Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
CO-FIRING BIOMASS WITH COAL IN BUBBLING FLUIDIZED BED COMBUSTORS
A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
OF MIDDLE EAST TECHNICAL UNIVERSITY
BY
ZUHAL GÖĞEBAKAN
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY IN
CHEMICAL ENGINEERING
SEPTEMBER 2007
Approval of the thesis:
CO-FIRING BIOMASS WITH COAL IN BUBBLING FLUIDIZED BED COMBUSTORS
submitted by ZUHAL GÖĞEBAKAN in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Chemical Engineering Department, Middle East Technical University by, Prof. Dr. Canan Özgen __________ Dean, Graduate School of Natural and Applied Sciences Prof. Dr. Nurcan Baç __________ Head of Department, Chemical Engineering Prof. Dr. Nevin Selçuk Supervisor, Chemical Engineering Dept., METU __________ Prof. Dr. Ekrem Selçuk Co-Supervisor, Metallurgical & Materials Engineering Dept., METU__________ Examining Committee Members: Prof. Dr. Yavuz Samim Ünlüsoy Mechanical Engineering Dept., METU _______________ Prof. Dr. Nevin Selçuk Chemical Engineering Dept., METU _______________ Assist. Prof. Dr. Murat Köksal Mechanical Engineering Dept., Hacettepe University _______________ Prof. Dr. Nurcan Baç Chemical Engineering Dept., METU _______________ Assist. Prof. Dr. Görkem Külah Chemical Engineering Dept., METU _______________
Date: 06.09.2007
I hereby declare that all information in this document has been obtained and
presented in accordance with academic rules and ethical conduct. I also declare
that, as required by these rules and conduct, I have fully cited and referenced
all material and results that are not original to this work.
Name, Last Name : Zuhal Göğebakan
Signature :
iii
ABSTRACT
CO-FIRING BIOMASS WITH COAL
IN BUBBLING FLUIDIZED BED COMBUSTORS
Göğebakan, Zuhal
Ph.D., Department of Chemical Engineering
Supervisor: Prof. Dr. Nevin Selçuk
Co-Supervisor: Prof. Dr. Ekrem Selçuk
September 2007, 275 pages
Co-firing of biomass with coal in fluidized bed combustors is a promising alternative
which leads to environmentally friendly use of coal by reducing emissions and
provides utilization of biomass residues. Therefore, effect of biomass share on
gaseous pollutant emissions from fluidized bed co-firing of various biomass fuels
with high calorific value coals have extensively been investigated to date. However,
effect of co-firing of olive residue, hazelnut shell and cotton residue with low
calorific value lignites having high ash and sulfur contents has not been studied in
bubbling fluidized bed combustors to date.
In this thesis study, co-firing of typical Turkish lignite with olive residue, hazelnut
shell and cotton residue in 0.3 MWt METU Atmospheric Bubbling Fluidized Bed
Combustion (ABFBC) Test Rig was investigated in terms of combustion and
emission performance and ash behavior of different fuel blends.
iv
The results reveal that co-firing of olive residue, hazelnut shell and cotton residue
with lignite increases the combustion efficiency and freeboard temperatures
compared to those of lignite firing with limestone addition only. O2 and CO2
emissions are not found sensitive to increase in olive residue, hazelnut shell and
cotton residue share in fuel blend. Co-firing lowers SO2 emissions considerably
while increasing CO emissions. Co-firing of olive residue and hazelnut shell has no
significant influence on NO emissions, however, reduces N2O emissions. Co-firing
cotton residue results in higher NO and N2O emissions. Regarding to major, minor
and trace elements partitioning, co-firing lignite with biomasses under consideration
shifts the partitioning of these elements from bottom ash to fly ash. No chlorine is
detected in both EDX and XRD analyses of the ash deposits. In conclusion, olive
residue, hazelnut shell and cotton residue can easily be co-fired with high ash and
sulfur containing lignite without agglomeration and fouling problems.
Keywords: Co-firing, biomass, lignite, bubbling fluidized bed combustor.
v
ÖZ
BİYOKÜTLE İLE KÖMÜRÜN KABARCIKLI AKIŞKAN YATAKLI
YAKICILARDA BİRLİKTE YAKILMASI
Göğebakan, Zuhal
Doktora, Kimya Mühendisliği Bölümü
Tez Yöneticisi: Prof. Dr. Nevin Selçuk
Ortak Tez Yöneticisi: Prof. Dr. Ekrem Selçuk
Eylül 2007, 275 sayfa
Biyokütle ve kömürün akışkan yataklı yakıcılarda birlikte yakılması, emisyonları
düşürerek kömürün çevre dostu olarak kullanılmasını ve biyokütle artıklarının
değerlendirilmesini sağlayan umut verici bir alternatiftir. Bu nedenle, biyokütle
miktarının kirletici gaz emisyonlarına etkisi, çeşitli biyokütleler ve yüksek kalorili
kömürleri birlikte yakan akışkan yataklı yakıcılarda günümüze kadar yaygın bir
şekilde incelenmiştir. Ancak, zeytin artığı, fındık kabuğu ve pamuk artığının düşük
kalorili, yüksek kükürt ve kül içeren linyitle kabarcıklı akışkan yataklı yakıcılarda
birlikte yakılmasının etkisi üzerine bir çalışma bulunmamaktadır.
Bu tez çalışmasında, tipik Türk linyiti ile zeytin artığı (prina), fındık kabuğu ve
pamuk artığının birlikte yakılması ODTÜ 0.3 MW ısıl gücündeki kabarcıklı akışkan
yataklı yakıcıda farklı yakıt karışımlarının yanma ve emisyon performanslarına ve
kül davranışına etkisi açısından incelenmiştir.
vi
Elde edilen sonuçlar, birlikte yanmanın yanma verimini ve serbest bölge
sıcaklıklarını kireçtaşı ilaveli linyit yakılmasına kıyasla artırdığını göstermiştir. O2 ve
CO2 emisyonlarının karışımdaki artan zeytin artığı, fındık kabuğu ve pamuk artığı
miktarına duyarlı olmadığı görülmüştür. Birlikte yanma SO2 emisyonlarını azaltırken
CO emisyonlarını artırmıştır. Zeytin artığı ve fındık kabuğu ile linyitin birlikte
yakılması NO emisyonlarında önemli değişikliğe yol açmazken, N2O emisyonlarını
azaltmıştır. Pamuk artığının linyit ile birlikte yakılması NO ve N2O emisyonlarını
artırmıştır. Majör, minör ve iz element dağılımlarına ilişkin olarak, linyit ve söz
konusu biyokütlelerin birlikte yakılması bu elementlerin dağılımlarını alt külden
uçucu küle değiştirmiştir. Hem EDX hemde XRD analizlerinde külde klor
bulunmadığı tesbit edilmiştir. Sonuç olarak, zeytin artığı, fındık kabuğu ve pamuk
artığı yüksek kül ve kükürt içeren linyit ile bitişme ve bırakıntı problemleri
olmaksızın kolayca birlikte yakılabilir.
Anahtar Kelimeler : Birlikte yanma, biyokütle, linyit, kabarcıklı akışkan yataklı
yakıcı
vii
To Yusuf
viii
ACKNOWLEDGMENTS
I wish to express my deepest gratitude to my supervisor, Prof. Dr. Nevin
Selçuk for her guidance, advice and encouragement throughout the development of
this study. I would like to show my appreciation to my co-supervisor Prof. Dr.
Ekrem Selçuk, for his valuable suggestions and support.
I am also grateful to Prof. Dr. Yavuz Samim Ünlüsoy and Asst. Prof. Dr.
Görkem Külah for their precious discussions and comments during the progress of
this work.
I also thank to the AFBC research team members for their support during the
experiments; Dr. Olcay Oymak, Dr. Yusuf Göğebakan, Asst. Prof. Dr. Görkem
Külah, Ümit Kırmızıgül, Mehmet Moralı, Aykan Batu, Dr. Ahmet Bilge Uygur,
Zeynep Serinyel, Güzide Aydın and İlker Soner.
I am very grateful for the technical assistance provided by Kerime Güney for
chemical analyses and Cengiz Tan for SEM and EDX analyses.
Financial supports provided by The Scientific and Technical Research
Council of Turkey (TUBİTAK) through MAG-104M200 project and by Middle East
Technical University (METU) through BAP-2006-07-02-00-01 project are gratefully
acknowledged.
I also want to thank to Cem Özdemiroğlu from Alfer Engineering Co. Ltd. for
construction and installation of the stack of the test rig.
I would like to thank to my family Nazife & Dr. Mehmet Coşkun, Gülizar &
İsmail Göğebakan, Dr. Nezihe & Derya Baltalı for their support and encouragement.
My special thanks go to my mother, Nazife Coşkun for her great kindness,
endless understanding and unshakable faith in me.
Finally, my warmest thanks go to my husband, Dr. Yusuf Göğebakan for his
understanding, endless support and encouragement throughout this study.
ix
TABLE OF CONTENTS
ABSTRACT………………………………………………………………....….. iv
ÖZ……………………………………………………………………..………... vi
DEDICATION…………………………………………………...……………... viii
ACKNOWLEDGEMENTS………………………………………………...…... ix
TABLE OF CONTENTS……………………………………………...………... x
LIST OF TABLES…………………………………………………...…………. xiv
LIST OF FIGURES………………………………………………………..…… xix
CHAPTER
1 INTRODUCTION……………………………………...………………... 1
1.1 General………………………………………………...……………. 1
1.2 Aim and Scope of the Thesis……………………………………….. 4
2 BACKGROUND AND LITERATURE REVIEW…………………….... 6
2.1 General……………………………………………………………… 6
2.2 Biomass Firing in FBC Systems……………………………………. 6
2.2.1 Biomass Characteristics…………………………………....…. 6
2.2.2 Biomass Combustion…………………………………………. 8
2.2.3 Operational Problems Associated with Biomass Ash………… 9
2.2.3.1 Bed Agglomeration…………………..………………… 11
2.2.3.2 Ash Deposition and Corrosion………………………… 13
2.3 Literature Review…………...……………………………………… 17
2.3 2.3.1 Biomass Firing Studies……………………………………….. 17
2.3.2 Biomass and Coal Co-firing Studies………………………….. 45
x
3 EXPERIMENTAL SET-UP, PROCEDURE AND CONDITIONS…….. 63
3.1 General……………………………………………………………… 63
3.2 METU 0.3 MWt ABFBC Test Rig…………………………………. 63
3.2.1 The Combustor………………………………………….……. 66
3.2.2 Air and Flue Gas System………………………………….….. 67
3.2.3 Solids Handling System…………………………………….… 68
3.2.4 Cooling Water System………………………………….…….. 70
3.2.5 Gas Sampling and Analysis System…………………….……. 70
3.2.5.1 Gas Sampling Probe...…………………………….….. 70
3.2.5.2 Gas Sampling and Conditioning System………….….. 73
3.2.5.3 Analytical System………………………………….…. 74
3.2.6 Deposit Sampling System………………………………….…. 77
3.2.7 Instrumentation and Control System……………………….… 77
3.3 Pre-Experimental Modifications……………………………………. 82
3.3.1 Feeding System……………………………………….………. 82
3.3.2 Air and Flue Gas System……………………………….…….. 83
3.3.3 Gas Sampling and Analysis System………………….………. 84
3.3.4 Instrumentation and Control System…………………….…… 84
3.4 Operating Procedures……………………………………………….. 85
3.4.1 Procedures before Cold Start-Up………………………….….. 85
3.4.2 Cold Start-Up………………………………………….……… 85
3.4.3 Procedure during Runs………………………………….……. 87
3.4.4 Shutdown……………………………………………….…….. 87
3.4.5 Post Shutdown………………………………………….…….. 88
3.5 Experimental Conditions…………………………………………… 88
3.5.1 Lignite, Biomass and Limestone Characteristics………….….. 88
xi
3.5.2 Operating Conditions…………………….…………………… 98
4 RESULTS AND DISCUSSION…………………………………………. 101
4.1 General……………………………………………………………… 101
4.2 Particle Size Distributions………………………………………….. 101
4.3 Ash Balance, Split and Discharge Compositions…………………... 107
4.3.1 Ash Balance and Ash Split…………...………………….…… 107
4.3.2 Ash Compositions………………….………………….……… 107
4.3.2.1 Major and Minor Elements……………………………. 107
4.3.2.2 Trace Elements…………………………….…………… 112
4.4 Partitioning of Major, Minor and Trace Elements………………….. 116
4.5 Combustion Efficiencies……………………………………………. 123
4.6 Temperature Profiles……………………………………………….. 126
4.7 Concentration Profiles……………………………………………… 131
4.7.1 O2, CO2 and CO Concentration Profiles……………….……... 131
4.7.2 SO2 Concentration Profiles……………………………….…... 136
4.7.3 NO and N2O Concentration Profiles…………………….……. 136
4.8 Emissions…………………………………………………………… 141
4.9 Agglomeration and Deposit Formation…………………………….. 147
5 CONCLUSIONS…………………………………………………………. 164
5.1 General……………………………………………………………… 164
5.2 Suggestions for Future Work……………………………………….. 166
REFERENCES…………………………………………………………………. 167
APPENDICES
A WORLWIDE CO-FIRING POWER PLANTS………………………….. 183
B TGA RESULTS………………………………………………………….. 196
C POINT VALUES OF MEASUREMENTS……………………………… 199
xii
D CHEMICAL ANALYSES OF ASH STREAMS………………………... 207
E TABULATED SIZE DISTRIBUTIONS………………………………… 211
F SIZE DISTRIBUTION GRAPHS……………………………………….. 237
G CALIBRATION CURVES OF SCREW FEEDERS……………………. 261
H DEPOSIT COMPOSITIONS…………………………………………….. 264
I SCREW FEEDER DESIGN CALCULATIONS………………………... 267
CURRICULUM VITAE………………………………………………………... 273
xiii
LIST OF TABLES
Table 1.1 World electricity generation in 2004 [5]…………………………. 2
Table 2.1 Proximate and ultimate analyses of some selected biomass and coal.………………………………………………………………. 7
Table 2.2 Influence of biomass ash on boiler performance [39]…………….. 10
Table 2.3 Operating conditions and system properties of rice husk firing FBCs……………………………………………………………… 18
Table 2.4 Operating conditions and system properties of straw firing FBCs.. 27
Table 2.5 Operating conditions and system properties of olive residue firing FBCs………………………………………………………... 34
Table 2.6 Operating conditions and system properties of forestry residue firing FBCs………………………………………………………... 39
Table 2.7 Operating conditions and system properties of straw and coal co-firing FBCs…………………………………………………….. 47
Table 2.8 Operating conditions and system properties of olive residue and coal co-firing FBCs……………………………………………….. 52
Table 2.9 Operating conditions and system properties of forestry residue and coal co-firing FBCs…………….…………………………….. 57
