mathematical modeling of municipal solid waste plasma gasification in a fixed-bed melting reactor
TRANSCRIPT
Mathematical modeling of municipal
solid waste plasma gasification in a
fixed-bed melting reactor
Qinglin Zhang
Doctoral Dissertation
Stockholm 2011
Royal Institute of Technology
School of Industrial Engineering and Management Department of Material 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，fredagen den 25 November 2011, kl. 10.00 i Lindstedtsvägen 5 Entreplan (D2), Kungliga Tekniska Högskolan, Stockholm.
ISRN KTH/MSE--11/37--SE+ENERGY/AVH ISBN 978-91-7501-141-7
Qinglin Zhang. Mathematical modeling of municipal solid waste plasma gasification in a fixed-bed melting reactor Royal Institute of Technology School of Industrial Engineering and Management Department of Material Science and Engineering Division of Energy and Furnace Technology SE-100 44 Stockholm Sweden ISRN KTH/MSE--11/37--SE+ENERGY/AVH ISBN 978-91-7501-141-7
© the author
Dedicated to my beloved parents
谨以此文献给我挚爱的父母
I
Abstract
The increasing yield of municipal solid waste (MSW) is one of the main by-products of modern society. Among various MSW treatment methods, plasma gasification in a fixed-bed melting reactor (PGM) is a new technology, which may provide an efficient and environmental friendly solution for problems related to MSW disposals. General objectives of this work are to develop mathematical models for the PGM process, and using these models to analyze the characteristics of this new technology.
In this thesis, both experimental measurement and numerical analysis are carried out to evaluate the performance of both air gasification and air&steam gasification in a PGM reactor. Furthermore, parameter studies were launched to investigate the effect of three main operation parameters: equivalence ratio (ER), steam feedstock mass ratio(S/F) and plasma energy ratio (PER). Based on the above analysis, the optimal suggestions aiming at providing highest syngas calorific value, as well as system energy efficiency, are given.
Six experimental tests were conducted in a demonstration reactor. These tests are classified into two groups: air gasification (case 1 and 2) and air&steam gasification (case 3 to 6). In all these cases, the plasma gasification and melting of MSW produced a syngas with a lower heating value of 6.0-7.0 MJ/Nm3. By comparing the syngas yield and calorific value, the study found out that the steam and air mixture is a better gasification agent than pure air. It is also discovered that the operation parameters seriously influence the operation of the PGM process.
A zero-dimensional kinetic free model was built up to investigate the influence of operation parameters. The model was developed using the popular process simulation software Aspen Plus. In this model, the whole plasma gasification and melting process was divided into four layers: drying, pyrolysis, char combustion&gasificaiton, and plasma melting. Mass and energy balances were considered in all layers. It was proved that the model is able to give good agreement of the syngas yield and composition. This model was used to study the influence of ER, S/F and PER on average gasification temperature, syngas composition and syngas yield. It is pointed out that a common problem for the PGM air gasification is the incomplete char conversion due to low ER value. Both increasing plasma power and feeding steam is helpful for solving this problem. The syngas quality can also be improved by reasonably feeding high temperature steam into the reactor.
In order to provide detailed information inside the reactor, a two-dimensional steady model was developed for the PGM process. The model used the Euler-Euler multiphase approach. The mass, momentum and energy balances of both gas and solid phases are considered in this model. The model described the complex chemical and physical processes such as drying, pyrolysis, homogeneous reactions, heterogeneous char reactions and melting of the inorganic components of MSW. The rates of chemical reactions are controlled by kinetic rates and physical transport theories. The model is capable of simulating the pressure fields, temperature fields, and velocity fields of both phase, as well as variations of gas and solid composition insider the reactor. This model was used to simulate both air gasification and air&steam gasification of MSW in the PGM reactor.
II
For PGM air gasification, simulated results showed that when ER varies from 0.043 to 0.077, both the syngas yield and cold gas efficiency demonstrated a trend of increasing. This is explained mainly by the increase of char conversion rate with ER. However, the increase of ER was restricted by peak temperature inside the fixed-bed reactor. Therefore, it is not suggested to use only air as gasification in the PGM process. The influence of plasma power is not obvious when PER varies from 0.098 to 0.138.
The positive influences of steam addition on cold gas efficiency and syngas lower-heating-value are confirmed by the simulation results of PGM air&steam gasification. The main effect of steam addition is the rouse of water shift reaction, which largely accelerates the char conversion and final yields of hydrogen and carbon dioxide. The effect of steam injection is affected by steam feeding rate, air feeding rate and plasma power.
Based on the above modeling work, Interactions between operation parameters were discussed. Possible operation extents of operation parameters are delimitated. The optimal points aiming at obtaining maximum syngas LHV and system CGE are suggested.
Key words: Mathematical modeling, plasma gasification, municipal solid waste, fixed-bed
III
Acknowledgment
First and foremost, I would like to express my sincere gratitude to my supervisors, Professor Wlodzimierz Blasiak and Docent Weihong Yang for their excellent guidance, continuous help, encouragement and support during my study in KTH.
I am very thankful to Mr. Liran Dor at Environmental Energy Resources Ltd. He has been a great support to me and a link to the real industrial scale PGM reactor. Liran is a very nice guy who always pleased to answer my questions. Many thanks for all the helps during the development of the numerical models.
I would like to thank Amit, Efthymios, Kentaro, Pawel, Lan and all other colleagues and friends in the Division of Energy and Furnace Technology. They are very helpful for my study at KTH. I have learned a lot from discussions with them.
This work is supported by the Environmental Energy Resources (Israel) Ltd., the inventor and owner of the PGM technology and the demonstration plant. The support from EER is very important for my work, and is grateful acknowledged.
I am grateful to China Scholarship Council for offering partial scholarship for my PhD study.
Last but not least, I would like to express my deepest thank to my girlfriend Wen for her support and love.
IV
List of paper included in the thesis
Supplement I Q. Zhang, L. Dor, D. Fenigshtein, W. Yang, W. Blasiak. Gasification of municipal solid waste in the Plasma Gasification Melting process. Applied Energy (2011), DOI:10.1016/j.apenergy.2011.01.041
Supplement II Q. Zhang, L. Dor, W. Yang, W. Blasiak. Properties and optimizing of a
plasma gasification & melting process of municipal solid waste. Paper #58 in the proceedings of International Conference of Thermal Treatment Technology & Hazardous Waste Combustors (IT3/HWC). May 17-20, 2010, San Francisco, California, USA.
Supplement III Q. Zhang, L. Dor, W. Yang, W. Blasiak. An eulerian model for
municipal solid waste gasification in a fixed-bed plasma gasification melting reactor. Energy Fuels, 2011, 25 (9), pp 4129–4137.
Supplement IV Q. Zhang, L. Dor, W. Yang, W. Blasiak. Modeling of steam plasma
gasification for municipal solid waste. Submitted to Fuel Processing Technology, in June 2011. Supplement V Q. Zhang, L. Dor, L. Zhang, W. Yang, W. Blasiak. Performance analysis
of municipal solid waste gasification with steam in a Plasma Gasification Melting reactor.
Submitted to Applied Energy, in July 2011.
V
List of papers not included in the thesis
1. Q. Zhang, A. Swiderski, W. Yang, W. Blasiak. Experimental and numerical studies of pulverized coal combustion with high-temperature air. 8th European Conference on Industrial Furnaces and Boilers, Vilamoura, Portugal, Match, 2008.
2. Q. Zhang, A. Swiderski, W. Yang, W. Blasiak. Properties of pulverized coal combustion in high temperature air/steam mixture. Finish-Swedish Flame Days. Naantali, Finland, January, 2009.
3. Q. Zhang, L. Dor, K. Umeki, W. Yang, W. Blasiak. Process modeling and performance analysis of a PGM gasifier. 10th Conference on Energy for a Clean Environment. Lisbon, Portugal, July, 2009.
4. Q. Zhang, L. Dor, W. Yang, W. Blasiak. CFD modeling of municipal solid waste gasification in a fixed-bed plasma gasification melting reactor. International Conference of Thermal Treatment Technology & Hazardous Waste Combustors. Jacksonville, Florida, USA, May, 2011.
5. L. Dor, Q. Zhang, W. Yang, W. Blasiak. Development of a new waste-to-energy system using plasma gasification & melting technology. International Conference of Thermal Treatment Technology & Hazardous Waste Combustors. Jacksonville, Florida, USA, May, 2011.
VI
List of figures
FIGURE 1. RELATIONSHIP BETWEEN COMBUSTION HEAT AND EXTERNAL ENERGY ............................................ 3
FIGURE 2. CONFIGURATIONS OF THREE DIFFERENT GASIFICATION PROCESSES. A) CONVENTIONAL
GASIFICATION B) NORMAL PLASMA GASIFICATION C) PLASMA GASIFICATION MELTING ...................... 4
FIGURE 3. THE DEMONSTRATION OF THE AREA OF STUDY IN THIS WORK ........................................................... 6
FIGURE 4. ILLUSTRATION OF THE FLOW SHEET OF THE DEMONSTRATION PLANT [68] ..................................... 17
FIGURE 5. THE SCHEME OF THE PGM REACTOR IN THE DEMONSTRATION PLANT ........................................... 18
FIGURE 6. SCHEME OF PGM GASIFICATION PROCESS ........................................................................................ 22
FIGURE 7. SCHEME OF THE CFD MODEL ............................................................................................................ 26
FIGURE 8. GEOMETRY AND MESH OF THE 2D MODEL ......................................................................................... 33
FIGURE 9. SYNGAS COMPOSITION OF CASES 1 AND 2 .......................................................................................... 37
FIGURE 10. SYNGAS CHARACTERISTICS OF CASES 1 AND 2 ................................................................................ 38
FIGURE 11. MEASURED TEMPERATURE DISTRIBUTIONS OF CASES 1 AND 2....................................................... 39
FIGURE 12. SYNGAS COMPOSITIONS OF CASES 2, 3 AND 4 .................................................................................. 40
FIGURE 13. SYNGAS CHARACTERISTICS OF CASES 2, 3 AND 4 ............................................................................ 40
FIGURE 14. SYNGAS COMPOSITIONS OF CASES 3, 5 AND 6 .................................................................................. 43
FIGURE 15. SYNGAS CHARACTERISTICS OF CASES 3, 5 AND 6 ............................................................................ 43
FIGURE 16. COLD-GAS EFFICIENCY ..................................................................................................................... 44
FIGURE 17. APPEARANCE OF SLAG AFTER COOLING .......................................................................................... 45
FIGURE 18. EFFECT OF PER ON GASIFICATION AND PYROLYSIS TEMPERATURE .............................................. 47
FIGURE 19. EFFECT OF PER ON SYNGAS COMPOSITION AND TAR YIELD ........................................................... 48
FIGURE 20. EFFECT OF PER ON TOTAL SYNGAS YIELD AND SYNGAS LHV ....................................................... 49
FIGURE 21. EFFECT OF ER ON SYNGAS COMPOSITION AND TAR YIELD ............................................................. 50
FIGURE 22. EFFECT OF ER ON SYNGAS LHV AND SYSTEM CGE ....................................................................... 51
FIGURE 23. EFFECT OF SAMR ON SYNGAS COMPOSITION AND TAR YIELD ....................................................... 52
FIGURE 24. TEMPERATURE DISTRIBUTION ALONG THE SHAFT HEIGHT OF THE BASE CASE 1 ........................... 53
FIGURE 25. GAS TEMPERATURE DISTRIBUTION (K) IN THE BASE CASE 1 ........................................................... 55
FIGURE 26 GAS TEMPERATURE DISTRIBUTIONS IN DIFFERENT HORIZONTAL SECTIONS IN THE BASE CASE 1 . 56
VII
FIGURE 27 SYNGAS COMPOSITIONS OF THE BASE CASE 1, (A) MOLAR FRACTION OF CO, (B) MOLAR FRACTION
OF H2, (C) MOLAR FRACTION OF LHCS, (D) MOLAR FRACTION OF CO2, (E) MOLAR FRACTION OF H2O,
(F) MASS FRACTION OF TAR ........................................................................................................................ 57
FIGURE 28. TEMPERATURE DISTRIBUTION ALONG THE SHAFT HEIGHT FOR DIFFERENT ER VALUES .............. 59
FIGURE 29. PREDICTED TEMPERATURE DISTRIBUTIONS FOR DIFFERENT ER ................................................... 60
FIGURE 30. ECR VALUES ALONG THE SHAFT HEIGHT FOR DIFFERENT ER VALUES ......................................... 61
FIGURE 31. TEMPERATURE DISTRIBUTIONS ALONG THE SHAFT HEIGHT FOR DIFFERENT PER ....................... 62
FIGURE 32. PREDICTED GAS TEMPERATURE (K) DISTRIBUTIONS FOR DIFFERENT S/F VALUES ........................ 64
FIGURE 33. EFFECT OF S/F ON Cη AND η AT ER= 0.06 AND PER= 0.118 ......................................................... 65
FIGURE 34. PREDICTED CONTENTS OF MAIN SPECIES IN GAS PHASE FOR DIFFERENT S/F VALUES. (A) H2
VOLUME FRACTIONS, (B) CO VOLUME FRACTIONS, (C)LHCS VOLUME FRACTIONS, (D) TAR MASS
FRACTIONS .................................................................................................................................................. 66
FIGURE 35. PREDICTED GAS TEMPERATURE (K) DISTRIBUTIONS FOR DIFFERENT ER VALUES ........................ 67
FIGURE 36 EFFECT OF ER ON Cη AT S/F= 0.167 AND PER= 0.118 .................................................................... 68
FIGURE 37. PREDICTED CONTENTS OF MAIN SPECIES IN GAS PHASE FOR DIFFERENT ER VALUES. (A) H2
VOLUME FRACTIONS, (B) CO VOLUME FRACTIONS, (C)LHCS VOLUME FRACTIONS, (D) TAR MASS
FRACTIONS .................................................................................................................................................. 69
FIGURE 38. PREDICTED GAS TEMPERATURE (K) DISTRIBUTIONS FOR DIFFERENT PER VALUES ..................... 70
FIGURE 39. PREDICTED CONTENTS OF MAIN SPECIES IN GAS PHASE FOR DIFFERENT PER VALUES. (A) H2
VOLUME FRACTIONS, (B) CO VOLUME FRACTIONS, (C) LHCS VOLUME FRACTIONS, (D) TAR MASS
FRACTIONS .................................................................................................................................................. 71
FIGURE 40. DEFINITION OF POSSIBLE OPERATION EXTENT OF PER AND ER IN THE PGM PROCESS ............... 72
FIGURE 41. DISTRIBUTIONS OF SYNGAS LHV IN REGION 1 ................................................................................ 74
FIGURE 42. DISTRIBUTIONS OF SYSTEM CGE IN REGION 1. .............................................................................. 74
FIGURE 43. DELIMITATION OF POSSIBLE OPERATION EXTENT OF SAMR AND ER IN THE PGM PROCESS...... 76
FIGURE 44. DISTRIBUTIONS OF SYNGAS LHV IN REGION 1’ .............................................................................. 77
VIII
List of tables
TABLE 1. OVERVIEW OF SUPPLEMENTS AND THEIR OBJECTIVES ......................................................................... 7
TABLE 2. MSW PROXIMATE AND ULTIMATE ANALYSES..................................................................................... 20
TABLE 3. OPERATION PARAMETERS FOR TRIAL CASES. ..................................................................................... 21
TABLE 4. KINETICS DATA FOR PRIMARY AND SECONDARY PYROLYSIS ............................................................. 30
TABLE 5. KINETIC RATES OF HOMOGENEOUS REACTIONS ................................................................................. 31
TABLE 6. EXPRESSION OF kk FOR HETEROGENEOUS REACTIONS ...................................................................... 32
TABLE 7. OPERATION CONDITIONS IN SERIES 1 .................................................................................................. 35
TABLE 8. OPERATION CONDITIONS IN SERIES 2 .................................................................................................. 36
TABLE 9. COMPARISON BETWEEN MEASURED AND PREDICTED RESULTS OF AIR AND STEAM GASIFICATION IN
THE PGM REACTOR (DRY BASIS) ............................................................................................................... 46
TABLE 10. SYNGAS YIELD AND COMPOSITIONS FOR THE BASE CASE 1 ............................................................... 53
TABLE 11. MEASURED AND PREDICTED SYNGAS YIELD AND MAIN COMPOSITIONS OF THE BASE CASE 2 ......... 63
IX
Content
1. INTRODUCTION ............................................................................................................................................. 1
1.1 BACKGROUND ............................................................................................................................................... 1
1.2 HEAT OF GASIFICATION ................................................................................................................................. 2
1.3 PLASMA GASIFICATION MELTING – AN INNOVATION TECHNOLOGY FOR MSW DISPOSAL ............................ 3
1.4 OUTLINE OF THIS WORK ................................................................................................................................ 5
1.5 SUPPLEMENTS ............................................................................................................................................... 6
2. LITERATURE REVIEW ................................................................................................................................. 9
2.1 EXPERIMENTAL STUDIES RELATED TO PLASMA GASIFICATION AND MELTING OF MSW ................................ 9
2.1.1 MSW gasification .................................................................................................................................. 9
2.1.2 Gasification and melting .................................................................................................................... 10
2.1.3 Application of Plasma in gasification ................................................................................................. 11
2.2 DEVELOPMENT OF GASIFICATION MODELS .................................................................................................. 13
2.3 REACTION RATES ........................................................................................................................................ 14
2.3.1 Drying ................................................................................................................................................. 14
2.3.2 Pyrolysis ............................................................................................................................................. 14
2.3.3 Heterogeneous char reactions ............................................................................................................ 15
2.3.4 Homogeneous reactions ..................................................................................................................... 16
3. METHODOLOGY ......................................................................................................................................... 17
3.1 TEST FACILITY............................................................................................................................................. 17
3.1.1 The demonstration plant ..................................................................................................................... 17
3.1.2 The PGM reactor ................................................................................................................................ 18
3.1.3 Measurement methods ........................................................................................................................ 19
3.1.4 Feedstock ............................................................................................................................................ 20
3.1.5 Test procedure .................................................................................................................................... 20
3.2 ZERO-DIMENSIONAL KINETICS-FREE MODEL ............................................................................................... 21
3.2.1 Drying ................................................................................................................................................. 23
3.2.2 Pyrolysis ............................................................................................................................................. 23
X
3.2.3 Char combustion&gasification ........................................................................................................... 24
3.2.4 Melting ................................................................................................................................................ 25
3.3 TWO-DIMENSIONAL CFD MODEL ................................................................................................................ 26
3.3.1 Conservation equations ...................................................................................................................... 26
3.3.2 Reaction model ................................................................................................................................... 29
3.3.2.1 Drying ........................................................................................................................................................... 29
3.3.2.2 Pyrolysis ....................................................................................................................................................... 29
3.3.2.3Homogeneous reactions ................................................................................................................................. 30
3.3.2.4 Heterogeneous char reactions ....................................................................................................................... 32
3.3.3 Geometry and boundary conditions .................................................................................................... 33
3.3.4 Simulated cases .................................................................................................................................. 34
4. RESULTS AND DISCUSSION ..................................................................................................................... 37
4.1 MEASURED RESULTS ................................................................................................................................... 37
4.1.1 Syngas quality in air gasification ....................................................................................................... 37
4.1.2 Syngas quality in air and steam gasification ...................................................................................... 40
4.1.2.1 Influence of steam feed rate .......................................................................................................................... 40
4.1.2.2 Influence of plasma power and ER ............................................................................................................... 42
4.1.3 Energy efficiency ................................................................................................................................ 44
4.1.4 Slag properties .................................................................................................................................... 45
4.2 RESULTS FROM ZERO-DIMENSIONAL KINETICS-FREE SIMULATION .............................................................. 46
4.2.1 Model validation ................................................................................................................................. 46
4.2.2 Effect of Plasma Power ...................................................................................................................... 46
4.2.3 Effect of ER ......................................................................................................................................... 49
4.2.4 Effect of SAMR ................................................................................................................................... 51
4.3 CFD RESULTS OF AIR GASIFICATION ........................................................................................................... 52
4.3.1 Analysis of the base case 1 ................................................................................................................. 52
4.3.1.1 Model validation ........................................................................................................................................... 52
4.3.1.2 Temperature profiles ..................................................................................................................................... 54
4.3.1.3 Nonuniformity of temperature distributions in horizontal sections .............................................................. 55
4.3.1.4 Composition profiles .................................................................................................................................... 57
XI
4.3.2 Influence of ER ................................................................................................................................... 58
4.3.2.1 Gas temperature distribution ......................................................................................................................... 58
4.3.2.2 Syngas composition ...................................................................................................................................... 59
4.3.2.3 Energy conversion ratio ................................................................................................................................ 60
4.3.3 Influence of PER ................................................................................................................................. 62
4.4 CFD RESULTS OF AIR AND STEAM GASIFICATION ........................................................................................ 63
4.4.1 Model validation ................................................................................................................................. 63
4.4.2 Effect of S/F ........................................................................................................................................ 64
4.4.3 Effect of ER ......................................................................................................................................... 67
4.4.4 Effect of PER ...................................................................................................................................... 70
4.5 OPTIMIZING OF THE PGM PROCESS ............................................................................................................. 72
4.5.1 Interactions between ER and PER ...................................................................................................... 72
4.5.2 Considering the oxygen equilibrium ................................................................................................... 75
5. CONCLUSIONS AND RECOMMENDATIONS ........................................................................................ 78
6. REFERENCE .................................................................................................................................................. 81
XII
Nomenclature
A Pre-exponential factor
vA Specific surface area (m-1)
C Molar concentration (kmol m-3)
pC Heat capacity (J kg-1 K)
D Diffusion coefficient of vapor in the bulk (m2 s-1)
sd Particle diameter (m)
E Activation energy (J kmol-1)
g Gravitational acceleration (m s-2)
G Gibbs energy (J)
H Height (m)
evaH Evaporation heat of moisture (J kmol-1)
gasiH Heat of gasification (J kg-1)
h Specific enthalpy (J kg-1)
K Interphase momentum exchange coefficient (kg m-3 s-1)
k Heat transfer coefficient (W m-2 K-1)
mk Mass transfer coefficient (m s-1)
rk Kinetic coefficient (m s-1)
M Molar weight (kg kmol-1)
M Mass flow rate (kg s-1)
m Mass transfer rate (kg m-3 s-1)
Nu Nusselt number
P Power (W)
Pr Prandtl number
p Pressure (Pa)
Q Intensity of heat exchange (W m-3)
q Heat flux (W m-2)
Re Reynolds number
r Reaction rate (kmol m-3 s-1)
kr Turbulent mixing rate (kmol m-3 s-1)
r Kinetic rate (kmol m-3 s-1)
XIII
S Source term
Sh Sherwood number
T Temperature (K)
t Time (s)
v Velocity (m s-1)
v Stoichiometric coefficient
x Thickness of reactor wall (m)
Y Mass fraction
Greek symbols α Volume fraction
λ Thermal conductivity (W m-1 K-1) ρ Density (kg m-3)
φ Angle of internal friction
τ Stress tensor (Pa)
µ Dynamic viscosity (Pa s)
Subscripts
agent Gasification agents
air Air
gasichar− Char gasification reactions
cel Cellulosic species
yxHC Light Hydrocarbons
CO Carbon monoxide
2CO Carbon dioxide
drying Drying
feedstock Feedstock
g Gas phase
gasi Gasification
XIV
2H Hydrogen
OH 2 Steam
i ith species
MSW MSW
moi Moisture
2O Oxygen
p Phase p
pla Plasma
pri Primary pyrolysis
pyro Pyrolysis
q Phase q
s Solid phase
sec Secondary pyrolysis
stoic Stoichiometric condition
1tar Primary tar
2tar Secondary tar
vol Volatiles
1
1. Introduction
1.1 Background
Municipal solid waste (MSW) is one of the main by-products of human society. In recent
decades, the development of economy in concurrence with changing lifestyle leads to a rapid
increase of MSW yield. According to a recent report by United Nations Environment
Programme (UNEP), the total amount of MSW generation globally in 2007 is about 2.12
billion tones. This number is still increasing at a rate of 7% annually [1].
The conventional MSW disposal method is landfill. Large amount of land is occupied by
landfill every year. Moreover, if landfill is carried out in an improper way, serious
environment problems related to air, water and soil can be aroused [2]. From this point of
view, landfill is not an environmental friendly waste disposal method.
The main components of MSW are food waste, wood, paper, cardboard, plastics, rubbers,
fabrics, metals and stones, and more than half of the MSW compositions are organic species,
which can be used as energy sources. Recently, the conception of energy recovery from MSW
has been a very hot topic, thus leading to a comprehensive study on waste-to-energy
technologies. Besides energy recovery, another advantage of waste-to-energy conception is
that it can sharply reduce the mass and volume of the original waste by 80-95% depending on
the composition of the MSW, since the organic components are consumed during waste-to-
energy processes.
Gasification is a thermal conversion process which converts solid fuels to a combustible gas
by partial combustion. In a sense, gasification is an ‘old’ technology since its first appearance
is about 100 years ago [3]. In recent decades, due to rapid increasing of energy usage, as well
as perceived potential shortages of oil and nature gas, it starts to revive the interest in solid
fuel gasification as an important process to produce gaseous fuels. The application of
gasification in waste-to-energy process is recognized as a promising method to provide a
successful solution for MSW energy usage [4-6].
2
1.2 Heat of gasification
Gasification is generally an endothermic process. The heat of gasification is defined as the
amount of heat required to gasify unit mass of a solid fuel into gaseous products, initially at
standard temperature and pressure. For an ideal gasification process, the heat of gasification
can be divided into three parts: heat required for releasing of moisture, heat required for
devolatilization of volatile species and heat required for char gasification:
gasicharpyrodryinggasi HHHH −++=
Generally, the heat of gasification can come from different sources:
• Reaction heat from partial combustion of feedstock;
• Sensible heat from external sources such as preheated gasification agents, hot sand,
heat pipe and heat radiant tubes;
• Other energy sources such as plasma and microwave;
For conventional gasification, the heat of gasification is mainly from partial combustion of
feedstock. When external energy sources (either sensible heat or other forms of energy) are
used, the heat of gasification can be provided mainly by external energy. In that case,
combustion of feedstock can be prevented.
Figure 1 shows schematically the relationship between combustion heat and external energy
in a gasification process. When no external energy is used, which represents the conventional
gasification, all the heat needed for gasification is provided by combustion of feedstock.
When external energy sources are introduced, the heat from external energy shows a linear
increase with the power of external energy. There exists a critical value of external energy,
where the external energy source supplies the energy needed for preheating, drying and
pyrolysis. If no heat is provided from combustion, the case is a standard pyrolysis. If the
energy required for char gasification can be provided by combustion, then the case turns to
gasification. The value of external energy can keep on increasing, and when value of external
energy reaches the heat of gasification, combustion of feedstock can be completely prevented.
In that case, pure steam gasification is available.
3
External energy
H
+
H
Hcri
dryi
ng
p
yro
Hch
ar-g
asi
Hear from combustion Hear from external sources
0
Figure 1. Relationship between combustion heat and external energy
The prevention of combustion can leads to two main benefits: Firstly, the total calorific value
of syngas increases. This syngas can be very good gaseous fuel and chemical engineering
feedstock. Secondly, the concentrations of combustible gases are enhanced since the dilution
by N2 in air can be prevented. As a result, the syngas lower-heating-value (LHV), as well as
total energy efficiency of gasification with external energy sources is higher than
conventional gasification.
1.3 Plasma Gasification Melting – an innovation technology for
MSW disposal
The study on plasma gasification has been very popular recently. [7-9]. There are two main
advantages of using thermal plasma in the gasification process. Firstly, thermal plasma
provides extra energy to the gasification system, thus receiving all benefits of preventing
combustion. Secondly, the high temperature plasma flow can melt the inorganic components
from MSW. As a result, problems caused by fly and bottom ash can be prevented. After
cooling down, the slag will turn to a vitrified solid, in which heavy metals are locked. This
vitrified solid can be used as good construction material.
4
a) Conventional
gasification b) Normal plasma
gasification c) PGM
Figure 2. Configurations of three different gasification processes. a) Conventional gasification
b) Normal plasma gasification c) Plasma Gasification Melting
Figure 2a shows the general configuration of conventional gasification. Since the heat of
gasification is provided by feedstock combustion, a large volume of air is needed. The
produced syngas has a large volume, but a low LHV due to dilution by a large fraction of N2.
The configuration of normal plasma gasification is shown in Figure 2b. In normal plasma
gasification processes, high temperature plasma flows are injected onto the solid fuel surface
from the top [10-12]. Since sensible heat is supplied by plasma generators, the request of
chemical heat from partial combustion decreases. The reduced combustion can be directly
reflected by decreasing equivalence ratio. As a result, the syngas LHV value and gasification
efficiency increase. However, it has to be point out that the syngas outlet temperature in this
configuration is very high. A relatively large plasma power is needed for normal plasma
gasification.
In order to further increase the energy efficiency, the plasma gasification melting (PGM)
process is developed. The configuration of normal plasma gasification is shown in Figure 2c.
The PGM reactor can be divided into two parts: the gasification shaft and the melting
chamber. In the melting chamber, several plasma Torches are settled. These plasma torches
5
ionizes the air (or any other gas) flowing through the torches, thus forming plasma jets which
extends beyond the tip of the torch. The plasma jets melt the inorganics of the MSW (also
known as ash), which enters the melting chamber. The actual melting/vitrifying of the
inorganics occurs at 1300 to about 2000°C. The hot gases with residual heat then flow into the
gasification shaft. The gasification shaft is a typical updraft fixed-bed gasifier. In this stage,
gasification of organic species in MSW happens, so a combustible gas mixture known as
syngas is produced. During the gasification process, the gases are further cooled by the MSW.
The temperature at the syngas outlet is about 200-400 °C.
By using the PGM technology, the following benefits can be achieved:
1. The required plasma power of PGM is lower than normal plasma gasification since the
flow rat e of ash in the melting chamber is much less than that of the raw MSW;
2. The syngas temperature at the outlet is much lower than normal plasma gasification,
thus leading to higher energy efficiency;
3. Syngas LHV can be enhanced due to less combustion;
4. Lower pollutant emission due to low reaction temperature.
1.4 Outline of this work
Since the PGM is an entirely new conception, the knowledge about this technology is still
poor. No relevant study is found in literatures. Before real industrial application of the PGM
technology, both experimental validation and numerical analysis of the PGM process are
needed.
This work provides comprehensive information of the characters of the PGM process, both
experimentally and numerically. Firstly, the results of a serious of experimental tests in an
industrial scale PGM reactor are analyzed. The results are used as fundamental of the
subsequent study. Then, a 0-dimentional model is used to simulate the PGM process, so as to
study the influences of different operation parameters. For further understanding of the
information inside the reactor, a computational fluid dynamic (CFD) model is then used to
6
simulate the exact behavior of the PEM reactor at different operation conditions. Both pure air
and air&steam mixture are used in above work, so as to find out the best gasification agent for
PGM. Based on the analysis, optimizing suggestions for PGM reactor designing are given.
Generally, the content of this thesis can be divided into four parts, which are listed below:
• Experimental study of a demonstration PGM reactor
• Process simulation of the PGM reactor (0-dimentional simulation)
• CFD simulation of the PGM reactor
• Optimizing of the operation parameters of PGM process
Figure 3. The demonstration of the area of study in this work
1.5 Supplements
The supplements in this thesis are illustrated in Table 1.
7
Table 1. Overview of supplements and their objectives
Supplements Event: Objective:
Experimental test of a demonstration PGM reactor
• Temperature distribution
• Syngas composition of PGM without steam
• Syngas composition of PGM with steam
0-dimentional simulation of the PGM process
• Developing a 0-dimentional model for PGM
• Influence of sensible heat from plasma
• Influence of steam feeding rates
CFD simulation of PGM air gasification
• Validation of the CFD model
• Temperature distributions and gas compositions in the reactor;
• Influence of plasma power
• Influence of ER
CFD study of PGM with steam addition
• Influence of ER in with steam addition
• Influence of steam feeding rates
• Influence of plasma power
Performance analysis and optimizing of the PGM reactor • Effect of single operation parameters
• Interaction between operation parameters
• Delimitation of possible operation condition
• Optimal operation conditions
In Supplement I, results from test runs of a demonstration industrial scale PGM reactor are
shown and analyzed. The temperature distributions and syngas compositions are demonstrated in
PGM both without and with steam injection. The energy efficiency of the PGM reactor is then
analyzed.
In Supplement II, a 0-dimentional model for PGM process is introduced. Cases for air PGM and
air&steam PGM are simulated. Attentions are paid to the syngas composition and energy
efficiency of PGM process. For air PGM, there exists a lower limit of air/MSW mass ratio for
100% conversion of MSW. When the air/MSW mass ratio exceeds the limitation, the syngas
LHV will descend by dilution of CO2 and N2. For air&steam PGM, high temperature steam as
gasification agent can reduce the limitation of air/MSW mass ratio, so further enhance the
syngas quality.
8
In Supplement III, a 2-dimentional CFD model is developed to simulate the PGM process using
Eulerian-Eulerian multiphase approach. The model considers the main chemical and physical
processes, such as drying, pyrolysis, homogeneous reactions, heterogeneous char reactions, and
melting of the inorganic components of MSW. The accuracy of this model is validated by the
experimental data demonstrated in Supplement I. Then, the characteristics of air PGM, such as
temperature distribution, syngas composition, tar yield, and energy conversion ratio at the
proposed condition are discussed.
In Supplement IV, the CFD model is further improved with more complex steam reaction
mechanisms. It is used to study the effects of steam addition in the PGM process. . It is found
that injection of high temperature steam is important for increasing the cold gas efficiency and
syngas lower-heating-value. The effect of steam injection is affected by steam feeding rate, air
feeding rate and plasma power. Based on the simulated results, an optimal condition is
suggested for air and steam gasification in the PGM reactor is given.
In Supplement V, optimizing of the operating conditions of PGM process is performed. Effects
of single operation parameters are analyzed. Then, the interactions between operation
parameters are discussed. Based on the above discussions, the possible operation condition of
PGM is delimitated. The optimal points aiming at obtaining maximum syngas LHV and
system CGE are given.
9
2. Literature review
As one of the promising MSW disposal methods, gasification has attracted more and more
attentions. MSW gasification has been frequently studied both experimental and numerical. The
main objective of this thesis is to develop mathematical models for fixed-bed plasma gasification
and melting of MSW. Therefore, at the first part of this literature review, the previous progresses
related to MSW gasification, especially plasma gasification and melting are introduced. Then,
attentions are paid to relevant literatures on mathematical modeling of gasification, as well as
relevant chemical and physical processes such as drying, pyrolysis, homogeneous reactions,
heterogeneous char reactions and melting of the inorganic components of MSW.
2.1 Experimental studies related to plasma gasification and melting
of MSW
2.1.1 MSW gasification
Experimental studies on gasification of both individual MSW components (such as food
waste [13-15], paper [16-17], cardboard [17-18], wood [19-21], and plastics [22-25]) and real
MSW samples [26-30] have been carried out by different researchers.
Ahmed and Gupta [13, 16-18] studied the gasification characteristics of various organic
components of MSW. Compared to pyrolysis, gasification was found to give better results in
terms of increased material destruction, and increased yields of combustible gases due to char
gasification. However, the time required for gasification was more as compared to pyrolysis.
Steam gasification provided higher energy efficiency and syngas LHV than air gasification.
The gasification temperature had a positive effect on both gasification speed and syngas yield.
It was also found that the inorganic constituents in food waste had a catalytic effect for
gasification.
Maitri et al. [27] performed MSW air gasification in a spout-fluid bed reactor. The result
showed decreasing trends of syngas higher-heating-value (HHV) and tar yield with increasing
primary air equivalence ratio (ER). The tar content in syngas was further reduced when
secondary air was supplied in the freeboard due to an increase in temperature. It was also
10
found that recirculation of carryover had a positive effect on both syngas HHV and
gasification efficiency.
Pinto et al. [28] studied gasification of mixtures of biomass and plastic wastes. They showed
that addition of plastics, especially the polyethylene (PE), clearly favored the release of
hydrogen and the decrease in CO content. The productions of light hydrocarbons were also
favored by plastics addition. The authors suggested that the steam/waste mass ratio should be
higher than 0.6 to ensure complete gasification. It was also suggested that gasification process
was strongly dependent on run temperature. A gasification temperature up to 900°C helped
the increase of hydrogen formation and reduced tars and hydrocarbons through thermal
cracking.
Dalai et al. [29] reported their experiment results on steam gasification of refuse derived fuels
(RDF) in a fixed-bed gasifier. In their paper, they confirmed the positive effects of
temperature on gasification speed. However, the gasification temperature was found to have a
negative effect on syngas LHV. The optimum gasification temperature for CO and H2
production was found to be 725 °C. The steam/waste mass ratio also showed a notable effect
on syngas LHV, and a ratio of 2 was suggested to be optimum in terms of syngas yield at 725
°C. The flow rate of carrier gas did not show any significant effect on products yield or their
distributions.
Anna et al. [30] studied the effect of feeding steam on the characters of waste gasification
with preheated gasification agent. The results confirmed that injection of steam has positive
effect on syngas calorific value. Meanwhile, a decrease of the total amount of detected tar in
response to steam addition is found. It was believed that the decrease in tar content was
attributed to steam reforming.
2.1.2 Gasification and melting
During the thermal treatment processes of MSW, the inorganic components may turn to fly
and bottom ash. The ash contains significant concentrations of heavy metals such as lead,
chromium, copper, zinc [31-32], as well as organic pollutants such as dioxins. High
temperature melting, which is also known as vitrification, is one of the most promising
solutions of ash problems.
11
Park et al. [33] reported the vitrification of fly ash along with the properties of the glasses and
leaching characteristics of heavy metal ions. It was pointed out that the produced glasses
showed Vickers hardness of 4000–5000MPa, bending strength of 60–90MPa and indentation
fracture toughness of about 0.9MPam1/2. Meanwhile, the glasses showed the excellent
resistance against leaching of heavy metal ions with Cd2+< 0.04 ppm, Cr3+< 0.02 ppm, Cu2+<
0.04 ppm and Pb2+< 0.2 ppm.
Jung et al. [34] investigated the behavior of metals in ash melting and gasification-melting of
municipal solid waste. Eight ash-melting and three MSW gasification-melting facilities with a
variety of melting processes and feedstock were selected in their study. The results showed
that the distribution ratio of metals could be predicted by the boiling point of each metal.
Metals with high boiling temperature were deposited to slag, while metals with low boiling
temperature might evaporate and exist in fly ash. The chlorine content in feedstock affected
the volatility of Cu and Pb by the formation of highly volatile chlorides. The volatility of Zn
was decreased in an oxidizing atmosphere by forming a non-volatile oxide compound.
Xiao et al. [35] studied the gasification and melting behavior of MSW. They found that the
combination of fluidized –bed gasification and swirl-melting produced a syngas with high
LHV value. Meanwhile, almost all the dioxins were decomposed, and most of the heavy
metals could be solidified. The solidification ratios of Ni, Cd, Cr, Cu, Pb and Zn were
respectively 95%, 48%, 75%, 54%, 43% and 83% approximately.
Calaminus et al. [36] developed a fixed-bed gasification and melting process called the
Thermoselect High Temperature Recycling process. This process combined slow degassing
with fixed-bed oxygen blown gasification and melting in a closed loop system. An industrial
scale demonstration plant of this process was been set up in northern Italy, and long term
operation of the plant had been performed.
The state-of-the-art of the MSW thermal treatment residue melting was summarized by Sakai
[37].
2.1.3 Application of Plasma in gasification
The application of thermal plasma in gasification has been an interesting topic since the end
of the 20th century.
12
Ivan et al. [38] studied the influence of different operation factors on solid fuel steam plasma
gasification. It was suggested that the ash content in feedstock, as well as the ER, strongly
influence the performance of the plasma gasification. There exists a temperature limit over
which the process does not proceed. They also suggested that the plasma gasification is also
affected by other factors.
