavaliere, "Cellulose slow pyrolysis products in a
pressurized steam flow reactor," Fuel, vol. 107, pp. 122–130, 2013.
[ 98 ] Haiping Yang, Rong Yan, Hanping Chen, Dong Ho Lee, and Chuguang Zheng,
"Characteristics of hemicellulose, cellulose and lignin pyrolysis," Fuel, vol. 86, vol. pp.
1781–1788, 2007.
[99] Shurong Wang, Xiujuan Guo, Tao Liang, Yan Zhou, and Zhongyang Luo, "Mechanism
research on cellulose pyrolysis by Py-GC/MS and subsequent density functional theory
studies," Bioresource Technology, vol. 104, pp. 722–728, 2012.
[ 100 ] Xiaolei Zhang, Weihong Yang, Wlodzimierz Blasiak. "Thermal decomposition
mechanism of levoglucosan during cellulose pyrolysis," Journal of Analytical and Applied
Pyrolysis, vol. 96, pp. 110–119, 2012.
[ 101 ] Xianglan Bai, Patrick Johnston, Sunitha Sadula, and Robert C. Brown, "Role of
levoglucosan physiochemistry in cellulose pyrolysis," Journal of Analytical and Applied
Pyrolysis, vol. 99, pp. 58–65, 2013.
[102] Pushkaraj R. Patwardhan, Justinus A. Satrio, Robert C. Brown, and Brent H. Shanks,
"Influence of inorganic salts on the primary pyrolysis products of cellulose," Bioresource
Technology, vol. 101, pp. 4646–4655, 2010
[103] Wu. Chao-Hsiung, Ching-Yuan Chang, Jyh-ping Lin, and Jian-Yuan Hwang, "Thermal
treatment of coated printing and writing paper in MSW: pyrolysis kinetics," Fuel, vol. 76,
pp. 1151–1157, 1997.
[104] Wu. Chao-Hsing, Chqing-Yuan Chang, and Chao-Heng Tseng, "Pyrolysis products of
uncoated printing and writing paper of MSW," Fuel, vol. 81, pp. 719–725, 2002.
[ 105 ] G.C. Galletti, P. Bocchini, M.E. Guandalix, G. Almendros, S. Camarero, and A.T.
Martinez, "Pyrolysis products as markers in the chemical characterization of paperboards
from waste paper and wheat straw pulps," Bioresource Technology, vol. 60, pp. 51–58, 1997.
[106] E.M. Grieco and G. Baldi, "Pyrolysis of polyethylene mixed with paper and wood:
interaction effects on tar, char and gas yields," Waste Management, vol. 32, pp. 833–839,
2012.
[107] Nobuhiro Tanigaki, Kazutaka Manako, and Morihiro Osada, "Co-gasification of municipal
solid waste and material recovery in a large-scale gasification and melting system," Waste
management, vol. 32, pp.667-675, 2012.
[108] Qinglin zhang, Liran Dor, Dikla Fenigshtein, Weihong Yang, and Wlodzmierz Blasiak,
"Gasification of municipals solid waste in the plasma gasification melting process,"
Applied Energy, vol.90, pp. 106-112, 2012.
[109] Ahmed I.I. and Gupta A.K., "Syngas yield during pyrolysis and steam gasification of
paper," Applied energy, vol. 86, pp. 1813-1821, 2009.
[110] Anna Ponzio, Sylwester Kalisz, Wlodzimierz Blasiak, "Effect of operating conditions on
tar and gas composition in high temperature air/steam gasification (HTAG) of plastic
containing waste," Fuel processing technology, vol. 87, pp. 223-233, 2006.
[111] Ximena A. Garca, Nelson A. Alarcon, and Alfredo L. Gordon, "Steam gasification of tars
using a CaO Catalyst," Fuel Processing Technology, vol.58, pp. 83–102, 1999.
123
[112] Yanwen Guan, Siyi Luo, Shiming Liu, Bo Xiao, and Lei Cai, "Steam catalytic gasification
of municipal solid waste for producing tar-free fuel gas," International Journal of Hydrogen
Energy, vol. 34, pp. 9341–9346, 2009.
[113] Georgios Taralas and Michael G. Kontominas, "Kinetic modeling of VOC catalytic steam
pyrolysis for tar abatement phenomena in gasification pyrolysis technologies," Fuel, vol.
83, pp. 1235–1245. 2004.
[114] A. DOrazio, A. Di Carlo, N. Dionisi, A. Dell Era, and F. Orecchini, "Toluene steam
reforming properties of CaO based synthetic sorbents for biomass gasification process,"
International Journal of Hydrogen Energy, vol. 38, pp. 13282-13292, 2013.