Table 3.1 Technical specifications of the baghouse filter…………………… 69
Table 3.2 Relative positions of the gas sampling probes...………………….. 71
Table 3.3 Gas analyzers……………………………………………………... 75
Table 3.4 Backup gas analyzers……..………………………………………. 75
Table 3.5 Relative positions of the thermocouples………………………….. 81
xiv
Table 3.6 Technical specifications of new air compressor………………….. 83
Table 3.7 Fuel analyses……………………………………………………… 90
Table 3.8 Fuel size distributions…………………………………………….. 91
Table 3.9 Fuel ash compositions…………………………………………….. 92
Table 3.10 Chlorine contents of the fuels, %..................................................... 93
Table 3.11 Trace element concentrations in lignites in Runs 2, 5, 8 and 10…. 94
Table 3.12 Trace element concentrations in biomass...………………………. 95
Table 3.13 Characteristics of Beypazarı limestone…………………………… 96
Table 3.14 Trace element concentrations in Beypazarı limestone……………. 97
Table 3.15 Durations of the Runs………………..…………………………… 98
Table 3.16 Operating conditions of Runs 1-10……………………………….. 99
Table 4.1 Ash balance, closure and split…………………………………….. 108
Table 4.2 Trace element concentrations in bottom ash of Runs 2, 5 and 8…. 113
Table 4.3 Trace element concentrations in cyclone ash of Runs 2, 5 and 8.... 114
Table 4.4 Trace element concentrations in filter ash of Runs 2, 5 and 8…..... 115
Table 4.5 Relative enrichment factors of bottom, cyclone and filter ashes…. 117
Table 4.6 Combustion efficiencies…………………………………………... 124
Table 4.7 Flue gas emission data……………………………………………. 142
Table 4.8 Fuel-N to NO conversion…………………………………………. 145
Table 4.9 Emission limits……………………………………………………. 146
Table 4.10 Bed agglomeration index of the fuel blends…………………..….. 148
Table 4.11 Melting temperatures of ternary systems…………………………. 149
Table 4.12 Alkali index of fuel blends………………………………………... 150
Table A.1 Worldwide samples of co-firing experienced pulverized fuel
power plant [13]…..………………………………………………. 184
Table A.2 Worldwide samples of co-firing experienced bubbling fluidized
xv
bed power plants [13].....………..………………………………… 192
Table A.3 Worldwide samples of co-firing experienced circulating fluidized
bed power plants [13].....………..………………………………… 193
Table C.1 Sampling probe readings of O2 concentrations of Runs 1-3, dry mole %....................................................................................... 199
Table C.2 Sampling probe readings of O2 concentrations of Runs 4-6, dry mole %....................................................................................... 199
Table C.3 Sampling probe readings of CO2 concentrations of Runs 1-6, dry mole %....................................................................................... 200
Table C.4 Sampling probe readings of CO concentrations of Runs 1-6, dry mole %....................................................................................... 201
Table C.5 Sampling probe readings of SO2 concentrations of Runs 1-6, dry mole %....................................................................................... 202
Table C.6 Sampling probe readings of NO concentrations of Runs 1-6, dry mole %....................................................................................... 203
Table C.7 Sampling probe readings of N2O concentrations of Runs 1-6, dry mole %....................................................................................... 204
Table C.8 Thermocouple readings of Runs 1-10, °C………………………… 205
Table C.9 Normalized temperatures of Runs 1-10 °C/°C…..………………... 206
Table D.1 Chemical analyses of bottom ashes of Runs 1-4.............................. 207
Table D.2 Chemical analyses of bottom ashes of Runs 5-7.............................. 208
Table D.3 Chemical analyses of bottom ashes of Runs 8-10............................ 208
Table D.4 Chemical analyses of cyclone ashes of Runs 1-10.......................... 209
Table D.5 Chemical analyses of baghouse filter ashes of Runs 1-10............... 210
Table E.1 Particle size distribution of bottom ash of Run 1…………………. 211
Table E.2 Particle size distribution of bottom ash of Run 2…………………. 212
xvi
Table E.3 Particle size distribution of bottom ash of Run 3…………………. 212
Table E.4 Particle size distribution of bottom ash of Run 4…………………. 213
Table E.5 Particle size distribution of bottom ash of Run 5…………………. 213
Table E.6 Particle size distribution of bottom ash of Run 6…………………. 214
Table E.7 Particle size distribution of bottom ash of Run 7…………………. 214
Table E.8 Particle size distribution of bottom ash of Run 8…………………. 215
Table E.9 Particle size distribution of bottom ash of Run 9…………………. 215
Table E.10 Particle size distribution of bottom ash of Run 10………………. 216
Table E.11 Particle size distribution of cyclone ash of Run 1...………………. 217
Table E.12 Particle size distribution of cyclone ash of Run 2...………………. 218
Table E.13 Particle size distribution of cyclone ash of Run 3...………………. 219
Table E.14 Particle size distribution of cyclone ash of Run 4...………………. 220
Table E.15 Particle size distribution of cyclone ash of Run 5...………………. 221
Table E.16 Particle size distribution of cyclone ash of Run 6...………………. 222
Table E.17 Particle size distribution of cyclone ash of Run 7...………………. 223
Table E.18 Particle size distribution of cyclone ash of Run 8...………………. 224
Table E.19 Particle size distribution of cyclone ash of Run 9...………………. 225
Table E.20 Particle size distribution of cyclone ash of Run 10……………….. 226
Table E.21 Particle size distribution of baghouse filter ash of Run 1...………. 227
Table E.22 Particle size distribution of baghouse filter ash of Run 2...………. 228
Table E.23 Particle size distribution of baghouse filter ash of Run 3...………. 229
Table E.24 Particle size distribution of baghouse filter ash of Run 4...………. 230
Table E.25 Particle size distribution of baghouse filter ash of Run 5...………. 231
Table E.26 Particle size distribution of baghouse filter ash of Run 6...………. 232
Table E.27 Particle size distribution of baghouse filter ash of Run 7...………. 233
Table E.28 Particle size distribution of baghouse filter ash of Run 8...………. 234
xvii
Table E.29 Particle size distribution of baghouse filter ash of Run 9...………. 235
Table E.30 Particle size distribution of baghouse filter ash of Run 10.………. 236
Table H.1 Deposit compositions……………………………………………... 264
Table I.1 Comparison of measured and calculated hazelnut shell flow rate... 268
Table I.2 Comparison of measured and calculated olive residue flow rate..... 269
xviii
LIST OF FIGURES
Figure 1.1 Share of total primary energy supply in world in year 2004 [5]…. 2
Figure 1.2 World electricity generation by fuel, (2004-2030) [7]…………… 3
Figure 2.1 Biomass ash compositions……………………………………….. 10
Figure 2.2 Agglomeration of bed material: Type 1- coating-induced, Type 2- melt-induced [40]……………………………………….. 12
Figure 2.3 Corrosion effects of alkalis in biomass combustion and co-combustion systems [48]……………………………………… 16
Figure 3.1 METU 0.3 MWt ABFBC Test Rig………………………………. 64
Figure 3.2 Flow sheet of METU 0.3 MWt ABFBC Test Rig………………... 65
Figure 3.3 Gas sampling probe………………………………………………. 72
Figure 3.4 GASS-II pre-conditioning system………………………..………. 73
Figure 3.5 Schematic of Perma Pure Nafion Membrane Dryer..……………. 74
Figure 3.6 Gas conditioning and analysis system……………………………. 76
Figure 3.7 Deposit sampling probe…………………………………………... 78
Figure 3.8 P & I diagram of METU 0.3 MWt ABFBC Test Rig……………. 79
Figure 3.9 Photographs of biomasses.……………………………………….. 89
Figure 4.1 Size distribution of all solid streams in Runs 1 and 2……………. 102
Figure 4.2 Size distribution of all solid streams in Runs 2, 3, 4 and 5………. 104
Figure 4.3 Size distribution of all solid streams in Runs 2, 6, 7 and 8………. 105
Figure 4.4 Size distribution of all solid streams in Runs 2, 9 and 10..………. 106
Figure 4.5 Bottom ash analyses of Runs 1-10……………………………….. 109
Figure 4.6 Cyclone ash analyses of Runs 1-10……...……………………….. 110
xix
Figure 4.7 Baghouse filter ash analyses of Runs 1-10…...………………….. 111
Figure 4.8 Recovery of major and minor elements in Runs 2, 5 and 8……… 119
Figure 4.9 Recovery of trace elements in Runs 2, 5 and 8…………...……… 120
Figure 4.10 Major and minor elements partitioning of Runs 2, 5 and 8……… 121
Figure 4.11 Trace elements partitioning of Runs 2, 5 and 8……………..…… 122
Figure 4.12 Temperature profiles of Runs 1 and 2……………………………. 127
Figure 4.13 Normalized temperature profiles of Runs 1 and 2.………………. 127
Figure 4.14 Temperature profiles of Runs 2, 3, 4 and 5………………………. 128
Figure 4.15 Normalized temperature profiles of Runs 2, 3, 4 and 5………….. 128
Figure 4.16 Temperature profiles of Runs 2, 6, 7 and 8………………………. 129
Figure 4.17 Normalized temperature profiles of Runs 2, 6, 7 and 8………….. 129
Figure 4.18 Temperature profiles of Runs 2, 9 and 10..………………………. 130
Figure 4.19 Normalized temperature profiles of Runs 2, 9 and 10..………….. 131
Figure 4.20 O2 concentration profiles of Runs 1-6……………………………. 132
Figure 4.21 CO2 concentration profiles of Runs 1-6……….…………………. 133
Figure 4.22 CO concentration profiles of Runs 1-6………..…………………. 135
Figure 4.23 SO2 concentration profiles of Runs 1-6.…………………………. 137
Figure 4.24 NO concentration profiles of Runs 1-6..…………………………. 139
Figure 4.25 N2O concentration profiles of Runs 1-6………………………….. 140
Figure 4.26 Appearance of deposit rings after olive residue/lignite co-firing runs……………………………………………………... 152
Figure 4.27 Appearance of deposit rings after hazelnut shell/lignite co-firing runs……………………………………………………... 153
Figure 4.28 Appearance of deposit rings after cotton residue/lignite co-firing runs……………………………………………………... 154
Figure 4.29 Rate of deposit build-up (RBU, g/m2h)…………………………... 155
xx
Figure 4.30 SEM micrograph of deposit of olive residue/lignite co-firing runs (a)…………………………………………………. 156
Figure 4.31 SEM micrograph of deposit of olive residue/lignite co-firing runs (b)…………………………………………………. 156
Figure 4.32 SEM micrograph of deposit of olive residue/lignite co-firing runs (c)…………………………………………………. 157
Figure 4.33 SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (a)…………………………………………………. 157
Figure 4.34 SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (b)…………………………………………………. 158
Figure 4.35 SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (c)…………………………………………………. 158
Figure 4.36 SEM micrograph of deposit of cotton residue/lignite co-firing runs (a)…………………………………………………. 159
Figure 4.37 SEM micrograph of deposit of cotton residue/lignite co-firing runs (b)…………………………………………………. 159
Figure 4.38 SEM micrograph of deposit of cotton residue/lignite co-firing runs (c)…………………………………………………. 160
Figure 4.39 Deposit compositions…………………………………………….. 161
Figure 4.40 X-ray diffraction patterns of the deposits………………………… 162
Figure B.1 TGA graph of lignite……………………………………………... 196
Figure B.2 TGA graph of olive residue………………………………………. 197
Figure B.3 TGA graph of hazelnut shell..……………………………………. 197
Figure B.4 TGA graph of cotton residue…………………………..…………. 198
Figure F.1 Particle size distribution of lignite fed in Run 1…………………. 237
Figure F.2 Particle size distribution of lignite fed in Run 2…………………. 238
xxi
Figure F.3 Particle size distribution of lignite fed in Run 3…………………. 238
Figure F.4 Particle size distribution of lignite fed in Run 4…………………. 239
Figure F.5 Particle size distribution of lignite fed in Run 5…………………. 239
Figure F.6 Particle size distribution of lignite fed in Run 6…………………. 240
Figure F.7 Particle size distribution of lignite fed in Run 7…………………. 240
Figure F.8 Particle size distribution of lignite fed in Run 8…………………. 241
Figure F.9 Particle size distribution of lignite fed in Run 9…………………. 241
Figure F.10 Particle size distribution of lignite fed in Run 10..………………. 242
Figure F.11 Particle size distribution of olive residue fed in Runs 3, 4 and 5... 242
Figure F.12 Particle size distribution of hazelnut shell fed in Runs 3, 4 and 5.. 243
Figure F.13 Particle size distribution of cotton residue fed in Runs 3, 4 and 5.. 243
Figure F.14 Particle size distribution of limestone fed in Runs 2, 3, 4 and 5…. 244
Figure F.15 Particle size distribution of limestone fed in Runs 6, 7 and 8……. 244
Figure F.16 Particle size distribution of limestone fed in Runs 9 and 10..……. 245
Figure F.17 Particle size distribution of bottom ash of Run 1………………… 246
Figure F.18 Particle size distribution of bottom ash of Run 2………………… 246
Figure F.19 Particle size distribution of bottom ash of Run 3………………… 247
Figure F.20 Particle size distribution of bottom ash of Run 4………………… 247
Figure F.21 Particle size distribution of bottom ash of Run 5………………… 248
Figure F.22 Particle size distribution of bottom ash of Run 6………………… 248
Figure F.23 Particle size distribution of bottom ash of Run 7………………… 249
Figure F.24 Particle size distribution of bottom ash of Run 8………………… 249