Galvita et al. [39] reported their results on coal gasification in steam and air atmosphere under
arc plasma conditions. It was found that for Podmoskovnyi brown coal, Kuuchekinski
bituminous coal and Canadian petro coke, the gasification degree to synthesis gas were
92.3%, 95.8 and 78.6% correspondingly in the plasma gasification process. The amount of
produced syngas was 30–40% higher in steam than in air gasification.
Kalinenko et al. [40] studied plasma-steam gasification of brown coals in an entrained-flow
reactor. The results showed that the degree of carbon gasification was 90.5-95%. Meanwhile,
the level of sulfur conversion into the gas phase was 94.3-96. 7%. They also found that
plasma steam gasification produced a high quality syngas, in which the concentration of CO
and H2 amounted to 84.7-85.7%.
Moustakas et al. [41-42] carried out a series of experiments in a demonstration plasma
gasification/vitrification reactor. Their aim was to examine the efficiency of plasma
gasification/ vitrification in dealing with hazardous waste. It was found that the plasma
gasification/vitrification had advantages in treatment of various waste, especially waste
having major organic part. Plasma gasification/vitrification resulted in significant volume
waste reduction, ranging from about 5:1 for ash input to maximum 50:1 for solid waste. They
also pointed out that the cost of the plasma system was high, so more work should be done on
the design of the plasma gasification/ vitrification system.
Despite the fact that these works are good references for the study on plasma gasification and
melting, rare work has been found on detailed performance study, or the process optimization
of a plasma gasification and melting process. The available experimental data on plasma
gasification melting, especially industrial-scale operational data, is very limited. This situation
serious has hindered the understanding and application of plasma gasification melting
technology.
13
2.2 Development of gasification models
Gasification model can be divided into different categories. Considering the time dependence,
gasification models can be divided into kinetic models and kinetic free models. Considering
the geometry dependent, gasification can be divided into zero-dimensional models, one-
dimensional models, two-dimensional model and three dimensional models.
The earliest and simplest gasification models are equilibrium models. The equilibrium models
are zero-dimensional kinetic free models, in which the gasification products are calculated by
equilibrium assumption. The equilibrium model was used by Manfrida et al. [43] for coal
gasification simulation and by Ruggiero et al. [44] and Zainal et al. [45] for biomass
gasification simulation. The drawback of the equilibrium models is that the accuracies of
these models are often inadequate since reality usually deviates from equilibrium predictions.
In order to overcome the drawbacks of equilibrium models, the stratified models are
developed. In stratified models, the gasification process is divided into several zones such as
drying, pyrolysis, char gasification and combustion. In each zone, different chemical reactions
are considered. Heat and mass balances are also simulated in every individual zone. The
stratified model was used by Vittorio et al. [46] to simulate updraft coal gasification. The
results showed that the accuracy of the stratified model was in the satisfactory level for
analyzing the gasification performance at different operation conditions.
Neither of above models considers reaction kinetics and transport phenomena during
gasification process, so they are kinetic free models. If reaction kinetics is considered in
gasification model, the variation of syngas composition and detail physical properties with
time can be simulated. Zero-dimensional kinetic models are used by Manurung et al. [47] and
Blasi et al. [48] to simulate downdraft gasifiers. At KTH, a zero-dimensional kinetic model
was used by Yang et al. [49] to simulate fixed-bed gasification with high preheated air.
In zero dimensional gasification models, the influence of reactor geometry cannot be
reflected. From 1990s, studies on one-dimensional gasification became popular [50-51]. In
these models, the vertical movements of both feedstock and gas are considered. The variation
of physical and chemical properties of both feedstock and gases along the reactor height can
be simulated with these models.
14
In recent years, CFD technology has been used as a powerful tool for the simulation of
gasification processes. The Euler-Lagrange discrete phase approach and Euler-Euler
multiphase approach were successfully used for entrained-flow gasification [52-54] and
fluidized-bed gasification [55-57].
For fixed-bed gasification, Rogel and Aguillon [58] developed a 1-D + 2-D method to
simulate the performance of a biomass stratified downdraft gasifier. In their model, the mass
and energy balances within particles were written for a one dimensional system, and the mass,
momentum and energy balances of gas phase was written for a two-dimensional system.
2.3 Reaction rates
Generally, the reactions in MSW gasification can be classified into four groups: drying,
pyrolysis, heterogeneous char reactions and homogeneous reactions.
2.3.1 Drying
Drying is the first process to take place during the gasification of feedstock. Despite its
seemly simplicity, drying of feedstock is a complex combination of three steps: evaporation
of free water, desorption and evaporation of absorbed water, and separation of chemically
bound water [59].
The global reaction rate of drying has been assumed to be diffusion limited [48, 60] or
kinetically controlled [61-63]. For fixed-bed gasification, most of the researchers adopted the
diffusion limited assumption.
2.3.2 Pyrolysis
Pyrolysis is the thermal decomposition of solid fuels in the absence of oxidizers. Due to the
complexity in both reaction paths and products generated, the detail kinetics of pyrolysis is
still unclear. Various empirical global models have been developed to describe the pyrolysis.
Generally, these models can be classified into three categories: one step pyrolysis models,
competing parallel pyrolysis models and pyrolysis models with secondary tar reactions.
15
The one step pyrolysis models are simplest pyrolysis model in which the pyrolysis reaction is
expressed by a single global reaction:
CharTarGasFeedstock r γβα ++→ (2-1)
The kinetic rate r is expressed using an Arrhenius expression. The one step pyrolysis models
are common used by researchers due to its simplification [49, 64].
The competing parallel pyrolysis models assumed that feedstock decomposes directly into
each product i by a series of independent reactions:
ir oductsFeedstock i Pr→ (2-2)
where ir is the kinetic rate of the reaction i . The competing parallel pyrolysis models are
available for gasification of fine particles where the secondary tar cracking is not significant
due to very limited residence time in high temperature area. For fixed-bed gasification of
MSW, it is not a good choice.
In the pyrolysis models with secondary tar reactions, the whole pyrolysis is divided into two
steps: At first, feedstock decomposes into primary tar, char and gases, and then the primary
tar decomposes into secondary tar and secondary gases by thermal cracking [65]. The
decomposition of feedstock can use either one step models or competing parallel models.
2.3.3 Heterogeneous char reactions
The word “char” indicates to the solid residual from pyrolysis, which is mainly a mixture of
carbon and ash.
The heterogeneous reactions involve two distinct phases. Thus, the mass transfer around the
feedstock particle has to be considered. Two different models are usually used to describe the
mass diffusion at the particle surface: the shrinking-core model and the ash-segregated model.
In the shrinking-core model, the reaction core is assumed to be surrounded by a shell of inert
material (the remaining ash in the reacted area). Therefore, the gaseous reactants have to
diffuse through the ash layer before reaching the reaction core. In the ash-segregated model,
16
as soon as the residual ash forms at the particle surface, it detaches and disintegrates into
small particles. As a result, the reaction core is always exposed to the gas environment [66].
The ash-segregated model is only suitable for feedstock with low ash content. For MSW
gasification, since the ash content is 10-20wt%, depending on the MSW source, the shrinking-
core model is more appropriate.
2.3.4 Homogeneous reactions
During the gasification process, reactive gas species are produced. Gas phase reactions occur
among these species (such as water-gas shift reaction, steam reforming of light hydrocarbons
and combustion of combustible species). The rates of these reactions should be calculated by
considering both the kinetic and turbulent mixing rates.
17
3. Methodology
3.1 Test facility
3.1.1 The demonstration plant
An industrial-scale PGM demonstration plant is located in Yblin Israel. A series of trial runs
were performed in this plant to investigate the characteristics of the PGM process.
The demonstration plant was constructed in 2007. The designed capacity of the plant is 20
tons of MSW per day. The process flow sheet is shown in Figure 4. MSW is fed into the
reactor through airtight feeding chambers placed at the upper part of the plasma chemical
reactor, where gasification reactions occur. Syngas produced from gasification flows into the
afterburner and is combusted there. The hot flue gas from combustion is sent to the boiler to
produce steam, which drives a steam turbine connected to an electrical generator. The
generated electricity, besides providing power for the plasma torches and the rest of the
system, can be sold to outside users. The fly ash is removed from the flue gas in the scrubber-
evaporator. SOx is absorbed in the reactor absorber and removed using a bag filter. The solid
residue from gasification is melted by the plasma jet and collected by the slag collectors.
Figure 4. Illustration of the flow sheet of the demonstration plant [68]
18
3.1.2 The PGM reactor
Figure 5. The scheme of the PGM reactor in the demonstration plant
The core of the PGM plant is the plasma chemical reactor, which is a typical fixed-bed
updraft gasification reactor. The scheme of the reactor is shown in Figure 5. Generally, the
reactor is a fixed-bed counter current gasification shaft, with a plasma melting chamber
located at the bottom of the shaft. Plasma torches are placed at the ceiling of the melting
chamber. Primary air flows into the melting chamber through the torches, where it is ionized
so forming plasma jets which extend beyond the tip of the torches. The temperature of the
plasma jets may reach up to 6000 K. The plasma jets supply the necessary heat to melt the
inorganic components of the feedstock, which reached the bottom of the reactor. Secondary
air nozzles are placed around plasma nozzles. Secondary air at room temperature is injected
through secondary air nozzles. The flow rate of secondary air is adjustable thus the feeding
rate of total air can be controlled. High temperature steam at 1000°C is fed into the reactor
from steam nozzles placed at the side wall of the melting chamber. An airtight feeding pipe is
19
placed at the top of the reaction shaft. MSW is fed into the reactor intermittently from the
shaft top every half an hour. The total height of the reaction shaft is 7.02m, and the height of
the fixed-bed is 6.11 m.
3.1.3 Measurement methods
To measure the temperature distributions inside the plasma chemical reactor, thermocouples
are placed both along the gasifier shaft and in the syngas conduit. The thermocouple positions
depend on their height above the reactor bottom, H. If H < 1.0 the thermocouples are placed
in the reactor wall, behind the refractory layer, to prevent damage to the thermocouples at
high temperature. If 1.0 ≤ H ≤ 2.0, the thermocouples are placed both behind the refractory
layer and inside the reactor. For H ≤ 2.0, thermocouples are placed inside the reactor. To
obtain the actual temperature inside the reactor, temperature compensation must be made for
the thermocouples placed behind the refractory layer. According to the heat conducting law,
the heat flux through the reactor wall can be written as:
2
122
1
011
)()(x
TTxTTq
∆−
=∆−
= λλ (3-1)
where λ1 is the average thermo conductivity of the reactor wall outside the refractory layer, λ2
is the thermo conductivity of the refractory layer, T0, T1, and T2 are temperatures at the outer
wall surface, behind the refractory layer and inside the reactor, respectively, Δx1 is the
thickness of reactor wall outside the refractory layer, and Δx2 is the thickness of the refractory
layer. We assume that the wall material of both the refractory layer and the reactor wall
outside the refractory layer are uniform. The ratio of λ1 and λ2 can be calculated from the
measured temperature at 1.0 ≤ H ≤ 2.0. The temperature inside the reactor at H < 1.0 range
can then be calculated as:
′+′∆
′−′′∆=′
1
12
01212
)(T
x
TTxT
λ
λ (3-2)
20
3.1.4 Feedstock
The feedstock used by the PGM gasifier is MSW collected in Israel. The proximate and
ultimate analyses of the MSW are given in Table 2. In the reality, the size of MSW particles
varies from 1-100 mm.
Table 2. MSW proximate and ultimate analyses Proximate analysis (in dry basis except moisture)
Moisture 20.0 %
Fixed carbon 10.7 %
Volatile 77.6 %
Ash 11.7 %
Ultimate analysis (in dry basis)
Carbon 50.5%
Hydrogen 5.6%
Oxygen 30.7%
Nitrogen 1.1%
Chlorine <0.1%
Sulphur 0.3%
LHV of raw MSW (MJ/kg) 12.89
3.1.5 Test procedure
In the trial runs, two groups of tests were carried out. The first were with air gasification of
MSW (Cases 1 and 2), and the second were with air and steam gasification (Cases 3–6). The
feed rate of MSW was set at 600 kg/h during all runs. Trial runs were conducted with
different operating parameters, such as plasma power, secondary air feed rate and steam feed
rate, as shown in Table 3. Before each run, the reactor was preheated for 12 hours with plasma
air.
21
Table 3. Operation parameters for trial cases.
Case
Number
MSW
Flow Rate
(kg/h)
Plasma
Power
(KW)
Plasma
Air
(kg/h)
Air
Injection
(kg/h)
Steam
Injection
(kg/h)
Steam
Temperature
(℃)
1 600 240 120 0 0 1000
2 600 240 120 60 0 1000
3 600 240 120 60 70 1000
4 600 240 120 60 100 1000
5 600 240 120 35 70 1000
6 600 260 130 13 70 1000
3.2 Zero-dimensional kinetics-free model
In this work, a zero-dimensional kinetics-free model for fixed-bed plasma gasification and
melting process was developed using Aspen Plus. The model schematized the PGM process
into four different sections: drying, pyrolysis, char gasification and combustion, and plasma
melting. Moisture, volatiles, fixed-carbon and ash were removed from feedstock in these
sections, respectively. The simplified scheme of the PGM gasifier model is shown in Figure
6.
22
Plasma Melting
MSWSyngas and tar
Slag Plasma air
Steam Char Gasification and Combustion
Pyrolysis
Drying
Figure 6. Scheme of PGM gasification process
The following model assumptions are used in this work:
• The system is zero dimensional. Material properties like temperature (of gas phase and
solid phase), gas composition and solid composition in each zone is expressed by
“mean” values, which are calculated from the mass and energy balance.
• The flow of solid is from top to the end, while the gas flow is from the bottom to the
top. No reflux for each phase is allowed.
• The ash-free fuel is composed of C, H and O. The gas-phase species included in this
model are CO, H2, CO2, H2O, CH4, C2H4, O2, N2 and tars (including primary tar from
cellulosic group, primary tar from plastic group and secondary tar).
• The heat loss of each section is calculated from the measured temperature layout of
gasifier wall surface and the gasifier structure.
23
3.2.1 Drying
In the drying section, raw MSW is heating up by hot syngas and decomposed into dry MSW
and steam. The energy balance of heat exchanger is described as:
++= ∫∫∑ ∫
−
−
−
−
−
−
−− OHeva
T
TOHpOH
T
TdryMSWpdryMSW
i
T
Tipi MHdTCMdTCMdTCM
outsyngas
inMSW
outMSW
inMSW
insyngas
outsyngas
222/,,,
(3-3)
Considering the impact of heat gradient inside MSW particles, the outlet temperature of
drying process is set to 120 ºC. The heat capacity of MSW is calculated using the correlation
given by IGT [69].
3.2.2 Pyrolysis
Compared with coal, MSW have higher content of volatiles. For an updraft gasifier model, the
pyrolysis process is especially important because most of the gas and tar yield in this section
will join the gas produced in the char gasification section and be released from the outlet of
the gasifier without further reactions.
The heterogeneous MSW composition determines the complication of pyrolysis. According to
the pyrolysis characteristics, the composition of MSW can be divided into two main groups:
cellulosic fractions (Wood, paper, vegetation and cardboard) and plastics (PE, PP, PVC and
rubber). In this model, the pyrolysis of each group was simulated separately. A two-step
pyrolysis model [65] was applied to both groups: feedstock decomposes into primary tar, char
and primary gases in the primary pyrolysis. Then, primary tar decomposes into secondary tar
and secondary gases by thermal cracking.
The primary pyrolysis reactions of both groups are written as:
CAshTarGasspeciesCellulosic pricelpricel δγβα +++→ ,, (3-4)
CAshTarGasspeciesPlastic priplapripla δγβα +′+′+′→ ,, (3-5)
24
The yields of the primary pyrolysis products, including the composition of produced gases
and tars are taken from literatures [70-71]. To simplify the model, all light hydrocarbons
except CH4 are considered as C2H4.
The cracking reactions of primary tars are written as:
sec,sec, celpricel GasTarTar ζε +→ (3-6)
sec,sec, plapripla GasTarTar ζε ′+′→ (3-7)
The yield of primary tar cracking of the cellulosic group is taken from Hla [70]. No literature
data is found for the secondary pyrolysis of plastic mixture, so the yield of primary tar
cracking of the plastic group is calculated from elementary balance. The composition of
secondary tar is assumed to be benzene.
It has been proved that for a fixed-bed gasifier, the tar production is sensitive to the pyrolysis
temperature. In this model, the extent of primary tar cracking is controlled by pyrolysis
temperature [72]:
))(exp( 0TTAY pyr −−= (3-8)
where T0=500°C. The constant A varies for different feedstock, and can be calculated from
test results. The Combustion values of MSW and tars are calculated based on their elementary
compositions, using the empirical correlation given by Boie [73].
3.2.3 Char combustion&gasification
Char coming from the pyrolysis zone will meet and react with gasification agents (H2O and
O2) in the gasification and combustion section. Lots of homogeneous and heterogeneous
reactions are involved in this process. Due to the high temperature in the char gasification and
combustion section, chemical equilibrium is assumed in this section, and the Gibbs free
energy theory is applied in this section.
In the Gibbs theory, the second law of thermodynamics can be expressed as:
( ) 0,, ≤mPTdG (3-9)
25
It states that the Gibbs function always decreases for a spontaneous, isothermal, isobaric
change of a fixed-mass system in the absence of all work effects except boundary work. This
principle allows us to calculate the equilibrium composition of a mixture at a given
temperature and pressure.
It can be expressed as:
( )[ ]∑∑ +== 00,, /ln ppRTggNG iTiTiimix (3-10)
where: iN is the number of moles of the ith species, Tig , is the Gibbs function of the pure
species. The superscript 0 means properties at standard pressure.
For fixed temperature and pressure, the equilibrium condition becomes
0=mixdG (3-11)
3.2.4 Melting
The inorganic components (ash) of the MSW coming from the gasification and combustion
zone were melted by high temperature plasma air in the plasma melting section.
The composition of the inorganic components is assumed according to the original
composition of the MSW. Based on the assumed composition, the heat capacity of the
inorganics is calculated as following:
∑=
=n
iipiashp CC
1,, ω (3-12)
The melting latent heat of the inorganics is calculated similarly to that of the heat capacity.
The heat loss of the plasma melting process is set to 30% of the total plasma energy, which is
summarized from the tested temperature distribution inside the melting chamber.
26
3.3 Two-dimensional CFD model
In this model the Eulerian multiphase approach was applied. The conservation equations of
mass, momentum and energy are solved for both gas and solid phases. Mass, momentum and
energy exchange is allowed between phases. The scheme of the model was shown in Figure 7.
Figure 7. Scheme of the CFD model
3.3.1 Conservation equations
Gas Phase
The Eulerian conservation equations for species mass, momentum and energy are solved for
the gas phase. The equations are written as follow [74]:
( ) ( ) iigiggigg SmvYYt
+=⋅∇+∂∂
ραρα
(3-13)
27
( ) ( ) ( ) sgsggssgggggggggggg vmvvKgpvvvt
+−++⋅∇+∇−=⋅∇+
∂∂ ραταραρα (3-14)
( ) ( ) sgsgsggggggggggggg hmQSqvtphvh
t
+++⋅∇−∇+
∂∂
−=⋅∇+∂∂ :ταραρα (3-15)
Solid Phase
( ) ( ) jjsjssjss SmvYYt
+=⋅∇+∂∂
ραρα (3-16)
The momentum equation of solid phase is written as:
( ) ( ) sgsgsssssssssss vmgpvvvt
−+⋅∇+−∇=⋅∇+
∂∂ ρατραρα (3-17)
The energy equation of the solid phase is written as:
( ) ( ) sgsggsssssssssssss hmQSqvtphvh
t
−++⋅∇−∇+
∂∂
−=⋅∇+∂∂ :ταραρα (3-18)
where sp and sτ denote the solid pressure and shear stress, which are defined to express the
normal and the shear stress parts of solid-phase stress. The solid-phase stress is a function of
solid volume fraction. At fixed-bed condition, the value of sp∇ , which is several orders of
magnitude larger than the fluid-solid stress, becomes the main driving force of granular flow
[75-76]. In other words, the influence of the fluid-solid stress on solid motion can be ignored
(the detailed numerical expressions of the solid-phase stress and fluid-solid stress are
introduced in the next section). This idea was used by Johnson and Jackson [77] to describe
non-reaction shearing granular flow. In the present work, the idea is also adopted so that the
fluid-solid stress term is disregarded in the solid phase momentum equations. This
simplification is very helpful for the convergence of the solid momentum equation since it
largely prevents the solution of interphase non-linear terms, which is the main cause of non-
convergence for Euler-Euler approach.
The energy equation of the solid phase is written as:
( ) ( ) sgsggsssssssssssss hmQSqvtphvh
t
−++⋅∇−∇+
∂∂
−=⋅∇+∂∂ :ταραρα (3-19)
28
Gas-solid stress
The gas-solid stress is ignored for the solid phase. However, it is considered in the momentum
equation of the gas phase. The gas-solid stress is simulated using the Ergun equation [78].
The interphase momentum exchange coefficient sgK is written as:
( )s
gssg
sg
ggssg d
vvd
K
−+
−=
αρ
αµαα
756.11
150 2 (3-20)
Solid phase stress
The solid phase stress is composed of two parts: the normal stress part and the shear stress
part. For fixed-bed gasification, the flow of solid phase should be treated as a plastic flow
[79]. The normal stress part is expressed by solid pressure sp [80]:
∗= pp ss α (3-21)
∗p is expressed by an empirical power law:
( )nggAp ∗∗ −= αα (3-22)
where ∗gα is the gas volume fraction at minimum fluidization. Empirical values of 2510=A
(Pa) and 10=n are used.
For the shear stress part, only the frictional viscosity is considered. Since the flow of solid
phase is dense flow, where the solid volume fraction for the solid phase is near the packing
limit, the Schaeffer’s formulation [81] of frictional viscosity is applied:
D
ss I
p
22sinφµ = (3-23)
Interphase heat transfer
The intensity of heat exchange between the solid and gas phases is assumed to be a function
of the temperature difference between solid and gas phase:
29
( )gssggssg TTkQQ −=−= (3-24)
The heat transfer coefficient is written as:
2
6
s
sgsgsg d
Nuk
αακ= (3-25)
Here sNu is the Nusselt number correlated by Gunn [82]:
( )( ) ( ) 33.07.0233.02.02 PrRe2.14.233.1PrRe7.015107 gsgggsggsNu αααα +−+++−= (3-26)
3.3.2 Reaction model
3.3.2.1 Drying
In this model, a drying model which is popular used in fixed-bed combustion or gasification
of MSW and biomass [83-87] is applied:
( )OHmoimv CCkAr21 −=
when CTs
100<
(3-27)
or
evpsg HQr /1 = when CTs100≥
(3-28)
The Mass transfer coefficient mk is calculated according to the Sherwood number [88-89]:
dShDkm = (3-29)
3.3.2.2 Pyrolysis
In this work, a two-step pyrolysis model is applied. The pyrolysis of MSW is divided into two
steps: at first, MSW decomposes into char, gas and primary tar. Then the primary tar
decomposes into gas and secondary tar by thermal cracking [65].
ChartarimaryGasMSW γβα ++→ Pr (3-30)
30
tarSecondaryGastarimary +→Pr (3-31)
It has been confirmed by experiments that the two-step pyrolysis model can correctly predict
the pyrolysis yields, especially tar yields at various conditions [90-91]. It is very appropriate
for modeling pyrolysis in the updraft fixed-bed gasification since the tar problem is the most
significant in this constitution.
Table 4. Kinetics data for primary and secondary pyrolysis Reaction Reaction rate (kmol m-3 s-1 Source )
Primary pyrolysis of cellulosic group
( ) 1
45 1060.1exp11020.3 v
sg T
r ρα
×−−×=
Chan et al. [93]
Primary pyrolysis of plastic group
( ) 2
6
1
,3,3 exp1 v
i s
iiig RT
EAYr ρα ∑
=
−−=
,
51,3
131,3 1034.2,103.9 ×=×= EA
Wu et al. [94]
52,3
122,3 1007.2,102.1 ×=×= EA
53,3
103,3 1084.1,103.6 ×=×= EA
54,3
104,3 1073.1,100.5 ×=×= EA
55,3
105,3 1080.1,105.9 ×=×= EA
56,3
126,3 1064.1,105.1 ×=×= EA
Secondary pyrolysis 1
44 1012.1exp1055.9 tar
gg T
r ρα
×−×=
Boroson et al. [90]
MSW is the mixture of different species. Generally, most of the organic components of MSW
can be divided into two groups: the cellulosic group (wood, paper, cardboard and textile et al.)
and plastics group (polystyrene, polypropylene, polyethylene, and polyvinyl chloride et al.
[92]). The pyrolysis characters of these two groups are different. In this work, the differences
are considered by using individually pyrolysis kinetics for each group (shown in Table 4).
The interactions between species are not considered in this model.
3.3.2.3Homogeneous reactions
The following homogeneous reactions are considered in this model:
OHOH r222
12/1 →+ (3-32)
31
( ) OHyxCOOyxHC ryx 22 2/4/2 2 +→++ (3-33)
2232/1 COOCO r→+ (3-34)
xCOHyxOxHHC ryx ++→+ 22 )2/(4 (3-35)
2225 COHOHCO r +→←+ (3-36)
Chemical reaction rates of (3-32)-(3-36) are considered by choosing the minimum of the
kinetic rates and turbulent mixing rates:
( )tkrkk rrr ,min= , 51−=k (3-37)
Turbulent mixing rates are calculated using the eddy dissipation model:
=
jj
j
ii
itk Mv
YMvY
kr ,min0.4 ερ (3-38)
where i and j denote reactants of reaction k .
Kinetic rates of homogeneous reactions are shown in Table 5.
Table 5. Kinetic rates of homogeneous reactions Kinetic rate (kmol m-3 s-1 Source )
( ) 1.11.14.81 22
/3670exp1056.3 OHggr CCTr −×= α Varma et al. [95]
( ) 8.07.07.112 2
/24369exp100.1 OHCggr CCTryx
−×= α
Dryer et al. [96]
( ) 5.05.0113 22
/62700exp103.1 OOHCOggr CCCTr −×= α Howard et al. [97]
( ) OHHCgggr CCTTryx 2
/15083exp100.3 84 −×= α
Jones et al. [98]
( ) ( )
−−−=
0265.0/7914exp
/7250exp03.0 22
25HCOg
OHCOggr
CCTCCTr α
Grebenshchikova [99], Yoon et al. [100]
32
3.3.2.4 Heterogeneous char reactions
Heterogeneous char reactions involved in this model include the following overall reactions:
22 222
21
6 COCOOC r
++
+→
++
+γ
γγγ
γ (3-39)
227 HCOOHC r +→+ (3-40)
COCOC r 282 →+ (3-41)
4292 CHHC r→+ (3-42)
In reaction (3-39) the ratio of produced CO and CO2 is calculated as [67]:
( )gTCOCO /6420exp2500/ 2 −= (3-43)
The heterogeneous reaction rates are estimated using the unreacted shrinking core model, in
which the real reaction rate depends on surface film diffusion and reaction kinetics [101]. For
all heterogeneous char reactions, first order of reaction is assumed with respect to gaseous
reactants.
km
iv
iik
kk
AMv
r 111
+
=
ρ 96 −=k (3-44)
where i is the gaseous reactants of reaction k .The expressions of kk are shown in Table 6.
Table 6. expression of kk for heterogeneous reactions
kk (m s-1) Source
( )ss TTk /9000exp685.06 −= Evans et al. [102]
( )ss TTk /15600exp714.57 −= Yoon et al. [100]
( )ss TTk /26801exp1089.5 28 −×=
Hobbs et al. [60]
( )ss TTk /26801exp1042.3 39 −×= −
Hobbs et al. [60]
33
3.3.3 Geometry and boundary conditions
Geometry used in this work should be a reflection of the real 3D geometry so that it can
capture most flow characteristics of a real PGM process. The geometry of the trial gasifier is
approximately symmetrical in the width direction, so the longitudinal section of the gasifier
can be used as the 2D geometry. Since the void fraction of the hillock of concretionary slag is
very small, the hillock was excluded from the flow field. The total number of mesh cells is
10107. In the areas associated with plasma injections and secondary air injections, the mesh
was refined.
Figure 8. Geometry and mesh of the 2D model
In order to express the feeding of MSW, a mass source of solid phase was defined as at the
top of the fixed-bed. Since this is a steady model, the feeding of MSW is assumed to be
continuous. For all cases, the feeding rate is 600kg/h. Mass flow inlet conditions are defined
at the corresponding positions of plasma air and secondary air inlets. The outlet of syngas is
defined as a pressure outlet. The relative pressure at the outlet is set to -700 Pa, which is the
measured result for the pilot reactor. The melting of unreacted solid residual is represented by
34
a mass and energy sink of solid phase. Another energy sink is defined in the melting
chambers to express the heat transfer from gas phase to slag. The motion of slag after melting
is ignored in this model. The reactor walls are defined as no slip walls. An empirical
temperature distribution is defined at the reactor wall to calculate the heat loss from the wall.
3.3.4 Simulated cases
Two series of CFD simulations were carried out in this work:
• Series 1: air gasification of the fixed-bed plasma gasification;
• Series 2: air and steam gasification of the fixed-bed plasma gasification.
In order to increase the versatility of the results, dimensionless numbers are used to
characterize and classify the operation parameters of PGM process.
The amount of available air per kilogram of MSW is represented by the equivalence ratio
(ER), defined as:
( )( )stoicMSWair
MSWair
MMMMER
//
= (3-45)
where MSWair MM / is the air/MSW mass flow ratio in the real cases and ( )stoicMSWair MM / is
the air/MSW mass flow ratio for a stoichiometric combustion where the fuel is fully
combusted.
The amount of plasma energy per kilogram of MSW is expressed by dimensionless plasma
energy ratio (PER), which is defined as:
MSWMSW
pla
MLHVP
PER⋅
= (3-46)
where plaP is the heat power of plasma generators, MSWLHV is the low heating value of raw
MSW, and MSWM is the mass flow rate of raw MSW.
35
The amount of steam feeding rate is expressed by steam air mass ratio (SAMR, in0
dimensional simulations) and steam feedstock mass ratio (S/F, in 2-dimensional simulations):
air
OH
MM
SAMR
2= (3-47)
MSW
OH
MM
FS
2/ = (3-48)
In each series, a simulation of a typical case named the base case was performed. The results
of the base case were compared with the measured results from the demonstration reactor to
evaluate the availability of the model. Then, groups of simulations were performed to study
the influence of operation parameters. The detailed operating conditions of each series are
shown in Table 7 and Table 8.
Table 7. Operation conditions in series 1
Base case 1
(Case 2 in measurement) Group 1 Group 2
Operation parameters
Plasma power (kw) 240 240 200 - 280
Air feeding rate (kg/h) 180 130 - 230 180
Steam feeding rate (kg/h) 0 0 0
After dimensionless treatments
ER 0.06 0.043 – 0.077 0.06
S/F 0 0 0
PER 0.118 0.118 0.98 - 0.138
36
Table 8. Operation conditions in series 2
Base case 2
(Case 3 in measurement) Group 3 Group 4 Group 5
Operation parameters
Plasma power (kw) 240 240 240 220 - 260
Air feeding rate (kg/h) 180 180 130 - 230 180
Steam feeding rate (kg/h) 70 0 - 150 100 100
After dimensionless treatments
ER 0.06 0.06 0.043 – 0.077 0.06
S/F 0.117 0 - 0.250 0.167 0.167
PER 0.118 0.118 0.118 0.108 - 0.128
37
4. Results and discussion
4.1 Measured results
4.1.1 Syngas quality in air gasification
Two tests were performed without steam feeds (Cases 1 and 2). In both cases, the plasma
power is 240kw. In Case 1, the secondary air flow rate was set to zero. In Case 2, the feed rate
of secondary air was calculated by assuming that the total air feed rate equals the
stoichiometric demand for converting all fixed carbon in the feedstock into CO. The results
from both cases are presented in Figure 9 and Figure 10.
Figure 9. Syngas composition of Cases 1 and 2
38
Figure 10. Syngas characteristics of Cases 1 and 2
Both cases showed good results in terms of the lower heating value (LHV) of the product
syngas, here 6–7 MJ/Nm3. This was mainly due to the low ER in the PGM process (0.04 in
Case 1 and 0.06 in Case 2). A low ER ratio prevents the dilution of syngas with nitrogen from
air; a substoichiometric oxygen level suppresses the formation of CO2, which is the other
main noncombustible gas in syngas. Due to the concentration of combustible gases in the
syngas, the total gas yields here (0.67 and 1.06 Nm3/kg MSW in Cases 1 and 2, respectively)
were lower than that of conventional gasification. In both cases, the H2
Despite the common features of the two cases, some important differences in syngas
composition and yield were found. Firstly, a significant increase of gas yield was observed
when a higher ER was used. This increase was partly due to the N
/CO molar ratio was
approximately 1.5, which is somewhat higher than that of common gasification processes,
mainly due to the high contents of hydrogen and oxygen in the feedstock.
2 content of the secondary
air, which led to the decrease of LHV in Case 2, and partly due to the cracking of tar favored
by the higher temperature associated with increasing ER.
39
Figure 11. Measured temperature distributions of Cases 1 and 2
Figure 11 is the measured temperature distribution along the reactor chamber of Cases 1 and
2. It can be found that the temperature of Case 2 is 100-200 ºC higher than that of Case 1,
with an exception near the bottom. The bottom temperature of Case 2 is lower than that of
Case 1 due to the additional low temperature air injection. Another reason for the increase in
gas yield from Case 1 to Case 2 was likely the insufficient carbon conversion in Case 1, as
shown by chemical-equilibrium calculations. A low ER ratio in the PGM ensures high syngas
quality. However, when the ER is too low, as in Case 1, the gasification agent cannot supply
enough oxygen to convert char into CO or CO2. Insufficient carbon conversion is an adverse
condition for gasification as it reduces both the gas yield and energy efficiency. In Case 2, the
feeding of secondary air solved this problem. The increased O2
Figure 9
enhanced the CO content in
syngas, as shown in . An interesting result from the two cases is that although the
syngas yield changed significantly, the volume fraction of H2 in the syngas was relatively
unchanged. This indicates an increase of H2 production with increasing ER. The positive
effect on H2
[103]
yield in response to increased ER is in accordance with the results by Pinto et al.
and Anna et al. [30]. The increased H2
[90]
production with increased ER was likely the
result of favorable conditions for the secondary pyrolysis of primary tar. According to the
Boroson’s theory , pyrolysis can be divided into two steps: primary pyrolysis and
secondary pyrolysis. H2 is mainly produced from the secondary pyrolysis step, which is
sensitive to pyrolysis temperature. A higher temperature due to increasing ER thus favored
secondary pyrolysis, and more H2
[104]
was produced. Another effect of increasing temperature
with ER was the decrease in total light hydrocarbon (THC) content. The relationship between
pyrolysis temperature and THC yield has been reported by Anh et al. and Li et al. [105],
and the mechanism was explained by Anthony et al.[106].
40
4.1.2 Syngas quality in air and steam gasification
4.1.2.1 Influence of steam feed rate
Along with Case 2, experiments for Cases 3 and 4 were performed to investigate the influence
of the steam feed rate. The plasma and air settings of Cases 3 and 4 were the same as for Case
2 but with different steam feed rates (70 and 100 kg/h, respectively). The results from Cases 2,
3, and 4 are presented in Figure 12 and Figure 13.
Figure 12. Syngas compositions of Cases 2, 3 and 4
Figure 13. Syngas characteristics of Cases 2, 3 and 4
We found that adding high-temperature steam is favorable for the PGM process. The total gas
yield increased significantly, and the gas LHV also increased with steam feeding. Generally,
41
it is believed that the increase of gas yield with steam feeding is due to the water-gas shift
reaction:
222 HCOOHCO +=+ . (4-1)
This reaction undoubtedly played some role in the yield increase, especially with excess
steam. However, the increase of LHV in our cases cannot be explained solely by this reaction.
Examining the composition of the syngas, we found that as the steam feed rate was increased
from Case 2 to 4, the THC content increased significantly. The CO content, in contrast,
increased from Case 2 to Case 3 but decreased from Case 3 to Case 4. In all three cases, the
fluctuation in H2 [107] content was very small. Similar results were reported by Blasiak et al. ,
who studied the high-temperature air and steam gasification of biomass in an updraft fixed-
bed gasifier. A possible explanation for this phenomenon is the steam reforming of tar at high
temperature. The mechanism and kinetics of tar steam reforming have been reported by Li et
al. [108]. The global reaction can be written as follows:
yxnm HdCcHbCOOaHHC ++=+ 22 , (4-2)
where nmHC represents tar and yxHC the light hydrocarbons.
A strict restriction of the steam-reforming reaction is that it can only occur at high
temperature. It was pointed out by Jess et al. [109] that at a temperature of approximately
1,200 ºC, the steam reforming of tar can go to completion in less than 10 s. As we measured
during the tests, the global gasification temperature was in the range of 1,000–1,200 ºC.
Considering the scale of the reactor, it is very likely that there was a strong steam-reforming
reaction during the air and steam gasification process. Therefore, the steam reforming of tar
and the water-gas shift reaction together resulted in the increased syngas yield.
Comparing the results of Cases 3 and 4, a notable difference is that the H2
[108]
/CO molar ratio
increased greatly. This may be due to the promotion of the water-gas shift reaction by the
excess steam in Case 4. According to Li et al. , the priority of tar steam reforming is
higher than the water-gas shift reaction at high temperature due to the occurrence of the
following reaction:
42
22 22 HnmCOmCOHC nm +=+ (4-3)
When there is insufficient steam for reforming, as in Case 3, the water-gas shift reaction is, in
a sense, suppressed. When steam is supplied in excess, the water-gas shift reaction then
becomes much more intensive, resulting in a high H2
4.1.2.2 Influence of plasma power and ER
/CO molar ratio.
The experiments in Cases 5 and 6 were conducted to investigate the influence of plasma
power and ER in air and steam gasification. In both cases, the steam feed rate was 70 kg/h, the
same as in Case 3; the plasma power and ER were then varied. The results from Cases 5 and 6
are presented in Figure 12 and Figure 13 with Case 3 for comparison.
43
Figure 14. Syngas compositions of Cases 3, 5 and 6
Figure 15. Syngas characteristics of Cases 3, 5 and 6
The results for Cases 3 and 5 were similar. The slight difference in gas yields can be
explained by the reduced air feed and lower tar cracking and reforming related to the reduced
combustion due to the lower ER. However, some significant differences in syngas
composition were found when comparing Case 6 with Cases 3 and 5 taken together. The
overall increase of combustible gases, especially of H2, may be mainly due to the sensitivity
of tar cracking and reforming to temperature. In the PGM process, more than half of the
energy need for gasification is from the plasma torches. The increased plasma power in Case
6 led to a significant increase of gasification temperature, which prompted the cracking and
reforming of tar. The reforming reaction of light hydrocarbons may have also taken place, as
in Eq. (4-3), enhancing the yield of H2
22 )/2( HyxxCOOxHHC yx ++=+
.
. (4-4)
44
4.1.3 Energy efficiency
The cold-gas efficiency (CGE) is a standard criterion frequently quoted to express the energy
efficiency of a gasification process. For a PGM process, the definition of CGE was modified
to
%100×++⋅
⋅=
plasmasteamfeedstockfeedstock
syngassyngas
PPLHVmLHVm
η , (4-5)
where syngasm and feedstockm denote the mass flow rates of the syngas and the feedstock,
respectively, while syngasLHV and feedstockLHV are their lower heating values on a mass basis.
steamP denotes the power used to heat the steam, and plasmaP is the plasma power.
The combustion value of the MSW was calculated from an empirical expression given by
Boie et al. [110]:
PClNOHCHHV 654.9158.253.27422.83 +=+−+= . (4-6)
Figure 16. Cold-gas efficiency
The CGE results of all six cases are shown in Figure 16. Here, the CGE varies from 30% to
60%. The energy efficiency of air gasification is lower than that of air and steam gasification.