[115] Jianfen Li, Shiyan Liao, Weiyi Dan, Kuile Jia, and Xiaorong Zhou, "Experimental study
on catalytic steam gasification of municipal solid waste for bioenergy production in a
combined fixed bed reactor," Biomass & Bioenergy, vol. 46, pp. 174-180, 2012.
[ 116 ] Siyi Luo, Yangmin Zhou, and Chuijie Yi, "Syngas production by catalytic steam
gasification of municipal solid waste in fixed-bed reactor," Energy, vol. 44, 391-395, 2012.
[117] Zhu Tingyu, Zhang Shouyu, Huang Jiejie, and Wang Yang, "Effect of calcium oxide on
pyrolysis of coal in a fluidized bed," Fuel Processing Technology, vol. 64, pp. 271–284, 2000.
[118] A. Aho, N. DeMartini, A. Pranovich, J. Krogell, N. Kumar, K. Eranen, et al., "Pyrolysis
of pine and gasification of pine chars-influence of organically bound metals," Bioresource
Technology, vol. 128, pp. 22–29, 2013.
[119] Xueyan Zhu, Byungho Song, and Dookie Kim, "Kinetic study on catalytic gasification of
a modified sludge fuel," Particuology, vol. 6, pp. 258–264, 2008.
[120] Umverto Arena, Lucio Zaccariello, and Maria Laura Mastellone, "Fluidized bed gasification
of waste-derived fuels," Waste Management, vol. 30, pp. 1212–1219, 2010.
[121] Mihai Brebu, Thallada Bhaskar, Kazuya Murai, Akinori Muto, Yusaku Sakata, and Md
Azhar Uddin, "Removal of nitrogen, bromine, and chlorine from PP/PE/PS/PVC/ABS-
Br pyrolysis liquid products using Fe- and C-based catalysts," Polymer Degradation Stability,
vol. 87, pp. 225–230, 2005.
[122] Toshiaki Yoshioka, Tomohiko Handa, Guido Grause, Zhigang Lei, Hiroshi Inomata, and
Tadaaki Mizoguchi, "Effects of metal oxides on the pyrolysis of poly(ethylene
terephthalate)," Journal of Analytical and Applied Pyrolysis, vol. 73, pp. 139–144, 2005.
[123] Ljubisa R. Radovic, Philip L. Walker Jr., and Robert G. Jenkins, "Effect of lignite pyrolysis
conditions on calcium oxide dispersion and subsequent char reactivity," Fuel, vol. 62, pp.
209–212, 1983.
[124] Zuo-liang Liu and He-hui Zhu, "Steam gasification of coal char using alkali and alkaline-
earth metal catalysts," Fuel, vol. 65, pp. 1334–1338, 1986.
[125] J. Rath, M.G. Wolfinger, G. Steiner, G. Krammer, F. Barontini, and V. Cozzani, "Heat of
pyrolysis," Fuel, vol. 82 , pp. 81–91, 2003.
[126] Alberto J. Tsamba, Weihong Yang, and Wlodzimierz Blasiak, "Pyrolysis characteristics and
global kinetics of coconut and cashew nut shells," Fuel Processing Technology, vol. 87, pp.
523–530, 2006.
124
[127] J. Jin, S.J. Chen, ang J. Zhang. "Non-isothermal crystallization kinetics of partially miscible
ethylene-vinyl acetate copolymer/low density polyethylene blends," Express polymer letters,
vol.4, pp.141-152, 2010.
[128] M. Avrami, "Kinetics of phase change I: General theory," Journal of Chemical Physics, vol. 7,
pp. 1103–1112, 1939.
[129] Chunguang Zhou, Weihong Yang, and Wlodzimierz Blasiak, "Characteristics of waste
printing paper and cardboard in a reactor pyrolyzed by preheated agents," Fuel Processing
Technology , vol. 116, pp. 63-71, 2013.
[130] C. David, S. Salvador, J-L. Dirion, and M. Quintar, "Determination of a reaction scheme
for cardboard thermal degradation using thermal gravimetric analysis," Journal of Analytical
and Applied Pyrolysis, vol.67, pp. 307-323, 2003.
[131] Wai-Chun R. Chan, Marcia Kelbon, Barbara B. Krieger, "Modelling and experimental
verification of physical and chemical processes during pyrolysis of a large biomass particle,
" Fuel, vol. 64, pp. 1505-1513, 1985.
[132] Sakae Yagi and Daizo Kunii, "Studies on effective thermal conductivities in packed bes,"
AIChE. Journal, vol. 3, pp. 373-381, 1957.