Figure F.25 Particle size distribution of bottom ash of Run 9………………… 250
Figure F.26 Particle size distribution of bottom ash of Run 10..……………… 250
Figure F.27 Particle size distribution of cyclone ash of Run 1...……………… 251
Figure F.28 Particle size distribution of cyclone ash of Run 2...……………… 251
xxii
Figure F.29 Particle size distribution of cyclone ash of Run 3...……………… 252
Figure F.30 Particle size distribution of cyclone ash of Run 4...……………… 252
Figure F.31 Particle size distribution of cyclone ash of Run 5...……………… 253
Figure F.32 Particle size distribution of cyclone ash of Run 6...……………… 253
Figure F.33 Particle size distribution of cyclone ash of Run 7...……………… 254
Figure F.34 Particle size distribution of cyclone ash of Run 8...……………… 254
Figure F.35 Particle size distribution of cyclone ash of Run 9...……………… 255
Figure F.36 Particle size distribution of cyclone ash of Run 10.……………… 255
Figure F.37 Particle size distribution of baghouse filter ash of Run 1………... 256
Figure F.38 Particle size distribution of baghouse filter ash of Run 2………... 256
Figure F.39 Particle size distribution of baghouse filter ash of Run 3………... 257
Figure F.40 Particle size distribution of baghouse filter ash of Run 4………... 257
Figure F.41 Particle size distribution of baghouse filter ash of Run 5………... 258
Figure F.42 Particle size distribution of baghouse filter ash of Run 6………... 258
Figure F.43 Particle size distribution of baghouse filter ash of Run 7………... 259
Figure F.44 Particle size distribution of baghouse filter ash of Run 8………... 259
Figure F.45 Particle size distribution of baghouse filter ash of Run 9………... 260
Figure F.46 Particle size distribution of baghouse filter ash of Run 10..……... 260
Figure G.1 Calibration curve for lignite flow rate……………………………. 261
Figure G.2 Calibration curve for olive residue flow rate…………………….. 262
Figure G.3 Calibration curve for hazelnut shell flow rate.…………………… 262
Figure G.4 Calibration curve for cotton residue flow rate…………………… 263
Figure G.5 Calibration curve for limestone flow rate………………………... 263
Figure H.1 EDX analysis graph of cotton residue and lignite co-firing
deposit……………………………………………………………. 265
Figure H.2 EDX analysis graph of cotton residue and lignite co-firing
xxiii
deposit……………………………………………………………. 265
Figure H.3 EDX analysis graph of cotton residue and lignite co-firing
deposit……………………………………………………………. 266
Figure I.1 Schematic description of ash feeder spiral……………………….. 267
Figure I.2 Comparison of measured and calculated hazelnut shell flow rate.. 269
Figure I.3 Comparison of measured and calculated olive residue flow rate… 270
Figure I.4 Schematic description of new bed feeder spiral………………….. 271
Figure I.5 Schematic drawing of the new bed feeder……………………….. 272
xxiv
CHAPTER 1
INTRODUCTION
1.1 General
Today, the demand for electric power continues to rise due to population growth,
technological and economical development. Coal is predicted to be the dominant
fossil fuel for energy production for decades with progressively improving clean coal
technologies and 909 billion tones of proved reserves [1]. Global environmental
impacts of fossil fuels used for power generation is a worldwide concerning topic. To
mitigate effects of fossil fuel combustion, share of renewables in energy production
is increasing. Among the renewable sources, biomass is the lowest risk and capital
required one to be used in energy generation. It is a renewable energy option due to
the fact that it can be considered as CO2-neutral fuel and contributes to the reduction
of SO2 and NOx emissions due to its low sulfur and nitrogen contents. Furthermore,
when burned instead of landfilled, it prevents CH4 release to atmosphere, which is a
more powerful greenhouse gas compared to CO2 [2].
In addition to the above mentioned advantages, biomass is the fourth largest energy
source after oil, coal and gas and that about 10 to 14 % of world’s energy is produced
from biomass. In Figure 1.1, the share of world’s total primary energy supply in year
2004 is shown. Africa obtains about 2/3 of its energy from biomass, Asia about 1/3,
and Latin America about 1/4 [3]. The role of biomass in European countries varies
from nearly 15 % in Finland and Sweden to less than 1 % in the UK. In longer term,
it is anticipated that biomass could contribute to 20 % of the current European Union
primary energy demand [4].
1
Figure 1.1: Share of total primary energy supply in world in year 2004 [5].
Biomass is a very versatile source of energy that can be readily stored and
transformed into electricity and heat [6]. Table 1.1 gives worldwide amounts of
electricity produced from different energy sources in year 2004. As can be seen from
the table, coal is the dominant source of electricity generation and renewable sources
have lower share. However, the share of biomass is expected to rise in the following
years due to increasingly strict legislations on emissions from fossil fuel sources. The
projections on electricity generation till year 2030 given in Figure 1.2 also show the
increase in electricity production from coal and renewable sources.
Table 1.1: World electricity generation in 2004 [5].
Fuel TWh % Coal 6944 39.7 Gas 3419 19.6 Oil 1170 6.7 Hydro 2889 16.5 Nuclear 2738 15.7 Biomass 149.8 0.9 Waste 77.5 0.4 Others 85.1 0.5 Total 17472 100.0
2
Figure 1.2: World electricity generation by fuel, (2004-2030) [7].
However, biomass combustion brings with it some operational problems when
burned alone. The most common problems encountered in industry and utility boilers
are severe fouling, slagging and corrosion which are mainly originated from high
alkali chlorides content of biomass ash. Ash deposits formed on heat transfer
surfaces deteriorate heat transfer and lead to loss in thermal efficiency and corrosion.
These problems in biomass firing combustion systems can be alleviated by either
leaching biomass with chemicals of various strengths to reduce the harmful
components of ash or co-firing biomass with coal. The former alternative was not
found economically feasible. However, the latter represents a near-term, low risk,
low cost, sustainable renewable energy option [8-11]. This has already been
demonstrated in over 100 coal-fired power plants. The capacity of typical power
stations where co-firing is applied varies in the range from 50 MWe to 700 MWe (a
few units are between 5 and 50 MWe). Information on worldwide coal-fired
3
pulverized fuel, bubbling and circulating fluidized bed power plants co-firing
biomass or waste is given in Appendix A. As can be seen from the Appendix, the
majority of co-firing applications are carried out in pulverized fuel-fired power
plants. This is considered to be due to the fact that most of the existing plants are
based on this conventional technology and are forced to comply with the CO2
reduction legislation.
For co-firing high volatile, moisture and low ash content biomass with low volatile,
moisture and high ash content coal, a fuel flexible clean combustion technology is
necessary. Within the available technologies, fluidized bed combustion (FBC)
technology is usually indicated to be the best choice, or sometimes the only choice.
This is confirmed by extensive experimental investigation carried out to date on
technical and economical feasibility and performance of different types of biomass in
FBC technology [12], and also by steadily growing number of industrial applications
(Appendix A [13]).
1.2 Aim and Scope of the Thesis
In Turkey, there exist 9.3 billion tons of indigenous lignite reserves characterized by
low calorific value and high ash and sulfur contents. Absence of studies on co-firing
of indigenous lignite/biomass blends in bubbling fluid bed combustors, on one hand,
and the recent trend in utilization of biomass with local reserves in industry and
utility boilers, on the other, necessitate investigation of co-firing lignite with
biomass.
The most available biomass sources in Turkey are olive residues, hazelnut shells and
cotton residues. Turkey is one of the main olive producers with 662 000 ha of olive
groves and 1 200 000 tons of production in year 2005 [14]. Hazelnut production in
Turkey accounts for 70 % of the worlds’ total production with 584 000 ha plantation
and 530 000 tons of production in year 2005 [15]. Moreover, Turkey is one of the
leader producers of cotton in the world with 546 880 ha plantation, 1 291 180 tons of
cotton lint and 863 700 tons of cotton seed productions in year 2005 [14].
4
Therefore, significant amounts of olive residues, hazelnut shells and cotton residues
are produced in Turkey to be used in co-firing applications.
Olive residue is a specific type of biomass from olive oil production process. It is the
remaining part of olives after pressing and extraction of olive oil. Annual production
of olive residue is about 682 993 tons [16]. Hazelnut shells, produced from crushing
of hazelnuts, have annual production of 453 184 tons [16]. Cotton residues are
produced from cotton oil production process. Cotton is grown mainly for the fibers
or lint, but cotton seeds having high amount of oil content are also very important.
Lint and fiber of the cotton are removed firstly by hand and then mechanically by
ginning process. The remaining part, cotton seed, is then extracted for cotton oil.
Annual production of cotton residues in Turkey is about 593 972 tons [16].
Availability of significant amounts of these residues and indigenous lignite reserves
together with gradual introduction of increasingly restrictive legislations on
emissions makes co-firing option attractive.
Therefore, the objective of this thesis study has been to investigate co-firing of
typical indigenous lignite with olive residue, hazelnut shell and cotton residue in the
METU 0.3 MWt Atmospheric Bubbling Fluidized Bed Combustion (ABFBC) Test
Rig in terms of combustion and emission performance and ash deposition tendencies.
5
CHAPTER 2
BACKGROUND AND LITERATURE REVIEW
2.1 General
Prior to presenting a detailed review of previous studies on biomass firing and
biomass and coal co-firing in fluidized bed combustion systems (FBC), various
issues associated with biomass characteristics and combustion are briefly
overviewed.
2.2 Biomass Firing in FBC Systems
2.2.1 Biomass Characteristics
The term biomass describes carbonaceous materials derived from plants. Biomass
fuels can be classified as agricultural residues, woody residues, dedicated energy
crops and industrial and municipal waste of plant origin [17]. Unlike conventional
fuels, some physical and chemical properties of biomass such as high volatile matter
and moisture content, low bulk density and low ash melting temperature complicate
its processing and combustion [18].
Table 2.1 displays proximate and ultimate analyses of some selected biomass and
coal. As can be seen from the table, biomass fuels are mainly characterized by their
high volatile matter and low ash contents.
6
LHV
(M
J/kg
)
19.5
0
17.2
3
11.7
0
12.3
4
17.7
0
17.9
9
15.9
5
17.3
0
11.7
0
7.99
15.2
0
25.5
9
25.4
0
12.2
6
Ash
8.00
6.22
22.6
0
14.6
0
1.30
5.22
0.88
4.80
0.30
0.32
0.92
18.1
0
14.8
0
42.2
0
Cl
0.01
1.05
0.08
0.00
0.05
0.00
0.00
0.00
0.00
0.00
2
0.00
0.00
0.00
0.00
O
32.8
1
39.7
0
35.3
6
41.1
2
43.8
5
38.6
7
45.7
1
42.0
0
46.0
0
41.3
8
44.1
8
8.25
12.5
0
12.4
0
S 0.30
0.22
0.06
0.04
0.05
0.00
0.67
0.00
0.10
0.10
0.10
0.71
0.60
2.70
N
1.38
0.50
0.30
0.36
0.20
1.23
0.22
0.30
0.20
0.10
0.20
1.34
1.20
1.40
H
5.80
5.36
4.70
5.46
5.97
5.59
5.76
5.50
6.10
6.00
6.10
3.73
3.70
3.20
Ulti
mat
e A
naly
sis (
dry,
wt %
)
C
51.7
0
46.9
5
36.9
0
38.4
2
48.5
8
49.2
9
46.7
6
47.4
0
47.3
0
52.1
0
48.5
0
67.8
7
67.2
0
38.1
0
Ash
7.28
5.74
20.0
2
12.9
9
1.24
4.90
0.77
4.22
0.20
0.15
0.80
17.1
1
14.5
0
36.4
0
FC
18.4
9
14.4
6
25.1
7
18.0
1
28.2
2
10.3
24.0
8
15.0
5
13.3
0
6.76
26.6
0
52.2
4
60.2
0
17.2
0
VM
65.3
0
72.0
5
43.4
1
58.0
0
65.5
4
78.7
0
62.7
0
68.7
3
51.6
0
39.7
9
59.6
0
25.1
6
23.0
0
32.7
0
Prox
imat
e A
naly
sis (
as re
ceiv
ed, w
t %)
Moi
stur
e
8.93
7.75
11.4
0
11.0
0
5.00
6.10
12.4
5
12.0
0
34.9
0
53.3
0
13.0
0
5.49
2.30
13.7
0
Tab
le 2
.1: P
roxi
mat
e an
d ul
timat
e an
alys
es o
f sel
ecte
d bi
omas
s and
coa
l.
Fuel
Bio
mas
s
Oliv
e re
sidu
e [1
9]
Whe
at st
raw
[19]
Ric
e st
raw
[20]
Ric
e hu
sk [2
1]
Bag
asse
[20]
Cot
ton
seed
cak
e [2
2]
Haz
elnu
t she
ll [2
3]
Bar
k [2
4]
Woo
d ch
ips [
25]
Saw
dust
[26]
Pine
seed
shel
l [27
]
Coa
l
Bitu
min
ous c
oal [
28]
Sout
h A
fric
an c
oal [
27]
Lign
ite [2
9]
7
It is possible to burn any type of biomass, but in practice combustion in fluidized bed
combustors is feasible only for biomass with moisture content less than 60 % unless
the biomass is pre-dried [6]. High moisture content can lead to poor ignition and
reduce combustion temperature and hence quality of combustion [18].