The CGE was the lowest for Case 1; Case 6 had the highest. For air gasification, increasing
ER was beneficial for increasing energy efficiency, whereas the influence of ER was not
obvious for air and steam gasification.
45
There are three main sources of energy loss in gasification: the chemical energy in the tar, the
sensible heat of the syngas, and the heat loss of the system. Because a PGM reactor is an
updraft fixed-bed reactor, the sensible heat of the syngas cannot be the main energy loss. The
normal system heat loss is approximately 2–5% of the total energy, so the main energy loss
for PGM should be chemical energy in the tar.
4.1.4 Slag properties
The inorganic components of MSW were melted to form a slag. The discharging of slag was
not continuous in the trial reactor. Instead, it was controlled by a valve placed at the exit of
the melting chamber. At most times, the valve is closed, and the volume of slag inside the
combustion chamber increases continually. When it reaches a certain level, the valve is
opened. The melted slag flows out of the melting chamber and into the slag collector. When
the collector is full, the valve is closed again, and the collector is transmitted into the open air
for cooling. When it exits the melting chamber, the slag is a glowing liquid. After four hours
of cooling, it becomes a black, vitreous solid. The appearance of the slag after cooling is
shown in Figure 17.
Figure 17. Appearance of slag after cooling
During these tests, the output of slag was approximately 25 kg/h. Due to the high density of
this slag (2,300 kg/m3), the volume ratio of slag to raw MSW is approximately 1:50. The
46
composition can vary with the feedstock, but the main contents should be SiO2
4.2 Results from zero-dimensional kinetics-free simulation
and CaO.
Undesirable materials such as heavy metals are locked in the slag, so that the slag is virtually
inert, meeting the most demanding TCLP Standards. This slag can be used as a good building
material.
4.2.1 Model validation
The comparison of measured results and predicted results of air and steam gasification in the
PGM demo-reactor are shown in Table 9. Results are shown in terms of syngas yield, syngas
LHV and the H2
Table 9. Comparison between measured and predicted results of air and steam gasification in the PGM reactor (dry basis)
/CO molar ratio. By comparing the predicted results with the measured
results, it was found that the results from modeling are in the acceptable ranges for analyzing
the character of PGM process.
Case number 3 4 5 6
Operation parameters
ER 0.060 0.060 0.052 0.048
PER 0.118 0.118 0.118 0.128
SAMR 0.389 0.556 0.452 0.490
Steam temperature (ºC) 1000 1000 1000 1000
Measured results
Syngas yield (Nm3/kg MSW) 1.36 1.38 1.26 1.29
Syngas LHV (MJ/Nm3) 8.23 8.43 8.24 8.70
H2/CO 1.24 1.53 1.45 1.70
Predicted results
Syngas yield (Nm3/kg MSW) 1.27 1.32 1.16 1.14
Syngas LHV (MJ/Nm3) 8.48 8.70 8.05 8.38
H2/CO 1.16 1.33 1.32 1.41
4.2.2 Effect of Plasma Power
The high-temperature plasma air injection is the most important significance of the PGM
process. It supplies heat for the melting of the inorganic component of MSW. After that, the
47
residual heat provides sensible heat to gasification. In this way, the power of plasma affects
the energy equilibrium of the whole gasification process, and directly influences the
temperature profile, syngas composition, tar yield and stability of the gasification process. A
serious of cases are simulated to investigate the effect of PER on gasification characters in
PGM process. In these cases, the values of ER and SAMR are set to 0.06 and 0.389
respectively, which are testified as “reasonable” values for PGM air and steam gasification by
previous test runs. The value of PER varies from 0.098 to 0.137.
The effect of PER on the average temperature in the gasification and pyrolysis zone is
illustrated in Figure 18. It was found that both gasification temperature and pyrolysis
temperature increase linear with PER. This is easy to understand since increasing PER
enhance the average temperature of feeding air, and increases the heat supply for gasification
and pyrolysis.
Figure 18. Effect of PER on gasification and pyrolysis temperature
Figure 19 shows the syngas composition, as well as tar yield with different PER. All the
gaseous species are shown in volume fractions on dry basis, and tar is shown by tar-MSW
mass ratio on dry basis. It was found that the volume fractions of all combustible gaseous
increase with PER, while the trends of CO2 and N2 are opposite. The increment of
combustible gases is mainly due to promoted tar cracking by increasing pyrolysis temperature.
At the same time, the total yields of incombustible gases like CO2 and N2 do not vary much.
Considering the increasing of combustible gases with PER, the decreasing trends of CO2 and
N2 volume fractions are understandable.
48
Figure 19. Effect of PER on syngas composition and tar yield
Figure 20 shows the effects of PER on syngas yield and syngas LHV, where both results are
calculated on dry basis. It was found that both the syngas yield and syngas LHV increase with
PER. This is not hard to understand since the increase of combustible gas yields by tar
cracking is profitable for both quantity and quality of syngas. When PER increase from 0.098
to 0.137, the syngas yield increases from 0.96 to 1.08 Nm3/kg MSW. At the same time the
syngas LHV varies from 7.32 to 9.31 MJ/Nm3
Figure 18
. It seems the positive effects of PER provides a
possible method for not only solving the tar problem, but also increase the syngas yield and
quality. However, it has to be noticed that beneficial effects of increasing PER is not
unlimited. As we can see in , the temperature inside the reactor also increases with
PER. A high PER value may leads to the formation of a high temperature zone in the
combustion and gasification section. Too high temperature challenges the thermostability of
the reactor wall. Furthermore, the low-melting-point components in the solid residual like
SiO2
Figure 18
may be melted in the gasification and combustion section if the temperature is too high.
The partial melting of solid residual will dramatically decrease the void fraction in the fixed-
bed, and leads to the occurrence of bridging. It can be found from that when PER=
0.26, the average temperature in the gasification section has reached 1330 ºC. This
temperature is already too high for an engineering application.
49
Figure 20. Effect of PER on total syngas yield and syngas LHV
4.2.3 Effect of ER
For a conventional gasifier, the energy needed for feedstock heating up, pyrolysis and char
gasification is mainly from the partial combustion of char. The equivalence ratio (ER) for
conventional gasifier should be around 0.3 to fulfill the need of energy. For PGM air and
steam gasification, heat can be supplied by plasma and high temperature steam, so the ER for
a PGM gasifier will be much lower (0.04-0.10). It is worthwhile to study the influence of ER
on the performance of a PGM gasifier.
Theoretically, the effects of ER on gasification process should be considered from two aspects.
On one side, higher ER provides more chemical heat by combustion. It is known that
increased heat supply is beneficial for both syngas yield and LHV value, so this effect of ER
is positive. On the other side, higher ER means more combustion in the reactor, which will
consume some combustible gases. Additionally, the increasing N2
A group of simulations with different ER was carried out to study the exact influence of ER
on PGM process. The values of SAMR and PER are set to 0.389 and 0.118, respectively. The
value of ER varies from 0.04 to 0.08.
introduced into the reactor
dilutes the content of combustible gases. From this point of view, the ER also has negative
effects on syngas production. The final influence of ER on PGM process should be a
combination of these two aspects.
Figure 21 shows the syngas composition and tar yield with different ER value. It was found
that when ER increases, the volume fractions of CH4, C2H4 and N2 increase, and the volume
50
fraction of H2 and tar yield decrease. The volume fraction of CO first increases and then
decreases. An opposite trend was found for CO2 volume fraction. The increase of CH4 and
C2H4 volume fractions can be understand as the result of positive effect of ER on tar cracking,
while the increase of N2 and decrease of H2 volume fractions are the results of negative
effects of ER. The variations of CO and CO2 volume fractions are affected by both aspects.
Figure 21. Effect of ER on syngas composition and tar yield
Figure 22 shows the variation of syngas LHV and system CGE with increasing ER. It was
found that the influence of ER on syngas LHV can be divided into two parts. When ER
increases from 0.04 to 0.07, the syngas LHV increases from 6.11 to 8.63MJ/Nm3. The
positive effect of ER is dominant. When ER increases from 0.07 to 0.08, the syngas LHV
keeps almost constant. It seems negative effect of ER starts to appear in this range, and
counterbalances the positive effect. For system CGE, however, the effect of increasing ER is
positive in all ER range.
51
Figure 22. Effect of ER on syngas LHV and system CGE
4.2.4 Effect of SAMR
The feeding of high temperature steam influences the PGM process from two aspects. Firstly,
steam is involved in chemical reactions such as water-gas reaction and water gas shift reaction.
In that case, it influences the chemical equilibrium of the PGM system. Secondly, the high
temperature steam changes the total mass and energy flow inside the reactor, and influence
the energy balance of the system.
As an example, the effect of different SAMR on syngas composition and tar yield for
ER=0.06 and PER=0.118 is shown in Figure 23. It was found that the most important effect of
increasing SAMR is the variation of H2, CO, and CO2 volume fractions. When the SARM
increases from 0 to 0.67, the volume fraction of H2 in syngas increases from 9.5% to 21.3%.
The volume fraction of CO2
222 COHOHCO +→←+
increases similarly from 12.4% to 20.2%, while the volume
fraction of CO decreases from 26.8% to 14.9%. The similar trends were also reported by other
researchers [111-117]. This phenomenon is the result of promoted water-gas shift reaction
( ) by increasing steam feeding rate. It was also found that the yield
of tar shown a slight decreasing trend when the SAMR increases. At the same time, the CH4
and C2H4
[118]
volume fractions increase slightly. It is believe that this phenomenon is the result of
steam preheating, which introduces extra energy to the PGM system, and increases the global
temperature of pyrolysis. It was reported by Lewis et al. that the critical steam
temperature for supporting energy supply in steam-only gasification process is above 1200 ºC.
In PGM process, due to the heat supply from plasma air and char combustion, the critical
steam temperature should be reduced. It was implied by the tar decreasing that the critical
52
steam temperature at the analyzed condition is lower than 1000 ºC. As a result of the extra
heat supply by high-temperature steam, the syngas yield and LHV increase slowly with
SAMR. It was also found that the effect of SAMR on syngas composition weakens with
increasing SAMR. The effect is most remarkable when SAMR varies from 0 to 0.1.
Figure 23. Effect of SAMR on syngas composition and tar yield
4.3 CFD results of air gasification
4.3.1 Analysis of the base case 1
4.3.1.1 Model validation
In order to evaluate the availability of the model, simulation data of temperature distribution
along the reaction shaft axis, as well as the syngas composition was compared with measure
data obtained from test runs of the base case 1. Results are shown in Figure 24 and Table 10.
The predicted and measured temperature profiles fit each other well. However, a slight
deviation was found at the reactor height 4.5 to 6.1 meter. Two reasons can explain this
deviation: firstly, the assumption of continuous feeding of MSW in the model leads to
imprecise temperature prediction near the fixed-bed top. Secondly, the uncertainty of
pyrolysis mechanisms also affects the accuracy of the temperature prediction in the pyrolysis
53
section. Considering the possible variation of MSW composition with time and area, the
disparities between predicted and measured results are acceptable.
0
1
2
3
4
5
6
7
8
0 500 1000 1500 2000 2500
T (K)
Rea
ctor
hei
ght y
(m)
Measured gas temperaturePredicted gas temperaturePredicted solid temperature
Figure 24. Temperature distribution along the shaft height of the base case 1
Table 10. Syngas yield and compositions for the base case 1
Syngas Predicted Measured Deviation
H2 Vol% (wet basis) 19.19 19.50 -0.016
CO Vol% (wet basis) 17.21 15.20 0.132
LHCs Vol% (wet basis) 7.22 6.90 0.046
Incombustible gases Vol% (wet basis) 56.68 58.40 -0.029
Syngas yield Nm3/kg MSW
(wet basis) 1.062 1.063 -0.001
54
Table 10 shows a comparison of predicted and measured syngas yields and composition for
the base case. It can be seen that the predicted yields and compositions of syngas are also in
good agreement with the measurements despite of an acceptable deviation related to CO. The
model slightly overestimates the CO volume fraction with the deviation equal to 0.132. It is
believed that this deviation is mainly caused by the overestimation of peak temperature due to
disregarding partial melting in the fixed-bed. Generally, the deviations of the predicted results
are in acceptable level for understanding the characteristics of PMG process.
4.3.1.2 Temperature profiles
The distribution of gas temperature in the base case is shown in Figure 25. It is found that
plasma air temperature reduces rapidly due to radiation and heat exchange with unmelted
inorganics. During this process, the plasma air also mixes with secondary air. The average gas
temperature at the gas-bed boundary is about 1800 K. When air flows into the fixed-bed, gas
temperature increases dramatically to around 2400 K due to char combustion. Then, gas
temperature rapidly decreases to around 1000K. This decrement can be explained by intense
heat exchange between phases and endothermic char gasification. Since the gasification agent
used in this case is air, the main char gasification here should be the boudouard reaction.
When gas temperature reaches 1000K, the temperature decreasing rate starts to slowdown
gradually. In this zone, the boudouard reaction generally stops. The heat transfer between gas
and solid phase also becomes slow due to the decrease of temperature differences (see Figure
24). At the reactor height 5.0-6.1 m, where pyrolysis and drying of MSW take place, gas
temperature starts to decrease visibly again from around 860K to 450K. After flowing out of
the fixed-bed area, no reaction or heat exchange happens for gases, so that the gas temperature
nearly keeps constant.
55
Figure 25. Gas temperature distribution (K) in the base case 1
4.3.1.3 Nonuniformity of temperature distributions in horizontal sections
In the PGM reactor, the temperature distribution in a horizontal section is not uniform. This
nonuniformity can be reflected from Figure 25.
Figure 26 shows detailed gas temperature distributions at different horizontal sections in the
base case. The nonuniformity of gas temperature is the most significant at the y=1.0 m section,
where the gas temperature varies from 1430 to 2080 K. The peak temperature in this section
appears at 0.40 < x < 0.55 m area, which is corresponding to the horizontal position of gas-
bed interface at the y=1.0 m section. The position of peak temperature denotes that char
combustion only occurs in a thin layer near the gas-bed interface. From engineering point of
view, high peak temperature should be prevented since it causes problems such as bridging
and damage of reactor wall. In order to prevent the problems caused by high peak temperature,
the intensity of char combustion have to be restrict the by controlling the ER value.
56
The temperature distribution in the y=2.0 m section shows a similar trend to the y=1.0 m
section. However, the difference between shaft axis temperature and peak temperature
dramatically decreases to about 150 K. It denotes that the nonuniformity of gas temperature
becomes weak with increasing height due to horizontal heat transfer. In the y=3.0 m and
y=4.0 m sections, the differences between axis temperature and peak temperature are not
visible. In all these four sections, the temperature decline near the reactor wall is caused by
heat loss from the reactor wall.
600
800
1000
1200
1400
1600
1800
2000
2200
0 0.1 0.2 0.3 0.4 0.5 0.6
Horizontal distance from the shaft axis x (m)
Tg (K
)
y=1.0 (m)y=1.5 (m)y=2.0 (m)y=4.0 (m)
Figure 26 Gas temperature distributions in different horizontal sections in the base case 1
57
4.3.1.4 Composition profiles
Figure 27 Syngas compositions of the base case 1, (a) molar fraction of CO, (b) molar fraction of
H2, (c) molar fraction of LHCs, (d) molar fraction of CO2, (e) molar fraction of H2O, (f) mass fraction of tar
Figure 27 (a) – (e) shows the volume fractions of CO, H2, LHCs, CO2 and H2O in the gas
phase. Since only air is used as gasification agent in this work, the water-gas reaction and
water-gas shift reaction are restrained. H2
Figure 27
can only be produced from the pyrolysis step. This
phenomenon is well presented in (a). A similar trend is also found for LHCs. CO is
generated from both char combustion and pyrolysis. As we can see in Figure 27 (c), the
volume fraction of CO reaches about 18% after heterogeneous char reactions. The volume
fraction of CO does not change much during pyrolysis. However, the yield of CO from
pyrolysis is still remarkable since the gas volume increases significantly during pyrolysis.
A significant of the PGM technology is that it produces a high quality syngas. As we can see
in Figure 27, the volume fractions of CO and H2 reaches around 20% at the syngas outlet, and
the volume fraction of LHCs is about 7%. The lower heating value (LHV) of the wet syngas
reaches about 6.79 MJ/Nm3 in the base case, which is a very high value for MSW gasification.
58
It is believed that the high LHV value is mainly due to low ER value in the PGM process,
which prevents the direct dilution of combustible gases from N2. Low ER also provides a
substoichiometric O2 environment, which suppress the formation of CO2
Tar yield is one of the main problems involved in the fixed-bed gasification, which reduces
the energy efficiency and causes blockage to the pipeline of syngas. The mass fraction of tar
in the gas phase is demonstrated in
.
Figure 27 (f). The tar mass fraction at the syngas outlet is
about 16.8%, which reflects a tar yield ratio of 0.193 kg/kg MSW.
4.3.2 Influence of ER
The ER is one of the most important operation parameters of gasification. It determines the
level of MSW partial combustion, and directly influences the temperature profile, syngas
composition and stability of the gasification process. The ER required for a typical PGM air
gasification varies from 0.05 to 0.10, while the ER for conventional gasification is about 0.3.
Few scientific works has been found on the characteristics of gasification in low ER
conditions. Studying the influence of ER on the performance of a PGM gasifier is of both
scientific and engineering values.
In this study, a group of cases with different ER values are simulated. The ER value varies
from 0.043 to 0.077. The PER values for all cases are set as 0.118.
4.3.2.1 Gas temperature distribution
Figure 28 shows the predicted gas temperature distributions at the shaft axis with different ER
values. It can be found that the gas temperature shows an increasing trend with ER. This
phenomenon is more significant in the lower part of the reaction shaft, where the pear
temperature increases from 1250 K to 2750 K. The temperature increase with ER is explained
by prompted char combustion due to increases O2 flow rate. According to chemical
equilibrium calculation, for 100% conversion of carbon in char, the ER value should be about
0.13, which is much higher than the ER values in the simulated cases. No doubt that the
increasing trend of gas temperature will continue if the ER keeps increasing. However, in
order to restrict the peak temperature under 2273 K, the ER value in PGM air gasification
59
should be controlled less than 0.067. This may results in insufficient combustion of char,
which leads to low energy efficiency.
Figure 28. Temperature distribution along the shaft height for different ER values
4.3.2.2 Syngas composition
Figure 29 shows the variation of syngas composition, as well as the tar-MSW mass ratio with
ER value. It is found that when ER increases, the volume fractions of H2, LHCs and CO2
decrease, and the volume fraction of CO increases. The increasing of CO volume fraction can
be explained by prompted char combustion with increasing ER, while the decreasing of H2
and LHCs volume fractions is explained by dilution of syngas by introduced N2. It is
interesting to find that the CO2 volume fraction also decrease when ER increases. This
phenomenon may be caused by increasing combustion temperature with ER, which prevents
the formation of CO2 during char combustion. Moreover, the tar-MSW mass ratio is also
increasing slightly with ER. This increasing is the result of increasing heating rate with ER in
the pyrolysis zone.
60
0
5
10
15
20
25
0.04 0.05 0.06 0.07 0.08
ER
Syng
as c
ompo
sitio
n (v
ol. %
,w
et b
asis
)
0.16
0.17
0.18
0.19
0.2
Tar-M
SW m
ass
ratio
CO H2 LHCs CO2 tar Figure 29. Predicted Temperature distributions for different ER
4.3.2.3 Energy conversion ratio
In order to quantify the energy conversion from MSW to syngas, the energy conversion ratio
(ECR) was defined and used:
%10022 ×+⋅
⋅+⋅+⋅=
plaMSWMSW
LHCLHCCOCOHH
PLHVMLHVMLHVMLHVM
ECR
(4-7)
The ECR is a very important process parameter which characterizes the combustion value of
the gas phase. It illustrates the variation of syngas composition during the gasification process,
and can be used as an index for the gas quality. The ECR value at the syngas outlet is the cold
gas efficiency (CGE).
61
Figure 30. ECR values along the shaft height for different ER values
Figure 30 shows the ECR value in the horizontal sections along the shaft height. In all cases,
the energy conversions mainly happen in two sections: char combustion and pyrolysis. It is
found that the ER has a positive effect on energy conversion in the char combustion section.
When ER varies from 0.043 to 0.077, the ECR will increase from 0.02 to about 0.06. No
doubt that this increasing is caused by prompted char combustion due to increasing O2. The
increasing of gas temperature with ER also has a positive effect on energy conversion since it
pushes reaction (20) to produce more CO rather than CO2
Volatiles take up to 77.6% of the total MSW mass. Most of the energy conversion happens in
the pyrolysis section, which is corresponding to the shaft height 4.5-5.5 m in the reactor. Only
a slight negative effect on energy conversion is found for ER. The proper reason is that the
enhanced heating rate caused by increasing ER, which results in an increasing tar yield. The
increasing of tar yield has been demonstrated in
.
Figure 29.
Since the influence of ER on ECR is more significant in the char combustion section, it is
possible to increase the system CGE by increasing ER. However, as discussed previously, the
increasing of ER is restricted by peak temperature. The ER value in PGM air gasification
should be lower than 0.067. A practical solution to this problem is to inject additional steam
into the reactor. The numerical study on air&steam gasification in PGM reactor will be
presented in the future.
62
4.3.3 Influence of PER
The most important significance of the PGM process is the high-temperature plasma air
injected from the bottom of the reactor. The high-temperature plasma flow supplies heat for
both gasification residual melting and reactions related to gasification process. The value of
PER may directly influence the temperature profile, syngas composition, tar yield and
stability of the gasification process. In this work, the influence of PER is investigated. The
value of PER varies from 0.098 to 0.138. The ER value for all cases is set as 0.060.
Figure 31. Temperature distributions along the shaft height for different PER
Figure 31 shows the predicted gas temperature distributions at the shaft axis with different
PER values. It was very interesting to find that the temperature distributions for all these cases
are similar. When PER increases from 0.098 to 0.138, the increment of peak temperature
along the shaft axis is less than 200K. A possible explanation for this phenomenon is the heat
loss in the melting chambers. Firstly, intense heat transfer exists between plasma air and slag
since the high-temperature and high-velocity plasma flow was directly injected into the slag
pool which exists at the bottom of the melting chamber. Secondly, the strong heat radiation
leads to large heat loss from the chamber wall. During the running of the pilot reactor, it is
found that heat loss from the chamber wall reaches about 30% of the total plasma power. It is
also found that the heat loss increases with PER value.
63
A thermodynamic calculation is done by authors to estimate the lower limit of PER to satisfy
the heat request for melting the inorganic components. The result shows that when the PER
value is larger than 0.09, the plasma flow can supply enough heat for the melting process.
From the view point of energy efficiency, the optimal PER value for PGM air gasification
should be about 0.09.
4.4 CFD results of air and steam gasification
4.4.1 Model validation
The simulated results were evaluated by comparison with measured results from the base case
2. The results are shown in Table 11. Generally, the predicted yield and composition of
syngas are in agreement with the measurements. However, a non-ignorable deviation exists
between predicted and measured CO. Meanwhile, an underestimation is found for CO2
Table 11. Measured and predicted syngas yield and main compositions of the base case 2
content. It is believed by authors that the deviation is mainly caused by overestimation of
fixed-bed peak temperature due to disregarding the melting in the fixed-bed. According to
previous researchers, melting of inorganic components in MSW starts when the solid
temperature reaches around 1800 K. The melting process is a highly endothermic, so that
further increase of solid temperature can be restrained. However, the possible melting in the
fixed-bed is not considered in the model, so the peak temperature may be overestimated. The
overestimation of solid temperature therefore influences the yield of char combustion, thus
overestimating the CO content in syngas. Despite the deviation between simulated and
measured results, the accuracy of this model is acceptable for analyzing the characters of air
and steam gasification in the PGM reactor.
Syngas Unit Predicted Measured Deviation
H Vol% (wet basis) 2 19.49 20.44 -0.046
CO Vol% (wet basis) 19.21 16.10 -0.193
LHCs (light hydrocarbons) Vol% (wet basis) 6.79 7.42 -0.085
CO Vol% (wet basis) 2 6.59 >10.0 -
Syngas yield Nm3 1.368 /kg MSW (wet basis) 1.359 -0.006
64
4.4.2 Effect of S/F
The gas temperature distribution at various S/F values when ER=0.06 and PER=0.118 is
shown in Figure 32. When S/F increases from 0 to 0.25, the temperature distribution along the
reaction shaft becomes more uniform. Meanwhile, the area of char reaction zone, where the
gas temperature is above 1000 K, is increasing. Significant advantages are obtained from
these variations: firstly, the uniformity of gas temperature prevents the formation of very high
temperature, which challenges the heat resistance of wall materials; secondly, the increase of
char reaction zone increases the reaction time of gasification agents with char. Another
advantage of increasing S/F is that increased steam feeding rate enhances the rate of water-
shift reaction, which is an important char gasification reaction.
Figure 32. Predicted gas temperature (K) distributions for different S/F values
In order to characterize the conversion of char, the char conversion efficiency Cη is defined as
the percentage of carbon in the MSW converted into gas species. The energy efficiency of
PGM is represented by cold gas efficiencyη .
The effects of S/F on Cη and η are indicated in Figure 33. It is found that steam injection has
a notable positive effect on both Cη and η . When steam is not injected, the value of Cη is only
about 23%, which is far from complete gasification of char. When S/F varies from 0 to 0.208,
the value of Cη increases dramatically from 22% to 96%. Further increase of S/F from 0.208
to 0.250, however has very limited effect on Cη . It is known that the incomplete conversion of
char is disadvantageous for gasification since it reduces the cold gas efficiencyη . It is found
65
that the variation of η with S/F has similar trends to that of Cη , which implies that the
enhance of Cη with S/F is the main cause for η variation. From this point of view, the point
S/F=0.208 is a critical point or optimizing of the air and steam gasification of a PGM reactor.
Figure 33. Effect of S/F on Cη and η at ER= 0.06 and PER= 0.118
Figure 34 shows the contents of main gaseous species inside the reactor at different S/F
values. When S/F increases from 0 to 0.208, the volume fractions of H2 and CO generally
show an increasing trend, especially at the bottom half of the reaction shaft where char
gasification reactions take place. This is mainly caused by promoted water-shaft reaction and
other char reactions due to increasing steam injection. When S/F further increases from 0.208
to 0.25, the volume fractions of CO seems decreasing. This phenomenon corresponds to the
variation of Cη with S/F. Since char conversion almost completes at S/F=0.208, further
increasing of S/F mainly promoted the water-gas shift reaction so the total yield of CO is
reduced. It is also found that most of the Light hydrocarbons (LHCs) are produced in the
pyrolysis section, while the effect of methanation reaction is not visible. The explanation of
this phenomenon may be the relatively high temperature in the gasification section which
accurate the reforming of LHCs. Finally, it is found that the overall tar yield shows a
decreasing trend with increasing S/F. this is mainly due to favored tar cracking and reforming
due to higher gas temperature in the pyrolysis section.
66
Figure 34. Predicted contents of main species in gas phase for different S/F values. (a) H2 volume
fractions, (b) CO volume fractions, (c)LHCs volume fractions, (d) tar mass fractions
67
4.4.3 Effect of ER
In this work, the effect of ER on air and steam gasification in the PGM reactor is studied at
S/F= 0.167 and PER= 0.118 condition. Figure 35 demonstrates the gas temperature
distributions at three typical ER values. It is found that increase of ER has a positive effect on
both overall temperature and peak temperature inside the reactor. The increasing of gas
temperature is the results of favored char combustion. It is known that increasing of
gasification temperature is favorable since it accelerates reaction rates, and influences the
energy equilibrium of endothermic gasification reactions. However, a high peak temperature
may challenge the hear resistance of wall materials. Moreover, the bridging problem may
happen at high temperature condition since part of the inorganic component of MSW may be
melted. When ER increases from 0.043 to 0.077, the value of peak gas temperature increases
from 2100 K to about 2500 K. Even after taking into account the overestimation of peak
temperature with the model, the peak temperature at ER=0.077 is still too high for practical
running of the PGM reactor.
Figure 35. Predicted gas temperature (K) distributions for different ER values
Figure 36 shows the variation of Cη with different ER values. The Cη increases all the way
with ER, and reaches about 100% at ER=0.08. This phenomenon can be explained by two
reasons. Firstly, the enhanced char combustion by increased ER directly favors char
conversion. Moreover, char combustion increases the global temperature inside the reactor,
thus accelerates the endothermic char gasification reactions such as water-shift reaction and
boudouard reaction. From this point of view, complete char conversion thus the highest cold
gas efficiency can be obtained at ER value of 0.077. However, considering the high peak
temperature at ER= 0.77, it is not suggested to use such high ER value. In real operation, it is
68
more applicable to use a relative low ER value like 0.06, while increase the S/F to increase the
Cη .
Figure 36 Effect of ER on Cη at S/F= 0.167 and PER= 0.118
Figure 37 shows the contents of main gaseous species inside the reactor at different ER
values. It is shown that the CO volume fraction increases evidently with ER. The increase of
CO content is explained by enhanced char combustion due to increasing ER. The total yield
of H2 is also enhanced by favored water-shift reaction due to temperature increase. However,
this positive effect is counteracted by the dilution from N2 due to enhanced ER. As a result,
the final volume fraction of H2 does not change much with ER. The yield of tar is sensitive to
the temperature in the pyrolysis section. As we can see in Figure 37 (d), the mass fraction of
tar reduces visible when ER increases. Cracking and reforming of tar also produces
combustible gases, thus increasing the η value. This is another positive effect of increasing
ER.
69
Figure 37. Predicted contents of main species in gas phase for different ER values. (a) H2 volume
fractions, (b) CO volume fractions, (c)LHCs volume fractions, (d) tar mass fractions
70
4.4.4 Effect of PER
The high-temperature plasma air injection is the most important significance of the PGM
process. Other than supplying heat for the melting of the inorganic component of MSW, the
plasma injection also preheats gasification agent to around 1700 K, thus influence the energy
balance inside the PGM reactor. The effect of PER value on gas temperature at ER=0.06 and
S/F=0.167 is shown in Figure 38. When PER increases from 0.108 to 0.128, the gas
temperature only increases slightly. This phenomenon can be explained by two reasons.
Firstly, the plasma flow is injected directly into the slag pool at the bottom of the melting
chamber. The outward slag flow take away much of increased sensible heat from plasma.
Secondly, heat loss from strong radiation in the melting chamber also increase with PER
value. As a result, only a small part of the increased plasma energy is introduced to the fixed-
bed.
Figure 38. Predicted gas temperature (K) distributions for different PER values
Figure 39 is the contents of main gaseous species inside the reactor at different PER values.
With the increase of PER, both H2 and CO shows a slight increasing trend. As introduced
previously, the enhancement of gas temperature favors char gasification reactions and also
cracking and reforming of tar. However, this positive effect is also not significant.
71
Figure 39. Predicted contents of main species in gas phase for different PER values. (a) H2
volume fractions, (b) CO volume fractions, (c) LHCs volume fractions, (d) tar mass fractions
72
4.5 Optimizing of the PGM process
4.5.1 Interactions between ER and PER
In the PGM process, the required heat for MSW gasification comes from two sources: the
sensible heat of plasma air and chemical heat from char combustion. In other words, the
energy equilibrium of PGM gasification is controlled by both PER and ER. From this point of
view, the effects of PER and ER are connected to each other. When study the energy
equilibrium of the PGM process, the interaction between PER and ER should be considered.
The SAMR value was set to 0.389 in this study.
Figure 40. Definition of possible operation extent of PER and ER in the PGM process
Figure 40 shows the delimitation of possible operation extent of PER and ER in the PGM
process. Three curves are defined to restrict the logical area for PGM:
• ERpla, min shows the minimum of ER requested for generating plasma flow. In PGM
process, air is used as the carrier of sensible heat from plasma generators. The
relationship between PER and ERpla, min is linear. The gradient of the ERpla, min
denotes the ratio between MSW LHV and the thermal enthalpy of plasma air:
( )pla
MSWstoicMSWair h
LHVmmk /= (4-8)
73
ERgasif, min shows the ER needed for complete gasification (i.e. no solid carbon
residual and enough temperature for gasification and pyrolysis). In this work, the
request for complete gasification is satisfied when the syngas temperature at the outlet
is higher than 120 ºC.
• ERtem, max shows the maximum of ER to prevent too high temperature in the char
combustion and gasification section. If this temperature is too high, the wall material
of the reactor might be damaged. In this study, the maximum of the temperature is set
to 1300 ºC.
• PERmel, min shows the minimum of PER required for melting all the solid residual.
Four different regions are divided by these curves:
• Region 1: In this region, the PGM process can operate normally. The energy supplied
by plasma and char combustion is enough for MSW gasification to take place. The
temperature in the gasification and combustion zone is not too high to damage the
reactor wall.
• Region 2: In this region, the energy supplied by plasma, char combustion and High
temperature steam is not enough for supporting MSW gasification.
• Region 3: In this region, the temperature of the char combustion and gasification
section is higher than 1300 ºC. In other words, the temperature in char combustion and
gasification section may damage the reactor wall.
• Region 4: In this region, the energy supplied by plasma flow is not enough for melting
of solid residual from MSW gasification.
It can be found from Figure 40 that when PER is less than 0.045, the plasma energy is not
enough for entirely melting of inorganic components in MSW. When PER increases from
0.045 to 0.13, the energy require for inorganic components melting is satisfied. The extent of
available ER is limited by ERtem, max and ERgasif, min. In other words, the minimum of available
ER is restricted by entire energy supply, and the maximum of available ER is controlled by
gasification and combustion temperature. When PER further increases from 0.13 to 0.14, the
lower limit of ER does not exist anymore, which means the energy supply is enough for PGM
74
even the secondary air feeding is set to 0. If PER is higher than 0.14, the PGM is not available
because the temperature at the char gasification and combustion section is too high. Generally
speaking, the available PER extent is 0.045-0.14. Increase of PER narrows the variation range
of ER.
Figure 41. Distributions of syngas LHV in Region 1
Figure 42. Distributions of system CGE in Region 1.
75
The distribution of syngas LHV, as well as system CGE in region 1 is demonstrated in Figure
41 and Figure 42. It was found that the maximum syngas LHV in region 1 is about 9.5, while
the minimum is about 4.0. It has been discussed previously that the LHV variation is mainly
caused by thermal cracking of primary tar. The large difference between maximum and
minimum syngas LHV illustrates that the extent of tar cracking is a very important factor
which determines the quality of syngas in PGM process. Furthermore, it is obvious that the
effect of PER on syngas LHV is stronger than that of ER. The positive effect of ER on syngas
LHV is due to promoted primary tar cracking caused by chemical heat from combustion.
However, the ER still have some negative effects on syngas LHV. For example, increased
combustion by increasing ER consumes some combustible gases in syngas. Additionally, the
introduced N2 also dilutes the contents of combustible gases. These negative effects somehow
weaken the positive effect of ER. So the maximum LHV was found in the area with highest
PER value. The dependence of LHV on ER and PER has been confirmed by previous running
of the pilot PGM reactor.
From Figure 42 it was found that the maximum CGE in region 1 is about 0.62 and the
minimum is about 0.22. The maximum CGE appears when ER=0.08 and PER=0.10. The
large difference of CGE is also explained by the influence of the extent of tar cracking. The
influences of PER and ER on CGE have similar intensity. An interesting phenomenon found
in Figure 42 is that the effects of ER and PER on CGE shows a linear relation. It implies that
the influence of ER and PER can be further synthesized to a unified parameter. The further
correlation of ER and PER can be an interesting topic of our future work.
4.5.2 Considering the oxygen equilibrium
Steam and air are two popular gasification agents which supply oxygen for the gasification
process. As the material base of gasification, the oxygen supply directly influence the
conversion of C during gasification and combustion section. From this perspective, the ER
and SAMR may also have internal connecting with each other.
Figure 43 shows the delimitation of possible operation extent of SAMR and ER in the PGM
process at PER =0.118. Three curves are used to restrict the possible operation conditions for
SAMR and ER: ERpla, min, ERgasif, min, and ERtem, max.
76
Figure 43. Delimitation of possible operation extent of SAMR and ER in the PGM process
These curves defined 3 main regions with different operation conditions: In region 1’, the
PGM process can work continuously; in region 2’, the energy supplied by plasma and char
combustion is not enough for MSW gasification; in region 3’, the temperature of the char
combustion and gasification section is too high. It was found that when SAMR increases, the
maximum of possible ER increases and the minimum of possible ER in region 1’ decreases.
Increase of SAMR means enhanced oxygen supply from steam. In that case, the oxygen
equilibrium in the reactor is affected, and the requested air decreases. The increase of
maximum ER can be explained by the increases of total heat capacity with increasing SAMR,
which increases the uniformity of temperature distribution inside the reactor. This uniformity
is also beneficial to syngas LHV because the temperature difference between gasification and
pyrolysis will be reduced.
77
Figure 44. Distributions of syngas LHV in Region 1’
The distribution of syngas LHV in region 1’ is demonstrated in Figure 44. It was found that
the syngas LHV in region 1’ varies from 6.5 to 9.0 MJ/Nm3. The increase of SAMR has
positive effects on syngas LHV. This positive effect may be mainly due to the high
temperature of steam, which also introduce some heat into the PGM system. At the same
time, the decreased temperature difference between gasification and pyrolysis section by
increasing SAMR enhances the potential of LHV increase by larger energy supply. An
interesting phenomenon found in Figure 44 is that the effect of increasing ER on syngas LHV
changes when SAMR is larger than 0.55. In this area, the LHV first increase, and then start to
decrease when ER keep increasing. The maximum of LHV appears at about ER=0.055. This
result illustrate that the positive aspect of ER effect by increasing chemical heat is not always
dominant. The negative aspects such as consumption of combustible gas and dilution from N2
play important roles in high SAMR condition. The suggested ER in high SAMR condition is
0.055.
78
5. Conclusions and recommendations
Experimental tests have been performed to study the performance of air gasification
and air&steam gasification in the PGM reactor. The following are the main discoveries:
• The syngas produced from the PGM has a high LHV (6–7 MJ/Nm3).
• In air gasification, the syngas yield increased significantly with increasing ER,
whereas the LHV decreased slightly.
• Feeding high-temperature steam into the PGM reactor greatly increased syngas yield,
with even higher gas LHV. The feeding of high-temperature steam can further reduce
the air demand for gasification.
• The energy efficiency of air and steam gasification was much higher than that of air
gasification. The CGE of PGM air and steam gasification can reach approximately
60%. Tar formation represents the main energy loss for the PGM reactor.
A zero-dimensional kinetics-free gasification model was developed. The accuracy of this
model is confirmed by the measurements from the tests. The influence of operation
parameters are studied with this model:
• The performance of PGM reactors with high-temperature steam feeding is analyzed by
both test measurement and model prediction. The effects of three dimensionless
operation parameters are discussed. PER has positive effect for both syngas yield and
syngas LHV. The main reason for this effect is the favored tar cracking by increasing
heat supply.
• The ER has two contradictory effects on syngas LHV: the positive effect by increasing
chemical heat and the negative effect by syngas combustion and N2 dilution. When ER
is lower than 0.065, the positive effect is dominant; When ER is larger than 0.065, two
effects counterbalance each other. The effect of ER on CGE is positive in the studied
region.
79
• The SAMR mainly influence the equilibrium of water-gas shift reaction in the PGM
process. Steam at 1000 ºC can supply some heat for pyrolysis, so the SAMR also have
slight positive effect on syngas yield and LHV.
A two-dimensional Eulerian-Eulerian multiphase model was also introduced in this work. This model was proved to be good enough for prediction the performance of the PGM, when
air or air&steam mixtures are used as gasification agents.