[133] G. Breitbach and H. Barthels, "The radiant heat transfer in the high temperature reactor
core after failure of the after heat removal systems," Nuclear Technology, vol. 49, pp. 392-
399, 1980.
[134] W. B. Argo and J. M. Smith, "Heat transfer in packed beds- prediction of radial rates in
gas-solid beds," Chemical Engineering Progress, vol. 49, pp. 443-451, 1953.
[135] William Schotte, "Thermal conductivity of packed beds," AIChE Journal, vol. 6, pp. 63-67,
1960.
[136] John A.Dean, "Lange’s Handbook of Chemistry," 15th ed, University of Tennessee,
Knoxville, pp. 10.46, 1999.
[137] G. Pompe, U. Schulze, J. Hu, J. Pionteck, G.W.H. Höhne, "Separation of glass transition
and melting in polyethylene/poly (butyle-methacrylate-co-methyl-methacrylate)
interpenetrating polymer networks in TMDSC and DSC curves," Thermochimica Acta, vol.
337, pp. 179-186, 1999.
[ 138 ] Shu-Sing Chang, "Thermal analysis of vamas polycarbonate-polyetheylene blends,"
Thermochimica Acta, vol. 139, pp. 313-325, 1989.
[139] A. Dawson, M. Rides, J. Nottay, "The effect of pressure on the thermal conductivity of
polymer melts," Polymer Testing, vol. 25, pp. 268-275, 2006.
[140] R.H. Perry and D.W. Green, "Perry’s chemical engineer’s handbook," (7th ed.) McGraw-
Hill, Singapore, 1997.
[141] N.A.S. Amin, and T.C. Yaw, "Thermodynamic equilibrium analysis of combined carbon
dioxide reforming with partial oxidation of methane to syngas," International Journal of
Hydrogen Energy, vol. 32, pp. 1789–1798, 2007.
[142] Y.H. Li, Y.Q. Wang, X.G. Zhang, and Z.T. Mi, "Thermodynamic analysis of autothermal
steam and CO2 reforming of methane," International Journal of Hydrogen Energy, vol. 33, pp.
2507–2514, 2008.
125
[143] Chun Wai Lee, Alan W. Scaroni, and Robert G. Jenkins, "Effect of pressure on the
devolatilization and swelling behavior of a softening coal during rapid heating," Fuel, vol.
70, pp. 957–965, 1991.
[ 144 ] Myung Won Seo, Sang Done Kim, See Hoon Lee, and Jae Goo Lee, "Pyrolysis
characteristics of coal and RDF blends in non-isothermal and isothermal conditions,"
Journal of Analytical and Applied Pyrolysis, vol. 88, pp. 160–167, 2010.
[145] Vladimir Strezov, John A. Lucas, and Terry F. Wall, "Effect of pressure on the swelling of
density separated particles," Fuel, vol. 84, pp. 1238–1245, 2005.
[146] H. Haykiri-Acma, S. Yaman, and S. Kucukbayrak, "Comparison of the thermal reactivities
of isolated lignin and holocellulose during pyrolysis," Fuel Processing Technology, vol. 91, pp.
759–764, 2010.
[147] Ivan Milosavljevic, Vahur Oja, and Eric M. Suuberg, "Thermal effects in cellulose
pyrolysis: relationship to char formation processes," Industrial and Engineering Chemistry
Research, vol. 35, pp. 653–662, 1996.
[148] John E. White, W. James Catallo, and Benjamin L. Legendre, "Biomass pyrolysis kinetics:
a comparative critical review with relevant agricultural residue case studies," Journal of
Analytical and Applied Pyrolysis, vol. 91, pp. 1–33, 2011.
[149] Xiaolei Zhang, Jun Li, Weihong Yang, and Wlodzimierz Blasiak, "Formation mechanism
of levoglucosan and formaldehyde during cellulose pyrolysis," Energy & Fuels, vol. 25, pp.
3739–3746, 2011.
[150] S. Li, J. Lyons-Hart, J. Banyasz, and K. Shafer, "Real-time evolved gas analysis by FTIR
method: an experimental study of cellulose pyrolysis," Fuel, vol. 80, pp. 1809–1817, 2001.
[151] D.L. Allara and D. Edelson, "A computational modeling study of the low-temperature
pyrolysis of n-alkanes; mechanisms of propane, n-butane, and n-pentane pyrolysis,"
International Journal of Chemical Kinetics, vol. VII, pp. 479–507, 1975.