Volatile matter content of biomass is greater than that of coal. High volatile content
leads to easy ignition and burning which may cause difficulties to control the
operation.
Biomass bulk densities are lower compared to that of coals which may lead to
transportation, storage, feeding and firing problems. Coal densities typically range
from 900 kg\m3 for low rank coals to 2330 kg\m3 for high density pyrolytic graphite
[30]. Biomass densities range from 100 kg\m3 for straw to 590 kg\m3 for olive
residue.
It is generally unfeasible to reduce biomass size as most of the biomass has fibrous
structure [8]. Blockage and discontinuous dosing problems may occur during feeding
biomass with screw feeders in fluidized bed combustors which can be prevented by
replacing them with pneumatic feeding systems [18].
Most of the biomasses are characterized by low ash contents and hence they require
addition of bed material in order to maintain the bed at a constant level during the
operation [9].
2.2.2 Biomass Combustion
Biomass has similar combustion mechanisms to coal such as drying, devolatilization
and combustion. During devolatilization they undergo thermal decomposition with
subsequent release of volatiles and char and tar formation [18]. Combustion is
composed of two phases which are volatile matter and char combustion phase. Coal
has short volatile and long char combustion phases whereas biomass has long
volatile and short char combustion phases. In biomass combustion devolatilization
8
starts at low temperatures and almost all combustion is completed in this phase.
Devolatilization of high volatile matter content can also result in a highly porous
char, thus accelerating char combustion as well [2].
In fluidized bed combustion of biomass, location of volatiles release and combustion
mainly depend on method of feeding and distribution of combustion air [18]. Over-
bed feeding generally leads to higher freeboard temperatures compared to under-bed
feeding as a consequence of less uniform spreading of the fuel along the combustor
[31]. Under-bed feeding provides better combustion of biomass and leads to lower
CO emissions [32].
Generally, biomass fuels have low melting temperature ash mainly composed of high
alkali oxides and salts which lead to many operational problems during combustion.
In the following part, problems associated with ash content of biomass are
summarized.
2.2.3 Operational Problems Associated with Biomass Ash
Operational problems in biomass combustion are mainly originated from inorganic
components of the fuel. Figure 2.1 shows ash compositions of some selected biomass
and coal. As can be seen from the figure, biomass ashes have higher portions of
potassium and sodium with respect to those of coals.
High sodium and potassium content of biomass ash cause bed agglomeration,
slagging on furnace walls and fouling of heat transfer surfaces which lead to reduced
reliability of electricity production from biomass. The successful design and
operation of a fluidized bed combustor depends on the ability to control and mitigate
these ash related problems [9]. Table 2.2 gives the possible operational problems
related with biomass ash. Biomass ash has relatively low ash fusion temperature
compared to that of coal.
9
Figure 2.1: Biomass ash compositions.
Table 2.2: Influence of biomass ash on boiler performance [39].
Fuel constituents Operational Problems
Alkali content (Sodium, Potassium)
Bed agglomeration Slagging Fouling of heat transfer surfaces Hot corrosion
Chlorine
Hot corrosion Fouling HCl emission Dioxin formation
Heavy metals Emissions Furnace corrosion Ash handling
10
The rate of slag formation on furnace walls tends to increase owing to the reduction
of ash fusion temperature by introduction of biomass ash into the system. The
problems associated with ash characteristics of biomass in fluidized bed combustion
of biomass are summarized in the following sections.
2.2.3.1 Bed Agglomeration
Agglomeration occurs when a part of fuel ash melts and causes adhesion of bed
particles [9]. The factors affecting agglomeration are bed temperature, ash
composition and alkali content of the fuel. Temperature has the most pronounced
effect on agglomeration tendency. Alkaline compounds resulting from high
potassium and sodium in biomass ash have very low melting points. In addition to
adhesion effect of sintered ash, alkali oxides or salts can react with silica compounds
of the bed material according to [18];
2 2 3 2 2 2
2 2 3 2 2 2
4 SiO + K CO K O . 4 SiO + CO (2.1)2 SiO + Na CO Na O . 2 SiO + CO (2.2)
⎯⎯→
⎯⎯→
Potassium and sodium form eutectic mixtures with melting points of 874 and 764 ºC,
respectively. At higher concentrations, melting point of these eutectic mixtures is as
low as 650-700 °C which is well below the normal operation temperature of a
fluidized bed combustor [9].
If sufficient amount of Fe2O3 is present in the ash of the fuels burned, formation rate
of agglomerates may be reduced by [18],
2 3 2 2 2 4
2 3 2 2 2 4
2 3 2 3 2 2 4 2
Fe O + K O K Fe O (2.3)Fe O + Na O Na Fe O (2.4)
Fe O + K CO K Fe O + CO
⎯⎯→
⎯⎯→
⎯⎯→
2 3 2 3 2 2 4 2
(2.5)Fe O + Na CO Na Fe O + CO (2.6)⎯⎯→
11
Figure 2.2 displays extreme agglomeration types. Type 1 is more commonly
observed in commercially operated fluidized bed combustors burning woody type
fuels and results from “coating-induced” agglomeration. Here, a uniform coating is
formed on the surface of the bed material grains. At certain critical conditions like
coating thickness or temperature, neck formation may occur between coatings of
individual grains which initiate agglomeration.
Agglomeration due to melt formation
Sintering of coatings
Bed grain
Bed grain
Gas phase ash
Molten ash particles
Coating
Bed grain
Figure 2.2: Agglomeration of bed material: Type 1- coating-induced,
Type 2- melt-induced [40].
The other type, results from “melt-induced” agglomeration. In this case, bed material
grains are glued together by a melt phase, which roughly matches the chemical
composition of the ash and is produced at normal operating temperature. If, however,
after first neck formation partial de-fluidization of the bed leads to local peak
temperatures, melt formation may occur and a combination of the two extremes can
be recognized in one sample. [40].
12
Agglomeration tendency of the fuel can be estimated by the ratio of iron oxides to
sum of potassium and sodium oxides in the fuel ash. Agglomeration occurs when the
ratio is less than 0.15 [9].
2 3
2 2 (2.7)
%(Fe O )Bed Agglomeration Index (BAI) = %(K O+Na O)
2.2.3.2 Ash Deposition and Corrosion
Slagging, fouling and corrosion of heat transfer surfaces are mostly influenced by
high alkaline, chlorine and low sulfur content of biomass. Ash constituents which
result from combustion processes react with flue gas or with each other to form
variety of compounds. Partially molten ash particles in the flue gas impact and
deposit on the heat transfer surfaces and furnace walls which are termed as fouling
and slagging, respectively.
Fouling tendency of the fuel can be estimated based on different indicators such as
the ratio of alkali metal oxides to silica oxide. The index higher than 1 indicates
severe fouling possibilities [39].
2
2 2 (2.8)%(K O+Na O)Alkali Index (AI) = %(SiO )
As indicated previously, inorganic compounds (mostly sodium, potassium and
chlorine) in easily vaporizing form lower the melting temperature of ash and
consequently expected to provoke serious ash deposition problems. Hot corrosion of
the boiler tubes becomes a problem with fuels having high chlorine. When the fuel
chlorine content exceeds 0.1 %, corrosion may occur on heat transfer surfaces [39].
Chlorine concentration often dictates the amount of alkali vaporized during
combustion more strongly than the alkali concentration in the fuel. In most cases
chlorine appears to play a shuttle role, facilitating the transport of alkali from the fuel
to surfaces where alkali often forms sulfates [41]. Removal of chlorine from fuels by
13
leaching is usually not economical. Corrosive deposits tend to damage the heat
transfer surfaces at metal temperatures >470 ºC and typical vapor temperatures of
modern steam boilers exceed 500 ºC [42].
Superheater corrosion and ash deposition can be reduced in biomass firing FBC
systems by;
• Leaching biomass to reduce its ash content.
• Using alternative bed material other than SiO2 (e.g. Fe2O3).
• Using additives (alumina silicates) to form alkali alumina silicates and
prevent the release of gaseous KCl and NaCl or react with them in the
gas phase and form less corrosive compounds (e.g. kaolin, bauxite).
• Using sulfur containing additives for sulfation of gaseous alkalis, KCl
and NaCl to form less corrosive sulfates, K2SO4 and Na2SO4.
• Co-firing biomass with coal.
Control of the rate of deposit formation in biomass combustion is associated with the
reactions between compounds containing chlorine, sulfur, aluminum and alkali. High
risk alkali chlorine compounds, NaCl and KCl, can be trapped by reactions with SO2
[43] and aluminum silicates [42, 44] to liberate gaseous HCl as shown in below
reactions;
2 2 2 2 4
2 2 2 2 4
2KCl(s)+SO (g)+1/2 O (g)+H O(g) K SO (s)+2HCl(g) (2.9) 2NaCl(s)+SO (g)+1/2 O (g)+H O(g) Na SO (s)+2HCl(g) (2.10)
→→
2 3 2 2 2 2 3 2
2 3 2 2 2 2 3 2
Al O 2SiO (s)+2KCl(s)+H O(g) K O Al O 2SiO (s)+2HCl(g) (2.11)Al O 2SiO (s)+2NaCl(s)+H O(g) Na O Al O 2SiO (s)+2HCl(g) (2.12)
⋅ → ⋅ ⋅⋅ → ⋅ ⋅
Corrosion of heat transfer surfaces can also be caused by acid components in flue
gases. Generally hydrogen sulfide (H2S) and hydrogen chloride (HCl) are the most
harmful gaseous species in flue gases. However, in the absence of reducing
14
conditions, high concentrations or high temperatures required to initiate corrosion
process, they are stable and leave the system with flue gas [45].
The S/Cl ratio in the feedstock has often shown to affect Cl deposition and corrosion.
Theoretically based in the sulfation reaction shown above, S/Cl ratio of 0.5 should be
enough to sulfate all alkali chlorides to alkali sulfates. However, in practice higher
ratios are required due to reactions with calcium compounds which consume
available SO2 from the flue gas. It has been suggested that if the S/Cl ratio of fuel is
less than 2, there is a high risk of superheater corrosion. When the ratio is at least 4,
the blend could be regarded as non-corrosive [46].
Inside the deposits, close to the heat transfer surface where temperature is below 600
ºC KCl is in solid form. Sulfation reaction can occur inside the deposits at anhydrous
conditions and produce free Cl2, instead of HCl, according to reaction [42];
2 2 2 4 22 KCl (s) + SO (g) + O (g) K SO (s) + Cl (g) (2.13)→
Free Cl2 may also form in the reaction between KCl and Fe2O3 which exists on
superheater tube surface according to the following reaction [42];
2 3 2 2 2 4 22 KCl (s) + Fe O (s) + 1/2 O (g) K Fe O (s) + Cl (g) (2.14)→
Chloride is highly reactive with iron and chromium, forming metal chlorides and in
long term these reactions will destroy heat exchange tubes, however, alkali sulfates
formed as result of reactions given in (2.8)-(2.11) are much less harmless to heat
exchange tubes.
Due to differences in elemental composition of biomass and coal, chlorine and alkali
metal behavior during co-firing will be different from both fuels because of the
interactions between volatile elements (sodium, potassium and chlorine) and other
mineral elements. Major ash forming elements of coal (aluminum and silicon) have
significant influence on this behavior [45]. Coal brings within itself protective ash
15
elements to the system which lowers gas phase alkali chlorides. Full-scale co-firing
of coal with high alkali content straw is reported to show little or no chlorine in the
deposits [47]. Protective behavior of ash forming elements in coal is illustrated in
Figure 2.3. In Case 1, biomass is burning alone. The ash of biomass fuels has high
alkaline metal content. When this is associated with high chlorine content, which is
often the case, these elements react to form alkali chlorides. This, in turn, induces
corrosion rates after deposition of these substances on the heat transfer surfaces. In
Case 2, sulfur and aluminum silicates from coal ash are able to form alkali silicates
and alkali sulfates. Hence, chlorine is released as HCl in flue gases and alkali metals
are bound in compounds that have a high melting point and no corroding effect [48].
Biomass
Combustion Biomass and Coal
Co-combustion
Case 1 Case 2
Figure 2.3: Corrosion effects of alkalis in biomass combustion
and co-combustion systems [48].
Due to these operational problems in biomass firing, co-firing of biomass with coal
becomes an attractive alternative from environmental and operational points of view.
16
2.3 Literature Review
The use of biomass in energy generation has importance as global warming is
concerned since biomass firing has the potential to be CO2 neutral. This is
particularly the case with regard to agricultural residues and energy plants which are
periodically planted and harvested [18]. They remove CO2 from atmosphere during
their growth and release again during combustion. Biomass types having high energy
potential include agricultural and forestry residues are rice husk, straw, olive residue,
sugar cane bagasse, pine barks and wood chips. In the following parts, review of
biomass firing and biomass and coal co-firing studies on agricultural and forestry
type biomass residues in FBC systems are presented.
2.3.1 Biomass Firing Studies
Rice husk is one of the high moisture and volatile matter content agricultural
residues. Rice husk alone is difficult to fluidize, but fluidization can be improved by
the addition of bed material. Details and operating conditions of FBC systems
burning rice husks are summarized in Table 2.3.
Combustion of rice husks was investigated in the study of Preto et al. [49] in a pilot
scale bubbling fluidized bed. The operating parameters in the experiments were
temperature, fluidizing velocity, rice husk feed rate and excess air. No feeding
problems were encountered during under bed feeding of rice hulls with screw feeder.
Due to high volatile matter content of rice husks, combustion mainly took place in
freeboard region. Over 97 % combustion efficiency was achieved. At fluidizing
velocities greater than 1.75 m/s, temperature peak was at the top of the freeboard
which opened up the potential of combustible losses. At low fluidizing velocities,
temperature gradient occurred in the bed which indicated improper mixing in bed.