For air gasification:
• Analysis of the base case 1 by means of CFD revealed that the horizontal temperature
distribution inside the reactor was non-uniform. In addition, maximum peak
temperature of the reactor was observed at the gas-solid interface. PGM air
gasification produced a syngas with a LHV of 6.79MJ/Nm3. The tar yield is around
0.193 kg/kg MSW.
• Further investigation of the PGM process by means of developed model revealed that
ER has positive influence on the calorific value of the syngas. Increase of ER from
0.043 to 0.077 showed around 5% increase in cold gas efficiency. However, the
maximum allowable ER for present gasification system was restricted to about 0.067
due to increase in peak temperature of the reactor.
• The influence of PER on PGM air gasification is not obvious. The optimal PER value
was considered as about 0.09 considering energy efficiency.
• Although PGM air gasification provided good calorific value of the syngas
(LHV=6.79MJ/Nm3) , detrimental effect on char conversion was observed.
Minimization of this problem will be addressed in our future research.
For air & steam gasification:
• Injection of high temperature steam has a positive effect on both syngas LHV value
and reactor cold gas efficiency. The main reason is that high temperature steam
supplies both oxygen atoms and sensible heat for char gasification, so that the char
convention ratio is highly enhanced. At the studied condition, the positive effect of
80
increasing S/F value is significant when 208.0/ ≤FS . When 208.0/ ≥FS , the effect
becomes very limited.
• The value of ER influences both chemical and energy balance inside the reactor.
Increasing ER promotes both the char combustion and water-gas reaction. Thus
increasing char conversion. At the studied condition, a theoretical maximum of cold
gas efficiency can be obtained at ER= 0.077, which corresponds to complete char
conversion ratio. However, this maximum value cannot be reached in reality since the
peak temperature at this condition is too high. An optimal ER value should be around
0.6 in reality.
• Increasing the plasma power also has a slight positive effect on syngas yield and LHV
value. However, the influence of PER is weaker than that of ER and S/F. From
economic point of view, the PER should be chosen as the minimum value which
satisfies the energy request for melting the inorganics of MSW.
Some optimizing work was done based on the proposed models:
• The available extent of PER and ER is defined at air/steam gasification conditions.
• The possible range for PER at the studied condition is 0.045-0.14. Increase of PER
narrows the variation range of ER. The optimal syngas LHV can be obtained when the
PER reaches its maximum. The effect of ER and PER on syngas CGE seems can be
synthesized to a unified parameter.
• The available extent of SAMR and ER is defined at PER=0.118. Increasing SAMR
broadens the available range of ER. When SAMR>0.6, the secondary air is not
necessary anymore. The optimal syngas LHV can be obtained at SAMR=0.8 and
ER=0.055.
81
6. Reference
[1]. UPEN year book, 2010. [2]. Belevi H., Baccini P., Long-term behavior of municipal solid waste landfills, Waste
Management 1989, 7(1): 43-56. [3]. Christopher, H.; Maarten, B., Gasification. Elsevier, 2008. [4]. Thomas M., Novel and innovative pyrolysis and gasification technologies for energy
efficient and environmentally sound MSW disposal, Waste Management 2004, 24(1): 53-79.
[5]. Ladislav Bebar, Petr Stehlik, Leos Havlen, Jaroslav Oral. Analysis of using gasification and incineration for thermal processing of wastes. Applied Thermal Engineering 2005, 25(7): 1045-1055.
[6]. Anna P., Yang W., Lucas C., Development of a Thermally Homogeneous gasifier system using high-temperature agents, Clean Air 2006, 7(4): 363-379.
[7]. Koutaro K., Tomonori A., Yoshihito K., Ryoji S., Melting municipal solid waste incineration residue by plasma melting furnace with a graphite electrode. Thin Solid Films 2001, 386(2): 183-188.
[8]. Ryo Y., Makoto N., Hiroshi M., Influence of ash composition on heavy metal emissions in ash melting process. Fuel 2002, 81(10): 1335-1340.
[9]. Shinichi S., Masakatsu H., Municipal solid waste incinerator residue recycling by thermal processes. Waste Management 2000, 20(2-3): 249-258.
[10]. Moustakas K., Fatta D., Malamis S., Haralambous K., Loizidou M., Demonstration plasma gasification/vitrification system for effective hazardous waste treatment. Journal of hazardous materials 2005, 123(1-3): 120-126.
[11]. Bert Lemmens, Helmut Elslander, Ive VanderreydT, Kurt Peys, Ludo Diels, Michel Oosterlinck, Marc Joos. Assessment of plasma gasification of high caloric waste streams. Waste Management 2007, 27(11): 1562-1569.
[12]. Hlína M., Hrabovský M., Kopecký V., Konrád M., Kavka T., Skoblja S., Plasma gasification of wood and production of gas with low content of tar. Czechoslovak Journal of Physics 2006, 56(2): 1179-1184.
[13]. Ahmed I.I., Gupta A.K., Pyrolysis and gasification of food waste: Syngas characteristics and char gasification kinetics. Applied Energy 2010, 87(1) 101–108.
[14]. Ko M.K., Lee W.Y., Kim S.B., Lee K.W., Chun H.S., Gasification of food waste with steam in fluidized bed. Chemistry and Materials Science 2001, 18(6):961-964.
[15]. Masaaki Tanaka, Hitoshi Ozaki, Akira Ando, Shinji Kambara, Hiroshi Moritomi. Basic characteristics of food waste and food ash on steam gasification. Ind. Eng. Chem. Res. 2008, 47(7): 2414-2419.
[16]. Ahmed I.I., Gupta A.K., Syngas yield during pyrolysis and steam gasification of paper. Applied Energy 2009, 86(9): 1813-1821.
[17]. Ahmed I.I., Gupta A.K., Characteristics of cardboard and paper gasification with CO2. Applied Energy 2009, 86(12): 2626-2634.
82
[18]. Ahmed I.I., Gupta A.K., Evolution of syngas from cardboard gasification. Applied Energy 2009, 86(9): 1732-1740.
[19]. Beck S.R., Wang M.J., Wood gasification in a fluidized bed. Ind. Eng. Chem. Process Des. Dev. 1980, 19(2), 312–317.
[20]. Eric R. Palmer. Gasification of wood for methanol production. Energy in Agriculture 1984, 3: 363-375.
[21]. Bhattacharya S.C., Md A.H., Siddique M.R., Pham H.L., A study on wood gasification for low-tar gas production. Energy 1999, 24(4): 285-269.
[22]. Xiao R., Jin B., Zhou H., Zhong Z., Zhang M., Air gasification of polypropylene plastic waste in fluidized bed gasifier. Energy Conversion and Management 2007, 48(3): 778-786.
[23]. Yoichi Kodera, Yumiko Ishihara. Novel process for recycling waste plastics to fuel gas using a moving-bed reactor. Energy and Fuels 2007, 20(1): 155-158.
[24]. Ooi N., Inoue M., Gasification technology of waste plastic. Journal of the Japan Institute of Energy 2010, 89(6): 516-521.
[25]. Zevenhoven R., Karlsson M., Hupa M., Frankenhaeuser M., Combustion and gasification properties of plastics particles. Journal of the Air and Waste Management Association 1997, 47(8): 861-870.
[26]. Chi Y., Zheng J., Jin Y., Mi H.B., Jiang X.G., Ni M.J., Experimental study on fluidized-bed gasification of simulated MSW. Proceedings of the Chinese Society of Electrical Engineering 2008, 28: 59-63.
[27]. Maitri Thamavithya, Animesh Dutta. An investigation of MSW gasification in a spout-fluid bed reactor. Fuel Processing Technology 2008, 89(10): 949-957.
[28]. Pinto F., Franco C., André R.N., Miranda M., Gulyurtlu I., Cabrita I., Co-gasification study of biomass mixed with plastic wastes. Fuel 2002, 81(3): 291-297.
[29]. Dalai A., Batta N., Eswaramoorthi I., Schoenau G., Gasification of refuse derived fuel in a fixed bed reactor for syngas production. Waste Management 2009, 29(1): 252-258.
[30]. Anna P., Sylwester K., Blasiak W. Effect of operating conditions on tar and gas composition in high temperature air/steam gasification (HTAG) of plastic containing waste. Fuel Processing Technology 2006, 87(3): 223-233.
[31]. Thomas W.M., Brendan P.M., William R.S., Steven J., Stanley E.M., Fate of heavy metals and radioactive metals in gasification of sewage sludge. Waste Manage 2004, 24(2):193-198.
[32]. Kirby C.S., Rimstidt J.D., Mineralogy and surface properties of municipal solid waste ash. Environ. Sci. Technol. 1993, 27: 652–660.
[33]. Park Y.J., Heo J., Vitrification of fly ash from municipal solid waste incinerator. J Hazard Mater 2002, 91(1-3):83-93.
[34]. Jung C.H., Matsuto T., Tanaka N., Behavior of metals in ash melting and gasification-melting of municipal solid waste (MSW). Waste Manage (Oxford) 2005, 25(3): 301-310.
[35]. Xiao G., Jin B., Zhong Z., Chi Y., Ni M., Cen K., Xiao R., Huang H., Experimental study on MSW gasification and melting technology. Journal of Environmental Sciences 2007, 19(11):1398-1403.
83
[36]. Bernd Calaminus, R Stahlberg. Continuous in-line gasification/vitrification process for thermal waste treatment: process technology and current status of projects. Waste Management 1998, 18(6-8):547-556.
[37]. Sakai S.I., Hiraoka M., Municipal solid waste incinerator residue recycling by thermal processes. Waste Manage (Oxford) 2000, 20(2-3):249-258.
[38]. Ivan B.G., Boris I.M., Some General conclusions from the results of studies on solid fuel steam plasma gasification. Fuel 1992, 71(8): 895-901.
[39]. Galvita V., Messerle V.E. Ustimenko A.B., Hydrogen production by coal plasma gasification for fuel cell technology. International Journal of Hydrogen Energy 2007, 32(16): 3899-3906.
[40]. Kalinenko R.A., Kuznetsov A.P., Levitsky A.A., Messerle V.E., Mirokhin Y.A., Polak L.S., Sakipov Z.B., Ustimenko A.B., Pulverized coal plasma gasification. Plasma Chemistry and Plasma Processing 1993, 13(1): 141-167.
[41]. Moustakas K., Datta D., Malamis K., Loizidou M., Demonstration plasma gasification/vitrification system for effective hazardous waste treatment. J Hazard Mater 2005, 123(1–3):120-126.
[42]. Moustakas K., Xydis G., Malamis S., Haralambous K. J., Loizidou M., Analysis of results from the operation of a pilot plasma gasification/ vitrification unit for optimizing its performance. Journal of Hazardous Materials 2008, 151 (2-3): 473-480.
[43]. Manfrida G., Bidini G., Trebbi G., Modeling coal gasification combined cycle (CGCC). A Future for Energy, Flowers’90, Pergamon Press, 1990.
[44]. Ruggiero M., Manfrida G., An equilibrium model for biomass gasification processes. Renewable Energy 1999, 16(1-4): 1106-1109.
[45]. Zainal Z.A., Ali R., Lean C.H., Seetharamu K.N., Prediction of performance of a downdraft gasifier using equilibrium modeling for different biomass materials. Energy Conversion and Management 2001, 42(12):1499-1515.
[46]. Vittorio T., Giorgio C., Process analysis and performance evaluation of updraft coal gasifier. In: Proceedings of the 3rd. International Conference on Clean Coal Technologies for Our Future; 2007 May 15-17; Cagliari, Italy.
[47]. Manurung R.K., Beenackers A.A.C.M., Modeling and simulation of an open-core downdraft moving bed rice husk gasifier, in A.V. Bridgewater (Ed.), Advances in Thermochemical Biomass Conversion, London, 1994, pp. 288-309.
[48]. Blasi C.D., Dynamic behavior of stratified downdraft gasifier, Chem. Eng. Sci. 2000, 55: 2931-2944.
[49]. Yang, W.; Ponzio, A.; Lucas, C.; Blasiak, W. Effect of operating conditions on tar and gas composition in high temperature air/steam gasification (HTAG) of plastic containing waste. Fuel Proc. Technol. 2006, 87(3), 235–245.
[50]. Syamlal M., Bissett L., METC Gasifier Advanced Simulation (MGAS) Model; Morgantown Energy Technology Center: Morgantown, WV, 1992.
[51]. Bryden K.M., Ragland K.W., Numerical modeling of a deep, fixed bed combustor, Energy Fuels 1996, 10: 269-275.
84
[52]. Chen C., Horio M., Kojima T., Numerical simulation of entrained flow coal gasifiers. Part I: modeling of coal gasification in an entrained flow gasifier. Chem. Eng. Sci. 2000, 55(18): 3861–3874.
[53]. Shi S.P., Zitney S.E., Shahnnam M., Syamlal M., Rogers W.A., Modelling coal gasification with CFD and discrete phase method. J. Energy Inst. 2006, 79 (4): 217-221.
[54]. Slezak A., Kuhlman J.M., Shadle L.J., Spenik J., Shi S., Powder Technol. 2010, 203(1): 98-108.
[55]. Gomez-Barea A., Leckner B., Prog. Energy Combust. Sci. 2010, 36(4): 444-509. [56]. Gerber S., Behrendt F., Oevermann M., Fuel 2010, 89(10): 2903-2917. [57]. Shi S., Guenther C., Orsino S., Proceeding of Power 2007; American Society of
Mechanical Engineers: San Antonio, Texas, 2007. [58]. Rogel A., Aguillon J., The 2D eulerian approach of entrained flow and temperature in a
biomass stratified downdraft gasifier, Am. J. Appl. Sci. 2006, 3: 2068-2075. [59]. Peters B., Thermal Conversion of solid fuels (Developments in heat transfer); WIT Press:
Billerica, MA, 2002; Vol. 15. [60]. Hobbs M.L., Radulovic F.T., Smoot L.D., Modeling fixed-bed coal gasifiers. AIChE
Journal 1992, 38(5):681-702. [61]. Chan W.B., Kelbon M., Krieger B.B., Modelling and experimental verification of
physical and chemical processes during pyrolysis of a large biomass particle. Fuel, 1985, 64: 1505-1513.
[62]. Krieger-Brockett B., Glaister D.S., Wood devolatilization-sensitivity to feed properties and process variables. In A. V. Bridgewater, editor, International Conference on Research in Thermochemical Biomass Conversion 1988, 127-142.
[63]. Bryden K.M., Computational model of wood combustion. PhD thesis, University of Wisconsin-Madison, 1998.
[64]. Smoot L. D., Pratt D.T., Pulverized coal combustion and gasification, Plenum press, 1979. [65]. Colomba D.B., Modeling wood gasification in a counter current fixed-bed reactor. AIChE
Journal 2004, 50(9): 2306-2319. [66]. Marcio L. de Souza-Santos. Solid fuels combustion and gasification: modeling, simulation,
and equipment operation. Marcel Dekker Inc. 2004. [67]. Arthur J.A., Reactions between carbon and oxygen. Transactions of the Faraday Society
1951; 47:164-178. [68]. www.eer-pgm.com. [69]. Coal Conversion Systems Technical Data Book. Virginia: Springfield, 1978. [70]. Hla SH. A theoretical and experimental study on a stratified downdraft biomass gasifier.
PhD thesis, University of Melbourne, 2004. [71]. Williams E.A., Williams P.T., The pyrolysis of individual plastics and a plastic mixture in
a fixed bed reactor. J Chem Technol Biotechnol 1997, 70(1): 9-20. [72]. Fagbemi L., Khezami L., Capart R., Pyrolysis products from different biomasses:
Application to the thermal cracking of tar. Applied Energy 2001, 69(4): 293-306. [73]. Boie W. Energietechnik 3. [74]. Ansys Fluent 12.0 theory guide; ANSYS: USA, 2009.
85
[75]. Kuipers J.A.M., Duin K.J.V., Beckum F.P.H., Swaaij W.P.M., A numerical model of gas-fluidized beds. Chem. Eng. Sci. 1992, 47(8), 1913-1924.
[76]. Gldaspow D., Ettehadleh B., Fluidization in two-dimensional beds with a jet. 2. Hydrodynamic modelling. Ind. Eng. Chem. Fundamen. 1983, 22 (2), 193–201.
[77]. Johnson P.C., Jackson R. J., Frictional-collisional constitutive relations for granular materials, with application to plane shearing. Fluid. Mech. 1987, 176: 67–93.
[78]. Ergun S., Fluid flow through packed columns. Chem. Eng. Prog. 1952, 48(2), 89–94. [79]. Cowin S.C., A theory for the flow of granular materials. Powder Technol. 1974, 9(2-3):
61-69. [80]. Syamlal M., Rogers W, O’Brien T. J., MFIX Documentation: Volume 1, Theory Guide;
National Technical Information Service: Springfield, VA, 1993. [81]. Schaeffer S., Balakrishnan L., ICASE Report 90-18: Application of a Reynolds-Stress
Turbulence Model to the Compressible Shear Layer; NASA: USA, 1990. [82]. Gunn D. J., Transfer of heat or mass to particles in fixed and fluidized beds. Int. J. Heat
Mass Transfer 1978, 21: 467–476. [83]. Yang Y.B., Goh Y.R., Zakaria R., Nasserzadeh V., Swithenbank J., Mathematical
modelling of MSW incineration in a travelling bed. Journal of Waste Management 2002, 22(4), 369-380.
[84]. Yang Y.B., Nasserzadeh V., Goodfellow J., Goh Y.R., Swithenbank J., Parameter study on the incineration of MSW in packed beds. J. Inst. Energ. 2002, 75: 66-80.
[85]. Yang Y.B., Yamauchi H., Nasserzadeh V., Swithenbank J., Effects of fuel devolatilization on the combustion of wood chips and incineration of simulated municipal solid wastes in a packed bed. Fuel 2003, 82(18): 2205-2221.
[86]. Yang Y.B., Sharifi V.N., Swithenbank J., Effect of air flow rate and fuel moisture on the burning behaviors of biomass and simulated solid waste in a packed bed. Fuel 2004, 83(11-12): 1553-1562.
[87]. Yang W., Ponzio A., Lucas C., Blasiak W., Performance analysis of a fixed-bed biomass gasifier using high-temperature air. Fuel Process Technol. 2006, 87(3): 235-245.
[88]. Ranz W.E., Marshall W.R., Evaporation from drops, Part I. Chem. Eng. Prog. 1952 48(3): 141–146.
[89]. Ranz W.E., Marshall W.R., Evaporation from drops, Part II. Chem. Eng. Prog. 1952 48(4): 173–180.
[90]. Boroson M.L., Howard J.B., Longwell J.P., Peter W.A., Product Yields and Kinetics from Vapor Phase Cracking of Wood Pyrolysis Tars. AIChE J. 1989, 35(1): 120-128.
[91]. Liden A.G., Berruti F., Scott D.S., Heat transfer controlled pyrolysis kinetics of a biomass slab, rod or sphere. Chem. Eng. Commun. 1988, 65(1): 207-221.
[92]. Sorum L., Gronli M. G., Hustad J. E., Pyrolysis characteristics and kinetics of municipal solid wastes. Fuel 2001, 80(9): 1217-1227.
[93]. Chan W.R, Kelbon M., Krieger B.B., Modeling and experimental verification of physical and chemical processes during pyrolysis of large biomass particle. Fuel 1985, 64(11): 1505-1513.
[94]. Wu C., Chang C., Hor J., On the thermal treatment of plastic mixture of MSW: pyrolysis kinetics. Waste Manage. (Oxford) 1993, 13(3): 221-235.
86
[95]. A.K Varma, A.U. Chatwani, F.V. Bracco, Studies of premixed laminar hydrogen-air flames using elementary and global kinetics models, Combust. Flame 64 (1986) 233-236.
[96]. F.L. Dryer, I. Glassman, High temperature oxidation of CO and CH4, 14th Symposium on Combustion, 1973, pp. 987-1003.
[97]. Howard J.B., William G.C. and Fine D.H. Kinetics of carbon monoxide oxidation in postflame gases, Symposium (International) on Combustion 1973, 14: 975–986.
[98]. Jones W.P., Lindstedt R.P., Global reaction schemes for hydrocarbon combustion, Combust. Flame 1988, 73: 233-249.
[99]. Grebenshchikova G.B., Podzemnaya Gazifikatsiya, 1957, 2: 54–57. [100]. Yoon H., Wei J., Denn M.M., A model for moving-bed coal gasification reactors, AIChE
Journal 1978, 24: 885-903. [101]. Hobbs M.L., Radulovic F.T., Smoot L.D., Combustion and gasification of coals in fixed
beds. Prog. Energy Combust. Sci. 1993, 19(6): 505-86. [102]. Evans D.D., Emmons H.W., Combustion of wood charcoal. Fire Res. 1977, 1:57-66. [103]. Pinto F., Franco C., André R.N., Miranda I., Gulyurtlu I., Cabrita I.. Co-gasification study
of biomass mixed with plastic wastes. Fuel 2002 81(3): 291–297. [104]. Anh N.P., Changkook R., Vida N.S., Jim S., Characterization of slow pyrolysis products
from segregated wastes for energy production. J Anal Appl Pyrolysis 2008, 81(1): 65-71. [105]. Li A.M., Li X.D., Li S.Q., Ren Y., Shang N., Chi Y., Yan J.H., Cen K.F., Experimental
studies on municipal solid waste pyrolysis in a laboratory-scale rotary kiln. Energy 1999, 24(3): 209-218.
[106]. Anthony D., Sylvie V., Pierre C., Sebastien T., Guillaume B., Andre Z., Glaude P.A., Mechanisms and kinetics of methane thermal conversion in a syngas. Ind Eng Chem Res 2009, 48(14): 6564-6572.
[107]. Blasiak W., Szewczyk D., Lucas C. Reforming of biomass wastes into fuel gas with high temperature air and steam. In Pyrolysis & Gasification of Biomass & Waste. Strasbourg, France, 2002.
[108]. Li C., Kenzi S., Tar property, analysis, reforming mechanism and model for biomass gasification—an overview. Renew Sustainable Energy Rev 2008, 13(3): 594-604.
[109]. Jess A., Mechanisms and kinetics of thermal reactions of aromatic hydrocarbons from pyrolysis of solid fuels. Fuel 1996, 75(12): 1441-1448.
[110]. Boie W., Energietechnik 3 (1953), 309-16. [111]. Lucas C., Szewczyka D., Blasiaka W., Mochidab S., High-temperature air and steam
gasification of densified biofuels. Biomass Bioenergy 2004, 27(6): 563–575. [112]. Okada T., Tojo Y., Tanaka N., Matsuto T., Recovery of zinc and lead from fly ash from
ash-melting and gasification-melting processes of MSW – Comparison and applicability of chemical leaching methods. Waste Manage (Oxford) 2007, 27(1): 69-80.
[113]. Park Y.J., Heo J., Vitrification of fly ash from municipal solid waste incinerator. Journal of Hazardous Materials 2002, 91(1-3):83-93.
[114]. Sakai S.I., Hiraoka M. Municipal solid waste incinerator residue recycling by thermal processes, Waste Manage (Oxford) 2000, 20(2-3): 249-258.
87
[115]. Ecke H., Sakanakura H., Matsuto T., Tanaka N., Lagerkvist A., State-of-the-art treatment processes for municipal solid waste incineration residues in Japan. Waste Manage Research 2000, 18(1): 41-51.
[116]. Jung C.H., Matsuto T, Tanaka N. Behavior of metals in ash melting and gasification-melting of municipal solid waste (MSW). Waste Manage (Oxford) 2005, 25(3): 301-310.
[117]. Blasiak W., Szewczyk D., Lucas C., Reforming of biomass wastes into fuel gas with high temperature air and steam. Pyrolysis &Gasification of Biomass & Waste Conference, Strasbourg, France, 2002.
[118]. Levis F.M., Swithenbank J., Hoecke D.A., Russell N.V., Shabangu S.V., High temperature, steam-only gasification and steam reforming with ultra-superheated steam. 5th International Symposium on High Temperature Air Combustion and Gasification, October 28-30, Yokohama Japan, 2002.
SUPPLEMENT I
Qinglin Zhang, Liran Dor, Dikla Fenigshtein, Weihong Yang, Wlodzimierz Blasiak.
Gasification of municipal solid waste in the Plasma Gasification
Melting process
Appl Energy (2011), DOI:10.1016/j.apenergy.2011.01.041
Gasification of municipal solid waste in the Plasma Gasification Melting process
Qinglin Zhang b,⇑, Liran Dor a, Dikla Fenigshtein a, Weihong Yang b, Wlodzmierz Blasiak b
a Environmental Energy Resources Ltd., 7 Jabotinsky Street, Ramat-Gan 52520, Israelb Energy and Furnace Technology, Royal Institute of Technology, Sweden
a r t i c l e i n f o
Article history:Received 30 September 2010Received in revised form 13 December 2010Accepted 18 January 2011Available online xxxx
Keywords:GasificationPlasmaMeltingWasteSyngas
a b s t r a c t
A new waste-disposal technology named Plasma Gasification Melting (PGM) was developed. A pilot PGMreactor was constructed in northern Israel. The reactor is an updraft moving-bed gasifier, with plasmatorches placed next to air nozzles to heat the incoming air to 6000 �C. The inorganic substances of thefeedstock are melted by the high-temperature air to form a vitrified slag in which undesirable materialssuch as heavy metals are trapped. The residual heat in the air supplies additional heat for the gasificationprocess.A series of tests were conducted to study the performance of PGM gasification. The plasma power was
varied from 2.88 to 3.12 MJ/kg of municipal solid waste (MSW), and the equivalence ratio (ER) was variedfrom 0.08 to 0.12. For air and steam gasification, the maximum steam/MSW mass ratio reached 0.33.The composition of the syngas product was analyzed in all tests; the lower heating value (LHV) of the
syngas varied from 6 to 7 MJ/Nm3. For air gasification, the syngas LHV decreased with increasing ER,whereas the gas yield and energy efficiency increased with ER. When high-temperature steam was fedinto the reactor, the overall gas yield was increased significantly, and the syngas LHV also increasedslightly. The positive effect may be attributed to the steam reforming of tar. In air and steam gasification,the influence of increased ER on syngas LHV was negative, while the effect of increased plasma powerwas positive. The maximum energy efficiency of the tests reached 58%. The main energy loss was dueto the formation of tar.
� 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Rapid economic development has lead to an annual increase inmunicipal solid waste (MSW) production. According to the conceptof sustainable waste disposal, a successful treatment of MSWshould be safe, effective, and environmentally friendly [1]. How-ever, existing waste-disposal methods cannot achieve this goal.Conventional waste landfills occupy large amounts of land andlead to serious environment problems [2]. Incineration technologywas developed to reduce the total volume of waste and make useof the chemical energy of MSW for energy generation [3,4]. How-ever, the emissions of pollutant species such as NOx, SOx, HCl,harmful organic compounds [5,6], and heavy metals [7,8] are highin the incineration process. Another problem with MSW incinera-tion is the serious corrosion of the incineration system by alkalimetals in solid residues and fly ash [7]. Furthermore, due to thelow incineration temperature related to the low energy densityof MSW, the energy efficiency of MSW incineration is relativelylow [9,10].
The development of an environmentally benign gasificationtechnology for processing MSW has been a topic of much research
in the past decade, as summarized by Thomas [11]. Among variousgasification technologies, high-temperature agent gasification (Hi-TAG) has been proven to be an efficient technology. The character-ization of HiTAG for different waste materials has been studied[12–19], and several positive features have been reported. Experi-mental work by Lucas et al. showed that preheating the gasifica-tion agent can sharply reduce the air demand in a gasificationprocess, so the concentration of noncombustible gases (N2 andCO2) in the syngas product can be reduced correspondingly [12].Another advantage of preheating the gasification agent is thatthe tar yield can be significantly reduced due to the high temper-ature [13]. Using a high-temperature gasification agent has otherbenefits including greater system stability. The syngas quality be-comes less sensitive to variations in the particle size, heating valueand moisture content of the MSW [14]. The characteristics of Hi-TAG are more significant if high-temperature steam is used asthe gasification agent [15–17,19].
In a HiTAG process, most of the heavy and alkali metals (withthe exception of mercury, zinc and lead, which can vaporize at hightemperatures and be retained in fly ash and syngas [20]) are re-tained in the bottom ash produced during gasification [21]. To pre-vent secondary pollution from the bottom products, a meltingtechnology has been widely applied in MSW incineration plants[22–24]. During the melting process, the solid residues are melted
0306-2619/$ - see front matter � 2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.apenergy.2011.01.041
⇑ Corresponding author. Tel.: +46 46 87906545.E-mail address: [email protected] (Q. Zhang).
Applied Energy xxx (2011) xxx–xxx
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier .com/locate /apenergy
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
to form a vitreous slag in which heavy metals and other harmfulcomponents are locked. After cooling, the slag becomes a vitrifiedsolid that can be used as a building material [25]. The same advan-tages are obtained if this melting technology is applied to MSWgasification. Various methods have been used in the solid-residuemelting process of MSW incineration/gasification [26–28], amongwhich plasma technology has proven successful. The state of theart of thermal plasma applications in MSW treatment was re-viewed by Gomez et al. [29].
The combination of plasma melting and HiTAG leads to a newMSW disposal technology, named Plasma Gasification Melting(PGM). A demonstration Plasma-Gasification-Melting system wasbuilt by Moustakas et al. The basic performance behaviors of thesystem were determined, and the influences of operating parame-ters were discussed [30,31]. However, reports on the performanceanalysis of the PGM process remain rare. The available experimen-tal data on Plasma Gasification Melting, especially industrial-scaleoperational data, is very limited. This situation serious has hin-dered the understanding and application of Plasma GasificationMelting technology.
An industrial-scale PGM plant was constructed in Israel by theEnvironmental Energy Resources, Israel Ltd., (EER). A series of trialruns was performed in this plant to investigate the characteristicsof the PGM process. This study aimed to provide detailed resultsand analysis of the trial runs.
2. Test method and materials
2.1. The PGM system
The demonstration plant was constructed in Yblin Israel in2007. The designed capacity of the plant is 20 tons of MSW perday. The process flow sheet is shown in Fig. 1. MSW is fed intothe reactor through airtight feeding chambers placed at the upperpart of the plasma chemical reactor, wherein gasification reactionsoccur. Syngas produced from gasification flows into the after-burner and is combusted there. The hot flue gas from combustionis sent to the boiler to produce steam, which drives a steam turbineconnected to an electrical generator. The generated electricity, be-sides providing power for the plasma torches and the rest of thesystem, can be sold to outside users. The fly ash is removed fromthe flue gas in the scrubber-evaporator. SOx is absorbed in the reac-tor absorber and removed using a bag filter. The solid residue fromgasification is melted by the plasma jet and collected by the slagcollectors.
The core of the PGM plant is the plasma chemical reactor, whichis a typical fixed-bed updraft gasification reactor. The scheme of thereactor is shown in Fig. 2. MSW is fed into the reactor from the top.The gasification agents, which are air and high-temperature steam
(1000 �C), are injected into the lower part of the reactor from vari-ous nozzles. A part of the air, called plasma air, is injected into thereactor from four plasma torches, which are embedded into thereactor at the upper surface of the melting chamber. Electrical arcsare formed between electrodes at the tip of the plasma torches, sothat air flowing through the arc is ionized and forms a plasma jetthat extends beyond the tips of the torches. The temperature ofthe plasma jet may reach up to 6000 �C. The power of the plasmatorches can be controlled by the central control system. Additionalair, known as secondary air, is fed into the reactor from nozzles sur-rounding the plasma torches. High-temperature steam nozzles arelocated at the lower part of the gasifier. The feed rates of both sec-ondary air and steam are adjustable and controlled by the centralcontrol system.
2.2. Measurements
An isokinetic probe is located in the syngas conduit between thereactor and the afterburner to obtain syngas samples. The hot syn-gas sample is first sent to a water-cooled tar collector and then to acontinuously operating gas analyzer to determine its composition.The total flow rate of syngas is measured by a flowmeter. Becausethe syngas goes to the afterburner without a cooling step, the flowrate is measured here on a wet basis (i.e., steam and some tar is in-cluded in the flow rate). The yield of dry syngas is calculated froman Aspen model developed specially for the PGM process.
To measure the temperature distributions inside the plasmachemical reactor, thermocouples are placed both along the gasifiershaft and in the syngas conduit. The thermocouple positions de-pend on their height above the reactor bottom, H(m). If H < 1 thethermocouples are placed in the reactor wall, behind the refractorylayer, to prevent damage to the thermocouples at high tempera-ture. If 1 6 H 6 2, the thermocouples are placed both behind therefractory layer and inside the reactor. For H 6 2 thermocouplesare placed inside the reactor. To obtain the actual temperature in-side the reactor, temperature compensation must be made for thethermocouples placed behind the refractory layer. According to theheat conducting law, the heat flow through the reactor wall can bewritten as:
/ ¼ k1ðT1 � T0Þ
Dx1¼ k2
ðT2 � T1ÞDx2
; ð1Þ
where k1 is the average heat-conduction coefficient of the reactorwall outside the refractory layer, k2 is the heat-conduction coeffi-cient of the refractory layer, T0, T1, and T2 are temperatures at theouter wall surface, behind the refractory layer and inside the reac-tor, respectively, Dx1 is the thickness of reactor wall outside therefractory layer, and Dx2 is the thickness of the refractory layer.
We assume that the wall material of both the refractory layerand the reactor wall outside the refractory layer are uniform. Theratio of k1 and k2 can be calculated from the measured temperatureat 1 6 H 6 2. The temperature inside the reactor at H 6 1 rangecan then be calculated as:
T 02 ¼ k1Dx02ðT 0
1 � T 00Þ
k2Dx01þ T 0
1: ð2Þ
2.3. Feedstock properties
MSW is fed into the reactor from airtight feeding chambers. Thefeedstock used in the trial runs was MSW collected in Israel, with acomposition as shown in Table 1. Proximate and ultimate analyseswere made on a sample of this MSW; the results are shown inTable 2.Fig. 1. Illustration of the PGM plant.
2 Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
2.4. Test procedure
In the trial runs, two groups of tests were carried out. The firstwere with air gasification of MSW (Cases 1 and 2), and the secondwere with air and steam gasification (Cases 3–6). The feed rate ofMSW was set at 300 kg/h during all runs. Trial runs were con-ducted with different operating parameters, such as plasma power,secondary air feed rate and steam feed rate, as shown in Table 3.Before each run, the reactor was preheated for 12 h with plasmaair.
3. Results and analysis
3.1. Syngas quality in air gasification
As basic cases, two tests were performed without steam feeds(Cases 1 and 2). In both cases, the plasma power is 2.88 MJ/kgMSW. In Case 1, the secondary air flow rate was set to zero. In Case2, the feed rate of secondary air was calculated by assuming thatthe total air feed rate equals the stoichiometric demand for con-verting all fixed carbon in the feedstock into CO. The results fromboth cases are presented in Figs. 3 and 4. Here, the amount of avail-able air per kilogram of MSW is represented by the equivalence ra-tio (ER), defined as:
ER ¼ ðA=FÞðA=FÞstoic
; ð3Þ
where A/F is the air/fuel mass ratio in the real cases and (A/F)stoic isthe air/fuel mass ratio for a stoichiometric combustion where thefuel is fully combusted.
Both cases showed good results in terms of the lower heatingvalue (LHV) of the product syngas, here 6–7 MJ/Nm3. This wasmainly due to the low ER in the PGM process (0.08 in Case 1 and0.12 in Case 2). A low ER ratio prevents the dilution of syngas withnitrogen from air; a substoichiometric oxygen level suppresses theformation of CO2, which is the other main noncombustible gas insyngas. Due to the concentration of combustible gases in the syn-gas, the total gas yields here (0.67 and 1.06 Nm3/kg MSW in Cases1 and 2, respectively) were lower than that of traditional gasifica-tion. In both cases, the H2/CO ratio was approximately 1.5, which issomewhat higher than that of common gasification processes,mainly due to the high contents of hydrogen and oxygen in thefeedstock.
Despite the common features of the two cases, some importantdifferences in syngas composition and yield were found. Firstly, asignificant increase of gas yield was observed when a higher ERwas used. This increase was partly due to the N2 content of the sec-ondary air, which led to the decrease of LHV in Case 2, and partlydue to the cracking of tar favored by the higher temperature asso-ciated with increasing ER. Fig. 5 is the measured temperature dis-tribution along the reactor chamber of Cases 1 and 2. It can befound that the temperature of Case 2 is 100–200 �C higher thanthat of Case 1, with an exception near the bottom. The bottom
Fig. 2. Typical schematics of a PGM gasifier.
Table 1Feedstock composition.
Component Weight percentage
Paper 50.0%Sawdust 3.0%Wood 11.0%Vegetation 3.7%Cloth 5.5%Plastics (polyethylene, PET, PVC) 10.0%Rubber 3.5%Ion-exchange resins 1.0%Electrical and electronic components 3.0%Debris 4.5%Glass 4.0%
Table 2Feedstock proximate and ultimate analyses.
Proximate analysis
Moisture 30.0%Fixed carbon 10.7%Volatiles 77.6%Ash 11.7%
Ultimate analysis (dry basis)Carbon C 50.5%Hydrogen H 5.6%Oxygen O 30.7%Nitrogen N 1.1%Chlorine Cl <0.1%Sulfur S 0.3%
Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx 3
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
temperature of Case 2 is lower than that of Case 1 due to the addi-tional low temperature air injection. Another reason for the in-crease in gas yield from Case 1 to Case 2 was likely the
insufficient carbon conversion in Case 1, as shown by chemical-equilibrium calculations. A low ER ratio in the PGM ensures highsyngas quality. However, when the ER is too low, as in Case 1,the gasification agent cannot supply enough oxygen to convertchar into CO or CO2. Insufficient carbon conversion is an adversecondition for gasification as it reduces both the gas yield and en-ergy efficiency. In Case 2, the feeding of secondary air solved thisproblem. The increased O2 enhanced the CO content in syngas, asshown in Fig. 3. An interesting result from the two cases is thatalthough the syngas yield changed significantly, the volume frac-tion of H2 in the syngas was relatively unchanged. This indicatesan increase of H2 production with increasing ER. The positive effecton H2 yield in response to increased ER is in accordance with theresults by Pinto et al. [32] and Anna et al. [13]. The increased H2
production with increased ER was likely the result of favorableconditions for the secondary pyrolysis of primary tar. Accordingto the Boroson’s theory [33], pyrolysis can be divided into twosteps: primary pyrolysis and secondary pyrolysis. H2 is mainly pro-duced from the secondary pyrolysis step, which is sensitive topyrolysis temperature. A higher temperature due to increasingER thus favored secondary pyrolysis, and more H2 was produced.Another effect of increasing temperature with ER was the decreasein total light hydrocarbon (THC) content. The relationship betweenpyrolysis temperature and THC yield has been reported by Anhet al. [34] and Li et al. [35], and the mechanism was explained byAnthony et al. [36].
3.2. Syngas quality in air and steam gasification
3.2.1. Influence of steam feed rateAlong with Case 2, experiments for Cases 3 and 4 were per-
formed to investigate the influence of the steam feed rate. Theplasma and air settings of Cases 3 and 4 were the same as for Case2 but with different steam feed rates (70 and 100 kg/h, respec-tively). The results from Cases 2, 3, and 4 are presented in Figs. 6and 7.
We found that adding high-temperature steam is favorable forthe PGM process. The total gas yield increased significantly, andthe gas LHV also increased with steam feeding. Generally, it is be-lieved that the increase of gas yield with steam feeding is due tothe water–gas shift reaction:
COþH2O ¼ CO2 þH2: ð4ÞThis reaction undoubtedly played some role in the yield in-
crease, especially with excess steam. However, the increase ofLHV in our cases cannot be explained solely by this reaction. Exam-ining the composition of the syngas, we found that as the steamfeed rate was increased from Case 2 to 4, the THC content increasedsignificantly. The CO content, in contrast, increased from Case 2 toCase 3 but decreased from Case 3 to Case 4. In all three cases, thefluctuation in H2 content was very small. Similar results were re-ported by Blasiak et al. [37], who studied the high-temperatureair and steam gasification of biomass in an updraft fixed-bed gas-ifier. A possible explanation for this phenomenon is the steam
Table 3Operating parameters for trial cases.