[152] G.L. Browne and W.K.Tang. U.S. Dept. of Agr. Forest Ser. Report. NO. TP-118(1961)
[153] D.K. Shen, S. Gu, "The mechanism for thermal decomposition of cellulose and its main
products," Bioresource Technology, vol. 100, pp. 6496-6504, 2009.
[154] Piskorz, J., Radlein, D., Scott, D.S., "On the mechanism of the rapid pyrolysis of
Cellulose, " Journal of Analytical and Applied Pyrolysis, vol. 9, pp. 121–137, 1986.
[155] Piskorz, J., Radlein, D.S.A.G., Scott, D.S., Czernik, S., "Pretreatment of wood and
cellulose for production of sugars by fast pyrolysis," Journal of Analytical and
Applied Pyrolysis, vol. 16, pp. 127–142, 1989.
[156] Yang C., Lu X., Lin W., Yang X., Yao J., "TG-FTIR study on corn straw pyrolysis-
inluence of minerals," Chemical Research in Chinese Universities, vol. 22, pp. 524-532, 2007.
[157] J.O. Allen, N.M. Dookeran, K.A. Smith, A.F. Sarofim, K. Taghizadeh, and A.L. Lafleur,
"Measurement of polycyclic aromatic hydrocarbons associated with size-segregated
atmospheric aerosols in Massachusetts," Environmental Science and Technology, vol. 30, p.
1023, 1996.
[158] A. Dufour, E. Masson, P. Girods, Y. Rogaume, and A. Zoulalian, "Evolution of aromatic
tar composition in relation to methane and ethylene from biomass pyrolysis-gasification.
" Energy & Fuels, vol. 25, pp. 4182–4189, 2011.
126
[159] G.A.Byrne, D.Gardner and F.H. Holmes, "The pyrolyssi of cellulose and the action of
flame-retardants, " Journal of Applied Chemistry, vol. 16, pp. 81-88, 1966.
[160] F. Shafizadeh and Y.Z. Lai, "Thermal degradation of 1,6-anhydro-β-d-glucopyranose,"
Journal of Organic Chemistry, vol. 37, pp. 278–284, 1972.
[ 161 ] R. Cypres, "Aromatic hydrocarbons formation during coal pyrolysis," Fuel Processing
Technology, vol. 15, pp. 1–15, 1987.
[162] S.C. Moldoveanu, "Pyrolysis of Organic Molecules with Applications to Health and
Environmental Issue," (first ed.)Elsevier, Amsterdam, 2010.
[163 ] Henning Richer and Jack B. Howard, "Formation and consumption of single-ring
aromatic hydrocarbons and their precursors in premixed acetylene, ethylene and benzene
flames," Physical Chemistry Chemical Physics, vol. 4, pp. 2038–2055, 2002.
[164] Thomas E. McGrath, W. Geoffrey Chan, Mohammad R. Hajaligol, "Low temperature
mechanism for the formation of polycyclic aromatic hydrocarbons from the pyrolysis of
cellulose," Journal Analysis and Applied Pyrolysis, vol. 66, pp. 51–70, 2003.
[165] Cullis CF, Hirschler MM, "The combustion of organic polymers," Oxford Clarendon
Press, 1981.
[166] R.W.J. Westerhout, J. Waanders, J.A.M. van Swaaij, "Development of a continuous
rotating cone reactor pilot plant for the pyrolysis of polyethene and polypropene,"
Industrial Engineering Chemical Research, vol. 37, pp. 2293-2300, 1998.
[167] G.Elordi, G. Lopez, M. Olazar, R.Aguado, J.Bilbao, "Product distribution modelling in
the thermal pyrolysis of high density polyethylene," Journal of Hazardous Materials, vol. 144,
pp.708-714, 2007.
[168 ] Ian C. McNeill, Livia Memetea, William J. Coke, "A study of the products of PVC
thermal degradation," Polymer Degradation and Stability, vol. 49, pp. 181-191, 1995.
[169] Nino Dimitrov, Ljerka Kratofil Krehula, Anita Pticek Sirocic, and Zlata Hrnjak-Murgic,
"Analysis of recycled PET bottles products by pyrolysis-gas chromatography," Polymer
Degradation and stability, vol. 98, pp. 972-979, 2013.
[170] Toshiaki Yoshioka, Guido Grause, Shoichi Otani, and Akitsugu Okuwaki, "Selective
production of benzene and naphthalene from poly(butylene terephthalate) and
poly( ethylene naphthalene-2,6-dicarboxylate) by pyrolysis in the presence of calcium
hydroxide," Polyer degradation and stability, vol. 91, pp. 1002-1009, 2006.