17
Table 2.3: Operating conditions and system properties of rice husk firing FBCs.
Reference Fuel Operating conditions System Bed material
Preto et al. [49]
Rice husks
Bed temperature: 650-900 °C u0: 0.4-2.2 m/s
Feeding rate: 22-70 kg/h Excess air: 30-95 %
BFBC 1 MWt 0.380 ×0. 406 m cross section & 4.8 m height
Sand
Guanyi et al. [11]
Rice husks Bed temperature: 750-850 °C
u0: 1-2 m/s Feeding rate: 20-40 kg/h
CFBC 1 MWt0.784 m2
cross section & 6 m height
Silica sand
550 μm
Armesto et al. [50]
Rice husks Bed temperature: 840-880 °C u0: 1-1.2 m/s
BFBC 0.3 MWt CIEMAT
Sand
Fang et al. [51] Rice husks Bed temperature: 750-850 °C
Excess air coefficient: 1.1-1.2
CFBC 1 MW
0.04 m2 cross section & 6 m height
Silica sand
550 μm
Permchart &
Kouprianov [52]
Rice husk Saw dust
Sugar cane bagasse
Excess air: 20, 40, 60, 80, 100 %
Silica sand 300-500 μm
Kurpianov et al. [53]
Rice husks Sugar cane
bagasse
Bed temperature: 700-800 °C Excess air: 40, 60, 80, 100 %
Feeding rate: 82.8 kg/h
BFBC Conical lower section (40º
cone angle), 0.9 m inner
diameter & 3 m total height
Silica sand
0.45 mm
Albina [32] Rice husks
Under-bed, over-bed feeding Primary to secondary air ratio:
40, 50, 60 % Excess air: 10, 20, 30 %
BFBC multiple-
spout/ spout
0.49 m2 cross section
Silica sand
300-1000 μm
Shimizu et al. [54]
Rice husks Bed temperature: 850 °C u0: 0.39 m/s
BFBC 0.053 m
inner diameter & 1.3 m height
Quartz sand
270 μm, Porous alumina 690 μm
Natarajan et al. [55]
Rice husk Bagasse Cane trash Olive flesh
6 % excess air under normal FBC conditions BFBC
Quartz sand or
lime
Liu et al. [56] Rice husks u0: 0.8 m/s
BFBC 0.255 m2 cross sectional area
Sand
Skrifvars et al. [57]
Rice husks Eucalyptus
bark
EFR flue gas temperature: 900 ºC
Deposit probe temperature: 500 ºC
BFBC 157 MWt & EFR
0.18 m inner diameter & 6 m height
Sand
Bakker et al. [58]
Rice straw Wood
Flue gas temperature: 900 ºC Excess air: 30-50 %
Secondary air: 13-20 % Deposit probe temperature:
500 ºC
Lab scale BFBC
42×10-4 m2 cross section & 0.915 m
height
Alumina-silicate grains
18
However, high ash softening temperature of rice husk ash (~1400 ºC) which have
significant portion of SiO2 ensured acceptable operation. CO emission varied from
200 to 5000 ppm. CO levels showed dependence on both temperature and fluidizing
velocity. The effect of the velocity is obvious as the higher the velocity, the lower is
the residence time. However, effect of temperature was opposite of the expected
behavior which was explained to be the outcome of masking effect of the velocity.
The pollutant emissions ranged from 50 to 150 ppm for SO2 and from 100 to 180
ppm for NOx. The rice husk ash was very fine and easily elutriated out of the bed.
The majority of the ash was collected as cyclone product.
Generally pneumatic feeding systems are preferred to avoid blocking in rice husk
burning fluidized bed combustors. In 1 MWt CFBC system of Guanyi et al. [11]
where ignition and combustion characteristics of rice husks were investigated, the
most serious problem during operation was bridging in the feed bin and blocking
along the screw. Therefore, screw feeder was replaced by pneumatic conveyor to
avoid blockage during fuel feeding. Operating parameters were bed temperature,
fluidization velocity and rice husk feed rate. Secondary air was injected to the system
1.2 m above the distributor to prolong the retention time of rice husk in bed and to
improve combustion. Compared to coal, ignition stage of rice husk was characterized
by lower ignition temperature (340 °C) and easier ignition, fast devolatilization,
intensive combustion and rapid temperature rise and slower temperature drop in
terminal stage and more time for complete burn-out of coke. Similar to the previous
studies, major portion of the rice husk combustion took place in the freeboard region.
About 97 % combustion efficiency was achieved. The emission levels were low with
SO2 ranging from 50 to 150 ppm and NOx ranging from 150-220 ppm.
The effect of temperature and fluidization velocity on emissions and combustion
performance of highly volatile (~74 %) rice husk combustion was investigated by
Armesto et al. [50] in 0.3 MWt CIEMAT bubbling fluidized bed combustor. Due to
high volatile matter content, rice husk was easier to ignite and burn with respect to
coal however; its combustion was rapid and difficult to control. The temperature
range was from 840 to 880 °C and fluidization velocity range was from 1 to 1.2 m/s.
19
Fuel was fed pneumatically from the bottom of the bed. Rice husk was defined as a
problematic fuel during handling and transportation due to its low density. The main
elements of rice husk ash were silica, potassium and phosphorus which could have
detrimental effect on the ash melting properties leading to agglomeration. Increase in
fluidization velocity resulted in reduction of combustion efficiency. Increase in bed
temperature improved the combustion efficiency to about 97 %. CO emissions were
greater than 1000 mg/Nm3. NOx emissions were between 200 and 300 mg/Nm3. Due
to low sulfur content of rice husks SO2 emissions were very low. Potassium and
calcium silicates, which played important role in agglomeration, were detected in the
bed ash, so that during experiments the control of potassium content of the bed was
significant for avoiding agglomeration.
Low CO, SO2 and NOx emissions were reported in the study of Fang et al. [51] who
studied combustion of high volatile content rice husks in 1 MWt circulating fluidized
bed. Rice husk alone was difficult to fluidize but it was well fluidized with silica
sand bed material. Some difficulties such as bridging in the feed bin and blocking at
the outlet occurred in feeding the rice husks with the screw feeder. The situation was
improved by supplying secondary air through the feeding port and by vibrating the
hopper. The ignition temperature of the rice husk was obtained as 340 °C. Due to its
high volatile content, major portion of the rice husk combustion took place in the
freeboard. Secondary air was injected 1.2 m above the distributor plate in order to
improve the combustion. The proper air split was obtained as 7:3 at fluidizing
velocity of 1.2 m/s. Combustion efficiency was 97 %. CO emission varied from 200
to 800 ppm. SO2 emission ranged from 50 to 100 ppm and NOx emission ranged
from 150 to 220 ppm.
The effect of excess air and combustor load on combustion performance and
emissions of some agricultural residues such as rice husk, sawdust and sugar cane
bagasse were investigated by Permchart and Kouprianov [52] in a bubbling fluidized
bed combustor with a conical bed. As received bagasse, was pre-dried from moisture
content of 48.8 % to 14.4 %. Fuel feeding rate and excess air percentage were taken
as independent variables. For maximum and minimum combustor loading, feed rates
20
were 81.5 kg/h and 35 kg/h for sawdust, 82.4 kg/h and 37.3 kg/h for rice husk, 70
kg/h and 31 kg/h for bagasse. The fuels were burnt at 20, 40, 60, 80, 100 % excess
air. The highest temperature in the combustor was observed for saw dust. Despite the
higher heating value of bagasse compared to rice husk, temperature profile of
bagasse was lower due to its high moisture content and smaller feed rate.
Temperatures in the freeboard section were found to have increasing tendency at
higher excess air levels. Increase in excess air form 20 % to 100 %, increased the
temperature at the top of combustor (2.75 m) by 60-80 °C for rice husk and bagasse,
and by 160 °C for saw dust. The maximum oxygen consumption was observed in the
bed section for all the fuels. CO formation rate of rice husk was much greater than
that for other fuels due to its higher fuel ash concentration and coarser char particles
which were 200 μm for rice husk, 10 μm for bagasse and 5 μm for saw dust. These
factors resulted in higher bed hold-up and led to higher values for CO emission from
rice husk combustion. NO emission was expected to originate from fuel-N [18, 59].
Basically, fuel-NO could be formed through combustion of HCN, NH3 with volatile
matter and oxidation of nitrogen retained in the char. These reactions resulted in
rapid NO formation in the bed region. In the upper region of the combustor as
oxygen concentration decreased, NO reduction was seen through its reaction with
NH3, followed by formation of N2 and H2O vapor. During rice husk combustion,
high fuel-N and high amount of coarser ash particles resulted in higher NO formation
in the bed region with respect to saw dust and bagasse. NO was reduced in freeboard
region. N2O formation in biomasses was found negligible. CO emission mostly
depended on the excess air level. When excess air in rice combustion was smaller
than 50 %, CO emission increased however, when the excess air was greater than 50
%, CO emission decreased. For the maximum combustor load and excess air of 50 to
100 %, over 99 % combustion efficiency was achieved when firing saw dust and
bagasse. Lower combustion efficiencies (80-86 %) were obtained for rice husk
combustion owing to high amounts of unburned carbon.
In more recent study of Kuprianov et al. [53] single firing of rice husks and co-firing
of rice husk with sugarcane bagasse were carried out in the same test unit to
investigate the effect of rice husk fraction, excess air on combustion efficiency and
21
emissions. Temperature during single firing of rice husk ranged from 700 to 800 ºC
within the combustor. Increase in the sugarcane bagasse fraction having 48.8 %
moisture content resulted in reduced temperatures at around 550-650 ºC. Increasing
excess air from 40 to 100 % did not show a certain effect on bed temperature. The
axial O2 profiles for various blends were similar to single combustion of rice husk.
CO emission was reduced from 2000 ppm to about 500 ppm with increasing excess
air from 40 to 100 %; however, the effect of excess air on CO emission was
weakened for higher values. NO emission was found to increase with increasing
excess air from about 120 to 200 ppm. Combustion efficiency of single firing rice
husk tests were about 96 % in the excess air range of 40 to 100%. Combustion
efficiency of co-firing of rice husk having greater than 60 % fraction on energy base
was similar to single rice husk firing and thus, found to be feasible.
The effect of excess air, primary to secondary air ratio and over-bed and under-bed
feeding on CO and CO2 emissions from rice husk combustion were investigated in
the study of Albina [32] in the multiple spouted and spouted fluidized beds. High
volatile matter and ash content of rice husks influenced the combustion and emission
characteristics. Due to its lower ignition temperature it was difficult to control the
operation compared to coal combustion. During over-bed feeding in multiple-
spouted bed, combustion efficiency peaked to 92 % at 10 % excess air and reduced
to 83 % when additional excess air was supplied. CO emission during under-bed
feeding was less compared with the emission with over-bed feeding. However, CO2
emission did not change significantly for both methods of feeding. Combustion
efficiency during under-bed feeding ranged from 92 to 94 % showing a peak value at
20 % excess air and resulted in lower CO emission. Highest values were obtained at
the primary to secondary air ratio of 40/60 for both over and under-bed feeding. To
determine the influence of different nozzle designs, some experiments were
performed in spout-fluid bed. CO emissions were lower at 10 % excess air due to
high combustion efficiency. Over or under-bed feeding had no significant influence
on CO and CO2 emissions in the given excess air range. Combustion efficiency
ranged from 95 to 98 % during both over and under-bed feeding. Lower CO
emissions were measured in spouted bed compared to that of multi-spouted bed. This
22
result was attributed to higher combustion efficiency attained by spouted fluidized
bed.
Using alternative bed materials instead of silica sand was considered to improve the
emission performance of biomass combustion systems. The influence of quartz sand
and porous alumina on CO and NOx emissions from combustion of rice husks was
investigated by Shimizu et al. [54] in a batch scale fluidized bed combustor. Quartz
sand was used as a conventional bed material and alternatively porous alumina was
employed. For quartz sand high CO emission was observed at O2 concentration
lower than 6 %. For alumina bed, the increase in CO emission was observed at O2
concentration lower than 4 %. Porous alumina bed was found to suppress CO
emission. The enhanced volatile matter combustion and reduced CO emissions was
explained by hydrocarbon capture of porous particles. The capture of volatile matter
in the pores prolonged its residence time in dense bed, thus enhanced the burn up of
the volatile matter. Another effect of volatile matter capture was, enhanced
horizontal mixing of carbonaceous material in the volatile matter. Carbon in solid
was dispersed horizontally by solids mixing. Both effects were effective in CO
emission reduction. NOx emissions for porous alumina were found to be lower (250-
550 ppm) than that of the quartz sand bed (150-400 ppm). One possible mechanism
of NOx reduction could be carbon retention within the pores which is known to
reduce NOx to N2 under fluidized bed conditions. Thus the increased carbon loading
in the bed material by volatile matter suppressed NOx emission. N2O emissions were
below 10 ppm for both bed materials. During the experiments with silica sand bed
material defluidization took place however, with alumina bed material no such
problem observed.
Prediction of agglomeration tendencies of agricultural residues are mainly obtained
by ASTM standard ash fusion test [60]. However, this test has been reported to be a
poor indicator of ash related problems in fluidized beds [61, 62] but, it is still used
since there is no reliable alternative method. To compare the predictions of ASTM
standard ash fusion test with agglomeration tendencies of residues in fluidized bed
Natarajan et al. [55] performed experiments on rice husk, bagasse, cane trash and
23
olive flesh by a method based on controlled fluidized bed agglomeration test. The
most important parameter for bed agglomeration is the actual bed temperature.