Case number MSW flow rate (kg/h) Plasma power (KW) Plasma air (kg/h) Air injection (kg/h) Steam injection(kg/h) Steam temperature (�C)
1 300 240 120 0 0 10002 300 240 120 60 0 10003 300 240 120 60 70 10004 300 240 120 60 100 10005 300 240 120 35 70 10006 300 260 130 13 70 1000
Fig. 3. Syngas composition of Cases 1 and 2.
Fig. 4. Syngas characteristics of Cases 1 and 2.
Fig. 5. Measured temperature distributions of Cases 1 and 2.
4 Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
reforming of tar at high temperature. The mechanism and kineticsof tar steam reforming have been reported by Li et al. [38]. The glo-bal reaction can be written as follows:
CmHn þ aH2O ¼ bCOþ cH2 þ dCxHy; ð5Þ
where CmHn represents tar and CxHy the light hydrocarbons.A strict restriction of the steam-reforming reaction is that it can
only occur at high temperature. It was pointed out by Jess [39] thatat a temperature of approximately 1200 �C, the steam reforming oftar can go to completion in less than 10 s. As we measured duringthe tests, the global gasification temperature was in the range of1000–1200 �C. Considering the scale of the reactor, it is very likelythat there was a strong steam-reforming reaction during the airand steam gasification process. Therefore, the steam reforming oftar and the water–gas shift reaction together resulted in the in-creased syngas yield.
Comparing the results of Cases 3 and 4, a notable difference isthat the H2/CO ratio increased greatly. This may be due to the pro-motion of the water–gas shift reaction by the excess steam in Case4. According to Li et al. [38], the priority of tar steam reforming ishigher than the water–gas shift reaction at high temperature dueto the occurrence of the following reaction:
CmHn þmCO2 ¼ 2mCOþ n=2H2: ð6ÞWhen there is insufficient steam for reforming, as in Case 3, the
water–gas shift reaction is, in a sense, suppressed. When steam issupplied in excess, the water–gas shift reaction then becomesmuch more intensive, resulting in a high H2/CO ratio.
3.2.2. Influence of plasma power and ERThe experiments in Cases 5 and 6 were conducted to investigate
the influence of plasma power and ER in air and steam gasification.In both cases, the steam feed rate was 70 kg/h, the same as in Case3; the plasma power and ER were then varied. Detailed informa-tion for each case is given in Table 3. The results from Cases 5and 6 are presented in Figs. 8 and 9 with Case 3 for comparison.
The results for Cases 3 and 5 were similar. The slight differencein gas yields can be explained by the reduced air feed and lower tarcracking and reforming related to the reduced combustion due tothe lower ER. However, some significant differences in syngas com-position were found when comparing Case 6 with Cases 3 and 5 ta-ken together. The overall increase of combustible gases, especiallyof H2, may be mainly due to the sensitivity of tar cracking andreforming to temperature. In the PGM process, more than half ofthe energy need for gasification is from the plasma torches. The in-creased plasma power in Case 6 led to a significant increase of gas-ification temperature, which prompted the cracking and reformingof tar. The reforming reaction of light hydrocarbons may have alsotaken place, as in Eq. (7), enhancing the yield of H2.
CxHy þ xH2O ¼ xCOþ ðxþ 2=yÞH2 ð7Þ
3.3. Energy efficiency
The cold-gas efficiency (CGE) is a standard criterion frequentlyquoted to express the energy efficiency of a gasification process.For a PGM process, the definition of CGE was modified by Qinglinet al. [40] to.
Fig. 6. Syngas compositions of Cases 2, 3 and 4.
Fig. 7. Syngas characteristics of Cases 2, 3 and 4.
Fig. 8. Syngas compositions of Cases 3, 5 and 6.
Fig. 9. Syngas characteristics of Cases 3, 5 and 6.
Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx 5
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
g ¼ _msyngas � LHVsyngas
_mfeedstock � LHVfeedstock þ Psteam þ Pplasma� 100%; ð8Þ
where _msyngas and _mfeedstock denote the mass flow rates of the syngasand the feedstock, respectively, while LHVsyngas and LHVfeedstock aretheir lower heating values on a mass basis. Psteam denotes the powerused to heat the steam, and Pplasma is the plasma power.
The combustion value of the MSW was calculated from anempirical expression given by Boie [41]:
HHV ¼ 83:22Cþ 274:3H� 25:8Oþ 15N ¼ 9:4Clþ 65P: ð9ÞThe CGE results of all six cases are shown in Fig. 10. Here, the
CGE varies from 30% to 60%. The energy efficiency of air gasifica-tion is lower than that of air and steam gasification. The CGE wasthe lowest for Case 1; Case 6 had the highest. For air gasification,increasing ER was beneficial for increasing energy efficiency,whereas the influence of ER was not obvious for air and steamgasification.
There are three main sources of energy loss in gasification: thechemical energy in the tar, the sensible heat of the syngas, and theheat loss of the system. Because a PGM reactor is an updraft fixed-bed reactor, the sensible heat of the syngas cannot be the main en-ergy loss. The normal system heat loss is approximately 2–5% ofthe total energy, so the main energy loss for PGM should be chem-ical energy in the tar.
4. Slag properties
The inorganic components of MSW were melted to form a slag.The discharging of slag was not continuous in the trial reactor. In-stead, it was controlled by a valve placed at the exit of the meltingchamber. At most times, the valve is closed, and the volume of slaginside the combustion chamber increases continually. When itreaches a certain level, the valve is opened. The melted slag flowsout of the melting chamber and into the slag collector. When thecollector is full, the valve is closed again, and the collector is trans-mitted into the open air for cooling. When it exits the meltingchamber, the slag is a glowing liquid. After four hours of cooling,it becomes a black, vitreous solid. The appearance of the slag dur-ing tap-out and after cooling is shown in Figs. 11a and 11b.
During these tests, the output of slagwas approximately 25 kg/h.Due to the high density of this slag (2300 kg/m3), the volume ratio ofslag to raw MSW is approximately 1:50. The composition can varywith the feedstock, but the main contents should be SiO2 and CaO.Undesirable materials such as heavy metals are locked in the slag,so that the slag is virtually inert, meeting themost demanding TCLPStandards. This slag can be used as a good building material.
5. Conclusions
The syngas produced from the PGM has a high LHV (6–7 MJ/Nm3).In air gasification, the syngas yield increased significantlywith increasing ER, whereas the LHV decreased slightly.
Feeding high-temperature steam into the PGM reactor greatlyincreased syngas yield, with even higher gas LHV. The feeding ofhigh-temperature steam can further reduce the air demand forgasification.
The energy efficiency of air and steam gasification was muchhigher than that of air gasification. The CGE of PGM air and steamgasification can reach approximately 60%. Tar formation representsthe main energy loss for the PGM reactor.
References
[1] Sakai S, Sawell SE, Chandler AJ, Eighmy TT. World trends in municipal solidwaste management. Waste Manage 1996;16(5–6):341–50.
[2] Belevi H, Baccini P. Long-term behavior of municipal solid waste landfills.Waste Manage 1989;7(1):43–56.
[3] Kristina H. Role of a district-heating network as a user of waste-heat supplyfrom various sources - the case of Goteborg. Appl Energy2006;83(12):1351–67.
[4] Marcelo RH, Jose A, Perrella B. Cogeneration in a solid-wastes power-station: acase-study. Appl Energy 1999;63(2):125–39.
Fig. 10. Cold-gas efficiency.
Fig. 11a. Appearance of slag during tap-out.
Fig. 11b. Appearance of slag after cooling.
6 Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
[5] Gordon M. Dioxin characterisation formation and minimisation duringmunicipal solid waste (MSW) incineration: review. Chem Eng J2002;86(3):343–68.
[6] Suksankraisorn K, Patumsawad S, Fungtammasan B. Combustion studies ofhigh moisture content waste in a fluidised bed. Waste Manage2003;23(5):433–9.
[7] Carlton CW. Municipal solid waste combustion ash: State-of-the-knowledge. JHazard Mater 1996;47(1–3):325–44.
[8] Floyd H, Anthony L. Analysis of heavy metal emission data from municipalwaste combustion. J Hazard Mater 1996;47(1–3):77–102.
[9] Lawrence AR. Energy from municipal solid waste: a comparison with coalcombustion technology. Prog Energy Combust Sci 1998;24(6):545–64.
[10] Suehiro O, Yasufumi M, Atsushi T. Estimation of energy recovery and reductionof CO2 emission in municipal solid waste power generation. Sci Total Environ1997;198(2):123–33.
[11] Thomas M. Novel and innovative pyrolysis and gasification technologies forenergy efficient and environmentally sound MSW disposal. Waste Manage2004;24(1):53–79.
[12] Lucas C, Szewczyka D, Blasiaka W, Mochidab S. High-temperature air andsteam gasification of densified biofuels. Biomass Bioenergy2004;27(6):563–75.
[13] Anna P, Sylwester K, Blasiak W. Effect of operating conditions on tar and gascomposition in high temperature air/steam gasification (HTAG) of plasticcontaining waste. Fuel Process Technol 2006;87(3):223–33.
[14] Anna P, Yang W, Lucas C. Development of a thermally homogeneous gasifiersystem using high-temperature agents. Clean Air 2006;7(4):363–79.
[15] Ahmed I, Gupta AK. Evolution of syngas from cardboard gasification. ApplEnergy 2009;86(9):1732–40.
[16] Nipattummakul N, Ahmed I, Kerdsuwan S, Gupta AK. High temperature steamgasification of wastewater sludge. Appl Energy 2010;87(12):3729–34.
[17] Ahmed I, Nipattummakul N, Gupta AK. Characteristics of syngas fromco-gasification of polyethylene and woodchips. Appl Energy2011;88(1):165–74.
[18] Ahmed I, Gupta AK. Characteristics of cardboard and paper gasification withCO2. Appl Energy 2009;86(12):2626–34.
[19] Young NC, Seong CK, Kunio Y. Pyrolysis gasification of dried sewage sludge in acombined screw and rotary kiln gasifier. Appl Energy 2011;88(4):1105–12.
[20] Okada T, Tojo Y, Tanaka N, Matsuto T. Recovery of zinc and lead from fly ashfrom ash-melting and gasification-melting processes of MSW – Comparisonand applicability of chemical leaching methods. Waste Manage2007;27(1):69–80.
[21] Thomas WM, Brendan PM, William RS, Steven J, Stanley EM. Fate of heavymetals and radioactive metals in gasification of sewage sludge. Waste Manage2004;24(2):193–8.
[22] Park YJ, Heo J. Vitrification of fly ash from municipal solid waste incinerator. JHazard Mater 2002;91(1–3):83–93.
[23] Sakai SI, Hiraoka M. Municipal solid waste incinerator residue recycling bythermal processes. Waste Manage 2000;20(2–3):249–58.
[24] Ecke H, Sakanakura H, Matsuto T, Tanaka N, Lagerkvist A. State-of-the-arttreatment processes for municipal solid waste incineration residues in Japan.Waste Manage Res 2000;18(1):41–51.
[25] Jung CH, Matsuto T, Tanaka N. Behavior of metals in ash melting andgasification-melting of municipal solid waste (MSW). Waste Manage2005;25(3):301–10.
[26] Koutaro K, Tomonori A, Yoshihito K, Ryoji S. Melting municipal solid wasteincineration residue by plasma melting furnace with a graphite electrode. ThinSolid Films 2001;386(2):183–8.
[27] Ryo Y, Makoto N, Hiroshi M. Influence of ash composition on heavy metalemissions in ash melting process. Fuel 2002;81(10):1335–40.
[28] Shinichi S, Masakatsu H. Municipal solid waste incinerator residue recyclingby thermal processes. Waste Manage 2000;20(2-3):249–58.
[29] Gomez E, Amutha RD, Cheeseman CR, Deegan D, Wise M, Boccaccini AR.Thermal plasma technology for the treatment of wastes: a critical review. JHazard Mater 2009;161(2–3):614–26.
[30] Moustakas K, Datta D, Malamis K, Loizidou M. Demonstration plasmagasification/vitrification system for effective hazardous waste treatment. JHazard Mater 2005;123(1–3):120–6.
[31] Moustakas K, Xydis G, Malamis S, Haralambous KJ, Loizidou M. Analysis ofresults from the operation of a pilot plasma gasification/vitrification unit foroptimizing its performance. J Hazard Mater 2008;151(2-3):473–80.
[32] Pinto F, Franco C, André RN, Miranda I, Gulyurtlu I, Cabrita I. Co-gasificationstudy of biomass mixed with plastic wastes. Fuel 2002;81(3):291–7.
[33] Boroson ML, Howard JB, Longwell JP, Peters WA. Product yield and kineticsfrom the vapour phase cracking of wood pyrolysis tars. AIChE J1989;35(1):120–8.
[34] Anh NP, Changkook R, Vida NS, Jim S. Characterisation of slow pyrolysisproducts from segregated wastes for energy production. J Anal Appl Pyrol2008;81(1):65–71.
[35] Li AM, Li XD, Li SQ, Ren Y, Shang N, Chi Y, et al. Experimental studies onmunicipal solid waste pyrolysis in a laboratory-scale rotary kiln. Energy1999;24(3):209–18.
[36] Anthony D, Sylvie V, Pierre C, Sebastien T, Guillaume B, Andre Z, et al.Mechanisms and kinetics of methane thermal conversion in a syngas. Ind EngChem Res 2009;48(14):6564–72.
[37] Blasiak W, Szewczyk D, Lucas C. Reforming of biomass wastes into fuel gaswith high temperature air and steam. In Pyrolysis & Gasification of Biomass &Waste. Strasbourg, France; 2002.
[38] Li C, Kenzi S. Tar property, analysis, reforming mechanism and model forbiomass gasification—an overview. Renew Sustain Energy Rev2008;13(3):594–604.
[39] Jess A. Mechanisms and kinetics of thermal reactions of aromatichydrocarbons from pyrolysis of solid fuels. Fuel 1996;75(12):1441–8.
[40] Qinglin Z, Liran D, Kentaro U, Yang W, Blasiak W. Process modelling andperformance analysis of a PGM gasifier. In: 10th Conference on Energy for aClean Environment, Lisbon, Portugal; 2009.
[41] Boie W. Energietechnik 1953;3:309–16.
Q. Zhang et al. / Applied Energy xxx (2011) xxx–xxx 7
Please cite this article in press as: Zhang Q et al. Gasification of municipal solid waste in the Plasma Gasification Melting process. Appl Energy (2011),doi:10.1016/j.apenergy.2011.01.041
SUPPLEMENT II
Qinglin Zhang, Liran Dor, Weihong Yang, Wlodzimierz Blasiak.
Properties and optimizing of a plasma gasification & melting
process of municipal solid waste
Paper #58 in the Proceedings of International Conference of Thermal Treatment
Technology & Hazardous Waste Combustors (IT3/HWC). San Francisco, USA, May,
2010. May 10-13, 2011, in Jacksonville, FL
Properties and Optimizing Of a Plasma Gasification
& Melting Process of Municipal Solid Waste
Paper 58
Qinglin Zhang1, Liran Dor2, Dikla Fenigshtein2, Weihong Yang1, Wlodzimierz Blasiak1
1Energy and Furnace Technology Division, Royal Institute of Technology,Brinellvägen 23, S-10044 Stockholm, Sweden2Environmental Energy Resources Ltd, 7 Jabotinski St., 52520 Ramat-Gan, Israel
ABSTRACT
A new solid waste treatment method called Plasma Gasification & Melting (PGM) has been developed by Environmental Energy Resources Ltd. (EER). In this technology, high temperature plasma air and steam are used to convert the waste into high-quality combustible syngas and vitreous benign slag. Due to the special features of the technology it is applicable for various stream of the solid waste field such as MSW, Medical Waste (MW) and Low Level Radioactive Waste (LLRW), where the technology was derived from. The aim of this study is to discuss the characteristics of this technology, and find out the optimal operation condition for a PGM plant.
A simulation model of the PGM process was built up and validated by the test results of a PGM demonstration plant. The result shows that the syngas LCV of PGM is much higher than that of traditional gasification. For air gasification, there exists a lower limit of air/MSW mass ratio for 100% conversion of MSW. When the air/MSW mass ratio exceeds the limitation, the syngas LCV will descend by dilution of CO2 and N2. The tar yield will decrease, because of higher pyrolysis temperature. For air/steam gasification, high temperature steam as gasification agent can reduce the limitation of air/MSW mass ratio, so further enhance the syngas quality. The influence of plasma power will be more prominent for air/steam gasification than air gasification. Based on above discussion, an optimizing conception design aiming at producing syngas with high LCV and energy efficiency of a PGM process is suggested.
INTRODUCTION
Increasing municipal solid waste (MSW) yield is one of the main by-products of economical development of human society. Among various methods of waste disposal,
gasification is a promising technology. Compared to direct incineration, gasification can prevent the formation of dioxins, as well as other gaseous contaminations likeNOx, SOx, and HCl 1. Another advantage of gasification is that it can produce a combustible gas mixture called syngas, which can be widely used for energy generation and chemical engineering 2. The volume of waste can be sharply reducedby gasification since the organic components release from waste during the gasification process. It has been confirmed that gasification is an advanced waste disposal which has advantages in both environment protection and energy generation aspects 1-4.
The application of plasma in gasification has been a hot topic in recent years 5-7.When plasma technology is applied in gasification, a number of unique advantagescould be found. For example, the sensible heat of plasma flow can provide high temperature conditions to accelerate the gasification reactions; some reactions which would not appear in conventional gasification can take place; the request of gasification air can be reduced so that the syngas quality can be improved. If plasma flow is injected from the exit of gasification residual, the residual can be melted, and forming a vitrified slag in which undesirable materials like heavy metals will betrapped 8-11. Due to these special features, plasma gasification is especially applicable for the treatment of various stream of the solid waste such as MSW, Medical Waste (MW) and Low Level Radioactive Waste (LLRW). The start-of-the-art of the plasma gasification technology for waste treatment was introduced by Gomez etc 12. Some laboratory scale result of plasma gasification was reported by Moustakas etc13.
An industrial scale plasma gasification plant was built up in Israel by the Environmental Energy Resources Israel Ltd. (EER). The aim is to demonstrate the features of a new plasma gasification technology called Plasma Gasification Melting (PGM). This design differs most significantly from other plasma gasification processes in its structure. As we can see in Figure 1, the PGM gasifier is an updraft Fixed-bed gasifier, with plasma air injected into the gasifier from the bottom part,whereas other designs are mainly fluidized bed. The advantage of this structure is obvious: the system is more compact; no milling of feedstock is needed before feeding; the energy efficiency of the PGM process is higher than that of fluidized bed since the temperature at the exit of syngas is much lower. Some information about the PGM gasification technology has been reported in previous publications 14-16.
Figure 1. Structure of the PGM gasifier
As famous process simulation software, Aspen Plus has been widely used to simulate different solid-fuel thermal chemical processes like fluidized bed combustion 17,fluidized bed gasification 18, co-generation 19, waste incineration 20, and fixed-bed gasification 21-22. These successful works ensure the capability of Aspen Plus in simulation and optimization of PGM process. In this work, a new fixed-bed gasification model was built to simulate the characteristics of a PGM plant. The model is validated by the measured data. Then, the influences of important operation parameters for PGM process were analyzed. The optimum operation conditions were determined by both the modeling results and feasibility in a real gasification process.
KEY OPERATION PARAMETERS IN GASIFICATION
The aim of optimization of a gasification process should be not only find out theoperating conditions under which the best-quality syngas can be produced, but also considering the practicality of the numerically best conditions in a industrial gasification process. So the understanding of the influence and relationship of key operation parameters is the basis of the optimization of a gasification process. In the PGM process, there are three basic parameters which determine the characteristics. They are equivalence ratio (ER), steam feedstock mass ratio (SFMR), and the sensible heat input per kilo feedstock (SH).
The gasification characteristics and syngas composition are mainly influenced by operation parameters and feedstock properties. Among various operation parameters, the equivalence ratio is the most basic parameter of a gasification process. It is commonly used to indicate quantitatively the extent of partial oxidization. The definition of equivalence ratio is as:
Equation 1. Definition of equivalence ratio
� � stoicFA
FA/
/�.
where:
� FA / = the mass ratio of air and feedstock in the real gasification process.
� stoicFA / = the mass ratio of air and feedstock for a stoichiometric oxidization
For an air gasification process, the ER should be controlled in an appropriate range. Ifit is too low, there will be no enough agents to ensure full conversion of fixed-carbon in feedstock into gases by partial oxidization. The partial oxidization also suppliesenergy for pyrolysis and drying of feedstock. However, if the ER is too large, the char combustion is more preferable to form carbon dioxide rather than carbon monoxide,so that the energy efficiency and syngas LHV may decrease. Nitrogen in feeding air will also dilute the syngas. In an optimal gasification process, the ER ratio should be around the point where 100% fixed-carbon conversion is satisfied, so as to ensure best gasification gas production.
Feeding steam into the system is a promising method to improve air gasification process. Since both air and steam are used as gasification agents in this process, it can be named air&steam gasification. The advantage of feeding steam is obvious: two moles of fuel gas could be produced from one mole of carbon by the water shift reaction. Steam is easy to remove from syngas, and no dilution of syngas will be made. Since the characteristics of air&steam gasification is also highly related to steam feeding rate, The mass ratio of steam and feedstock (SFMR) can be seem as another key factor if steam is feed into gasification system.
Preheating the gasification agents up to a high temperature has been a promisingtechnology to enhance both syngas quality and gasification efficiency. It has been proved that when the gasification agents are preheated to a relatively high temperature, significant advantages can be achieved, such as less tar production, higher energy concentration in syngas and higher energy efficiency. Moreover, the gasification system becomes relatively insensitive to the variations in particle size, heating value and moisture content typically associated with feedstock. The sensible heat input from agents is the parameter which is used to express the sensible heat carried by feeding agent. In the PGM process, this parameter includes two parts: the sensible heat of plasma flow, and the sensible heat of preheated steam. In this work, it is expressed by sensible heat input per kilo feedstock (SH).
THE MATHEMATIC MODEL
In this work, the PGM process was simulated with a steady model with Aspen Plus.
The whole PGM process was divided into four modules: drying, pyrolysis, gasification and combustion, and melting. Moisture, volatiles, char and ash components are removed from the solid phase by these modules, respectively. The summary of the flowsheet of the model is shown in Figure 2.
Figure 2. GPM gasification flowsheet
Drying
The drying process is modeled using two main blocks: a heat exchanger and an evaporator. Original MSW is heated by exchanging the sensible heat of the syngas in the exchanger, and then separated into steam and dry MSW in the evaporator. DryMSW is sent to the pyrolysis zone, while steam mixes with the syngas from pyrolysissection.
Temperature change of MSW and gases is calculated from mass and energy conservation equations. The mass balance is described as following:
Equation 2. Mass balance of the drying process
steamdryMSWwetMSW mmm �� ��
where:
The ratio of dryMSWm � to steamm is given by the proximate analysis of MSW.
Energy balance of heat exchanger is described as:
Equation 3. Energy balance of the drying process
���
����
���� �
�
�
�
�
�
�
�� steam
T
Tsteampsteam
T
TdryMSWpdryMSW
i
T
Tipi LdTCmdTCmdTCm
outsyngas
inMSW
outMSW
inMSW
insyngas
outsyngas
,,,
Where:
ipC , = the specific heat capacity of the component i
steamL = the latent heat of the component i
inMSWT � = the temperature of MSW flowing intothe drying zone
insyngasT � = the temperature of syngas flowing into the drying zone
outMSWT � = the temperature of dry MSW flowing out of the drying zone
outsyngasT � = the temperature of syngas flowing out of the drying zone
In this equation, left hand and right hand describe the change of the sensible heat of product gas and MSW.
According to the definition of drying, outMSWT � should not be too higher than 100 C,
because the boiling temperature of water is 100 C at the pressure of 1 bar. Considering the impact of heat gradient inside MSW particles, we assume that the temperature of dried MSW and steam is 120 C.
Pyrolysis
Compared with coal, MSW have higher content of volatiles. For an updraft gasifier model, the pyrolysis process is especially important because most of the gas and tar yield in this section will join the gas produced in the char gasification section and be released from the outlet of the gasifier without further reactions.
The composition of MSW can be divided into two main series: cellulosic fractions (Wood, paper, vegetation and cardboard) and plastics (PE, PP, PVC and rubber). The pyrolysis characters of each series are different. In this model, the pyrolysis of each series was simulated with different submodel, as was shown in Figure 3. For each series, a two-step pyrolysis process was used.
Figure 3a. Pyrolysis model for cellulosic fractions
Figure 3b. Pyrolysis model for Plastics
It has been proved that for an updraft gasifier, the tar production is sensitive to the pyrolysis temperature. In this model, the extent of secondary pyrolysis is controlledby pyrolysis temperature:
Equation 4. The extent of secondary pyrolysis
))(exp( 0TTAYT ���
where:T = pyrolysis temperature
Plastics
Char Primary tarCmH2m
Primary gasesCO, H2, CH4, C2H4
Second tarC6H6
Secondary gasesCH4, C2H4
Cellulosic fractions
Char Primary tarC6H10,71O3,264
Primary gasesCO, CO2, H2O
Second tarC6H6
Secondary gasesCO, CO2, H2, CH4
CT �5000 � =the temperature at which the maximum yield of tar will be produced
Char was considered as a mixture of ash and fixed carbon, so the yield and composition of char can be calculated directly from the ultimate and proximate analysis of MSW.
Heating values of MSW, tar and char are calculated with the HCOALGEN model,which includes a number of empirical correlations for heat of combustion, heat of formation and heat capacity, while densities of MSW and char are calculated using the DCOALIGT model.
Gasification and Combustion
Char coming from the pyrolysis zone will meet and react with gasification agents
( OH2 and 2O ) in the gasification and combustion section. Lots of chemical
reactions are involved in this process.
For a moving bed gasification process, the residual time for solid fuel is very long, so the chemical reactions occurring in this zone can be considered as at chemicalequilibrium.
The second law of thermodynamics can be expressed as:
Equation 5. Second law of thermodynamics
� 0,, #mPTdG
It states that the Gibbs function always decreases for a spontaneous, isothermal, isobaric change of a fixed-mass system in the absence of all work effects except boundary work. This principle allows us to calculate the equilibrium composition of a mixture at a given temperature and pressure.
The can be expressed as:
Equation 6. Gibbs function for a mixture of ideal gases
� $ %�� ��� 00,, /ln PPTRggNG iuTiTiimix
where:
iN = the number of moles of the ith species
Tig , = the Gibbs function of the pure species, the superscript 0 means properties at
standard pressure
For fixed temperature and pressure, the equilibrium condition becomes
Equation 7. The equilibrium condition for fixed temperature and pressure
0�mixdG
In our char gasification reaction, the main species to participate in the reaction
are OH2 , 2H , CO , 2CO , 4CH , 2O and C . C can be treated as the naturally
occurring element, and the other species can be treated as ideal gases.
Plasma Melting
The inorganic components (ash) of the MSW coming from the gasification and combustion zone were melted by high temperature plasma air in the plasma melting zone. The temperature of slag flowing out of the gasifier was calculated from energy balance by setting an appropriate temperature difference between air and slag. Nochemical reaction was considered in the melting process.
FEEDSTOCK PROPERTIES
Feedstock used by in the trial runs is MSW collected in Israel. Proximate and ultimate analysis has been made on the MSW sample, and the result is shown in Table 1.
Table 1. Feedstock propertiesProximate analysisMoisture 20.0 %Fixed carbon (dry basis)
10.7 %
Volatile (dry basis) 77.6 %Ash (dry basis) 11.7 %
Ultimate analysis (dry basis)Carbon C 47.9%Hydrogen H 6.0%Nitrogen N 1.2%Chlorine Cl <0.1%Sulphur S 0.3%Oxygen O 32.9%
MODEL VALIDATION
In order to validate the proposed model, 4 sets of data from the trial runs of the demonstrate plant in different operation conditions (see Table 2) were used. The simulation results in terms of syngas yields, volume content of H2, THC, and CO in syngas (wet basis) are compared with the measured results. Detailed results are presented in Figure 4 and Figure 5, respectively.
Table 2. Operation conditions of validation casesCase number ER Plasma power (kW) Steam injection (kg/h)1 0.060 240 702 0.060 240 1003 0.052 240 704 0.048 260 70
It was found that the model can relatively predict the gasification results for PGM process. The trend of syngas yields is just the same as measured values, but the syngas yield is a little predicted, but the differences between model results and measured values are less than 10%. This might be due to the equilibrium assumption in the gasification section, since full equilibrium can not be reached in a real gasification process. The difference of combustible components volume fractions with measured results are also mostly less than 10%.
Figure 4. Measured results of the gas yield and composition of the PGM gasification
Figure 5. Modelling results of the gas yield and composition of the PGM gasification
RESULT AND DISCUSSION
Influence of sensible heat from plasma
The most important significance of the PGM gasification, which differentiates it from other fixed-bed gasification process is the high-temperature plasma air injected from the bottom of the system. The influence of the sensible heat from plasma air is investigated using the above model. The plasma energy varies from 0 to 3MJ/kg feedstock. All calculations were carried out at 100% carbon conversion point.
Figure 6. Influence of SE of plasma on minimum ER
Figure 6 is the variation of ER number needed for 100% fixed-carbon conversion with sensible heat input. It could be found that with the increase of sensible heat (SH) from plasma, the change of ER can be divided into two different parts. When SH increases from 0 to 1.5 MJ/kg MSW, the ER decreases approximately linearly from 0.24 to 0.15.When the sensible heat keeps increasing, the ER will keep constant. This phenomenonmay be explained by low gasification temperature with low plasma power (see Figure 7). When the SE of plasma is lower than 1.5 MJ/kg MSW, the gasification is only 600-800 C. According to the report by J. R. Arthur, the ratio of CO and CO2 in char combustion is the function of temperature. When the temperature ranges from 450 to 900 C, the ratio can be expressed by the following equation23:
Equation 8. The function of CO and CO2 molar ratio
���
������
TCOCOr 6420exp2500
2
Increasing SE leads to increasing gasification, so that the char combustion trends to form CO, rather than CO2. The oxygen needed for combustion is then decreases. When SE exceeds 1.5 MJ/kg MSW, the gasification is higher than 800ºC. In that case, the partial oxidization of char will form mostly CO, so the ER required will be constant.
It can also be found that the increasing of gasification temperature can also be divided in to two parts by the point SE=1.5 MJ/kg MSW, this is mainly due to the difference of reaction heat of two combustion reactions. During the variation of SE, the syngas temperature does not change much. The syngas temperature nearly keep constant, with the exception at high SE (2.5-3.5), where there is a slightly increase.
Figure 7. Influence of SE of plasma on gasification and syngas temperature
The trend of syngas yield, together with syngas composition with different SE of plasma are shown in Figure 7 and Figure 8. It could be found that SE= 1.8 MJ/kg MSW is an important transition point where the trends of both syngas yield and LHV change. When SE increases from 0 to 1.8 MJ/kg MSW, the syngas yield decreasesfrom 1.6 to 1.2 Nm3/kg MSW. At the same time, syngas LHV increases from 1.44 to 3.30 MJ/Nm3. Looking at the syngas composition, the volume fraction of CO increase dramatically, while the fraction of CO2 decreases from 16% to 10%. The content of H2 also increases slightly. The decreasing air feeding may be the main reason for that.When SE exceeds 1.8 MJ/kg MSW, the syngas yield start to increase. When SE reaches 3.6, the syngas yield is about 1.5 Nm3/kg MSW. During this process, the syngas LHV increases from 3.30 to 7.61 MJ/Nm3. this could mainly be explained by cracking of tar with increasing pyrolysis temperature (see Figure 9).
It has been clear that compared to conventional gasification, PGM can significantly increases the syngas LHV. From the trend of the LHV, we can know that if the SH goes on increasing, the syngas LHV can be higher. However, this is not practical in a real gasification process. When SH equal to 3.6MJ/kg MSW, the gasification has been over 2000 ºC. This temperature has exceeds the upper limit of materials of the furnace. Obviously, simply increasing sensible heat to air gasification is not the best solution for the PGM optimizing.
Figure 8. Influence of SE of plasma on syngas yield and LHV
Figure 9. Influence of SE of plasma on syngas composition (dry basis)
Influence of steam feedstock mass ratio
It is known that the characteristics of PGM can be improved by addition of high temperature steam into the gasification section. The influences of steam / MSW mass ratio is investigated in this work.
Figure 10 is the relationship of ER requested for 100% carbon conversion and SFMR with different plasma sensible heat (1.8, 2.4, 3.0, and 3.6 MJ/kg MSW). It can be found that the influence of SFMR on ER value is negative. Theoretically, the air request can finally reduce to zero with increasing steam feeding. However, in a real PGM plant, the plasma energy has to be carried by air, so there exists a lower limitation of ER, depending on how much plasma energy is fed into the system. This relationship can be expressed as:
Equation 9. The relationship between ER and SH
SHER '� �
where � is a constant. The value of � varies with different feedstock.
Looking at the behaviour of different SH values, it can be found that when SE=1.8 MJ/kg MSW, the ER decreases linearly with SFMR, the ER reaches its lower limit (about 0.05) when SFMR=0.25. However, when SE=3.6 MJ/kg MSW, the ER will drop dramatically to its lower limit (about 0.1) when SFMR=0.05. The slope of SE= 3.0 MJ/kg MSW is similar to SE=3.6 MJ/kg MSW, while the other series (SE=2.4 MJ/kg MSW) is something between.
Figure 10. Influence of steam feedstock mass ratio on optimal ER value
Figure 11. Influence of steam feedstock mass ratio on gasification temperature
Figure 11 is the variation of gasification with steam feedstock mass ratio. It was found that generally, the influence of steam feeding on gasification temperature is negative. For high SH value (SH=3.6 and 3.0 MJ/kg MSW), this can be explained by the dilution of high temperature gas by steam, whose temperature is relatively low (1000 ºC). For SH= 2.4, and 1.8 MJ/kg MSW, the gasification temperature is lower than 1000 ºC, while the increasing of steam feeding also results in lower gasification temperature. It was believed that this phenomenon can be explained by water shift reaction aroused by steam feeding:
Equation 10. The water shift reaction
22 HCOOHC ��� KmolMJ /131�
It is also found that although the steam feeding can reduce the gasification temperature, the superfluous SH value is not appropriate. When SH=3.6 MJ/kg MSW, the gasification is about 1700-1900 ºC. Obviously no furnace material can endure such high temperature. The best gasification temperature should be get with the SH value between 2.4 and 3.0 MJ/kg MSW.
Figure 12. Influence of steam feedstock mass ratio on syngas yield
Figure 13. Influence of steam feedstock mass ratio on syngas LHV
The influence of SFMR on syngas yield and LHV value are shown in Figure 12 and Figure 13, respectively. It is obvious that for both syngas yield and syngas LHV, higher SH value is more advisable. This is mainly due to the tar cracking in high pyrolysis temperature (see Figure 14). Despising the decrease of syngas yield due to the reduction of ER at low steam feeding rate, it can be seen that the syngas yield will increase with increasing SFMR value. A possible explanation for this phenomenon is the steam reforming of tar in high temperature. The mechanism and kinetics of steam reforming of tar has been reported by Chunshan et al. 24. The global reaction can be written as following:
Equation 11. The steam reforming of tar
yxnm HdCcHbCOOaHHC ���� 22
where nmHC stands for tar, while yxHC stands for light hydrocarbons.
Figure 14. Influence of steam feedstock mass ratio on tar yield
Figure 15. Influence of steam feedstock mass ratio on energy efficiency
Figure 15 is the relation ship of SFMR on the electric efficiency of PGM process. Considering that MSW is a free feedstock, here the electric efficiency is defined as:
Equation 12. The definition of electric efficiency
� %100-
'�'�
plasma
gensteamsyngassyngas
PPLHVm �
��
where:
syngasm� = the mass flow rates of syngas
syngasLHV = the low heating values of syngas on mass basis
steamP = the heat used to heat up steam
plasmaP = the power of plasma
gen� = the efficiency of syngas in generating electricity, and here the gen� value is
assumed to be 50%.
It is found that the SH is the main parameter which influence the electric efficiency. For SH= 3.0MJ/kg MSW, despising the dramatic drop with low steam feeding level, the electric efficiency increases with steam feeding rate. The maximum electricefficiency can be obtained at SFMR=0.33, and the value of electric efficiency can reach 145%. This means that electricity generated from syngas combustion is more than that required by plasma generation. The PGM system can produce extra
electricity for outside users.
CONCLUSION
For air gasification, the injecting of plasma flow can reduce the ER required for 100% carbon conversion. When the SH value is 2.4 MJ/kg MSW, the ER required can be reduced by about 50%. At the same time, the syngas LHV can increase from 1.5 MJ/Nm3 to about 6 MJ/Nm3. the gas yield, then, is at the same level as conventional gasification. The reason is mainly tar cracking aroused by sensible heat from plasma flow. From the model results, further increasing of SH value can still enhance both syngas yield and LHV. However, this is not practical due to exorbitant gasification temperature.
Feeding of steam into PGM process can further improve the gasification properties. For SH=3.0 MJ/kg MSW, the ER requested can be reduced to 0.08, when SFMR value is over 0.06. Both syngas yield and syngas LHV will increase when SFMR value enhances. When SFMR=0.35, it seems both syngas yield and LHV can reach a maximum value, which are 1.28 Nm3/kg MSW, and 8.00 MJ/ Nm3, respectively. Thegasification temperature in this case is about 1250 C.
An electric efficiency is defined for the PGM process. In the optimal operation condition described before, the electric efficiency can reach 145%. The PGM plant can supply all electricity required by itself, and supply extra electricity for outside users.
REFERENCES
1. Thomas, M. Waste Manage. 2004, 24, 53-79.2. Sakai, S.; Sawell, S.E.; Chandler, A.J.; Eighmy, T.T. Waste Manage. 1996, 16,
341-350.3. Filippis, P.; Borgianni, C.; Paolucci, M.; Pochetti, F. Waste Manag. 2004, 24,
633-639.4. Choy, K.K.H.; Porter, J.F.; Hui, C.W.; McKay, G. Chem Eng. J. 2004, 105, 31-41.5. Koutaro, K.; Tomonori, A.; Yoshihito, K.; Ryoji, S. Thin Solid Films 2001, 386,
183-188.6. Ryo, Y.; Makoto, N.; Hiroshi, M. Fuel 2002, 81, 1335-1340.7. Shinichi, S.; Masakatsu, H. Waste Manage. 2000, 20, 249-258.8. Park, Y.J.; Heo, J. J. Hazard. Mater. 2002, 91, 83-93.9. Sakai, S.I.; Hiraoka, M. Waste Manage. 2000, 20, 249-258.10. Ecke, H.; Sakanakura, H.; Matsuto, T.; Tanaka, N.; Lagerkvist, A. Waste Manage.
Res. 2000, 18, 41-51.11. Jung, C.H.; Matsuto, T.; Tanaka, N.; Waste Manage. 2005, 25, 301-310.
12. Gomez, E.; Amutha, R.D.; Cheeseman, C.R.; Deegan, D.; Wise, M.; Boccaccini,A.R. J. Hazard. Mater. 2009, 161, 614-626.
13. Moustakas, K.; Xydis, G.; Malamis, S.; Haralambous, K.J.; Loizidou, M. J.Hazard Mater 2008, 151, 473-480.
14. Liran, D.; Qinglin, Z.; Dikla, F.; Weihong, Y.; Wlodzmierz, B. Gasification of Municipal Solid Waste in the Plasma Gasification Melting Process. ICAE 2010, Singapore, 2010.
15. Qinglin, Z.; Liran, D.; Kentaro, U. Weihong, Y.; Wlodzmierz, B. Process Modelling and Performance Analysis of A PGM Gasifier. 10th Conference on Energy for a Clean Environment, Lisbon 2009.