[171 ] Donald F. Arseneau, "Competive reactions in the thermal decomposition of cellulose,"
Canadiam Journal of Chemistry, vol. 49, pp. 632-638, 1971.
[172] Shiyi Chen, Dong Wang, Zhipeng Xue, Xiaoyan Sun, Wenguo Xiang, "Calcium looping
gasification for high-concentration hydrogen production with CO2 capture in a novel
compact fluidized bed: simulation and operation requirements, " International Journal of
Hydrogen Energy, vol. 36, pp. 4887–4899, 2011.
[173] Long Han, Qinhui Wang, Zhongyang Luo, Nai Rong, Guangyi Deng, "H2 rich gas
production via pressurized fluidized bed gasification of sawdust with in situ CO2 capture,
" Applied Energy, vol. 109, pp. 36–43, 2013.
127
[174] Yujiang Fan, Haruo Nishida, Tomokazu Mori, Yoshihito Shirai, and Takeshi. Endo,
"Thermal degradation of poly(L-lactide): effect of alkali earth metal oxides for selective
L,L-lactide formation," Polymer, vol. 45, pp. 1197–1205, 2004.
[175] Jasmin Shah, M. Rasul Jan, Fazal Mabood, and Farah Jabeen, "Catalytic pyrolysis of
LDPE leads to valuable resource recovery and reduction of waste problems," Energy
Conversion Management, vol. 51, pp. 2791–2801, 2010.
[176] Jale Yanik, Md. Azhar Uddin, Kazuo IKeuchi, Yusaku Sakata. "The catalytic effect of
Red Mud on the degradation of poly(vinyl chloride) containing polymer mixture into fuel
oil," Polymer Degradation Stability, vol. 73, pp. 335–346, 2001.
[177] D.W. Mckee. "Mechanisms of the alkali metal catalyzed gasification of carbon, " Fuel, vol.
62, pp. 170–175, 1983.
[ 178 ] S.G. Chen, R.T. Yang. "Unified mechanism of alkali and alkaline earth catalyzed
gasification reactions of carbon by CO2 and H2O, " Energy Fuels, vol. 11, pp. 421–427,
1997.
[179] Jie Wang, Mingquan Jiang, Yihong Yao, Yanmei Zhang, Jianqin Cao. "Steam gasification
of coal char catalyzed by K2CO3 for enhanced production of hydrogen without
formation of methane," Fuel, vol. 88, pp. 1572–1579, 2009.
[180] Jie Wang, Yihong Yao, Jianqian Cao, Mingquan Jiang. "Enhanced catalysis of K2CO3 for
steam gasification of coal char by using Ca(OH)2 in char preparation," Fuel, vol. 89, pp.
310–317, 2010.
[181] J.Y. Kim, P.J. Reucroft, V.R. Pradhan, A.G. Comelli. "Surface analysis of catalyst used in
co-processing of coal and waste materials," Fuel Processing Technology, vol. 50, pp. 49–55,
1997.
[182] A.M. Huang, G.G. Xia, J.Y. Wang, S.L. Suib, Y.J. Hayashi, H. Matsumoto, "CO2
reforming of CH4 by atmospheric pressure ac discharge plasmas," Journal of Catalysis, vol.
189, pp. 349–359, 2000.
[183] T. Wurzel, S. Malcus, L. Mleczko, "Reaction engineering investigations of CO2 reforming
in a fluidized-bed reactor," Chemical Engineering Science, vol. 55, pp. 3955–3966, 2000.
[184] D.J. Wilhelm, D.R. Simbeck, A.D. Karp, R.L. Dickenson, "Syngas production for gas-to-
liquids applications: technologies, issues and outlook," Fuel Processing Technology, vol. 71, pp.
139–148, 2001.
[185] B.Y. Huang, X.J. Li, S.G. Ji, B. Lang, F. Habimana, C.Y. Li, "Effect of MgO promoter on
Ni-based SBA-15 catalysts for combined steam and carbon dioxide reforming of
methane," Journal of Natural Gas Chemistry, vol. 17, pp. 225–231, 2008.
[186] C.S. Song, "Tri-reforming: a new process for reducing CO2 emissions," Chemical Innovation,
vol. 31, pp. 21–26, 2001.
[ 187 ] M.K. Drayton, A.V. Saveliev, L.A. Kennedy, A.A. Fridman, Yao-En Li. "Syngas
production using superadiabatic combustion of ultra-rich methane-air mixtures," Twenty-
Seventh Symposium (International) on Combustion, The Combustion Institute, pp.
1361–1367, 1998.