However, the particle temperature in fluidized bed combustion is an unknown
variable which may exceed the bed temperature [63, 64]. To avoid the uncertainty in
the surface temperatures of the particles in the controlled bed agglomeration tests,
experiments were performed without actual combustion but in the similar
combustion atmosphere. In the experiments, the bed was charged with the bed
material (quartz sand or lime). Then fuel was fed to the combustor and burnt with 6
% excess air under normal FBC conditions to produce the required ash content for
further agglomeration test. When 10 % ash is formed in the bed, fuel feeding
stopped and combustion atmosphere is maintained by air pre-heater in order to avoid
localized high temperature around particles and to maintain isothermal conditions in
the bed. Then the bed temperature was raised by 3 °C/min until the agglomeration
was encountered. Agglomeration was indicated by the differential pressure drop in
the bed due to cohesion of ash particles. From the ASTM ash fusion tests, initial
deformation temperatures were obtained as 1600 °C for rice husks, 1200 °C for
bagasse, 900 °C for cane trash and 1100 °C for olive flesh. But during FBC of these
residues significant melting of the ash occurs far below the initial deformation
temperature. Initial agglomeration temperatures were 1009 and 1020 °C for rice
husks, 1020 and 1020 °C for bagasse, 890 and 905 °C for cane trash and 933 and
1020 °C for olive flesh in the case of the bed materials quartz sand and lime,
respectively. The use of lime instead of quartz sand as the bed material improved the
agglomeration temperature. Cane trash and olive flesh showed higher bed
agglomeration tendency than rice husk and bagasse.
Slagging and agglomeration characteristics of rice husks were studied by Liu et al.
[56] in a bubbling fluidized bed combustor. FBC is given as a recommended way for
burning rice husk with excellent heat and mass transfer characteristics. Chemical
reaction kinetics of rice husk was obtained on the basis of thermogravimetric
experiments. Agglomeration and slagging characteristics of rice husks were studied
in crucibles heated in electrical furnace for 2 hours at 950, 1000 and 1050 °C. Ash
fusion temperature of rice husk was measured to be higher than 1500 °C which
24
implied safe operation under FBC conditions. Experiments were carried out with an
accelerated surface area and porosimetry system to examine the microscopic
structure change of rice husk. Pore volume and specific surface area of rice husk
reached maximum at 850 °C but decreased when temperature exceeded 900 °C.
During FBC of rice husk release of volatiles and burning of carbon occurred almost
separately which complicated its burnout. High silica content of rice husk ash (95 %)
resulted in combination of silica crystals with carbon at high temperatures so that
carbon did not oxidize even at 1200 °C. Therefore, high temperature was not
indicative of high efficiency in the case of rice husk. The optimum temperature range
for fluidized bed combustion of rice husk was given from 800 to 850 °C.
Bubbling fluidized bed boilers generally are very suitable for biomass fuels and fuel
mixtures. However, all combustion systems burning biomass are quite sensitive to
fuel ash behavior. Fouling and corrosion of heat exchange surfaces are common
problems in units burning biomass. Slagging and fouling behavior of rice husks
having high ash content (20 wt %) were investigated by Skrivars et al. [57] in pilot
and full scale units. The investigation was based on short term (3-10 h) deposits
samples taken with air-cooled deposit probes in the superheater region of a large
scale (157 MWt) bubbling fluidized bed boiler burning rice husks and eucalyptus
bark. From previous experiences, it was stated that for biomass fired boilers 2-3 h
sampling time was sufficient to give the first indication of fouling. Pilot scale tests
were performed in an entrained flow reactor which was built to simulate conditions
that fly ash particles experience around the superheater area in bubbling fluidized
bed boilers. Thus, flue gas temperature was set to 900 ºC and deposit probe
temperature was set to 500 ºC. Rice husk ash was composed of 95 wt % SiO2 with
the remainder being mostly K, Ca and P and produced coarse fly ash particles. When
fired alone, rice husk particles did not stick onto heat exchange surfaces and caused
no significant fouling or slagging problems in both pilot and full scale tests. In co-
firing tests, the presence of rice husk ash particles in fly ash kept the tube surface
unfouled, even if a fouling fuel such as eucalyptus bark was fired together with rice
husk.
25
Leaching of inorganic constituents from biomass prior to combustion has been
shown to be of substantial benefit in improving combustion properties and reducing
fireside fouling [65, 66]. Rice straw ash content is greater than rice husk ash content
and rice straw ash differs from rice husk ash in that it contains lower silica and
higher potassium. Therefore, it involves higher possibility to cause agglomeration,
slagging and fouling problems. In the study of Bakker et al. [58] rice straw
combustion tests were performed in a laboratory scale fluidized bed to asses fouling,
slagging and agglomeration properties of rice straw due to leaching. Leaching
extracted large amount of alkali metals and chlorine from rice straw ash so that
potassium and chlorine contents were reduced from 16.60 to 1.99 % and 1.15 to 0.03
%, respectively. During leached rice straw experiments, good fluidization is achieved
and no evidence of agglomeration was observed. A slight light brown deposit
accumulated on the air cooled deposit probes. However, in untreated rice straw
combustion tests, extensive bed agglomeration occurred. Inspection of the reactor
revealed that a large section of reactor tube was completely filled with agglomerated
straw ash and bed media. Most of the agglomerates consisted of straw ash and char
with bed particles adhering to the char. A bright white deposit consisting of very fine
particles had accumulated on the air cooled deposit probes, and some larger black
ash particles were adhering to this fine deposit.
Straw is another mostly utilized agricultural residue in fluidized bed combustors.
Similar to rice husk case, addition of silica sand bed material is needed to have better
fluidization during straw combustion. Details and operating conditions of FBC
systems burning straw are summarized in Table 2.4.
Main operational problem is agglomeration during straw combustion. High
potassium content of straw causes formation of agglomerates and eventually
defluidization. In combustion processes, potassium containing compounds are prone
to remain in the bed and form low melting potassium-silicates. The molten ashes coat
the bed material, promoting agglomeration and defluidization in fluidized beds [67].
26
Table 2.4: Operating conditions and system properties of straw firing FBCs.
Reference Fuel Operating conditions System Bed material
Lin et al. [68]
Straw
Bed temperature: 805-930 °C
Stoichiometric factor: 0.95-2.58
Quartz sand
Lin et al. [67]
Wheat straw
Bed temperature: 725-930 °C
Stoichiometric factor: 1-2.6
Gas flow rate: 14-28 Nl/min
BFBC 0.068 m inner diameter & 1.2
m height Quartz sand 275-460 μm
Skrifvars et al. [69]
Wheat straw Olive flesh
Peat, Sugar cane
bagasse Cane trash
Forest residue Grass Bark RDF
ASTM Ash fusion test Sintering test
Laboratory scale FBC test N.A.
Öhman et al. [70]
Wheat straw Sugar cane
bagasse Forest residue
Grass Bark
Quartz sand
Öhman & Nordin
[71]
Wheat straw Bark
Bed temperature: 760 ºC, The agglomeration test was
at 650 ºC (for straw) Quartz sand & 10 wt % Kaolin (200
µm)
Brus et al. [72]
Wheat straw Olive residue
Bark Peat
Grass
Bed temperature: 800 ºC For straw and grass 650 ºC
40 h operation
BFBC 5 kW
0.1, 0.2 m inner
diameter in bed and
freeboard, respectively
& 2 m height
Quartz sand & blast-
furnace slag (106-125
µm)
Arvelakis et al. [19]
Wheat straw Olive residue Bed temperature: 800 ºC
BFBC 0.0996 m
inner diameter &
2.55 m height
Sand
Nielsen et al. [73]
Straw
Exposure of superheater tube alloys to synthetic
flue gas (6 vol % O2, 12 vol %
CO2, 400 ppmv HCl, 60 ppmv SO2, balance N2) at
550 ºC
Electrically heated oven N.A.
N.A: Not available
27
Temperature has strong influence on agglomeration. Agglomeration tests for straw
were performed by Lin et al. [68] in a laboratory scale bubbling fluidized bed. The
influence of temperature and stoichiometric factor on defluidization time was
investigated. Straw was manufactured as pellets of 1-10 mm to avoid feeding
problems. Agglomeration of the bed material was marked by significant pressure
drop over the bed. Increasing the temperature accelerated the agglomeration time.
Increasing the stoichiometric factor also increased the agglomeration tendency
however; its effect was negligible compared to that of temperature. The ash was
mainly composed of K2O and SiO2 which formed low melting point eutectic
mixtures below 800 °C. The adhesive efficiency of the particles increased
significantly with increasing the temperature after exceeding a critical point and a
low viscous molten phase occurred [74]. This caused the accumulation of the
particles and formation of agglomerates in the bed during higher temperature
combustion at 930 ºC. During the lower temperature experiments at 805 ºC, it was
found that agglomerates have already been formed long before the pressure drop.
When the particles began to agglomerate, they form multi-size particles which tend
to segregate with bigger or heavier particles at the bottom (jetsam) and smaller or
lighter particles at the top (flotsam). In straw combustion, the agglomerates mostly
acted as jetsam particles. Increasing agglomerate levels could cause segregation and
a layer of defluidized agglomerates would form.
The influence of bed temperature, gas velocity, particle size and stoichiometric factor
of bed material on agglomeration tendency during Danish wheat straw (in 1-10 mm
pellet form) combustion were investigated by Lin et al. [67] in the same laboratory
scale BFBC system mentioned above. Defluidization occurred in all the combustion
experiments which were indicated by a sudden decrease in the pressure drop over the
bed. Temperature profile was another indicator of defluidization. When the bed was
in normal fluidization state, bed temperature was very uniform. Just before
defluidization, difference between temperature in the bottom of the bed and 2 cm
above (bed temperature) became larger due to poor mixing. It was noticed that
pressure drop declined slowly before defluidization occurred at relatively low
combustion temperatures, suggesting a segregation of large agglomerates to the
28
bottom of the bed. Temperature influence on defluidization time was obtained to be
very significant. Straw ash started to melt already at 750 °C. Potassium compounds
in straw ash remained in bed and little amount was evaporated at 810-900 °C.
Decrease in temperature led to increasing defluidization time. On the other hand,
increase in temperature caused decrease in ash melt viscosity and coatings were
formed which accelerate defluidization. Doubling the gas flow rate provided longer
defluidization time due to better mixing of the particles and increased force acting on
agglomerates. The better mixing and higher breaking rate of formed agglomerates
prolonged defluidization time about 30 %. Increasing the particle size of the bed
material caused shorter defluidization times because of the particles having smaller
outer surface area that resulted in thicker coating layer. The effect of stoichiometric
factor on agglomeration was not found to be significant.
Agglomeration tendency of ten biomasses during combustion in a fluidized bed
boiler were predicted by three different methods in the study of Skrifvars et al. [69].
ASTM ash fusion test, a sintering test based on compression strength measurements
of ash pellets and a combustion test in a laboratory scale fluidized bed furnace in
which bed agglomeration was achieved in a controlled manner were used to predict
agglomeration tendency. A Danish wheat straw rich in chlorine (3.7 %), olive flesh
from olive oil processing waste, sugar cane bagasse, cane trash mainly the leaves
from sugar cane, reed canary and Lucerne grass, peat, forest residue, bark from
barking process of a pulp mill material and refuse derived fuel (RDF) from source-
separated community waste were used in agglomeration tests. In all cases, except for
wheat straw, the initial deformation temperature was above 1000 ºC, the highest
being 1550 ºC for grass ash. For the wheat straw initial deformation temperature was
900 ºC. Ash from Lucerne grass showed the lowest sintering temperature as 625-650
ºC while peat ash showed the highest sintering temperature as 1000-1100 ºC.
Sintering temperature of wheat straw was about 700 ºC. In combustion tests the bed
agglomeration was detected from bed pressure drop. Agglomeration temperatures of
wheat straw, olive flesh, bagasse and bark were detected as 740, 930, 1020 and 988
ºC, respectively. In all cases the ASTM ash fusion test failed to predict the bed
agglomeration temperature. Many processes in the ash affecting bed agglomeration
29
such as sintering and melting had already started at far lower temperatures than ash
fusion test detected. Although compression strength tests gave more accurate results
than as fusion tests, it also had limits. This test only detected ash particle to ash
particle or ash particle to gas phase interactions. Possible interactions with bed
material were not detected. Adding bed material would mainly give a diluting effect
on ash pellets. The controlled bed agglomeration tests gave the accurate predictions
of possible bed agglomeration problems caused by ash.
In the study of Öhman et al. [70], in-bed behavior of ash forming elements in FBC of
wheat straw, wood, peat, cane trash, grass and bark was examined by SEM/EDS
analyses of samples collected during controlled agglomeration tests. The fuels were
in pellet form having a diameter of 6-8 mm and length of 5-15 mm. Fuels were
combusted in a 5 kW bench scale fluidized bed reactor and compared with bed ash
samples collected from biomass fired full-scale fluidized bed boilers. The
agglomeration test was initialed by loading the bed with a certain ash to bed material
ratio, under normal FBC conditions. Bed temperature was maintained at 760 ºC for
all fuels except for straw. To avoid agglomeration during the ash forming procedure,
650 ºC bed temperature was used for straw. At an ash amount corresponding to 6 wt
% ash in the bed, feeding stopped and operation was switched to external heating.
Several initial runs showed that 1.5 wt % of ash in bed were sufficient for
agglomeration to occur. The onset of agglomeration was determined by pressure and
temperature differences in the bed. The SEM examinations of the bed material
samples showed that a coating with 10-50 µm was formed around the bed particles.
Similar SEM images were observed for 18 MWt CFB boiler samples. From the
elemental analyses of bulk bed samples, silicon was found to dominate in ash
forming elements. Although a major fraction of ash forming elements introduced by
fuel was retained in the bed, they were depleted by some elements such as sulfur and
chlorine. In the case of full-scale boilers, total concentrations of these elements in the
bulk bed were also low. The vaporization during combustion in bench scale unit was
found to increase linearly with increasing bed temperature. In addition, no
vaporization was found during external heating period, even when the temperature
was increased to 960 ºC. Thus, coating characteristics seem to be preserved during
30
this period to determine reliable coating agglomeration temperatures. The results of
SEM/EDS analyses indicated that layers covering the bed particles were
homogeneous, but elemental distributions in the coatings varied significantly
between bed samples of different fuels. Samples collected from firing wood, bark,
cane trash and wheat straw showed that overall compositional distributions of major
fraction of bed particle coatings are mainly limited to ternary system K2O-CaO-SiO2.