16. Sotudeh, G.R.; Legros, R.; Chaouki, J.; Paris, J. Fuel 1998, 77, 327-337.17. Mansaray, K.G.; Al-Taweel, A.M.; Ghaly, A.E.; Hamdullahpur, F.; Ugursal, V.I.
Energy Sources 2000, 22, 83-98.18. Zheng, L.; Edward, F. Energy Convers. Manage.2003, 44, 1845-1851.19. Silvano, C.; Marina, P.; Diego, B. Waste Manage. 2005, 25, 171-175.20. Rogers, R. Hydrogen Production by Gasification of Municipal Solid Waste.
Technical report of Lawrence Livermore National Lab., CA (United States), 1994.21. Vittorio, T.; Giorgio, C. Process Analysis and Performance Evaluation of Updraft
Coal Gasifier. Proceedings of the 3rd. International Conference on Clean Coal Technologies for Our Future; 15-17 May, 2007, Cagliari, Italy.
22. Christopher, H.; Maarten, B. Gasification. Elsevier, 2008.23. Arthur, J.R. Trans. Faraday Soc. 1951, 47, 164-178.24. Li, C.; Kenzi, S. Renew. Sustain. Energy Rev. 2008, 13, 594-604.
SUPPLEMENT III
Qinglin Zhang, Liran Dor, Weihong Yang, Wlodzimierz Blasiak.
An eulerian model for municipal solid waste gasification in a fixed-
bed plasma gasification melting reactor
Energy Fuels, 2011, 25 (9), pp 4129–4137
Published: July 07, 2011
r 2011 American Chemical Society 4129 dx.doi.org/10.1021/ef200383j | Energy Fuels 2011, 25, 4129–4137
ARTICLE
pubs.acs.org/EF
Eulerian Model for Municipal Solid Waste Gasification in a Fixed-BedPlasma Gasification Melting ReactorQinglin Zhang,*,† Liran Dor,‡ Weihong Yang,† and Wlodzimierz Blasiak†
†Energy and Furnace Technology Division, Royal Institute of Technology, Brinellv€agen 23, S-10044 Stockholm, Sweden‡Environmental Energy Resources, Limited, 7 Jabotinski Street, 52520 Ramat-Gan, Israel
ABSTRACT: Plasma gasification melting (PGM) is a promising waste-to-energy process, which provides many features superior tothose of conventional gasification. In this work, a steady Euler�Euler multiphase model is developed to predict the performance ofmunicipal solid waste (MSW) gasification inside a PGM reactor. The model considers the main chemical and physical processes,such as drying, pyrolysis, homogeneous reactions, heterogeneous char reactions, and melting of the inorganic components of MSW.The model is validated by one experimental test of a pilot reactor. The characteristics of PGM gasification, such as temperaturedistribution, syngas composition, tar yield, and energy conversion ratio (ECR, chemical energy of the gas phase divided by the totalenergy input), at the proposed condition are discussed. A total of nine cases are used to investigate the effects of the equivalence ratio(ER) and plasma power with a fixed flow rate of MSW. It is found that the ER has a positive effect on the cold gas efficiency of PGMgasification. However, the increase of the ER is restricted by the peak temperature. The influence of the plasma power then is notobvious for PGM gasification.
1. INTRODUCTION
Municipal solid waste (MSW) is one of the major potentialenergy sources. Among various waste-to-energy processes,gasification is recognized as an important technology.1 Inrecent years, various MSW gasification technologies have beendeveloped,2�4 while plasma gasification melting (PGM) is one ofthe promising technologies. In the PGM, gasification is com-bined with thermal plasma technology. High-temperature plas-ma torches are applied to provide both heat for melting theinorganic components and sensible heat for gasification. Incomparison to conventional gasification, the PGM technologyhas advantages, such as higher syngas lower heating value (LHV)and higher energy efficiency. Moreover, the PGM technologyprovides a wise solution for problems related to bottom ashbecause the inorganic components of MSW are melted and forma vitrified slag.
The characteristics of PGM were studied experimentally in apilot PGM reactor located in Yblin, Israel.5 The results show thatthe PGM is promising in treating MSW and producing high-quality syngas. However, it was also found that the syngas yieldand quality highly depend upon the operation parameters, suchas the air feeding rate and power of plasma torches. Consideringthe high expenses of running the pilot reactor, it is not practical tostudy the detail influences of different operation parameters only byexperiments. Essentially, the PGM reactor is an updraft fixed-bedgasifier. From this point of view, gasification models can be usedto predict the performance of PGM, thus helping the designingand optimizing of a PGM reactor. Several numerical models offixed-bed gasification have been found in the literature.6�10 Mostof them are one-dimensional models that do not consider theinfluence of the reactor structure and shape.6�9 Rogel andAguillon10 developed a one-dimensional (1D) + two-dimensional(2D) model, which solves the conservation equations of the gasphase in a 2D system and solves the conservation equations of
the solid phase in a 1D system. In the authors’ opinion, full 2Dsimulation for fixed-bed gasification was not investigated indetail.
The object of this work is to develop a 2D fixed-bed gasifica-tion model for the PGM process. This work is based on aprevious model study on fixed-bed gasification with high pre-heated air.11 In the presented model, the Euler�Euler multi-phase method is applied. The main chemical and physicalprocesses, such as drying, pyrolysis, homogeneous reactions,heterogeneous char reactions, and melting of the inorganiccomponents of MSW, are considered in this model. Ansys Fluent12.1 is used as a tool to do the above work.
Initially, simulation results are examined with measuredresults. Thereafter, the model is used as a tool to analyze theperformance of the system by varying different operation param-eters, such as the air feeding rate and plasma power. Attentionsare focused on the syngas composition, tar yield, and systemenergy efficiency.
2. PILOT REACTOR
A pilot PGM reactor is located in Yblin, Israel, with adesigned capacity of 12�20 tons of MSW per day. The mainscheme of the reactor is shown in Figure 1. The reactor iscomposed of two main parts: the reaction shaft and themelting chambers.
Plasma torches are placed at the ceiling of the meltingchamber. Primary air flows into the melting chamber throughthe torches, where it is ionized, therefore forming plasma jets,which extend beyond the tip of the torches. The temperature ofthe plasma jets reaches up to 6000 K. The plasma jets supply the
Received: March 11, 2011Revised: June 29, 2011
4130 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
necessary heat to melt the inorganic components of the MSWin the bottom of the reactor. Secondary air nozzles are placedaround plasma torches. Secondary air is injected throughsecondary air nozzles at room temperature. The flow rateof secondary air is adjustable; thus, the feeding rate of totalair can be controlled. An airtight feeding pipe is placed at thetop of the reaction shaft. MSW is fed into the reactor inter-mittently from the shaft top every half an hour. The total heightof the reaction shaft is 7.02 m, and the height of the fixed bedis 6.11 m.
To measure the temperature distributions inside the reactor,thermocouples were placed both along the gasifier shaft and inthe syngas conduit. A probe was placed in the syngas conduit toobtain syngas samples, which were sent to a gas analyzerfor composition analysis. The total syngas yield was then mea-sured by a flowmeter placed in the syngas conduit. Detailedinformation about the reactor can be found in our previouspublication.5
3. NUMERICAL MODEL
The Euler�Euler multiphase approach is applied in this work. Theconservation equations of mass, momentum, and energy are solved forboth gas and solid phases. Mass, momentum, and energy exchangesbetween phases are allowed. To ensure convergence of the model, theapproach is simplified by disregarding the gas�solid stress terms in themomentum equation of the solid phase. The turbulence of the gas phaseis simulated with the standard k�ε turbulence model.A total of 11 gaseous species and 5 solid species are defined in the
model. The main chemical and physical processes, such as drying,pyrolysis, homogeneous reactions, heterogeneous char reactions, andmelting of the inorganic components of MSW, are considered in thismodel. For heterogeneous char reactions, the reaction rates are calcu-lated using an unreacted shrinking core model. The reaction rates ofhomogeneous reactions are determined by considering both the kineticrates and turbulent mixing rates.The flow of the slag is ignored. The melting of inorganic components
of solid is simulated by setting mass and energy sinks for the solid phaseat the melting area, which is determined by experience.
3.1. Governing Equations. 3.1.1. Gas Phase. The Eulerian con-servation equations for species mass, momentum, and energy are solvedfor the gas phase. The equations are written as follows:12
∂
∂tðRgFgYiÞ þ ∇ðRgFgYi vBgÞ ¼ _mi þ Si ð1Þ
∂
∂tðRgFg vBgÞ þ ∇ðRgFg vBg vBgÞ ¼ �Rg∇p þ ∇¼
τg þ RgFg gB
þ Ksgð vBs � vBgÞ þ _msg vBsg ð2Þ
∂
∂tðRgFghgÞ þ ∇ðRgFg vBghgÞ ¼ �Rg
∂p∂t
þ ¼τg : ∇ vBg �∇ qBg
þ Sg þ Q sg þ _msghsg ð3Þ
3.1.2. Solid Phase. The solid phase is treated as an interpenetratingcontinuous phase. The continuity equation for the jth species in the solidphase is similar to that in the gas phase.
∂
∂tðRsFsYjÞ þ ∇ðRsFsYj vBsÞ ¼ _mj þ Sj ð4Þ
The momentum equation of the solid phase is written as
∂
∂tðRsFs vBsÞ þ ∇ðRsFs vBs vBsÞ
¼ �∇ps þ ∇¼τs þ RsFs gB� _msg vBsg ð5Þ
where ps and τCs denote the solid pressure and shear stress, which aredefined to express the normal and shear stress parts of solid-phase stress.The solid-phase stress is a function of the solid volume fraction. At afixed-bed condition, the value of rps, which is several orders ofmagnitude larger than the fluid�solid stress, becomes the main drivingforce of granular flow.13,14 In other words, the influence of thefluid�solid stress on solid motion can be ignored (the detailednumerical expressions of the solid-phase stress and fluid�solid stressare introduced in the next section). This idea was used by Johnson andJackson15 to describe nonreaction shearing granular flow. In the presentwork, the idea is also adopted, so that the fluid�solid stress term isdisregarded in the solid-phasemomentum equations. This simplificationis very helpful for the convergence of the solid momentum equationbecause it largely prevents the solution of interphase nonlinear terms,which is the main cause of nonconvergence for the Euler�Eulerapproach.
The energy equation of the solid phase is written as
∂
∂tðRsFshsÞ þ ∇ðRsFs vBshsÞ ¼ �Rs
∂p∂t
þ ¼τs : ∇ vBs �∇ qBs
þ Ss þ Q gs � _msghsg ð6Þ
3.1.3. Gas�Solid Stress. The gas�solid stress is ignored for thesolid phase. However, it is considered in the momentum equation ofthe gas phase. The gas�solid stress is simulated using the Ergunequation.16 The interphase momentum exchange coefficient Ksg iswritten as
Ksg ¼ 150Rsð1� RgÞμg
Rgds2 þ 1:756
FgRsj vBs � vBgjds
ð7Þ
3.1.4. Solid-Phase Stress. The solid phase stress is composed of twoparts: the normal stress part and the shear stress part. For fixed-bedgasification, the flow of the solid phase should be treated as a plasticflow.17 The normal stress part is expressed by solid pressure ps.
18
ps ¼ Rsp� ð8Þ
Figure 1. Scheme of the PGM reactor in the demonstration plant.
4131 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
p* is expressed by an empirical power law
p� ¼ AðRg � Rg�Þn ð9Þwhere Rg* is the gas volume fraction at minimum fluidization. Empiricalvalues of A = 1025 Pa and n = 10 are used.
For the shear stress part, only the frictional viscosity is considered.Because the flow of the solid phase is dense flow, where the solid volumefraction for the solid phase is near the packing limit, Schaeffer’sformulation19 of frictional viscosity is applied.
μs ¼ps sin ϕ
2ffiffiffiffiffiffiI2D
p ð10Þ
3.1.5. Interphase Heat Transfer. The intensity of heat exchangebetween the solid and gas phases is assumed to be a function of thetemperature difference between the solid and gas phases.
Q sg ¼ �Q gs ¼ ksgðTs�sTgÞ ð11ÞThe heat-transfer coefficient is written as
ksg ¼ 6kgRsRgNusds
2 ð12Þ
Here, Nus is the Nusselt number correlated by Gunn.20
Nus ¼ ð7� 10Rg þ 5Rg2Þð1 þ 0:7Res
0:2Prg0:33Þ
þ ð1:33� 2:4Rg þ 1:2Rg2ÞRes0:7Prg0:33 ð13Þ
3.2. Reaction Rates. 3.2.1. Drying. Drying is the first process totake place for feedstock during gasification. Despite its seemly simplicity,drying of feedstock is a complex combination of three steps: evaporationof free water, desorption and evaporation of absorbed water, andseparation of chemically bound water.21 In our work, a drying modelthat is popularly used in fixed-bed combustion or gasification of MSWand biomass22�26 is applied
r1 ¼ AvkmðCmoi � CH2OÞ when Ts < 100oC ð14Þ
or
r1 ¼ Qsg=Hevp when Ts g 100 oC ð15Þ
Themass-transfer coefficient km is calculated according to the Sherwoodnumber.27,28
km ¼ ShDd
ð16Þ
3.2.2. Pyrolysis. Pyrolysis is the thermal decomposition of solid fuels inthe absence of oxidizers. Because of the complexity in both reactionpaths and products generated, the detail kinetics of pyrolysis is stillunclear. Therefore, overall kinetics expressions are commonly used inpyrolysis models. In this work, a two-step pyrolysis model is applied. Thepyrolysis of MSW is divided into two steps. At first, MSW decomposesinto char, gas, and primary tar. Then, the primary tar decomposes intogas and secondary tar by thermal cracking.9
MSW f Rgas þ βprimary tar þ γchar ð17Þ
primary tar f gas þ secondary tar ð18ÞIt has been confirmed by experiments that the two-step pyrolysis modelcan correctly predict the pyrolysis yields, especially tar yields at variousconditions.29,30 It is very appropriate for modeling pyrolysis in theupdraft fixed-bed gasification because the tar problem is the mostsignificant in this constitution.
MSW is the mixture of different species. Generally, most of theorganic components of MSW can be divided into two groups: thecellulosic group (wood, paper, cardboard, textile, etc.) and the plasticgroup (polystyrene, polypropylene, polyethylene, polyvinyl chloride,31
etc.). The pyrolysis characters of these two groups are different. In thiswork, the differences are considered using individual pyrolysis kineticsfor each group (shown in Table 1). The interactions between species arenot considered in this model.
3.2.3. Heterogeneous Char Reactions.Heterogeneous char reactionsinvolved in this model include the following overall reactions:
C þ γ þ 1γ þ 2
O2 sfr1 2γ
γ þ 2CO þ γ
γ þ 2CO2 ð19Þ
C þ H2O sfr2
CO þ H2 ð20Þ
C þ CO2 sfr3
2CO ð21ÞFor reaction 20, the ratio of produced CO/CO2 γ is calculated as34
γ ¼ CO=CO2 ¼ 2500 expð�6420=TgÞ ð22ÞThe heterogeneous reaction rates are estimated using the unreactedshrinking core model, in which the real reaction rate depends uponsurface film diffusion and reaction kinetics.35
ri ¼ 1vi, jMj
!AvFj
1km
þ 1kri
i ¼ 1� 3, j ¼ O2, H2O,CO2 ð23Þ
Table 1. Kinetics Data for Primary and Secondary Pyrolysis
reaction reaction rate reference
primary pyrolysis of the cellulosic groupr ¼ 3:20� 105ð1� RgÞexp � 1:60� 104
Ts
!Fv1
32
primary pyrolysis of the plastic group r ¼ ð1� RgÞ ∑6
i¼ 1YiA3, i exp
� E3, iRTs
� �� �Fv2
A3,1 = 9.3 � 1013, E3,1 = 2.34 � 105
33
A3,2 = 1.2 � 1012, E3,2 = 2.07 � 105
A3,3 = 6.3 � 1010, E3,3 = 1.84 � 105
A3,4 = 5.0 � 1010, E3,4 = 1.73 � 105
A3,5 = 9.5 � 1010, E3,5 = 1.80 � 105
A3,6 = 1.5 � 1012, E3,6 = 1.64 � 105
secondary pyrolysisr ¼ 9:55� 104Rg exp
� 1:12� 104
Tg
!Ftar1
29
4132 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
For all heterogeneous char reactions, a first order of reaction is assumedwith respect to gaseous reactants. The kinetic rate is calculated as
kri ¼ AiTs exp�EiRTs
� �ð24Þ
The values of pre-exponent factors and activation energy of hetero-geneous reactions are shown in Table 2.3.2.4. Homogeneous Gas-Phase Reactions. Because of the low ER
value in a typical PGM process, most of the oxygen will be consumed inthe char oxidation section; therefore, combustion of combustible gasescan be ignored. The water�gas shift reaction, as well as cracking andreforming of light hydrocarbons (LHCs) is considered in this work.
CO þ H2OTr4H2 þ CO2 ð25Þ
CxHy þ xH2O sfr5 ðx þ y=2ÞH2 þ xCO ð26Þ
Chemical reaction rates are considered by choosing the minimum of thekinetic rates and turbulent mixing rates.
ri ¼ minðrri, rtiÞ, i ¼ 4�5 ð27ÞTurbulent mixing rates are calculated using the eddy dissipation model
rti ¼ 4:0Fε
kmin
Yjvi, jMj
!ð28Þ
where j denotes any reactant of reaction i.The water�gas shift reaction is considered as a reversible reaction.
The kinetics of the water�gas shift reaction is taken from the work byGrebenshchikova.38
rr4 ¼ RgA4 exp�E4RTg
!CCOCH2O
MCOMH2O�RgA4 exp
�E4RTg
!
Kc
CCO2CH2
MCO2MH2
ð29Þwhere Kc is the equilibrium constant. The value of Kc is given byBensen.39
Kc ¼ exp1
RTg11321� 31:08Tg þ 3Tg ln Tg2:8� 10�4Tg
2 � 91500Tg
!24
35
ð30ÞThe kinetics of the steam-reforming reaction is taken from Jones et al.40
rr5 ¼ RgA5Tg exp�E5RTg
!CCH4CH2O ð31Þ
3.3. Geometry and Boundary Conditions. Geometry used inthis work should be a reflection of the real three-dimensional (3D)geometry, so that it can capture most flow characteristics of a real PGMprocess. The geometry of the trial gasifier is approximately symmetrical
in the width direction; therefore, the longitudinal section of the gasifiercan be used as the 2D geometry. Because the void fraction of the hillockof concretionary slag is very small, the hillock was excluded from theflow field. The total number of mesh cells is 10 107. In the areasassociated with plasma injections and secondary air injections, the meshwas refined.
To express the feeding of MSW, a mass source of the solid phase wasdefined at the top of the fixed bed. Because this is a steady model, thefeeding of MSW is assumed to be continuous. For all cases, the feedingrate is 600 kg/h. Mass flow inlet conditions are defined at thecorresponding positions of plasma air and secondary air inlets. Theoutlet of syngas is defined as a pressure outlet. The relative pressure atthe outlet is set to �700 Pa, which is the measured result for the pilotreactor. The melting of unreacted solid residual is represented by a massand energy sink of the solid phase. Another energy sink is defined in themelting chambers to express the heat transfer from the gas phase to slag.The motion of slag after melting is ignored in this model. The reactorwalls are defined as no-slip walls. An empirical temperature distributionis defined at the reactor wall to calculate the heat loss from the wall.
Altogether nine cases with different plasma powers and air feedingrates are simulated with this model. To increase the versatility of theresults, dimensionless numbers are used to characterize and classify theoperation parameters of the PGM process.
The amount of available air per kilogram of MSW is represented bythe equivalence ratio (ER), defined as
ER ¼ ð _Mair= _MMSW Þð _Mair= _MMSW Þstoic
ð32Þ
where _Mair/ _MMSW is the air/MSW mass flow ratio in the real cases and(Mair/MMSW)stoic is the air/MSW mass flow ratio for a stoichiometriccombustion where the fuel is fully combusted.
The amount of plasma energy per kilogram of MSW is expressed bythe dimensionless plasma energy ratio (DPER), which is defined as
DPER ¼ PplaLHVMSW _MMSW
ð33Þ
where Ppla is the heat power of plasma generators, LHVMSW is the lowheating value of rawMSW, and _MMSW is the mass flow rate of rawMSW.
Despite the common boundary conditions, the detailed boundaryconditions for all cases are expressed by the above dimensionlessnumbers, as found in Table 3. The base case is a standard case, in whichthe operation parameters are set to a “reasonable” value. The choice ofthe operation parameters for the base case is based on experiences from aseries of trial runs. To study the influence of ER and DPER, all cases aredivided into two groups. The first group includes cases 1�5, which aims
Table 2. Pre-exponent Factors and Activation Energy ofHeterogeneous and Homogeneous Reactions
value of i Ai Ei (J mol�1) reference
1 87.1 m s�1 K�1 1.13� 105 36
2 5.71 m s�1 K�1 1.30� 105 37
3 589 m s�1 K�1 2.23� 105 35
4 0.03 m3 kmol�1 s�1 6.03� 104 38
5 3.0 � 108 m3 kmol�1 s�1 K�1 1.254� 105 40
Table 3. Boundary Conditions of Simulated Cases withDimensionless Numbers
case number
plasma
power
(kW)
plasma
air (kg/h)
secondary
air (kg/h) ER DPER
1 240 120 10 0.043 0.118
2 240 120 40 0.053 0.118
3 (base case) 240 120 60 0.060 0.118
4 240 120 80 0.067 0.118
5 240 120 110 0.077 0.118
6 200 100 80 0.060 0.098
7 220 110 70 0.060 0.108
8 260 130 50 0.060 0.128
9 280 140 40 0.060 0.138
4133 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
at studying the influence of ER. The second group (including cases 6�9,as well as the base case) is used to study the influence of DPER.3.4. MSW Properties. Feedstock used by the trial runs is MSW
collected in Israel. Proximate and ultimate analyses have been made on asample of this MSW. The results are shown in Table 4.
4. RESULTS AND ANALYSIS
4.1. Analysis of the Base Case. 4.1.1. Comparison of Pre-dicted and Measured Results. To evaluate the availability of themodel, simulation data of the temperature distribution alongthe reaction shaft axis as well as the syngas composition werecompared to measure data obtained from test runs of the basecase. Results are shown in Figure 2 and Table 5. The predictedand measured temperature profiles fit each other well. However,a slight deviation was found at the reactor height of 4.5�6.1 m.Two reasons can explain this deviation. First, the assumption ofcontinuous feeding of MSW in the model leads to imprecisetemperature prediction near the fixed-bed top. Second, theuncertainty of pyrolysis mechanisms also affects the accuracyof the temperature prediction in the pyrolysis section. Consider-ing the possible variation of MSW composition with time andarea, the disparities between predicted and measured results areacceptable.Table 5 shows a comparison of predicted andmeasured syngas
yields and composition for the base case. It can be seen that thepredicted yields and compositions of syngas are also in good
agreement with the measurements, despite an acceptable devia-tion related to CO. The model slightly overestimates the COvolume fraction, with the deviation equal to 0.132. It is believedthat this deviation is mainly caused by the overestimation of thepeak temperature because of the disregarding partial melting inthe fixed bed. Generally, the deviations of the predicted resultsare in an acceptable level for understanding the characteristics ofthe PMG process.4.1.2. Temperature Profiles. The distribution of the gas
temperature in the base case is shown in Figure 3. It is foundthat the plasma air temperature reduces rapidly because ofradiation and heat exchange with unmelted inorganics. Duringthis process, the plasma air also mixes with secondary air. Theaverage gas temperature at the gas�bed boundary is about 1800K. When air flows into the fixed bed, the gas temperatureincreases dramatically to around 2400 K because of char com-bustion. Then, the gas temperature rapidly decreases to around1000 K. This decrement can be explained by intense heatexchange between phases and endothermic char gasification.Because the gasification agent used in this case is air, the mainchar gasification here should be the Boudouard reaction. Whenthe gas temperature reaches 1000 K, the temperature decreasingrate starts to slow gradually. In this zone, the Boudouard reactiongenerally stops. The heat transfer between the gas and solidphases also becomes slow because of the decrease of temperaturedifferences (see Figure 2). At the reactor height of 5.0�6.1 m,
Table 4. MSW Proximate and Ultimate Analyses
Proximate Analysis (on a Dry Basis, except Moisture)
moisture (%) 20.0
fixed carbon (%) 10.7
volatile (%) 77.6
ash (%) 11.7
Ultimate Analysis (on a Dry Basis)
carbon (%) 50.5
hydrogen (%) 5.6
oxygen (%) 30.7
nitrogen (%) 1.1
chlorine (%) <0.1
sulfur (%) 0.3
LHV of raw MSW (MJ/kg) 12.89
Figure 2. Temperature distribution along the shaft height of thebase case.
Table 5. Syngas Yield and Compositions for the Base Case
syngas predicted measured deviation
H2 (vol %, wet basis) 19.19 19.50 �0.016
CO (vol %, wet basis) 17.21 15.20 0.132
LHCs (vol %, wet basis) 7.22 6.90 0.046
incombustible gases
(vol %, wet basis)
56.68 58.40 �0.029
syngas yield [N m3
(kg of MSW)�1, wet basis]
1.062 1.063 �0.001
Figure 3. Gas temperature distribution (K) in the base case.
4134 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
where pyrolysis and drying of MSW take place, the gas tempera-ture starts to decrease visibly again from around 860 to 450 K.After gas flows out of the fixed-bed area, no reaction or heatexchange happens for gases, so that the gas temperature nearlykeeps constant.4.1.3. Non-uniformity of Temperature Distributions in Hor-
izontal Sections. In the PGM reactor, the temperature distribu-tion in a horizontal section is not uniform. This non-uniformitycan be reflected from Figure 3. Figure 4 shows detailed gastemperature distributions at different horizontal sections in thebase case. The non-uniformity of the gas temperature is mostsignificant at the y = 1.0 m section, where the gas temperaturevaries from 1430 to 2080 K. The peak temperature in this section
appears at the 0.40 < x < 0.55 m area, which is corresponding tothe horizontal position of gas�bed interface at the y = 1.0 msection. The position of the peak temperature denotes that charcombustion only occurs in a thin layer near the gas�bed inter-face. From an engineering point of view, a high peak temperatureshould be prevented because it causes problems, such as bridgingand damage of the reactor wall. To prevent the problems causedby a high peak temperature, the intensity of char combustion hasto be restricted by controlling the ER value. The influence of theER on PGM gasification will be discussed later.The temperature distribution in the y = 2.0 m section shows a
similar trend to the y = 1.0 m section. However, the differencebetween the shaft axis temperature and peak temperaturedramatically decreases to about 150 K. It denotes that the non-uniformity of the gas temperature becomes weak with anincreasing height because of horizontal heat transfer. In the y =3.0 m and y = 4.0 m sections, the differences between the axistemperature and peak temperature are not visible. In all of thesefour sections, the temperature decline near the reactor wall iscaused by heat loss from the reactor wall.4.1.4. Composition Profiles. Parts a�e of Figure 5 show the
volume fractions of CO, H2, LHCs, CO2, and H2O in the gasphase. Because only air is used as a gasification agent in this work,the water�gas reaction and water�gas shift reaction are re-strained. H2 can only be produced from the pyrolysis step. Thisphenomenon is well-presented in Figure 5a. A similar trend isalso found for LHCs. CO is generated from both char combus-tion and pyrolysis. As in Figure 5c, the volume fraction ofCO reaches about 18% after heterogeneous char reactions.
Figure 4. Gas temperature distributions in different horizontal sectionsin the base case.
Figure 5. Syngas compositions of the base case: (a) molar fraction of CO, (b) molar fraction of H2, (c) molar fraction of LHCs, (d) molar fractionof CO2, (e) molar fraction of H2O, and (f) mass fraction of tar.
4135 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
The volume fraction of CO does not changemuch during pyrolysis.However, the yield of CO from pyrolysis is still remarkablebecause the gas volume increases significantly during pyrolysis.A significant aspect of the PGM technology is that it produces
a high-quality syngas. As we can see in Figure 5, the volumefractions of CO and H2 reach around 20% at the syngas outletand the volume fraction of LHCs is about 7%. The LHV of thewet syngas reaches about 6.79 MJ N�1 m�3 in the base case,which is a very high value for MSW gasification. It is believed thatthe high LHV value is mainly due to a low ER value in the PGMprocess, which prevents the direct dilution of combustible gasesfrom N2. A low ER also provides a sub-stoichiometric O2
environment, which suppresses the formation of CO2.The tar yield is one of the main problems involved in the fixed-
bed gasification, which reduces the energy efficiency and causesblockage to the pipeline of syngas. The mass fraction of tar in thegas phase is demonstrated in Figure 5f. The tar mass fraction atthe syngas outlet is about 16.8%, which reflects a tar yield ratio of0.193 kg/kg of MSW.4.2. Influence of ER. The ER is one of the most important
operation parameters of gasification. It determines the level ofMSWpartial combustion and directly influences the temperatureprofile, syngas composition, and stability of the gasificationprocess. The ER required for a typical PGM air gasificationvaries from 0.05 to 0.10, while the ER for conventional gasifica-tion is about 0.3. Few scientific works has been found on thecharacteristics of gasification in low ER conditions. Studying theinfluence of ER on the performance of a PGM gasifier is of bothscientific and engineering value.In this study, five cases with different ER values are simulated.
From cases 1�6, the ER value varies from 0.043 to 0.077. TheDPER value for all cases is set as 0.118.4.2.1. Gas Temperature Distribution. Figure 6 shows the
predicted gas temperature distributions at the shaft axis withdifferent ER values. From cases 1�5, the gas temperature showsan increasing trend with ER. This phenomenon is more sig-nificant in the lower part of the reaction shaft, where thetemperature increases from 1250 to 2750 K. The temperatureincrease with ER is explained by prompted char combustionbecause of increases of the O2 flow rate. According to chemicalequilibrium calculation, for 100% conversion of carbon in char,the ER value should be about 0.13, which is much higher than theER values in the simulated cases. No doubt that the increasing
trend of the gas temperature will continue if the ER keepsincreasing. However, to restrict the peak temperature under2273 K, the ER value in PGM air gasification should be con-trolled less than 0.067. This may result in insufficient combustionof char, which leads to low energy efficiency.4.2.2. Syngas Composition. Figure 7 shows the variation of
syngas composition, as well as the tar/MSWmass ratio, with theER value. It is found that, when the ER increases, the volumefractions of H2, LHCs, and CO2 decrease and the volume fractionof CO increases. The increasing CO volume fraction can beexplained by prompted char combustion with an increasing ER,while the decreasing H2 and LHC volume fractions are explainedby dilution of syngas by introduced N2. It is interesting to findthat the CO2 volume fraction also decreases when ER increases.This phenomenon may be caused by an increasing combustiontemperature with ER, which prevents the formation of CO2
during char combustion. Moreover, the tar/MSW mass ratio isalso increasing slightly with ER. This increase is the result of anincreasing heating rate with ER in the pyrolysis zone.4.2.3. Energy Conversion Ratio (ECR). To quantify the energy
conversion fromMSW to syngas, the ECR was defined and used.
ECR ¼_MH2LHVH2 þ _MCOLHVCO þ _MLHCLHVLHC
_MMSWLHVMSW þ Ppla
�100% ð34ÞThe ECR is a very important process parameter that charac-terizes the combustion value of the gas phase. It illustrates thevariation of syngas composition during the gasification processand can be used as an index for the gas quality. The ECR value atthe syngas outlet is named cold gas efficiency (CGE),40 which iswidely used as a standard criteria for the energy efficiency ofgasification. In the PGM process, the original definition of CGEwas modified slightly.5
Figure 8 shows the ECR value in the horizontal sections alongthe shaft height for cases 1�5. In all cases, the energy conversionsmainly happen in two sections: char combustion and pyrolysis. Itis found that the ER has a positive effect on energy conversion inthe char combustion section. When ER varies from 0.043 to0.077, the ECR will increase from 0.02 to about 0.06. No doubtthat this increasing is caused by prompted char combustionbecause of increasing O2. The increase of the gas temperaturewith ER also has a positive effect on energy conversion because itpushes reaction 20 to produce more CO rather than CO2.
Figure 6. Temperature distribution along the shaft height for differentER values.
Figure 7. Predicted temperature distributions for different DPER values.
4136 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
Volatiles take up to 77.6% of the total MSWmass. Most of theenergy conversion happens in the pyrolysis section, which iscorresponding to the shaft height of 4.5�5.5 m in the reactor.Only a slight negative effect on energy conversion is found forER. The proper reason is that the enhanced heating rate is causedby an increasing ER, which results in an increasing tar yield. Theincrease of the tar yield has been demonstrated in Figure 7.Because the influence of ER on ECR is more significant in the
char combustion section, it is possible to increase the systemCGE by increasing ER. However, as discussed previously, theincrease of ER is restricted by the peak temperature. The ERvalue in PGM air gasification should be lower than 0.067. Apractical solution to this problem is to inject additional steaminto the reactor. The numerical study on air and steam gasifica-tion in a PGM reactor will be presented in the future.4.3. Influence of DPER. The most important significance of
the PGM process is the high-temperature plasma air injectedfrom the bottom of the reactor. The high-temperature plasmaflow supplies heat for both gasification residual melting andreactions related to the gasification process. The value of DPERmay directly influence the temperature profile, syngas composi-tion, tar yield, and stability of the gasification process. In thiswork, the influence of DPER is investigated. Together with thebase case, five cases with different DRER values are simulated(cases 3 and 6�9). The value of DPER varies from 0.098 to0.138. The ER value for all cases is set as 0.060.
Figure 9 shows the predicted gas temperature distributions atthe shaft axis with different DPER values. It was very interestingto find that the temperature distributions for all of these cases aresimilar. When DPER increases from 0.098 to 0.138, the incre-ment of the peak temperature along the shaft axis is less than 200K. A possible explanation for this phenomenon is the heat loss inthemelting chambers. First, an intense heat transfer exists betweenplasma air and slag because the high-temperature and high-velocityplasma flow was directly injected into the slag pool, which exists atthe bottom of the melting chamber. Second, the strong heatradiation leads to large heat loss from the chamber wall. During therunning of the pilot reactor, it is found that heat loss from thechamber wall reaches about 30% of the total plasma power. It isalso found that the heat loss increases with the DPER value.A thermodynamic calculation is performed by authors to
estimate the lower limit of DPER to satisfy the heat request formelting the inorganic components. The result shows that, whenthe DPER value is larger than 0.09, the plasma flow can supplyenough heat for the melting process. From the viewpoint ofenergy efficiency, the optimal DPER value for PGM air gasifica-tion should be about 0.09.
5. CONCLUSION
(1) The PGM systemwas examined for air gasification by a 2DEuler�Euler multiphase model. The model considered all majorchemical and physical processes involved during PGM gasifica-tion. The model showed significant agreement with the measure-ment from the base case. (2) Analysis of the base case by meansof computational fluid dynamics (CFD) revealed that thehorizontal temperature distribution inside the reactor was non-uniform. In addition, a maximum peak temperature of the reactorwas observed at the gas�solid interface. PGM air gasification pro-vided a higher calorific value of the syngas (LHV = 6.79 MJ N�1
m�3). The tar yield is around 0.193 kg/kg of MSW. (3) A furtherinvestigation of the PGM process by means of the developedmodel revealed that ER has a positive influence on the calorificvalue of the syngas. An increase of ER from 0.043 to 0.077 showedaround a 5% increase in CGE. However, the maximum allowableER for the present gasification systemwas restricted to about 0.067because of the increase in the peak temperature of the reactor. (4)The influence of DPER on PGM air gasification is not obvious.The optimal DPER value was considered as about 0.09, consider-ing energy efficiency. (5) Although PGM air gasification provideda higher calorific value of the syngas (LHV = 6.79MJN�1 m�3), adetrimental effect on char conversion was observed. Minimizationof this problem will be addressed in our future research.
’AUTHOR INFORMATION
Corresponding Author*Telephone: 0046-8790-8405. Fax: 0046-8207-681. E-mail:[email protected].
’NOMENCLATUREAi = pre-exponential factor of reaction iAv = specific surface area (m�1)C = mass concentration (kg m�3)d = particle diameter (m)D = diffusion coefficient of vapor in the bulk (m2 s�1)Ei = activation energy of reaction igB = gravitational acceleration (m s�2)
Figure 8. ECR values along the shaft height for different ER values.
Figure 9. Temperature distributions along the shaft height for differentDPER values.
4137 dx.doi.org/10.1021/ef200383j |Energy Fuels 2011, 25, 4129–4137
Energy & Fuels ARTICLE
h = specific enthalpy (J kg�1)Heva = evaporation heat of the solid material (J kg�1)k = heat-transfer coefficient (W m�2 K�1)km = mass-transfer coefficient (m s�1)kri = kinetics rate of heterogeneous reaction i (m s�1)K = interphase momentum exchange coefficient (kg m�3 s�1)_m = mass-transfer rate (kg m�3 s�1)M = molar weight (kg kmol�1)_M = mass flow rate (kg s�1)p = pressure (Pa)P = power (W)q = heat flux (W m�2)Q = intensity of heat exchange (W m�3)r = reaction rate (kmol m�3 s�1)rri = kinetic rate of homogeneous reaction i (kmol m�3 s�1)rti = turbulent mixing rate of reactants involved in homogeneous
reaction i (mol m�3 s�1)R = universal gas constant (J mol�1 K�1)S = source termSh = Sherwood numbert = time (s)T = temperature (K)vB = velocity (m s�1)vi,j = stoichiometric coefficient of reactant j in reaction iYi = mass fraction of the ith species
Greek SymbolsR = volume fractionF = density (kg m�3)τC = stress tensor (Pa)μ = dynamic viscosity (Pa s)ϕ = angle of internal frictionk = thermal conductivity (W m�1 K�1)ε = turbulent dissipation rate (m2 s�3)
Subscriptsair = airc = fixed carbonCO = carbon monoxideCO2 = carbon dioxideCxHy = light hydrocarbonsg = gas phaseH2 = hydrogenH2O = steammoi = moistureMSW = municipal solid wasteO2 = oxygenpla = plasmas = solid phasetar1 = primary tarv1 = volatile from cellulosic speciesv2 = volatile from plastic species
’REFERENCES
(1) Malkow, T. Waste Manage. 2004, 24 (1), 53–79.(2) Filippis, P. D.; Borgianni, C.; Paolucci, M.; Pochetti, F. Waste
Manage. 2004, 24 (6), 633–639.(3) Choy, K. K. H.; Porter, J. F.; Hui, C. W.; McKay, G. Chem. Eng. J.
2004, 105 (1�2), 31–41.(4) Thamavithya, M.; Dutta, A. Fuel Process. Technol. 2008, 89 (10),
949–957.(5) Zhang, Q.; Dor, L.; Fenigshtein, D.; Yang, W.; Blasiak, W. Appl.
Energy 201110.1016/j.apenergy.2011.01.041.
(6) Syamlal, M.; Bissett, L. METC Gasifier Advanced Simulation(MGAS) Model; Morgantown Energy Technology Center: Morgan-town, WV, 1992.
(7) Bryden, K. M.; Ragland, K. W. Energy Fuels 1996, 10, 269–275.(8) Hla, S. S. A theoretical and experimental study on a stratified
downdraft biomass gasifier. Ph.D. Thesis, The University of Melbourne,Melbourne, Victoria, Australia, 2004.
(9) Blasi, C. D. AIChE J. 2004, 50 (9), 2306–2319.(10) Rogel, A.; Aguillon, J. Am. J. Appl. Sci. 2006, 3 (10), 2068–2075.(11) Yang, W.; Ponzio, A.; Lucas, C.; Blasiak, W. Fuel Proc. Technol.
2006, 87 (3), 235–245.(12) ANSYS, Inc.ANSYS FLUENT 12.0 Theory Guide; ANSYS, Inc.:
Canonsburg, PA, 2009.(13) Kuipers, J. A. M.; Duin, K. J. V.; Beckum, F. P. H.; Swaaij,
W. P. M. Chem. Eng. Sci. 1992, 47 (8), 1913–1924.(14) Gldaspow, D.; Ettehadleh, B. Ind. Eng. Chem. Fundam. 1983, 22
(2), 193–201.(15) Johnson, P. C.; Jackson, R. J. Fluid Mech. 1987, 176, 67–93.(16) Ergun, S. Chem. Eng. Prog. 1952, 48 (2), 89–94.(17) Cowin, S. C. Powder Technol. 1974, 9 (2�3), 61–69.(18) Syamlal, M.; Rogers, W.; O’Brien, T. J. MFIX Documentation:
Theory Guide; National Technical Information Service: Springfield, VA,1993; Vol. 1.
(19) Schaeffer, S.; Balakrishnan, L. ICASE Report 90-18: Applicationof a Reynolds�Stress Turbulence Model to the Compressible Shear Layer;National Aeronautics and Space Administration (NASA): Washington,D.C., 1990.
(20) Gunn, D. J. Int. J. Heat Mass Transfer 1978, 21, 467–476.(21) Peters, B. Thermal Conversion of Solid Fuels (Developments in
Heat Transfer); WIT Press: Billerica, MA, 2002; Vol. 15.(22) Yang, Y. B.; Goh, Y. R.; Zakaria, R.; Nasserzadeh, V.; Swithenbank,
J.Waste Manage. 2002, 22 (4), 369–380.(23) Yang, Y. B.; Nasserzadeh, V.; Goodfellow, J.; Goh, Y. R.;
Swithenbank, J. J. Inst. Energy 2002, 75 (504), 66–80.(24) Yang, Y. B.; Yamauchi, H.; Nasserzadeh, V.; Swithenbank, J.
Fuel 2003, 82 (18), 2205–2221.(25) Yang, Y. B.; Sharifi, V. N.; Swithenbank, J. Fuel 2004, 83
(11�12), 1553–1562.(26) Yang, W.; Ponzio, A.; Lucas, C.; Blasiak, W. Fuel Process.
Technol. 2006, 87 (3), 235–245.(27) Ranz,W. E.;Marshall,W. R.Chem. Eng. Prog. 1952, 48 (3), 141–146.(28) Ranz, W. E.; Marshall, W. R. Chem. Eng. Prog. 1952, 48 (4),
173–180.(29) Broson, M. L.; Howard, J. B.; Longwell, J. P.; Peter, W. A.
AIChE J. 1989, 35 (1), 120–128.(30) Liden, A. G.; Berruti, F.; Scott, D. S.Chem. Eng. Commun. 1988,
65 (1), 207–221.(31) Sorum, L.; Gronli, M. G.; Hustad, J. E. Fuel 2001, 80 (9),
1217–1227.(32) Chan, W. R.; Kelbon, M.; Krieger, B. B. Fuel 1985, 64 (11),
1505–1513.(33) Wu,C.; Chang, C.;Hor, J.WasteManage. 1993, 13 (3), 221–235.(34) Arthur, J. A. Trans. Faraday Soc. 1951, 47, 164–178.(35) Hobbs, M. L.; Radulovic, F. T.; Smoot, L. D. Prog. Energy
Combust. Sci. 1993, 19 (6), 505–586.(36) Desai, P. R.; Wen, C. Y. Computer Modeling of Morgantown
Energy Research Center’s Fixed Bed Gasifier; Department of ChemicalEngineering, West Virginia University: Morgantown, WV, 1978.
(37) Yoon, H. W.; Denn, M. M. AIChE J. 1978, 24 (5), 885–903.(38) Grebenshchikova, G. B. Podzemn. Gazif. Uglei 1957, 2, 54–57.(39) Lowry, H. H. Chemistry of Coal Utilization; John Wiley and
Sons, Inc.: New York, 1981; Supplementary Vol. 2.(40) Reimert, R. Gas Production in Ullmann’s Encyclopedia of In-
dustrial Chemistry, 5th ed.; VCH Publishers: Weinheim, Germany, 1989;Vol. A12.
SUPPLEMENT IV
Qinglin Zhang, Liran Dor, Weihong Yang, Wlodzimierz Blasiak.
Modeling of steam plasma gasification for municipal solid waste
Manuscript submitted to Fuel Processing Technology, in June 2011
Modeling of steam plasma gasification for municipal solid waste
Qinglin ZHANGa*, Liran DORb, Amit BISWASa, Weihong YANGa, Wlodzmierz BLASIAKa
a Royal Institute of Technology (KTH), School of Industrial Engineering and Management, Department of Material Science and Engineering, Division of Energy and Furnace Technology, SE 100 44, Stockholm, Sweden
b Environmental Energy Resources Ltd, Israel
* Corresponding author: Qinglin ZHANG; Email address: [email protected]; Telephone: +46 8 790 6545; Fax: +46 8 207 681
Abstract Plasma gasification melting (PGM) is a promising gasification technology aiming at providing sustainable disposal for various wastes. In this work, an Euler-Euler multiphase model was developed to study the characteristics of air and steam gasification of municipal solid waste in a PGM reactor. The model is validated by measurement data from a demonstration PGM reactor. With this model, three groups of simulations were performed to study the influences of operating conditions. It is confirmed that injection of high temperature steam is important for increasing the cold gas efficiency and syngas lower-heating-value. The effect of steam injection is affected by steam feeding rate, air feeding rate and plasma power. Based on the simulated results, an optimal condition is suggested for air and steam gasification in the PGM reactor.
Keywords: Gasification, Melting, MSW, Plasma, Modeling
1. Introduction Waste-to-energy has drawn significant attentions to policymakers and industrial community. In recent years, a new gasification technology named Plasma Gasification Melting (PGM) was developed [1]. In the PGM, updraft fixed-bed gasification is combined with thermal plasma technology. High temperature plasma torches are placed at the air inlet to produce high temperature plasma flow. The high temperature plasma air melts the inorganic components of waste, and reacts with organic components as a gasification agent. Compared to conventional gasification, the PGM has advantages such as better syngas heating value, higher energy efficiency and prevention of bottom ash problems. Additionally, no milling or pretreatment of feedstock is needed. It is very suitable for low-calorific-value feedstock such as municipal solid waste (MSW), medical waste and industrial waste.
Since gasification air is preheated by plasma torches, the equivalence ratio (ER) required by a typical PGM reactor is much lower than that of conventional gasifier. A low ER alleviates char combustion and dilution of syngas by N2. However, it also leads to decrease of gasification agents. To solve this problem, high temperature steam is fed into the PGM reactor as an additional gasification agent. The effect of steam addition into gasification has been widely investigated in open literatures [2-4]. It is believed that use of steam in gasification process can largely enhance the H2 yield which in turn increases the syngas lower-heating-value (LHV). In our previous research, the effect of steam injection on gasification characteristics in a PGM reactor has been studies [1]. The result shows that injection of high temperature steam largely increase the syngas yield. In addition, the syngas LHV is also improved. However, due to the high expense for running the PGM demonstration reactor, it is not practical to study in detail the effect of steam addition in PGM process only by experiments.
Essentially, the PGM reactor is an updraft fixed-bed gasifier. From this point of view, gasification models can be used for analyzing the performance of PGM air and steam gasification. The earliest gasification models are equilibrium models, which predict gasification products by equilibrium assumption. The accuracies of these models are often inadequate since reality usually deviates from equilibrium predictions. Kinetics models were then developed by considering the effects of reaction rates and transport phenomena [5-6]. In these models, the whole reactor is usually divided into several layers, in which different reactions are considered. However, the effect of reactor geometry was not included, so these models are still zero-dimensional. One-dimensional gasification models are popular studied in 1990s [7-8]. In these models, the vertical movements of both feedstock and gas are considered. These modes are further developed by Rogel and Aguillon [9]. In their work, the conservation equations of gas phase are solved in a two-dimensional system and of the solid phase in a one-dimensional system. However, full two-dimensional simulation for fixed-bed gasification was not found by authors in open literatures.
In this work, an Euler-Euler multiphase model is used to simulate the performance of a PGM reactor with steam injection. The model considers the main chemical and physical processes such as drying, pyrolysis, homogeneous reactions, heterogeneous char
reactions and melting of the inorganic components of MSW. The model was previously used to study the characteristics of PGM air gasification [10]. Now it is further improved with more complex steam reaction mechanisms. In this paper, the model is used to study the characters of air and steam gasification of a PGM reactor.
2. Methodology
2.1. Facility A PGM demonstration reactor has been built up in Northern Israel, with a capacity of 12 to 20 tons per day. The scheme of the demonstration reactor is shown in Figure 1. Generally, the reactor is a fixed-bed counter current gasification shaft, with a plasma melting chamber located at the bottom of the shaft. Plasma torches are placed at the ceiling of the melting chamber. Primary air flows into the melting chamber through the torches, where it is ionized so forming plasma jets which extend beyond the tip of the torches. The temperature of the plasma jets may reach up to 6000 K. The plasma jets supply the necessary heat to melt the inorganic components of the feedstock, which reached the bottom of the reactor. Secondary air nozzles are placed around plasma nozzles. Secondary air at room temperature is injected through secondary air nozzles. The flow rate of secondary air is adjustable thus the feeding rate of total air can be controlled. High temperature steam at 1273 K is fed into the reactor from steam nozzles placed at the side wall of the melting chamber. An airtight feeding pipe is placed at the top of the reaction shaft. MSW is fed into the reactor intermittently from the shaft top every half an hour. The total height of the reaction shaft is 7.02m, and the height of the fixed-bed is 6.11 m.
Thermocouples are placed along the gasifier shaft to measure the temperature distribution inside the reactor. Additionally, a probe is placed in the syngas outlet to obtain syngas samples, which are sent to a gas analyzer for composition analysis. More information about the demonstration reactor is introduced previously [1].
Syngas outlet
Plasma torches
Slag outlets
Reaction shaft
10°
Melting chamber
Concretionary slag
Secondary air
SteamInlet
Figure 1. The scheme of the PGM reactor in the demonstration plant
2.2. Feedstock properties The feedstock used by the PGM gasifier is MSW collected in Israel. The proximate and ultimate analyses of the MSW are given in Table 1. In the reality, the size of MSW particles varies from 1-100 mm. For simplification, an average particle size of 10 mm is assumed in this work.
Table 1 MSW proximate and ultimate analyses
Proximate analysis (in dry basis except moisture)
Moisture 20.0 %
Fixed carbon 10.7 %
Volatile 77.6 %
Ash 11.7 %
Ultimate analysis (in dry basis)
Carbon 50.5%
Hydrogen 5.6%
Oxygen 30.7%
Nitrogen 1.1%
Chlorine <0.1%
Sulphur 0.3%
LCV of raw MSW (MJ/kg) 12.89
2.3. Operating conditions
The performance of the PGM reactor is directly influenced by the choice of operating conditions such as air feeding rate, steam feeding rate and plasma power. In this work, simulation of a typical case named the base case was performed. The results were compared with the measured results from the demonstration reactor to evaluate the availability of the model. Then, three groups of simulations were performed to study the influence of these operation parameters. The detailed operating conditions are presented in Table 2. To ensure the applicability of the results in the practical industrial design, all studied operation parameters are characterized to dimensionless numbers, which are shown below.
ER
( )( )stoicMSWair
MSWair
MMMMER
//
= (1)
Steam feedstock mass ratio (S/F)
MSW
OH
MM
FS
2/ = (2)
Dimentionless plasma energy ratio (DPER)
MSWMSW
pla
MLHVP
DPER⋅
= (3)
Table 2 Operation conditions in this study
Base case Group 1 Group 2 Group 3
Operation parameters
Plasma power (kw) 240 240 240 220 - 260
Air feeding rate (kg/h) 180 180 130 - 230 180
Steam feeding rate (kg/h) 70 0 - 150 100 100
After dimensionless treatments
ER 0.06 0.06 0.043 – 0.077 0.06
S/F 0.117 0 - 0.250 0.167 0.167
DPER 0.118 0.118 0.118 0.108 - 0.128
3. Numerical model An Euler-Euler multiphase model for fixed-bed gasification is applied in this work. This model is developed originally for studying the characters of PGM air gasification [10]. In the presented work, the model is further improved with more complicated steam reaction mechanisms.
The model is based on Euler-Euler multiphase theory. Gas and solid are treated as two continuous phases. The conservation equations of mass, momentum and energy are solved for both phases, respectively. Mass, momentum and energy transfer between phases are allowed. The turbulence of the gas phase is simulated with the standard k-ε turbulence model.
The main chemical and physical processes such as drying, pyrolysis, homogeneous reactions, heterogeneous char reactions and melting of the inorganic components of MSW are simulated in this model. The rates of homogeneous reactions are determined by consider both the kinetic rates and turbulent mixing rates. The reaction rates of char heterogeneous reactions are calculated with the unreacted shrinking core model. The melting of solid residual is simulated by setting mass and energy sinks for the solid phase at the gas-bed boundary in the melting chamber.
3.1. Flow equations The Eulerian continuity equation is solved for each chemical species. Momentum and energy conservation equations are solved for each phases. The equations are shown below [11].
Continuity equation:
( ) ( ) iiqiqqiqq SmvYYt
+=⋅∇+∂∂
ραρα
(4)
Momentum equation:
( ) ( ) ( ) pqpqqppqqqgqqqqqqqqq vmvvKgpvvvt
+−++⋅∇+∇−=⋅∇+
∂∂ ραταραρα (5)
Energy equation:
( ) ( ) pqpqpqqqggqqqqqqqq hmQSqvtphvh
t
+++⋅∇−∇+
∂∂
−=⋅∇+∂∂ :ταραρα (6)
In the fixed-bed, the solid stresses arise mainly because of Coulomb friction between particles in enduring contact. This kind of granular flow can be treated as plastic flow [12]. A solid pressure is calculated independently and used for the pressure gradient term
sp∇ in the solid-phase momentum equation 13:
( )nggs Ap ∗−= αα (7)
where ∗gα is the gas volume fraction at minimum fluidization. Empirical values of
2510=A (Pa) and 10=n are used.
When the solid volume fraction approaches the solid packing limit, the value of the solid pressure becomes tremendous [14, 15]. The solid pressure gradient sp∇ becomes the main driven force of granular flow. In other words, the influence of the interphase momentum exchange term ( )sggs vvK
− on solid motion can be ignored at the fixed-bed condition. Johnson and Jackson [16] used the plastic flow model to simulate non-reaction shearing granular flow, and good results were obtained. In this work, this model is adopted. The interphone momentum exchange term is disregarded in the solid momentum equations. It is found by authors that this simplification is very helpful for the convergence of the solid momentum equations since it largely prevents the solution of interphase non-linear terms. For the gas phase momentum equations, then, the interphase momentum exchange term is retained. The interphase momentum exchange coefficient sgK is written as:
( )s
gssg
sg
ggssg d
vvd
K
−+
−=
αρ
αµαα
756.11
150 2 (8)
The intensity of heat exchange between the solid and gas phases is assumed to be a function of the temperature difference between solid and gas phase:
( )gssggssg TTkQQ −=−= (9)
The heat transfer coefficient is written as:
2
6
s
sgsgsg d
Nuk
αακ= (10)
Here sNu is the Nusselt number correlated by Gunn [17]:
( )( ) ( ) 33.07.0233.02.02 PrRe2.14.233.1PrRe7.015107 gsgggsggsNu αααα +−+++−= (11)
3.2. Reaction models
3.2.1. Drying According to the previous researchers, moisture can be driven out from solid particles by the mass diffusion or evaporation at different temperature. The rate of drying can be calculated as [18-20]:
( )OHmoimv CCkAr2
−= when CTs
100<
(12)
or
evpsg HQr /= when CTs100≥
(13)
The Mass transfer coefficient mk is calculated according to the Sherwood number [21-22]:
sm d
ShDk = (14)
3.2.2. Pyrolysis
Selection of pyrolysis model is considered to be decisive due to complexity of the pyrolysis process. Among various pyrolysis models, the two step pyrolysis model is considered to be more appropriate for present study due to the nature of producing higher tar in updraft gasification process. This model divides pyrolysis into two steps: at first, MSW decomposes into char, gas and primary tar; then the primary tar decomposed into gas and secondary tar by thermal cracking and reforming [23]-[25]. Equations (16) and (17) represent the two step model.
ChartarimaryGasFeedstock ++→ Pr (15)
tarSecondaryGastarimary +→Pr (16)
MSW is a complex mixture of different components which can be divided into two main groups: Cellulosic groups (wood, paper, cardboard and textile) and Plastic. Due to its difference in components, each group has been model with its corresponding kinetic parameters. Table 3 shows kinetic data both for primary and secondary pyrolysis.
Table 3 Kinetics data for primary and secondary pyrolysis of MSW
Reaction Reaction rate Source
Primary pyrolysis of cellulosic group
( ) 1
45 1060.1exp11020.3 v
sg T
r ρα
×−−×= Chan et
al. [26]
Primary pyrolysis of plastic group
( ) 2
6
1
,3,3 exp1 v
i s
iiig RT
EAYr ρα ∑
=
−−= ,
51,3
131,3 1034.2,103.9 ×=×= EA
Wu
[27]et
al.
52,3
122,3 1007.2,102.1 ×=×= EA
53,3
103,3 1084.1,103.6 ×=×= EA
54,3
104,3 1073.1,100.5 ×=×= EA
55,3
105,3 1080.1,105.9 ×=×= EA
56,3
126,3 1064.1,105.1 ×=×= EA
Secondary pyrolysis 1
44 1012.1exp1055.9 tar
gg T
r ρα
×−×=
Boroson et al. [23]
3.2.3. Homogeneous reactions.
The following homogeneous reactions are considered in this model:
OHOH r222
12/1 →+ (17)
( ) OHyxCOOyxHC ryx 22 2/4/2 2 +→++ (18)
2232/1 COOCO r→+ (19)
xCOHyxOxHHC ryx ++→+ 22 )2/(4 (20)
2225 COHOHCO r +→←+ (21)
Chemical reaction rates of (18)-(22) are considered by choosing the minimum of the kinetic rates and turbulent mixing rates:
( )tkrkk rrr ,min= , 51−=k (22)
Turbulent mixing rates are calculated using the eddy dissipation model:
=
jj
j
ii
itk Mv
YMvY
kr ,min0.4 ερ (23)
where i and j denote reactants of reaction k .
Kinetic rates of homogeneous reactions are shown in Table 4.
Table 4 Kinetic rates of homogeneous reactions
Kinetic rate (kmol m-3 s-1 Source )
( ) 1.11.14.81 22
/3670exp1056.3 OHggr CCTr −×= α Varma et al. [28]
( ) 8.07.07.112 2
/24369exp100.1 OHCggr CCTryx
−×= α Dryer et al. [29]
( ) 5.05.0113 22
/62700exp103.1 OOHCOggr CCCTr −×= α Howard et al. [30]
( ) OHHCgggr CCTTryx 2
/15083exp100.3 84 −×= α
Jones et al.[31]
( ) ( )
−−−=
0265.0/7914exp
/7250exp03.0 22
25HCOg
OHCOggr
CCTCCTr α
Grebenshchikova [32], Yoon et al. [33]
3.2.4. Heterogeneous reactions
The following heterogeneous char reactions are considered in this work:
22 222
21
6 COCOOC r
++
+→
++
+γγ
γγ
γγ
(24)
227 HCOOHC r +→+ (25)
COCOC r 282 →+ (26)
4292 CHHC r→+ (27)
The heterogeneous reaction rates are estimated using the unreacted shrinking core model, in which the real reaction rate is determined by surface film diffusion and reaction kinetics. First order of reaction is assumed with respect to gaseous reactants [34]:
km
iv
iik
kk
AMv
r 111
+
=
ρ 96 −=k (28)
where i denotes the gaseous reactants of reaction k .
In reaction (25), the γ is the ratio of produced CO and CO2 [35] are it is calculated as :
( )gT/6420exp2500 −=γ (29)
The expressions of kk are shown in Table 5.
Table 5 Expression of kk for heterogeneous reactions
kk (m s-1 Source )
( )ss TTk /9000exp685.06 −= Evans et al. [36]
( )ss TTk /15600exp714.57 −= Yoon et al.[37]
( )ss TTk /26801exp1089.5 28 −×= Hobbs et al. [38]
( )ss TTk /26801exp1042.3 39 −×= − Hobbs et al. [38]
4. Results and discussions
4.1. Evaluation of model
The simulated results were evaluated by comparison with measured results from the base case. The results are shown in Table 6. Generally, the predicted yield and composition of syngas are in agreement with the measurements. However, a non-ignorable deviation exists between predicted and measured CO. Meanwhile, an underestimation is found for CO2 content. It is believed by authors that the deviation is mainly caused by overestimation of fixed-bed peak temperature due to disregarding the melting in the fixed-bed. According to previous researchers, melting of inorganic components in MSW starts when the solid temperature reaches around 1800 K. The melting process is a highly endothermic, so that further increase of solid temperature can be restrained. However, the possible melting in the fixed-bed is not considered in the model, so the peak temperature may be overestimated. The overestimation of solid temperature therefore influence the yield of char combustion (reaction 24), thus overestimating the CO content in syngas. Despite the deviation between simulated and measured results, the accuracy of this model is acceptable for analyzing the characters of air and steam gasification in the PGM reactor.
Table 6 Measured and predicted syngas yield and main compositions of the base case
Syngas Unit Predicted Measured Deviation
H Vol% (wet basis) 2 19.49 20.44 -0.046
CO Vol% (wet basis) 19.21 16.10 -0.193
LHCs (light hydrocarbons) Vol% (wet basis) 6.79 7.42 -0.085
CO Vol% (wet basis) 2 6.59 >10.0 -
Syngas yield Nm3/kg MSW (wet basis) 1.368 1.359 -0.006
4.2. Effect of S/F
In PGM technology, high temperature steam is used as an additional gasification agent. The effect of steam feeding rate is characterized by the mass ratio between steam and MSW (S/F). The gas temperature distribution at various S/F values when ER=0.06 and DPER=0.118 is shown in Figure 2. When S/F increases from 0 to 0.25, the temperature distribution along the reaction shaft becomes more uniform. Meanwhile, the area of char reaction zone, where the gas temperature is above 1000 K, is increasing. Significant advantages are obtained from these variations: firstly, the uniformity of gas temperature prevents the formation of very high temperature, which challenges the heat resistance of wall materials; secondly, the increase of char reaction zone increases the reaction time of gasification agents with char. Another advantage of increasing S/F is that increased steam feeding rate enhances the rate of water-shift reaction, which is an important char gasification reaction.
S/F=0 S/F=0.083 S/F=0.208 S/F=0.250
Figure 2. Predicted gas temperature (K) distributions for different S/F values
In order to characterize the conversion of char, the char conversion efficiency Cη is defined as the percentage of carbon in the MSW converted into gas species. The energy efficiency of PGM is represented by cold gas efficiency η , which is defined as follow:
%1002
×++⋅
⋅=
plaOHMSWMSW
syngassyngas
PPLHVMLHVM
η (30)
The effects of S/F on Cη and η are indicated in Figure 3. It is found that steam injection has a notable positive effect on both Cη and η . When steam is not injected, the value of
Cη is only about 23%, which is far from complete gasification of char. When S/F varies from 0 to 0.208, the value of Cη increases dramatically from 22% to 96%. Further increase of S/F from 0.208 to 0.250, however has very limited effect on Cη . It is known that the incomplete conversion of char is disadvantageous for gasification since it reduces the cold gas efficiencyη . It is found that the variation of η with S/F has similar trends to that of Cη , which implies that the enhance of Cη with S/F is the main cause for η variation. From this point of view, the point S/F=0.208 is a critical point or optimizing of the air and steam gasification of a PGM reactor.
20
40
60
80
100
0 0.05 0.1 0.15 0.2 0.25S/F
(%)
η ηc
Figure 3. Effect of S/F on Cη and η at ER= 0.06 and DPER= 0.118.
Figure 4 shows the contents of main gaseous species inside the reactor at different S/F values. When S/F increases from 0 to 0.208, the volume fractions of H2
Cη
and CO generally show an increasing trend, especially at the bottom half of the reaction shaft where char gasification reactions take place. This is mainly caused by promoted water-shaft reaction and other char reactions due to increasing steam injection. When S/F further increases from 0.208 to 0.25, the volume fractions of CO seems decreasing. This phenomenon corresponds to the variation of with S/F. Since char conversion almost completes at S/F=0.208, further increasing of S/F mainly promoted the water-gas shift reaction so the total yield of CO is reduced. It is also found that most of the Light hydrocarbons (LHCs) are produced in the pyrolysis section, while the effect of methanation reaction is not visible. The explanation of this phenomenon may be the relatively high temperature in the gasification section which accurate the reforming of LHCs. Finally, it is found that the overall tar yield shows a decreasing trend with increasing S/F. this is mainly due to favored tar cracking and reforming due to higher gas temperature in the pyrolysis section.
(a)
S/F=0 S/F=0.083 S/F=0.208 S/F=0.25
(b)
S/F=0 S/F=0.083 S/F=0.208 S/F=0.25
(c)
S/F=0 S/F=0.083 S/F=0.208 S/F=0.25
(d)
S/F=0 S/F=0.083 S/F=0.208 S/F=0.25
Figure 4. Predicted contents of main species in gas phase for different S/F values. (a) H2 volume fractions, (b) CO volume fractions, (c)LHCs volume fractions, (d) tar mass fractions.
4.3. Effect of ER
ER is an important parameter for any gasification process since it influences both the oxygen supply and energy balance inside the reactor. In this work, the effect of ER on air and steam gasification in the PGM reactor is studied at S/F= 0.167 and DPER= 0.118 condition. Figure 5 demonstrates the gas temperature distributions at three typical ER values. It is found that increase of ER has a positive effect on both overall temperature and peak temperature inside the reactor. The increasing of gas temperature is the results of favored char combustion. It is known that increasing of gasification temperature is favorable since it accelerates reaction rates, and influences the energy equilibrium of endothermic gasification reactions. However, a high peak temperature may challenge the hear resistance of wall materials. Moreover, the bridging problem may happen at high temperature condition since part of the inorganic component of MSW may be melted. When ER increases from 0.043 to 0.077, the value of peak gas temperature increases from 2100 K to about 2500 K. Even after taking into account the overestimation of peak temperature with the model, the peak temperature at ER=0.077 is still too high for practical running of the PGM reactor.
ER=0.043 ER=0.060 ER=0.077
Figure 5. Predicted gas temperature (K) distributions for different ER values
Figure 6 shows the variation of Cη with different ER values. The Cη increases all the way with ER, and reaches about 100% at ER=0.08. This phenomenon can be explained by two reasons. Firstly, the enhanced char combustion by increased ER directly favors char conversion. Moreover, char combustion increases the global temperature inside the reactor, thus accelerates the endothermic char gasification reactions such as water-shift reaction and boudouard reaction. From this point of view, complete char conversion thus the highest cold gas efficiency can be obtained at ER value of 0.077. However, considering the high peak temperature at ER= 0.77, it is not suggested to use such high ER value. In real operation, it is more applicable to use a relative low ER value like 0.06, while increase the S/F to increase the Cη .
40
60
80
100
0.04 0.05 0.06 0.07 0.08
ER
ηc (%
)
Figure 6. Effect of ER on Cη at S/F= 0.167 and DPER= 0.118.
Figure 7 shows the contents of main gaseous species inside the reactor at different ER values. It is shown that the CO volume fraction increases evidently with ER. The increase of CO content is explained by enhanced char combustion due to increasing ER. The total
yield of H2 is also enhanced by favored water-shift reaction due to temperature increase. However, this positive effect is counteracted by the dilution from N2 due to enhanced ER. As a result, the final volume fraction of H2 does not change much with ER. The yield of tar is sensitive to the temperature in the pyrolysis section. As we can see in Figure 7 (d), the mass fraction of tar reduces visible when ER increases. Cracking and reforming of tar also produces combustible gases, thus increasing the η value. This is another positive effect of increasing ER.
(a)
ER=0.043 ER=0.060 ER=0.077
(b)
ER=0.043 ER=0.060 ER=0.077
(c)
ER=0.043 ER=0.060 ER=0.077
(d)
ER=0.043 ER=0.060 ER=0.077
Figure 7. Predicted contents of main species in gas phase for different ER values. (a) H2 volume fractions, (b) CO volume fractions, (c)LHCs volume fractions, (d) tar mass fractions.
4.4. Effect of DPER The high-temperature plasma air injection is the most important significance of the PGM process. Other than supplying heat for the melting of the inorganic component of MSW, the plasma injection also preheats gasification agent to around 1700 K, thus influence the energy balance inside the PGM reactor. The effect of DPER value on gas temperature at ER=0.06 and S/F=0.167 is shown in Figure 8. When DPER increases from 0.108 to 0.128, the gas temperature only increases slightly. This phenomenon can be explained by two reasons. Firstly, the plasma flow is injected directly into the slag pool at the bottom of the melting chamber. The outward slag flow take away much of increased sensible heat from plasma. Secondly, heat loss from strong radiation in the melting chamber also increase with DPER value. As a result, only a small part of the increased plasma energy is introduced to the fixed-bed.
DPER=0.108 DPER=0.118 DPER=0.128
Figure 8. Predicted gas temperature (K) distributions for different DPER values
Figure 9 is the contents of main gaseous species inside the reactor at different DPER values. With the increase of DPER, both H2 and CO shows a slight increasing trend. As introduced in 4.3, the enhancement of gas temperature favors char gasification reactions and also cracking and reforming of tar. However, this positive effect is also not significant.
(a)
DPER=0.108 DPER=0.118 DPER=0.128 (b)
DPER=0.108 DPER=0.118 DPER=0.128
(c)
DPER=0.108 DPER=0.118 DPER=0.128
(d)
DPER=0.108 DPER=0.118 DPER=0.128
Figure 9. Predicted contents of main species in gas phase for different DPER values. (a) H2 volume fractions, (b) CO volume fractions, (c) LHCs volume fractions, (d) tar mass fractions.
5. Conclusions A 2D Euler-Euler Multiphase model for PGM air and steam gasification is developed for analyzing the performance characters of a typical PGM reactor. A good agreement is found between the model prediction and experimental measurement concerning syngas yield and composition.
The effect of operating conditions on reactor performance was discussed. The following are the main discoveries:
• Injection of high temperature steam has a positive effect on both syngas LHV value and reactor cold gas efficiency. The main reason is that high temperature steam supplies both oxygen atoms and sensible heat for char gasification, so that the char convention ratio is highly enhanced. At the studied condition, the positive effect of increasing S/F value is significant when 208.0/ ≤FS . When
208.0/ ≥FS , the effect becomes very limited.
• The value of ER influences both chemical and energy balance inside the reactor. Increasing ER promotes both the char combustion and water-gas reaction. Thus increasing char conversion. At the studied condition, a theoretical maximum of cold gas efficiency can be obtained at ER= 0.077, which corresponds to complete char conversion ratio. However, this maximum value cannot be reached in reality since the peak temperature at this condition is too high. An optimal ER value should be around 0.6 in reality.
• Increasing the plasma power also has a slight positive effect on syngas yield and LHV value. However, the influence of DPER is weaker than that of ER and S/F. From economic point of view, the DPER should be chosen as the minimum value which satisfies the energy request for melting the inorganics of MSW.
Nomenclature
vA Specific surface area (m-1)
C Molar concentration (kmol m-3)
D Diffusion coefficient of vapor in the bulk (m2 s-1)
sd Particle diameter (m)
g Gravitational acceleration (m s-2)
evaH Evaporation heat of moisture (J kmol-1)
h Specific enthalpy (J kg-1)
K Interphase momentum exchange coefficient (kg m-3 s-1)
k Heat transfer coefficient (W m-2 K-1)
mk Mass transfer coefficient (m s-1)
rk Kinetic coefficient (m s-1)
M Molar weight (kg kmol-1)
M Mass flow rate (kg s-1)
m Mass transfer rate (kg m-3 s-1)
Nu Nusselt number
P Power (W)
Pr Prandtl number
p Pressure (Pa)
Q Intensity of heat exchange (W m-3)
q Heat flux (W m-2)
Re Reynolds number
r Reaction rate (kmol m-3 s-1)
kr Turbulent mixing rate (kmol m-3 s-1)
rr Kinetic rate (kmol m-3 s-1)
S Source term
Sh Sherwood number
T Temperature (K)
t Time (s)
v Velocity (m s-1)
v Stoichiometric coefficient
Y Mass fraction
Greek symbols α Volume fraction κ Thermal conductivity (W m-1 K-1) ρ Density (kg m-3)
τ Stress tensor (Pa) µ Dynamic viscosity (Pa s)
Subscripts
air Air
yx HC Light Hydrocarbons
CO Carbon monoxide
2CO Carbon dioxide
g Gas phase
2H Hydrogen
OH 2 Steam
i i th species
MSW MSW
moi Moisture
2O Oxygen
p Phase p
pla Plasma
q Phase q
s Solid phase
stoic Stoichiometric condition
1tar Primary tar
6. References
[1] Q. Zhang, L. Dor, D. Fenigshtein, W. Yang, W. Blasiak, Gasification of municipal solid waste in the Plasma Gasification Melting process, Appl. Energy. In press.
[2] C. Lucas, D. Szewczyka, W. Blasiaka, S. Mochidab, High-temperature air and steam gasification of densified biofuels, Biomass Bioenergy. 27 (2004) 563–575.
[3] W. Blasiak, D. Szewczyk, C. Lucas, Reforming of biomass wastes into fuel gas with high temperature air and steam. Pyrolysis &Gasification of Biomass & Waste Conference, Strasbourg, France, 2002.
[4] P. Lv, Z. Yuan, L. Ma, C. Wu, Y. Chen, J. Zhu, Hydrogen-rich gas production from biomass air and oxygen/steam gasification in a downdraft gasifier, Renewable Energy, 32 (2007), 2173–2185.
[5] R.K. Manurung, A.A.C.M. Beenackers, (1994), Modeling and simulation of an open-core downdraft moving bed rice husk gasifier, in A.V. Bridgewater (Ed.), Advances in Thermochemical Biomass Conversion, London, pp. 288-309.
[6] C.D. Blasi, Dynamic Behavior of Stratified downdraft gasifier, Chem. Eng. Sci. 55, (2000) 2931-2944.
[7] M. Syamlal, L. Bissett, METC Gasifier Advanced Simulation (MGAS) Model; Morgantown Energy Technology Center: Morgantown, WV, 1992.
[8] K.M. Bryden, K.W. Ragland, Numerical modeling of a deep, fixed bed combustor, Energy Fuels. 10 (1996) 269-275.
[9] A. Rogel, J. Aguillon, The 2D eulerian approach of entrained flow and temperature in a biomass stratified downdraft gasifier, Am. J. Appl. Sci. 3 (2006) 2068-2075.
[10] Q. Zhang, L. Dor, W. Yang, W. Blasiak, CFD Modeling of Municipal Solid Waste Gasification in a Fixed-Bed Plasma Gasification Melting Reactor, International conference on thermal treatment technologies, Jacksonville, Florida, USA, 2011.
[11] ANSYS FLUENT 12.0 Theory Guide, ANSYS Inc. USA, 2009.
[12] J.T. Jenkins, S.C. Cowin, Theories for flowing granular materials, In S.C. Cowin (Ed.), Mechanics Applied to the Transport of Bulk Materials, 1979, pp. 79-89.
[13] M. Syamlal, W Rogers, T.J. O’Brien, MFIX Documentation: Volume 1, Theory Guide; National Technical Information Service, Springfield, VA, 1993.
[14] J.A.M. Kuipers, K.J.V. Duin, F.P.H. Beckum, W.P.M. Swaaij, A numerical model of gas-fluidized beds, Chem. Eng. Sci. 47 (1992) 1913-1924.
[15] D. Gidaspow, B. Ettehadleh, Fluidization in two-dimensional beds with a jet: 2. Hydrodynamic modeling, Ind. Eng. Chem. Fundamen. 22 (1983) 193–201.
[16] P.C. Johnson, R.J. Jackson. Frictional–collisional constitutive relations for granular materials, with application to plane shearing, Fluid. Mech. 176 (1987) 67–93.
[17] D.J. Gunn, Transfer of heat or mass to particles in fixed and fluidised beds, Int. J. Heat Mass Transfer, 21 (1978) 467–476.
[18] Y.B. Yang, Y.R. Goh, R. Zakaria, V. Nasserzadeh, J. Swithenbank, Mathematical modelling of MSW incineration on a travelling bed, Waste Manage. 22 (2002) 369-380.
[19] Y.B. Yang, V.N. Sharifi, J. Swithenbank, Effect of air flow rate and fuel moisture on the burning behaviours of biomass and simulated municipal solid wastes in packed beds, Fuel. 83 (2004) 1553-1562.
[20] W. Yang, A. Ponzio, C. Lucas, W. Blasiak, Performance analysis of a fixed-bed biomass gasifier using high-temperature air, Fuel Process Technol. 87 (2006) 235-245.
[21] W.E. Ranz, W.R. Marshall, Evaporation from drops. Part I, Chem. Eng. Prog. 48 (1952) 141–146.
[22] W.E. Ranz, W.R. Marshall, Evaporation from drops. Part II, Chem. Eng. Prog. 48 (1952) 173–180.
[23] M. L. Boroson, J.B. Howard, J.P. Longwell, W.A. Peters, Product yields and kinetics from the vapor phase cracking of wood pyrolysis tars, AIChE J. 35 (1989) 120-128.
[24] A.G. Liden, F. Berruti, D.S. Scott, A kinetic model for the production of liquids from the flash pyrolysis of biomass, Chem. Eng. Commun. 65 (1988) 207-221.
[25] C.D. Blasi, AIChE J. Modeling wood gasification in a countercurrent fixed-bed reactor, 50 (2004) 2306-2319.
[26] W.R. Chan, M. Kelbon, B.B. Krieger, Modelling and experimental verification of physical and chemical processes during pyrolysis of large biomass particle. Fuel. 64 (1985) 1505-1513.
[27] C. Wu, C. Chang, J. Hor, On the thermal treatment of plastic mixtures of MSW: Pyrolysis kinetics, Waste Manage. 13 (1993) 221-235.
[28] A.K Varma, A.U. Chatwani, F.V. Bracco, Studies of premixed laminar hydrogen-air flames using elementary and global kinetics models, Combust. Flame 64 (1986) 233-236.
[29] F.L. Dryer, I. Glassman, High temperature oxidation of CO and CH4
[30] J.B. Howard, G.C. William and D.H. Kinetics of carbon monoxide oxidation in postflame gases, Symposium (International) on Combustion, 14 (1973) 975–986.
, 14th Symposium on Combustion, 1973, pp. 987-1003.
[31] W.P. Jones, R.P. Lindstedt, Global reaction schemes for hydrocarbon combustion, Combust. Flame 73 (1988) 233-249.
[32] G.B. Grebenshchikova, Podzemnaya Gazifikatsiya, 2 (1957) 54–57.
[33] H. Yoon, J. Wei, M.M. Denn, A model for moving-bed coal gasification reactors, AIChE Journal 24 (1978) 885-903.
[34] M.L. Hobbs, P.T. Radulovic, L.D. Smoot, Combustion and gasification of coals in fixed-beds, Prog. Energy Combust. Sci. 19 (1993) 505-586.
[35] J.A. Arthur, Reaction between carbon and oxygen, Trans. Faraday Soc. 1951, 47, 164-78.
[36] D.D. Evans, H.W. Emmons, Combustion of wood charcoal, Fire Res. 1 (1977) 57-66.
[37] H.W. Yoon, M.M. Denn, A model for moving-bed coal. gasification reactors, AIChE J. 24 (1978) 885-903.
[38] M. L. Hobbs, P.T. Radulovic, L.D. Smoot, Modeling Fixed-Bed Coal. Gasifiers, AIChE J. 38 (1992) 681-702.
SUPPLEMENT V
Qinglin Zhang, Liran Dor, Lan Zhang, Weihong Yang, Wlodzimierz Blasiak.