The melting behavior was very sensitive to relative amounts of potassium and
calcium in the fuel. Coatings with a relatively high fraction of potassium and lower
fraction of calcium (wheat straw, wood) contained large amount of melt below 900
ºC. On the other hand, coatings with higher calcium and less potassium (bark, cane
trash) did not contain large amount of melt until temperatures well above 900 ºC.
To reduce agglomeration and fouling problems various kinds of mineral additives
can be used for alkali sorption to increase the melting temperature of the system [75].
In a comparative study between kaolinite, bauxite and emalthite, kaolinite was
proved to be the most efficient mineral for alkali sorption [76]. Kaolin addition to
fluidized bed combustion system burning various high alkali and chlorine content
fuels was found to reduce chlorine deposition at typical superheater temperatures of
modern fluidized bed boilers (500 ºC) [42]. The effect of kaolin addition on actual
agglomeration temperature of two troublesome biomass fuels, straw and bark having
high potassium content, and role of kaolin in bed agglomeration were investigated in
the study of Öhman and Nordin [71]. Several initial runs showed that 1.5 % of ash in
bed was sufficient for agglomeration to occur and that 50 % of the added kaolin were
retained in the bed during the experiments. When kaolin, Al2Si2O5(OH)4, was added
to the bed, it was first transformed to metakaolinite, Al2O3.2SiO2, and captured
potassium to form potassium alumina-silicates, K2O.Al2O3.2SiO2, not leaving
sufficiently available potassium oxide to form molten coatings, K2O.CaO.SiO2, on
bed particles. Hence, compositions of the coatings were changed toward higher
melting temperatures mainly because of their decreased potassium content. This
resulted in an increase of initial agglomeration temperature by over 100 °C and 10 °C
for straw and bark, respectively.
31
To prevent agglomeration during combustion of high alkaline fuels different bed
materials other than quartz sand can be used. In the study of Brus et al. [72] iron
blast-furnace slag (BFS) was used as bed material in controlled agglomeration
experiments of bark, olive residue, peat, straw, and reed canary grass. The raw
materials were in pellet form having 6-8 mm size. The initial agglomeration
temperature for quartz sand was about 25 to 60 ºC lower compared to BFS bed
material when bark or olive residue was burned. The formed agglomerates were
found to be more porous and the agglomeration process was extended when BFS bed
material was used. A significant difference in the time until agglomeration was
detected for combustion of reed canary grass at 800 ºC (210 min for sand, 380 min
for BFS), while no significant difference in bed agglomeration was detected for
straw. The bed particles of the quartz sand showed an inner attack layer more often
than those of the BFS. The BFS bed material showed a lower tendency to react with
ash forming elements from the fuel. The outer coating layer had similar thickness
and characteristics for both materials. SEM/EDS analyses of the agglomerates
showed that inner calcium-potassium-silicon attack layer was responsible for
agglomeration process using quartz sand bed material. The composition of
agglomerates formed when using BFS as bed material was similar to outer coating
layer.
Leaching is an efficient way of reducing ash content and agglomeration tendency of
high alkaline content straw ash. Leaching is the treatment of materials with tap water
at room temperature in order to reduce the ash content and change the ash chemistry
to minimize ash melting behavior of the materials [65, 77]. To investigate the
influence of leaching on agglomeration behavior, Arvelakis et al. [19] performed
combustion tests on Danish wheat straw and olive residue (1<L<1.4 mm and L<1
mm) in a laboratory scale bubbling fluidized bed. Fuel was fed to the combustor with
pneumatic transport. In the combustor, limestone with average size varying from 0.5
to 1 mm and density of 2650 kg/m3 was selected as the inert bed material since it
could act as desulphurization agent and have low tendency to react with alkali
metals. At the end of each test, bed material was collected and analyzed with SEM-
EDS technique to define the main inorganic elements forming the agglomerates.
32
Leaching was seen to be insufficient to prevent deposit formation and agglomeration
during fluidized bed combustion of wheat straw. Leaching improved ash thermal
behavior of olive residues securing unproblematic operation of the combustor. No
deposition and agglomeration problem was observed during the tests carried out with
leached olive residues.
Deposition problems such as slagging and fouling are caused by inorganic
constituents of biomass whereas problem with corrosion of superheater tubes is
related to deposition and presence of potassium chloride in the deposits and it is
traditionally avoided by keeping steam temperatures below 450 ºC. Presence of
potassium chloride in the deposits can cause selective chlorine corrosion of
chromium and iron and leaving a nickel skeleton behind. In the study of Nielsen et
al. [73], corrosion measurements were performed in the laboratory to investigate
corrosion of superheater tubes in straw-fired boilers under well defined conditions.
Experiments were conducted to determine the corrosion rate for test metals (ferric
×20 and austenitic AISI 347) which are the commercial superheater alloys. They
were exposed to a synthetic flue gas (6 vol % O2, 12 vol % CO2, 400 ppmv HCl, 60
ppmv SO2, balance N2) in 550 ºC electrically heated ovens for 1 week to 5 months.
The corrosion of the metal was quite uniform and corrosion products were mainly
consisted of iron and chromium oxides. The inner oxide layer contained both iron
and chromium oxides whereas the outer part primarily consisted of iron oxides. On
the top of the oxide layers, a dense and very distinct mixed layer of K2SO4 and iron
oxide (FexOy) was present. Surface of the deposits were covered by K2SO4 and KCl-
K2SO4 mixed layers. The corrosion rate was determined as the thickness of the oxide
layer measured in 3-6 points on each metal sample. Austenitic AISI 347 was shown
to be more resistant toward the corrosion from KCl deposit than ferric×20. A
corrosion mechanism for chlorine corrosion was also suggested in this study which
was based on gaseous chlorine attack where iron and chromium in the metal reacted
with gaseous chlorine and form metal chlorides. High partial pressure of the chlorine
close to the metal was previously believed to be originated from in-deposit sulfation
of KCl and K2SO4. The new part of the corrosion mechanism revealed that KCl
formed a melt with K2SO4 and Fe compounds, and the sulfation was fast in this melt.
33
At low temperatures, solid phase sulfation was slow and metal suffered only from
general oxidation. When metal temperature exceeded the lowest melting temperature
in the KCl/ K2SO4/Fe compounds system, KCl sulfated quickly in the melt,
generating a high partial pressure of Cl2/HCl. This caused accelerated oxidation and
possibly selective corrosion of the metal.
Olive residues are also utilized in fluidized bed combustors. Olive residue is a
specific type of biomass from olive oil production process. Olive residues produced
from pressing of olives and extraction of olive oil, have much lower moisture and oil
contents than the ones produced from centrifuging and then extracting of the oil.
Details and operating conditions of bubbling and circulating fluidized combustors
burning olive residues are summarized in Table 2.5.
Table 2.5: Operating conditions and system properties of olive residue firing FBCs.
Reference Fuel Operating conditions System Bed material
Khraisha et al. [78]
Olive cake Bed temperature: 850-900 °C
BFBC 0.15 m inner diameter & 1 m height
N.A.
Topal et al. [79]
Olive cake Bed temperature: 830-870 °C
Excess air ratio: 1.1-2.16 u0: 1.75-2.61 m/s
CFBC 0.125 m
inner diameter &
1.8 m height
Silica sand
Cammarota et al. [80]
Olive husk Bed temperature: 850-900 °C u0: 0.5 m/s
Quartz sand (212-
400 µm)
Scala & Chirone
[81] Olive husk
Bed temperature: 850-900 °C u0: 0.38-0.92 m/s
Excess air ratio: 1.17-2.1
BFBC 0.102 m
inner diameter & 2 m height
Quartz sand (212-
400 µm & 600-850 µm)
N.A.: Not available
Combustion tests of high moisture content olive residue so called olive cake was
fired in the study of Khraisha et al. [78] in a bubbling fluidized bed. The effect of
bed temperature, feeding rate, fluidization velocity and particle size on combustion
34
efficiency and flue gas composition was investigated. Olive cake samples were dried
in oven before feeding into the hopper. The temperature along the bed was fairly
uniform at about 900 ºC which indicated good mixing within the bed. The
combustion efficiency increased with increasing bed temperature due to increased
reaction rate of carbon with oxygen. The combustion efficiency increased with
increasing particle size due to reduction in elutriation losses for larger particles.
When particle size was greater than 600 μm, combustion efficiency stayed almost
constant at about 70 %. The combustion efficiency decreased with increasing fuel
feeding rate and fluidization velocity.
Combustion of olive cake volatiles mainly takes place in the upper regions of the
combustor so that freeboard temperatures are higher with respect to coal combustion.
In the study of Topal et al. [79] combustion characteristics of olive cake was
investigated in a circulating fluidized bed at different excess air ratios changing from
1.1 to 2.16. Feeding rate was 15.5 kg/h and fluidizing velocity was varied in the
range from 1.75 to 2.61 m/s. Freeboard temperatures were measured as 860-900 °C.
CO and hydrocarbon emissions at excess ratio of 1.25 were measured as 3000
mg/Nm3 and 2400 mg/Nm3, respectively. Increase in excess air ratio resulted in a
sharp decrease in both CO and hydrocarbon concentrations. Combustion efficiency
was 98.7 % at excess air ratio of 1.35 (36 %). When excess air ratio was higher than
60 %, CO and hydrocarbon emissions were increased due to insufficient residence
time in bed and incomplete combustion. The optimum value of excess air was given
as 36 %. NOx emission increased slightly with increasing excess air ratio due to lack
of air staging and small residence time in the combustor. Almost zero SO2 emission
was detected during the tests. The ash of the olive cake was mainly composed of
Na2O (50 wt %) and CaO (32 wt %), however no agglomeration of bed material was
noticed during combustion.
Agglomeration mostly occurs during burning olive residues with high potassium
content. The influence of temperature and excess air on agglomeration tendency
during combustion of high potassium content olive husk was investigated by
Cammarota et al. [80] in a stainless steel cylindrical fluidized bed. Fuel particle size
35
was in the range from 20 to 4000 µm. Feeding system consisted of a fuel hopper
mounted over a screw feeder that further delivered powder into a mixing chamber
where a swirled air flow pneumatically conveyed fuel to the bed. Steady combustion
tests were carried out at fluidization velocity of 0.5 m/s. Fuel feeding was started 750
°C. Bed temperature was kept fairly constant at either 850 or 900 °C. Bed pressure
increased with time during the experiment due to accumulation of ash in bed and
absence of a drain flow. Every 15-20 minutes elutriated material collected at the
cyclone was measured and analyzed for carbon concentration. The pressure variance
was used as onset of agglomeration and it was constant at 60 % for both
temperatures. Agglomeration occurred when critical ash content was reached
depending on bed temperature and irrespective of excess air value (10-90 %). Faster
agglomeration occurred with higher temperature and lower excess air. Potassium
enrichment in agglomerates confirmed that sand surface composition reached the
silica-potassium eutectic point which led to extensive melting.
In the study of Scala and Chirone [81], fluidized bed combustion of olive husk in
virgin and exhausted forms were investigated. Olive residue was having high
propensity for unwanted bed agglomeration problems as a consequence of the high
potassium content (26.56 % in exhausted and 41.65 % in virgin olive husk ash). The
influence of temperature, excess air, fluidization velocity and, bed particle size were
investigated. In contrast to other fuels, air-assisted in bed olive husk feeding was
fairly smooth due to high density (1000 kg/m3) and sphericity (0.84) of the particles.
CO, CH4, SO2 and NOx concentrations were measured at the outlet of the combustor
at different excess air ratios. As expected, CO and CH4 concentrations sharply
decreased as the excess air increased. SO2 concentration was relatively high because
the sulfur content of the fuel was considerable. As excess air was increased SO2
concentration decreased due to gas dilution. NOx concentration at the outlet exhibited
a minimum with excess air. Possible explanation of this situation might rely on two
conflicting effects given as tendency of NOx to decrease due to larger gas dilution
and tendency of NOx to increase due to higher oxygen concentration which promoted
nitrogen oxidation. Carbon elutriation rate decreased and combustion efficiency
increased as excess air increased. Total combustion efficiencies higher than 96 %
36
were found for all the experiments, and they were higher than 98 % when excess air
was above 1.5. Faster agglomeration occurred with higher temperature as
agglomeration was enhanced at higher temperatures where low melting point
eutectics at the bed particle surface were easily reached. Lower excess air and higher
fluidization velocity also led to fast agglomeration. When the experiments were
carried out with larger bed particles, defluidization time was almost doubled with
respect to smaller particles. As larger bed particles had more inertia they were
associated with more energetic collisions and hence, adhesion of the particles to form
agglomerates was difficult. Defluidization time for virgin residue was shorter than
the exhausted residue due to its higher alkali content ash. When critical ash content
was reached in the bed, defluidization occurred. Ash amount in bed was strongly
dependent on bed temperature, sand size and husk type.