Performance analysis of municipal solid waste gasification with
steam in a Plasma Gasification Melting reactor
Manuscript submitted to Applied Energy, in July 2011
1
Performance analysis of municipal solid waste gasification with
steam in a Plasma Gasification Melting reactor
Qinglin Zhanga, Liran Dorb, Lan Zhanga, Weihong Yanga, Wlodzimierz Blasiaka aEnergy and Furnace Technology Division, Royal Institute of Technology, Brinellvägen 23, S-10044 Stockholm,
Sweden b
Abstract
Environmental Energy Resources Ltd, 7 Jabotinski St., 52520 Ramat-Gan, Israel
Plasma Gasification Melting (PGM) is a novel gasification technology which offers a promising treatment of
low-heating-value fuels like municipal solid waste (MSW), medical waste (MW) and other types of waste. By
considering the differences in pyrolisys characteristics between cellulosic fraction and plastics in MSW, a
semi-empirical model was developed to predict the performance of the PGM process.
The measured results of MSW air and steam gasification in a PGM demo-reactor are demonstated and compared
with the model predicted results. Then, the effects of dimentionless opertation parameters (ER, PER and SAMR)
are discussed. It was found that all three numbers have pisitive effects on system cold gas efficiency. The reasons
can be attributed to promoted tar cracking by enhanced heat supply. The effects of PER and ASME on syngas LHV
are also positive. The influence of ER on syngas pyrolysis can be divided into two parts. When 0.04 < ER < 0.065,
the effect of ER is on LHV positive; when 0.065 < ER < 0.08, the effect of ER is positive. The is phenomenum
was explained by two contradictory effects of ER.
Interactions exist between opertation parameters. For example, increasing PER narrows the possible range of ER
while increasing SAMR broadens possible ER range. Detail extents for those opertation parameters are
demonstated and discussed in this paper. Finally, the optimal point aiming at obtaining maximun syngas LHV and
system CGE are given.
1. Introduction The increment of municipal solid waste (MSW) yield gives prominence to sustainable waste disposal.
Among various methods of waste disposal, gasification is one of the promising technologies. During gasification,
the chemical energy inside MSW can be recovered through production of a combustible syngas. Meanwhile, the
volume of solid waste can be sharply reduced [1]. Compared to direct incineration, MSW gasification prevented
largely dioxin formation and reduced thermal NOx formation due to low temperature and the reduction condition.
Moreover, Moreover, the volume of produced syngas was much lower than that of flue gas from incineration. The
reduction of gaseous volume produced positive reflects in a decreasing size of gas cleaning equipment[2]. The
state-of-art of MSW gasification technology was summarized by Thomas[3].
2
If additional sensible heat is provided to the gasification process, the efficiency of gasification can be
increased [4]. Meanwhile, other benefits like higher syngas quality, better system stability, and lower tar yield can
be obtained [5]-[6]. When the temperature of gasification residual reaches its melting temperature, the solid
residual would be melted and form vitrified slag. In that case, corrosion and emission by retaining heavy metals
(with the exception of mercury, zinc and lead, which can vaporize at high temperatures and be retained in fly ash
and syngas [7]) would be prevented since they were trapped by slag [8]-[11]. Based on above studies, a new MSW
gasification technology called Plasma Gasification Melting (PGM) has been developed. In this technology, MSW
gasification and plasma melting of gasification residual are achieved in a single fixed-bed reactor by a continuous
one-step process. By applying PGM technology, benefits like less investment and operation cost, reduced
emissions, and overall environmental friendliness can be achieved.
Steam is a widely used gasification agent which affects energy and mass balance of the gasification process.
The previous experimental study on the characteristics of steam added gasification [6], [12] showed that the
addition of steam favors the formation of H2 and CO2
In our previous work, experimental test has been performed and analysis has been carried out to study the
characteristics of a trial PGM reactor
, and restrains the CO formation by water-gas and
water-gas-shift reactions. Total syngas yield will decline since the addition of steam decreases the temperature
inside the fixed-bed. It was also discovered that the steam temperature has a positive effect on both syngas LHV
and Syngas yield, so high-temperature steam feeding is more favorable for gasification.
[13]. Several test runs were performed at different operation conditions
where both air and steam are used as gasification agents. For each test run, the temperature and pressure
distribution inside the reactor, as well as syngas composition, were measured. Due to the limitation of test
condition, it is not practical to test all possible operation conditions by experimental measurement. For further
understanding of the gasification characters of MSW in the PGM reactor, it is necessity of develop an accurate
model to predict the performance of PGM process under various operating conditions, and determine the optimal
operating conditions according to the desired target.
Process simulation is an important tool which has been widely applied in various energy-engineering
processes. For gasification, various models have been developed. For most models, a global chemical equilibrium
was assumed [14]-[19]. The equilibrium might be available for entrained-flow, fluidized-bed and downdraft
fixed-bed gasification, but not an appropriate approach for updraft fixed-bed gasification. Firstly, in updraft
fixed-bed gasification process, the pyrolysis gases go straight out of the reactor. The chemical equilibrium model
cannot correctly predict the yields of pyrolysis. Secondly, the equilibrium model always underestimates the yield
of light hydrocarbons from gasification [18]. Vittorio tried to simulating fixed-bed coal gasification by using
several individual reactors, and consider the whole gasification process as an assembly of there reactors. In his
work the pyrolysis process is simply assumed as a constant yield reaction, the influence of pyrolysis temperature
on pyrolysis yields is not considered [20]. When modelling MSW gasification, the pyrolysis mechanism is more
complicated than that of coal and biomass, because the composition of MSW is complex. In a common MSW
sample, the mass fraction of volatile species is 60% to 80%. An accurate simulation of the pyrolysis is the key for a
successful simulation of MSW gasification in a fixed-bed gasifier. However, very few works has been found on
this topic.
In this study, a semi-empirical model for the PGM process of MSW is developed using Aspen Plus. Results
from the test runs of air and steam gasification inside the PGM reactor are demonstrated, and compared with the
predicted results. The effects of operating parameters such as air feeding rate, steam feeding rate and plasma power
on characteristics of MSW gasification in the PGM process are discussed. The interactions between operating
parameters are also considered from view points of both energy and chemical equilibrium. Finally, the optimal
3
operation conditions by considering highest syngas lower heating value (LHV) and cold gas efficiency (CGE) are
suggested.
2. Methodology
2.1. Feedstock The feedstock used in this study is MSW collected in Israel. The main components of this MSW are paper,
wood, cloth vegetation material, plastics, rubber and debris. The proximate and ultimate analysis was performed
for a sample of this MSW, and the results are shown in Table 1.
2.2. The PGM reactor A PGM demonstration reactor has been built up in Northern Israel, with a capacity of 12 to 20 tons of MSW
per day. The PGM reactor is generally a moving-bed counter current updraft gasifier, with a melting chamber
placed at its bottom. . The scheme of the reactor is shown in Figure 1. Air is fed into the melting chamber through
plasma torches at high speed, and forms high temperature plasma jets which melt the inorganic components which
fall from the fixed-bed. Then, air with residual heat mixes with steam fed through steam nozzles placed at the side
wall of the melting chamber, and flows into the fixed-bed. The feeding rates of air and steam are controlled by
central control system. Feedstock is fed into the reactor from airtight feeding chambers located at the top of the
reactor. MSW is fed intermittently every half an hour.
In order to measure the temperature distribution inside the reactor, thermocouples are placed along the
gasifier shaft. Additionally, a probe is placed in the syngas outlet to obtain syngas samples, which are sent to a gas
analyzer for composition analysis. Detailed information about the reactor has been published previously [13].
1.1. Operation parameters
In this study, three dimensionless characteristic numbers are used to characterize the operating parameters of the PGM air and steam gasification.
Plasma flow supplies heat for gasification in the PGM process. The amount of plasma heat is characterized by plasma energy ratio (PER), which is defined as:
MSWMSW
pla
mLHVP
PER⋅
= (4)
The equivalence ratio (ER) is commonly used to indicate quantitatively the extent of combustion in the combustion/gasification processes:
( )( )stoicMSWair
MSWair
mmmmER
//
= (5)
Steam-air mass ratio (SAMR) is a dimensionless parameter which was usually used to characterize the steam feeding rate in air and steam gasification process. It was used in this work as the third dimensionless parameter.
4
2. Numerical model A steady state model of the PGM gasification process was developed using Aspen Plus. Considering the
real process in the PGM reactor, the model schematized the PGM process into four different sections: drying,
pyrolysis, char gasification and combustion, and plasma melting. Moisture, volatiles, fixed-carbon and ash were
removed from feedstock in these sections, respectively. The simplified scheme of the model is shown in Figure 2.
The following model assumptions are used in this work:
• The system is zero dimensional. Material properties like temperature (of gas phase and solid phase), gas
composition and solid composition in each zone is expressed by “mean” values, which are calculated
from the mass and energy balance.
• The flow of solid is from top to the end, while the gas flow is from the bottom to the top. No reflux for
each phase is allowed. • The ash-free fuel is composed of C, H and O. The gas-phase species included in this model are CO , 2H ,
2CO , OH2, 4CH , 42HC , 2O , 2N and tars (including primary tar from cellulosic group, primary
tar from plastic group and secondary tar). • The heat loss of each section is calculated from the measured temperature layout of gasifier wall surface
and the gasifier structure.
The flow sheet of the model is shown in Figure 3. The blocks shown in this figure is summarized and
described in Table 2.
2.1. Drying
In the drying section, raw MSW is heating up by hot syngas and decomposed into dry MSW and steam. The
energy balance of heat exchanger is described as:
++= ∫∫∑ ∫
−
−
−
−
−
−
−− steam
T
Tsteampsteam
T
TdryMSWpdryMSW
i
T
Tipi LdTCmdTCmdTCm
outsyngas
inMSW
outMSW
inMSW
insyngas
outsyngas
,,, (1)
Considering the impact of heat gradient inside MSW particles, the average temperature of dried MSW at the
DRYER outlet is set to 120 ºC. The heat capacity of MSW is calculated using the correlation given by IGT [21].
2.2. Pyrolysis
The heterogeneous MSW composition determines the complication of pyrolysis. According to the pyrolysis
characteristics, studied MSW composition can be divided into two main groups: cellulosic fractions (Wood, paper,
vegetation and cardboard) and plastics (PE, PP, PVC and rubber). In this model, the pyrolysis of each group was
simulated separately. The two-step pyrolysis model [22] was applied to both groups. The yields of the primary
pyrolysis products, including the composition of produced gases and tars are taken from literatures [23]-[24]. To
simplify the model, all light hydrocarbons except CH4 are considered as C2H4
[23]
. The yield of primary tar cracking
of the cellulosic group is taken from Hla . No literature data is found for the secondary pyrolysis of plastic
mixture, so the yield of primary tar cracking of the plastic group is calculated from elementary balance. The
composition of secondary tar is assumed to be benzene.
It has been proved that for a fixed-bed gasifier, the tar production is sensitive to the pyrolysis temperature. In
this model, the extent of primary tar cracking is controlled by pyrolysis temperature [25]:
5
))(exp( 0TTAY pyr −−= (2)
where CT 5000 = . The constant A varies for different feedstock, and can be calculated from test results. The
Combustion values of MSW and tars are calculated based on their elementary compositions, using the empirical
correlation given by Boie [26].
2.3. Char gasification and combustion
Char coming from the pyrolysis zone meets and reacts with gasification agents in the char gasification and
combustion section. At the same time, homogeneous gas phase reactions also take place. When high temperature
steam is injected into this section, as we did in the test runs, the reaction system will become more complex. Firstly,
the injection of high temperature steam activates the water shift reaction and water-gas shift reaction, which visibly
affects the chemical equilibrium in this section. Secondly, the introductions of additional mass and energy affects
the mass and energy balance inside PGM reactor. It is not practical to accurately simulate all reactions which occur
in this section. Instead, the Gibbs free energy theory is applied in this section. The products of this section is
calculated by minimizing the system Gibbs energy.
2.4. Plasma melting
The inorganic components (ash) of the MSW coming from the gasification and combustion zone were
melted by high temperature plasma air in the plasma melting section.
The composition of the inorganic components is assumed according to the original composition of the MSW.
Based on the assumed composition, the heat capacity of the inorganics is calculated as following:
∑=
=n
iipiashp CC
1,, ω (3)
The melting latent heat of the inorganics is calculated similarly to that of the heat capacity. The heat loss of
the plasma melting process is set to 30% of the total plasma energy, which is summarized from the tested
temperature distribution inside the melting chamber.
2.5. Dimensionless operation parameters
In this study, three dimensionless characteristic numbers are used to characterize the operating parameters of the PGM air and steam gasification.
Plasma flow supplies heat for gasification in the PGM process. The amount of plasma heat is characterized by plasma energy ratio (PER), which is defined as:
MSWMSW
pla
mLHVP
PER⋅
= (4)
The equivalence ratio (ER) is commonly used to indicate quantitatively the extent of combustion in the combustion/gasification processes:
( )( )stoicMSWair
MSWair
mmmmER
//
= (5)
6
Steam-air mass ratio (SAMR) is a dimensionless parameter which was usually used to characterize the steam feeding rate in air and steam gasification process. It was used in this work as the third dimensionless parameter.
3. Results and analysis
3.1. Measured results of air and steam gasification
The measured results [13] of air and steam gasification of MSW in the PGM demo-reactor are shown in
Table 3. Results are shown in terms of syngas yield, syngas LHV and the ratio of H2 and CO volume fractions
(H2
The predicted results at the same operation parameters as that of test measurement are also shown in
/CO). Despite the fact that the number of cases are very limited, some effects of operation parameter can still
be found. For example, the measured results for case 1 and case 2 show that increasing SAMR is beneficial for
syngas production. The results of case 2 and case 3 demonstrate a negative effect of decreasing ER on syngas yield
and LHV. In order to further understand the character of air/steam gasification in the PGM process, more results
from model prediction could be used.
Table 3.
By comparing the predicted results with the measured results, it was found that the results from modeling are in
the acceptable ranges for analyzing the character of PGM process. It has to be point out that the calculation method
for syngas LHV in this work is slightly updated from the one we used in the previous publication [13]. The LHCs
are assumed to be composed of CH4 and C2H4, instead of the previous assumption that LHCs were pure CH4
[5]
. The
new assumption is supported by Ponzio et al. .
3.2. Effect of Plasma Power
The high-temperature plasma air injection is the most important significance of the PGM process. It supplies
heat for the melting of the inorganic component of MSW. After that, the residual heat provide sensible heat to
gasification. In this way, the power of plasma affects the energy equilibrium of the whole gasification process, and
directly influences the temperature profile, syngas composition, tar yield and stability of the gasification process. A
serious of cases are simulated to investigate the effect of PER on gasification characters in PGM process. In these
cases, the values of ER and SAMR are set to 0.06 and 0.389 respectively, which are testified as “reasonable”
values for PGM air and steam gasification by previous test runs. The value of PER varies from 0.098 to 0.137.
The effect of PER on the average temperature in the gasification and pyrolysis zone is illustrated in Figure
4a. It was found that both gasification temperature and pyrolysis temperature increase linear with PER. This is
easy to understand since increasing PER enhance the average temperature of feeding air, and increases the heat
supply for gasification and pyrolysis.
Figure 4b shows the syngas composition, as well as tar yield with different PER. All the gaseous species are
shown in volume fractions on dry basis, and tar is shown by tar-MSW mass ratio on dry basis. It was found that the
volume fractions of all combustible gaseous increase with PER, while the trends of CO2 and N2 are opposite. The
increment of combustible gases is mainly due to promoted tar cracking by increasing pyrolysis temperature. At the
same time, the total yields of incombustible gases like CO2 and N2 do not vary much. Considering the increasing
of combustible gases with PER, the decreasing trends of CO2 and N2
Figure 4
volume fractions are understandable.
c shows the effects of PER on syngas yield and syngas LHV, where both results are calculated on
dry basis. It was found that both the syngas yield and syngas LHV increase with PER. This is not hard to
understand since the increase of combustible gas yields by tar cracking is profitable for both quantity and quality
7
of syngas. When PER increase from 0.098 to 0.137, the syngas yield increases from 0.96 to 1.08 Nm3/kg MSW. At
the same time the syngas LHV varies from 7.32 to 9.31 MJ/Nm3
Figure 4
. It seems the positive effects of PER provides a
possible method for not only solving the tar problem, but also increase the syngas yield and quality. However, it
has to be noticed that beneficial effects of increasing PER is not unlimited. As we can see in , the
temperature inside the reactor also increases with PER. A high PER value may leads to the formation of a high
temperature zone in the combustion and gasification section. Too high temperature challenges the thermostability
of the reactor wall. Furthermore, the low-melting-point components in the solid residual like SiO2
Figure 4
may be melted
in the gasification and combustion section if the temperature is too high. The partial melting of solid residual will
dramatically decrease the void fraction in the fixed-bed, and leads to the occurrence of bridging. It can be found
from that when PER= 0.26, the average temperature in the gasification section has reached 1330 ºC. This
temperature is already too high for an engineering application.
3.3. Effect of ER
For a traditional gasifier, the energy needed for feedstock heating up, pyrolysis and char gasification is
mainly from the partial combustion of char. The equivalence ratio (ER) for traditional gasifier should be around
0.3 to fulfill the need of energy. For PGM air and steam gasification, heat can be supplied by plasma and high
temperature steam, so the ER for a PGM gasifier will be much lower (0.04-0.10). It is worthwhile to study the
influence of ER on the performance of a PGM gasifier.
Theoretically, the effects of ER on gasification process should be considered from two aspects. On one side,
higher ER provides more chemical heat by combustion. It is known that increased heat supply is beneficial for
both syngas yield and LHV value, so this effect of ER is positive. On the other side, higher ER means more
combustion in the reactor, which will consume some combustible gases. Additionally, the increasing N2
A group of simulations with different ER was carried out to study the exact influence of ER on PGM process.
The values of SAMR and PER are set to 0.389 and 0.118, respectively. The value of ER varies from 0.04 to 0.08.
introduced
into the reactor dilutes the content of combustible gases. From this point of view, the ER also has negative effects
on syngas production. The final influence of ER on PGM process should be a combination of these two aspects.
Figure 5a shows the syngas composition and tar yield with different ER value. It was found that when ER
increases, the volume fractions of CH4, C2H4 and N2 increase, and the volume fraction of H2 and tar yield
decrease. The volume fraction of CO first increases and then decreases. An opposite trend was found for CO2
volume fraction. The increase of CH4 and C2H4 volume fractions can be understand as the result of positive effect
of ER on tar cracking, while the increase of N2 and decrease of H2 volume fractions are the results of negative
effects of ER. The variations of CO and CO2
Figure 5
volume fractions are affected by both aspects.
b shows the variation of syngas LHV and system CGE with increasing ER. It was found that the
influence of ER on syngas LHV can be divided into two parts. When ER increases from 0.04 to 0.07, the syngas
LHV increases from 6.11 to 8.63MJ/Nm3
3.4. Effect of SAMR
. The positive effect of ER is dominant. When ER increases from 0.07 to
0.08, the syngas LHV keeps almost constant. It seems negative effect of ER starts to appear in this range, and
counterbalances the positive effect. For system CGE, however, the effect of increasing ER is positive in all ER
range.
The feeding of high temperature steam influences the PGM process from two aspects. Firstly, steam is
involved in chemical reactions such as water-gas reaction and water gas shift reaction. In that case, it influences
the chemical equilibrium of the PGM system. Secondly, the high temperature steam changes the total mass and
8
energy flow inside the reactor, and influence the energy balance of the system.
As an example, the effect of different SAMR on syngas composition and tar yield for ER=0.06 and
PER=0.118 is shown in Figure 6. It was found that the most important effect of increasing SAMR is the variation
of H2, CO, and CO2 volume fractions. When the SARM increases from 0 to 0.67, the volume fraction of H2 in
syngas increases from 9.5% to 21.3%. The volume fraction of CO2
[6]
increases similarly from 12.4% to 20.2%,
while the volume fraction of CO decreases from 26.8% to 14.9%. The similar trends were also reported by other
researchers -[12]. This phenomenon is the result of promoted water-gas shift reaction
( 222 COHOHCO +→←+ ) by increasing steam feeding rate. It was also found that the yield of tar shown
a slight decreasing trend when the SAMR increases. at the same time, the CH4 and C2H4
[27]
volume fractions
increase slightly. It is believe that this phenomenon is the result of steam preheating, which introduces extra energy
to the PGM system, and increases the global temperature of pyrolysis. It was reported by Lewis et al. that the
critical steam temperature for supporting energy supply in steam-only gasification process is above 1200 ºC. In
PGM process, due to the heat supply from plasma air and char combustion, the critical steam temperature
should be reduced. It was implied by the tar decreasing that the critical steam temperature at the analyzed
condition is lower than 1000 ºC. As a result of the extra heat supply by high-temperature steam, the syngas
yield and LHV increase slowly with SAMR. It was also found that the effect of SAMR on syngas composition
weakens with increasing SAMR. The effect is most remarkable when SAMR varies from 0 to 0.1.
3.5. Interaction between operating parameters
PER, ER and SAMR generalize main operating parameters during PGM process. However, the effects of
different operating parameters are not independent, and internal connections exist between these parameters. For
deep understanding of the PGM process, these interactions between operating parameters should be considered.
3.5.1. Interactions between ER and PER
In the PGM process, the required heat for MSW gasification comes from two sources: the sensible heat of
plasma air and chemical heat from char combustion. In other words, the energy equilibrium of PGM gasification is
controlled by both PER and ER. From this point of view, the effects of PER and ER are connected to each other.
When study the energy equilibrium of the PGM process, the interaction between PER and ER should be
considered. The SAMR value was set to 0.389 in this study.
Figure 7 shows the delimitation of possible operation extent of PER and ER in the PGM process. Three
curves are defined to restrict the logical area for PGM:
• ERpla, min shows the minimum of ER requested for generating plasma flow. In PGM process, air is
used as the carrier of sensible heat from plasma generators. The relationship between PER and
ERpla, min is linear. The gradient of the ERpla, min
( )pla
MSWstoicMSWair h
LHVmmk /=
denotes the ratio between MSW LHV and the
thermal enthalpy of plasma air:
(6)
• ERgasif, min shows the ER needed for complete gasification (i.e. no solid carbon residual and enough temperature for gasification and pyrolysis). In this work, the request for complete
gasification is satisfied when the syngas temperature at the outlet is higher than 120 ºC.
9
• ERtem, max
• PER
shows the maximum of ER to prevent too high temperature in the char combustion and gasification section. If this temperature is too high, the wall material of the reactor might
be damaged. In this study, the maximum of the temperature is set to 1300 ºC.
mel, min
Four different regions are divided by these curves:
shows the minimum of PER required for melting all the solid residual.
• Region 1: In this region, the PGM process can operate normally. The energy supplied by plasma
and char combustion is enough for MSW gasification to take place. The temperature in the
gasification and combustion zone is not too high to damage the reactor wall.
• Region 2: In this region, the energy supplied by plasma, char combustion and High temperature
steam is not enough for supporting MSW gasification.
• Region 3: In this region, the temperature of the char combustion and gasification section is higher
than 1300 ºC. In other words, the temperature in char combustion and gasification section may
damage the reactor wall.
• Region 4: In this region, the energy supplied by plasma flow is not enough for melting of solid
residual from MSW gasification.
It can be found from Figure 7 that when PER is less than 0.045, the plasma energy is not enough for entirely
melting of inorganic components in MSW. When PER increases from 0.045 to 0.13, the energy require for
inorganic components melting is satisfied. The extent of available ER is limited by ERtem, max and ERgasif, min
The distribution of syngas LHV, as well as system CGE in region 1 is demonstrated in
. In
other words, the minimum of available ER is restricted by entire energy supply, and the maximum of available ER
is controlled by gasification&combustion temperature. When PER further increases from 0.13 to 0.14, the lower
limit of ER does not exist anymore, which means the energy supply is enough for PGM even the secondary air
feeding is set to 0. If PER is higher than 0.14, the PGM is not available because the temperature at the char
gasification&combustion section is too high. Generally speaking, the available PER extent is 0.045-0.14. Increase
of PER narrows the variation range of ER.
Figure 8. It was
found that the maximum syngas LHV in region 1 is about 9.5, while the minimum is about 4.0. It has been
discussed previously that the LHV variation is mainly caused by thermal cracking of primary tar. The large
difference between maximum and minimum syngas LHV illustrates that the extent of tar cracking is a very
important factor which determines the quality of syngas in PGM process. Furthermore, it is obvious that the effect
of PER on syngas LHV is stronger than that of ER. The positive effect of ER on syngas LHV is due to promoted
primary tar cracking caused by chemical heat from combustion. However, the ER still have some negative effects
on syngas LHV. For example, increased combustion by increasing ER consumes some combustible gases in syngas.
Additionally, the introduced N2
From
also dilutes the contents of combustible gases. These negative effects somehow
weaken the positive effect of ER. So the maximum LHV was found in the area with highest PER value. The
dependence of LHV on ER and PER has been confirmed by previous running of the pilot PGM reactor.
Figure 8 (b) it was found that the maximum CGE in region 1 is about 0.62 and the minimum is about
0.22. The maximum CGE appears when ER=0.08 and PER=0.10. The large difference of CGE is also explained by
the influence of the extent of tar cracking. The influences of PER and ER on CGE have similar intensity. An
interesting phenomenon found in Figure 8 (b) is that the effects of ER and PER on CGE shows a linear relation. It
implies that the influence of ER and PER can be further synthesized to a unified parameter. The further correlation
of ER and PER can be an interesting topic of our future work.
10
3.5.2. Considering the oxygen equilibrium
Steam and air are two popular gasification agents which supply oxygen for the gasification process. As the
material base of gasification, the oxygen supply directly influence the conversion of C during gasification and
combustion section. From this perspective, the ER and SAMR may also have internal connecting with each other.
Figure 9 shows the delimitation of possible operation extent of SAMR and ER in the PGM process at PER =0.118.
Three curves are used to restrict the possible operation conditions for SAMR and ER: ERpla, min, ERgasif, min, and
ERtem, max
These curves defined 3 main regions with different operation conditions: In region 1’, the PGM process can
work continuously; In region 2’, the energy supplied by plasma and char combustion is not enough for MSW
gasification; In region 3’, the temperature of the char combustion and gasification section is too high. It was found
that when SAMR increases, the maximum of possible ER increases and the minimum of possible ER in region 1’
decreases. Increase of SAMR means enhanced oxygen supply from steam. In that case, the oxygen equilibrium in
the reactor is affected, and the requested air decreases. The increase of maximum ER can be explained by the
increases of total heat capacity with increasing SAMR, which increases the uniformity of temperature distribution
inside the reactor. This uniformity is also beneficial to syngas LHV because the temperature difference between
gasification and pyrolysis will be reduced.
.
The distribution of syngas LHV in region 1’ is demonstrated in Figure 10. It was found that the syngas LHV
in region 1’ varies from 6.5 to 9.0 MJ/Nm3
Figure 10
. The increase of SAMR has positive effects on syngas LHV. This
positive effect may be mainly due to the high temperature of steam, which also introduce some heat into the PGM
system. At the same time, the decreased temperature difference between gasification and pyrolysis section by
increasing SAMR enhances the potential of LHV increase by larger energy supply. An interesting phenomenon
found in is that the effect of increasing ER on syngas LHV changes when SAMR is larger than 0.55. In
this area, the LHV first increase, and then start to decrease when ER keep increasing. The maximum of LHV
appears at about ER=0.055. This result illustrate that the positive aspect of ER effect by increasing chemical heat
is not always dominant. The negative aspects such as consumption of combustible gas and dilution from N2
4. Conclusions
play
important roles in high SAMR condition. The suggested ER in high SAMR condition is 0.055.
A semi-empirical model for air and steam gasification of MSW in the PGM process has been built up.
The performance of PGM reactors with high-temperature steam feeding is analyzed by both test
measurement and model prediction. The effects of three dimensionless operation parameters are discussed. PER
has positive effect for both syngas yield and syngas LHV. The main reason for this effect is the favored tar
cracking by increasing heat supply.
The ER has two contradictory effects on syngas LHV: the positive effect by increasing chemical heat and the
negative effect by syngas combustion and N2
The SAMR mainly influence the equilibrium of water-gas shift reaction in the PGM process. Steam at 1000
ºC can supply some heat for pyrolysis, so the SAMR also have slight positive effect on syngas yield and LHV.
dilution. When ER is lower than 0.065, the positive effect is
dominant; When ER is larger than 0.065, two effects counterbalance each other. The effect of ER on CGE is
positive in the studied region.
Interactions exist between PER and ER. The available extent of PER and ER is defined at air/steam
gasification conditions. The possible range for PER at the studied condition is 0.045-0.14. Increase of PER
narrows the variation range of ER. The optimal syngas LHV can be obtained when the PER reaches its maximum.
11
The effect of ER and PER on syngas CGE seems can be synthesized to a unified parameter.
The available extent of SAMR and ER is defined at PER=0.118. Increasing SAMR broadens the available
range of ER. When SAMR>0.6, the secondary air is not necessary anymore. The optimal syngas LHV can be
obtained at SAMR=0.8 and ER=0.055.
12
Nomenclature m Mass flow rate, kg h
pC-1
Heat capacity, J kg-1 ºC
T
-1
Temperature, ºC L Latent heat, J kgY
-1 Extent of primary tar cracking
ω Mass fraction P Power, W h Thermal enthalpy of plasma air, J kgLHV
-1 Lower heating value, J kg
-1
Subscripts
i Species i MSW Municipal solid waste
steam Steam ash Ash
pla Plasma
air Air stoic Stoichimetric reaction pyr Pyrolysis
13
References
[1] Hefa C, Yuanan H. Municipal solid waste (MSW) as a renewable source of energy: Current and future
practices in China. Bioresour Technol 2010; 101(11):3816-24.
[2] Ladislav B, Petr S, Leos H, Jaroslav O. Analysis of using gasification and incineration for thermal
processing of wastes. Appl Therm Eng 2005; 25(7):1045-55.
[3] Thomas M. Novel and innovative pyrolysis and gasification technologies for energy efficient and
environmentally sound MSW disposal. Waste Manage (Oxford) 2004; 24(1):53-79.
[4] Yang W, Lucas C, Blasiak W. Performance analysis of a fixed-bed biomass gasifier using high-temperature
air. Fuel Process Technol 2006; 87(3):235-45.
[5] Ponzio A, Kalisz S, Blasiak W. Effect of operation conditions on tar and gas composition in high
temperature air/steam gasification (HTAG) of plastic containing waste. Fuel Process Technol 2006;
87(3):223-33.
[6] Lucas C, Szewczyka D, Blasiaka W, Mochidab S. High-temperature air and steam gasification of densified
biofuels. Biomass Bioenergy 2004; 27(6):563–75.
[7] Okada T, Tojo Y, Tanaka N, Matsuto T. Recovery of zinc and lead from fly ash from ash-melting and
gasification-melting processes of MSW – Comparison and applicability of chemical leaching methods.
Waste Manage (Oxford) 2007; 27(1):69-80.
[8] Park YJ, Heo J. Vitrification of fly ash from municipal solid waste incinerator. J Hazard Mater 2002;
91(1-3):83-93.
[9] Sakai SI, Hiraoka M. Municipal solid waste incinerator residue recycling by thermal processes, Waste
Manage (Oxford) 2000; 20(2-3):249-58.
[10] Ecke H, Sakanakura H, Matsuto T, Tanaka N, Lagerkvist A. State-of-the-art treatment processes for
municipal solid waste incineration residues in Japan. Waste Manage Res 2000; 18(1):41-51.
[11] Jung CH, Matsuto T, Tanaka N. Behavior of metals in ash melting and gasification-melting of municipal
solid waste (MSW). Waste Manage (Oxford) 2005; 25(3):301-10.
[12] Blasiak W, Szewczyk D, Lucas C. Reforming of biomass wastes into fuel gas with high temperature air and
steam. Pyrolysis &Gasification of Biomass & Waste Conference, Strasbourg, France, 2002.
[13] Zhang Q et al. Gasification of municipal solid waste in the plasma gasification and melting process. Appl
Energy Forthcoming 2011.
[14] Vaezi M, Passandideh-Fard M, Moghiman M, Charmchi M. Gasification of heavy oils: A thermochemical
equilibrium approach. Fuel 2011; 90(2):878-85.
[15] Mansaray KG, Al-Taweel AM, Ghaly AE, Hamdullahpur F, Ugursal VI. Mathematical modelling of a
fluidized bed rice husk gasifier: Part I-Model development. Energy Sources 2000; 22 (1):83-98.
[16] Mehrdokht B N, Nader M. Simulation of biomass gasification in fluidized bed reactor using ASPEN Plus.
Biomass Bioenergy 2008; 32(12):1245-54.
[17] Wayne D, Anthony R, David K. The effect of air preheating in a biomass CFB gasifier using ASPEN Plus
simulation. Biomass Bioenergy 2009; 33(9):1158-67.
[18] Ilkka H, Esa K. A semi-empirical model for pressurised air-blow fluidized-bed gasification. Bioresour
Technol 2010; 101(12):4608-15.
[19] Raju A, Park CS, Norbeck JM. Synthesis gas production using steam hydrogasification and steam
reforming. Fuel Process Technol 2008; 90(2):330-6.
[20] Vittorio T, Giorgio C. Process analysis and performance evaluation of updraft coal gasifier. In: Proceedings
14
of the 3rd. International Conference on Clean Coal Technologies for Our Future; 2007 May 15-17; Cagliari,
Italy.
[21] Coal Conversion Systems Technical Data Book. Virginia: Springfield; 1978.
[22] Colomba DB. Modeling wood gasification in a countercurrent fixed-bed reactor. AIChE J 2004;
50(9):2306-19.
[23] Hla SH. A theoretical and experimental study on a stratified downdraft biomass gasifier. PhD thesis,
University of Melbourne, 2004.
[24] Williams EA, Williams PT. The pyrolysis of individual plastics and a plastic mixture in a fixed bed reactor.
J Chem Technol Biotechnol 1997; 70(1):9-20.
[25] Fagbemi L, Khezami L, Capart R. Pyrolysis products from different biomasses: Application to the thermal
cracking of tar. Appl Energy 2001; 69(4):293-306.
[26] Boie W. Energietechnik 3.
[27] Levis FM, Swithenbank J, Hoecke DA, Russell NV, Shabangu SV. High temperature, steam-only
gasification and steam reforming with ultra-superheated steam. 5th
International Symposium on Tigh
Temperature Air Combustion and Gasification, October 28-30, Yokohama Japan, 2002
15
Table 1 Proximate and ultimate analysis of MSW
Proximate analysis
Moisture 20.0 %
Fixed carbon (dry basis) 10.7 %
Volatile (dry basis) 77.6 %
Ash (dry basis) 11.7 %
Ultimate analysis
Carbon C 47.9%
Hydrogen H 6.0%
Nitrogen N 1.2%
Chlorine Cl <0.1%
Sulphur S 0.3%
Oxygen O 32.9%
16
Table 2 Description of unit operation blocks used in the model
Block ID Block type Description
DRYER Heat exchanger Exchanging heat between raw MSW and hot syngas
SEPARATR Splitter Separating dry MSW and steam in the drying section
MIXER2 Mixer Mixing syngas and steam from drying section
MIX1 Mixer Mixing dry MSW and gaseous products from the gasification section
PRI-PYRO Stoichiometric reactor Stoichiometric reactor for the primary pyrolysis
SEC- PYRO Stoichiometric reactor Stoichiometric reactor for the primary tar cracking
PYR-SEPR Separator Separating gas phase and solid phase in the pyrolysis section
STEAM-HT Heater Heating up the fed steam to 1273 K
CHAR-GAS Gibbs reactor Gibbs reactor for char gasification
SEPARATI Splitter Separating gaseous and solid products from gasification section
HEATER1 Heater/Cooler Adjust the temperature difference between gaseous and solid
products from gasification section HEATER2 Heater/Cooler
PLASMA-G Heater Heating up the primary air using plasma generator
MELT-H Cooler Calculating the heat usage in the solid residual melting
HEATEX2 Heat exchanger Exchanging heat between primary air and slag, ensuring the slag
temperature at the outlet is constant (1773 K)
MELTING Stoichiometric reactor Simulating the melting process of solid residual
AIR-MIX Mixer Mixing primary and secondary air
HEATLOSS Cooler simulating the system heat loss
17
Table 3 Comparison between measured and predicted results of air and steam gasification in the PGM reactor (dry basis)
Case number 1 2 3 4
Operation parameters
ER 0.060 0.060 0.052 0.048
PER 0.118 0.118 0.118 0.128
SAMR 0.389 0.556 0.452 0.490
Steam temperature (ºC) 1000 1000 1000 1000
Measured results
Syngas yield (Nm3 1.36 /kg MSW) 1.38 1.26 1.29
Syngas LHV (MJ/Nm3 8.23 ) 8.43 8.24 8.70
H2 1.24 /CO 1.53 1.45 1.70
Predicted results
Syngas yield (Nm3 1.27 /kg MSW) 1.32 1.16 1.14
Syngas LHV (MJ/Nm3 8.48 ) 8.70 8.05 8.38
H2 1.16 /CO 1.33 1.32 1.41
18
Figure 1. Scheme of the PGM reactor
19
Plasma Melting
MSWSyngas and tar
Slag Plasma air
Steam Char Gasification and Combustion
Pyrolysis
Drying
Figure 2. Scheme of PGM gasification process
20
Figure 3. GPM gasification flow sheet
21
a)
0
200
400
600
800
1000
1200
1400
0.09 0.10 0.11 0.12 0.13 0.14
PER
Tem
pera
ture
(ºC
)
Gasification temperature Pyrolysis temperature b)
0
10
20
30
40
0.09 0.10 0.11 0.12 0.13 0.14
PER
Syng
as c
ompo
sitio
n (v
ol. %
, dry
basi
s)
0
0.1
0.2
0.3
0.4
0.5
Tar-
MSW
mas
s rat
io
CO CH4 CO2 H2 N2 C2H4 tar c)
7.0
8.0
9.0
10.0
0.09 0.10 0.11 0.12 0.13 0.14
PER
Syng
as L
HV
(M
J/N
m3,
dry
basi
s)
0.94
0.96
0.98
1.00
1.02
1.04
1.06
1.08
Syng
as y
ield
(Nm
3/kg
MSW
,dry
basi
s)
Syngas LHV Syngas yield Figure 4. Effect of PER on a) gasification and pyrolysis temperature, b) syngas composition and tar yield, c) total
syngas yield and syngas LHV
22
a)
0
10
20
30
40
0.04 0.05 0.06 0.07 0.08
ER
Syng
as c
ompo
sitio
n (v
ol. %
, dry
basi
s)
0
0.1
0.2
0.3
0.4
0.5
0.6
Tar-
MSW
mas
s rat
io
CO CH4 CO2 H2 N2 C2H4 tar b)
0123456789
10
0.04 0.05 0.06 0.07 0.08
ER
LHV
of s
ynga
s (M
J/N
m3,
dry
basi
s)
0
10
20
30
40
50
60
70
CG
E (%
, LH
V b
asis
)
LHV CGE Figure 5. Effect of ER on a) syngas composition and tar yield, b) syngas LHV and system CGE
23
0
10
20
30
40
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
SAMR
Syng
as c
ompo
sitio
n (v
ol. %
, dry
basi
s)
0
0.1
0.2
0.3
0.4
0.5
Tar-
MSW
mas
s rat
io
CO CH4 CO2 H2 N2 C2H4 tar Figure 6. Effect of SAMR on syngas composition and tar yield
24
Figure 7. Definition of possible operation extent of PER and ER in the PGM process
25
(a)
(b)
Figure 8 Distributions of syngas LHV and system CGE in Region 1. (a) Syngas LHV, (b) System CGE.
26
Figure 9 Delimitation of possible operation extent of SAMR and ER in the PGM process
27
Figure 10 Distributions of syngas LHV in Region 1’