Sugar cane bagasse was also burned in fluidized beds [52, 53, 61, 69, 70]. Despite
its fibrous nature, low density and high moisture content, it can be fluidized by
mixing with bed material [82]. FBC of sugar cane bagasse was studied by Kuprianov
et al. [83] in a bubbling fluidized bed of 0.9 m inner diameter, 3 m total height and
having conical bed with cone angle of 40°. Since ash content of bagasse was low,
silica sand with particle size of 0.3-0.5 mm was used as bed material to ensure
sustainable ignition and combustion. Bagasse was dried from 48.8 % to 14.4 % by
natural ventilation. Feeding rate of bagasse was 31 and 70 kg/h. The excess air ratio
of 20, 60 and 120 % were tested. In the bed region temperature profiles were found
to be quite uniform and almost independent of excess air. Meanwhile, in freeboard
region temperature profiles were slightly diverged depending on feeding rate and
excess air ratio and simultaneously diminished along the combustor height due to
heat loss across the walls. The effect of combustor load on O2 and CO2
concentrations were negligible. Both NOx and CO axial profiles go through maxima
whose location divided conventionally the combustor volume into the region of the
predominant formation and that of the predominant decomposition of these
pollutants. Higher excess air led to higher NOx emissions. The maximum point of
NOx emission was mainly depending on feeding rate (115 ppm at 70 kg/h and 87
ppm at 31 kg/h) but independent of excess air ratio. The maximum level of CO
37
emission was strongly affected by combustor load as well as excess air ratio. Higher
excess air ratios led to lower CO emissions. Combustion efficiency changed between
96 and 99.7 %. Excess air ratio of 50-60 % was given as the optimum value to secure
high combustion efficiency and low emissions.
Forestry residues such as wood chips, tree barks and pine seed shells are favorable
fuels to be utilized in fluidized beds. Details and operating conditions of fluidized
bed combustion systems burning forestry residues are summarized in Table 2.6.
The ash behaviors of wood chips, peat and coal were investigated in 12 MW
circulating fluidized bed boiler in the study of Skrifvars et al. [84]. Deposit samples
were collected at the cyclone inlet where the flue gas temperature was 850 ºC and
from one location on the convective path where flue gas temperature was 680 ºC
with a sampling time varying from 15 minute to 21 hour. Sampling was carried out
using specially designed surface temperature controlled deposit probes. The probes
were also equipped with a removable ring for later SEM/EDX analyses. The main
ash elements in wood were calcium, potassium and magnesium. No heavy slagging
and fouling was experienced during the tests. Deposits were thin having 0.1-1 mm
thickness. No baghouse samples could be collected during wood combustion tests
since ash amount reaching the bag house was too low. In combustion tests chlorine
was enriched in deposits. In coal and peat firing the main ash components appeared
to be silicates while in wood combustion main components were alkali and alkaline
earth salts. Alkali salts seemed to be present in larger proportions in the deposits
when firing wood than when firing coal or peat, even if the wood itself contained low
amounts chlorine (0.05 wt %) and sulfur (0.03 wt %). The melting behavior or
stickiness of the deposits was evaluated in the temperature range of 500 to 900 °C. It
was assumed that alkali salt fraction was subject to melting and the remainder being
in solid state. In wood combustion, alkali salts appeared in the secondary cyclone
samples and the deposits. In the deposit probe at the convective path, the first
melting point was reached at 564 ºC in the front side and 582 ºC in the back side
deposits.
38
Table 2.6: Operating conditions and system properties of forestry residue firing
FBCs.
Reference Fuel Operating conditions System Bed material
Skrifvars et al. [84]
Wood chips Peat Coal
Bed temperature: 840-850 °C Feeding rate: 0.50-0.75 kg/s
Excess air ratio: 1.2
Quartz sand
(350 μm)
Lyngfelt & Leckner
[85] Wood chips
Bed temperature: 750-850 °C Total air ratio: 1.2
Combustor load: 50, 70, 100 %
CFBC 12 MW
2.5 m2 cross section &
13.5 m height
Sand
Preto [86] Tree bark
Bed temperature: 780-900 °C Fluidizing velocity: 2.1- 2.4 m/s
Excess air: 20-75 %
CFBC 0.8 MWt
CANMET 0.405 m
inner diameter & 6.7 m height
Sand (210-
500 μm)
Miccio et al. [87]
Pine seed shells
Bed temperature: 850 °C Fluidization velocity:
0.66- 0.97 m/s Excess air ratio: 1.23-1.35
BFBC 200 kWt 0. 37 m inner
diameter & 4.65 m height
Silica sand (300
or 725 μm)
Chirone et al. [88]
Pine seed shells
Bed temperature: 850-900 °C Fluidization velocity:
0.5- 0.7 m/s (bench-scale) 0.7-1.0 m/s (pilot-scale)
Excess air: 30-90 %
BFBC Bench-scale,
0.102 m inner
diameter & 1.625 m height BFBC
Pilot-scale, 0.37 m inner diameter &
4.65 m height
Quartz sand
(300-725 μm)
Tranvik et al. [89]
Wood mixture (40 % sawdust + 25 % bark
+ 25 % forest residue
+ 10 % peat)
Bed temperature: 850 °C CFBC 104 MWt
Quartz sand
(280 μm)
39
The alkali salt part of the deposit was completely molten at 724 ºC on the front side
and 754 ºC on the backside indicating roughly that 20 % of the deposit would be
molten. Fly ash entering the flue gas channel was problematic due to its high alkali
salt content. Although, coal and peat have higher ash forming elements than wood in
their structure, the resulting ash entering the flue gas channel was non-sticky and as a
result it was unlikely to have ash related problems.
Effect of air staging and excess air on NO and CO emissions from fluidized bed
combustion of highly volatile wood chips were investigated in the study of Lyngfelt
and Leckner [85]. In the case of coal combustion NO emissions were known to
depend on both temperature and excess air [90], however, for the case of biomass
combustion; NO emission was found to be insensitive to temperature and decreasing
the excess air had an influence on NO emission which was less pronounced than its
influence on that of coal [91]. By this method NO emissions were reduced, however,
CO emissions were increased due to unburned hydrocarbons. Under normal
conditions in a CFB with a cyclone, CO emission from biomass was lower than that
of coal [92]. As CO was produced from combustion of small char particles in the
cyclone during combustion of coal, this source was less important during combustion
of low char containing biomass. Air staging was also given as another way to reduce
NO emissions. However, conventional air staging in CFB was not effective for NO
reduction in biomass combustion except when secondary air was added as late as the
cyclone outlet for final burnout [92] which was called as reversed air-staging. The
aim of reversed air staging was to lower air ratio at the upper part of the furnace for
N2O reduction, while increasing the air ratio in the lower part for improved sulfur
capture. As for wood combustion sulfur capture was out of interest, the mode of
operation was called as late air staging.
The reference conditions were full-load, 850 ºC bed temperature and normal air
staging, which meant 60 % primary and 40 % secondary air introduced at a height of
2.2 m. Under these conditions CO emission was around 14 ppm and NO emission
was around 104 ppm. Decrease in load resulted in lower emissions. Increasing the
bed temperature from 750 to 850 ºC at full-load had no influence on NO emission,
40
but it had a pronounced reducing effect on CO emission. Reducing total air ratio
from 1.2 to 1.12 resulted in decreased NO emission (27 %) at the expense of high
CO emission. Additional test series that included late air staging at the cyclone exit
were carried out for full and 70 % and half loads. Increase in the flow of air added to
the cyclone exit resulted in an equal reduction in air flow from bottom and the total
air flow was kept constant. As combustor air ratio (air ratio in the riser and cyclone
inlet) was lowered, NO emission decreased and passed through a minimum, while
CO emission increased and went through a maximum.
The maximum reductions in NO emission were 25 % relative to normal air staging.
At the minimum NO emission, CO emission level was about 100 ppm. At 70 % load,
NO emissions decreased by 42 % relative to normal air staging at full-load. At the
minimum NO emission, CO emission level was below 100 ppm. At the half load
maximum NO reduction relative to normal air staging was 65 % and, CO emission
was 750 ppm. At full-load, introduction of secondary air from both cyclone inlet and
outlet resulted in a maximum decrease of 35 % in NO emission relative to normal air
staging and reduced CO emissions somewhat to about 54 ppm. At 70 % load,
introduction of secondary air from both cyclone inlet and outlet resulted in a
maximum decrease of 60 % in NO emission relative to normal air staging and CO
emissions were only few ppm as a consequence of a raised temperature at the exit
chamber. At the half load with introduction of secondary air at both cyclone inlet and
exit, reduction in NO emission was about 64 % but CO emission was increased to
162 ppm due to fall in temperature at the inlet and exit of cyclone. CO emission was
strongly correlated to exit chamber temperature that high temperatures resulted in
lower CO emissions while low combustor air ratios and low temperature resulted in
higher CO emission. At the bed temperature of 850 ºC, the temperature at the
chamber exit was typically 840-850 ºC during full load, 800-820 ºC during 70 % load
and 740-760 ºC during half load.
A wood processing residue, tree bark, was burned in a 0.8 MWt CANMET
circulating fluidized bed combustor by Preto [86]. Large amount of Ca in ash
provided Ca/S molar ratio above 3.8, so there was high intrinsic sulfur capture. Due
41
to high volatile content of the fuel, large part of combustion occurred in gas phase,
either in bed or in freeboard. Combustion efficiency was about 99 %. Bed
temperature was uniform but it increased above the bed. Temperature rise was
depending on several factors but moisture was likely to be an important factor.
Lower moisture content of fuel led to more uniform temperature throughout the
system. Emissions of NOx (100-130 ppm), N2O (<1 ppm) and SO2 (5-20 ppm) were
below the limits [93]. CO levels were greater than the proposed limits due to
fluctuations in the feeding system, variation of the fuel composition and small hang-
ups during the feeding process. Formation of chlorinated hydrocarbons, dioxins and
furans was not an issue for inland wood residues.
Highly volatile fuel particles have greater segregation tendency at the bed surface
during devolatilization. The released volatiles may form bubbles around
devolatilizing particles and lift the coarse particles to the bed surface [94]. So that
significant amount of volatile matter bypasses the bed and burns in freeboard section
and increases temperature in this region. Under-bed or over-bed feeding has
significant influence on segregation characteristics. Miccio et al. [87] worked on this
influence on pine seed shells in a 200 kWt bubbling fluidized bed. Experiments were
performed at different fluidization velocities and excess air ratios. No sorbent was
added to bed as the fuel was sulfur free. CO emission was less than 50 ppm due to
high reactivity of volatile matter and residual char. NO emissions were within the
range of 70 to 140 ppm due to low nitrogen content of the fuel and lower at low
excess air ratios. Carbon elutriation at the cyclones was negligible so that about 99 %
combustion efficiency was achieved. Higher freeboard temperature compared to that
of bed indicated that post combustion of the unburned species occurred in this region
due to insufficient residence time, incomplete mixing of oxygen and combustibles in
bed and segregation of fuel particles in the upper bed zone. Over-bed feeding caused
higher freeboard temperatures compared to under-bed feeding. Difference between
freeboard and bed temperatures was 113 °C at fluidization velocity of 0.97 m/s
during over-bed feeding. Radial profiles of gaseous species in the splashing zone
showed significant differences between under and over-bed feeding. Changes in
measured radial concentration along combustor diameter were related to the
42
segregation of combustible matter. This mechanism was enhanced by adopting
under-bed feeding as a consequence of less uniform spreading of fuel particles along
the combustor.
Formation of particle agglomerates in the bed was the most likely occurred
phenomena during woody biomass combustion. Extensive bed agglomeration
problems were experienced during combustion of these fuels as a consequence of
enrichment of sodium and potassium on the sand particles, in conjunction with high
temperature spots near burning char particles. Agglomeration behavior of pine seed
shells was investigated in the study of Chirone et al. [88] in a bench and a pilot scale
bubbling fluidized bed combustor. The runs were ended when agglomeration of the
bed occurred, as indicated by a jump in the temperature and pressure drop profiles
within the bed. Combustion efficiencies were higher than 96.8 % for the experiments
carried out in bench-scale fluidized bed, and higher than 99.9 % in the pilot-scale
fluidized bed. Both facilities eventually ended with bed defluidization. Upon
defluidization, temperatures in the lower section of the bed tended to decrease, while
those in the upper section of the bed increased due to segregated fuel combustion at
the upper section. Agglomerate formation was enhanced at higher temperatures,
where it was easier to reach low-melting point eutectics at the inert particle surface.
Using larger sand particle size resulted in doubled defluidization times with respect
to the corresponding experiments where the smaller sand was used. This was
considered to be due to higher inertia of larger particles associated with more
energetic collisions making the formation of agglomerates more difficult. Combustor
scale exerted a relevant role that defluidization time was 2.5 times higher in the
larger facility under similar operating conditions due to different bed-fluid dynamics
and presence of horizontal cooling tubes. It was also given that irrespective of the
excess air value and fluidization velocity, well-defined ash content was necessary to
defluidize the bed, depending on the bed temperature and scale of the combustor.
Deposition of bed material during woody biomass firing was investigated in the
study of Tranvik et al. [89] in a 104 MWt wood firing circulating fluidized bed
combustor. Bed material deposits collected from cyclone and riser were analyzed by
43
SEM/EDX. The deposits from bottom ash were also sieved to particle sizes of 125,
355, 500, 710, 1000 and 1400 μm and analyzed with respect to potassium content.
The content of potassium oxide increased with increasing particle size. Bed particle
layers were found in all bed samples. The accumulated material around the bed
particles consisted of two layers. The inner layers were thicker and more
homogeneous; the outer layers were thinner and more particle rich. Inner layer on the
quartz bed particles mainly consisted of silica, calcium, potassium and oxygen. The
composition of the outer layers was more complex and also similar to ash
composition. The samples were crashed mechanically and divided into shell and
kernel. Kernel was mainly concentrated with silica, and most of the other compounds
were enriched in the shell. For prevention of deposition of bed material in cyclones
of wood fired FBCs, the use of non-quartz bed material was suggested.
Co-combustion with coal or addition of chemical materials is known to be practical
and economic treatments to reduce sintering caused by alkaline compounds. The
effect of these treatments is to increase the melting point of biomass ash. In the study
of Llorente et al. [95] limestone was used as the bed material to eliminate tendency
of bed material agglomeration and sintering that normally occurs in plants that
operate with the traditional silica bed material. Combustion tests were performed in a
1 MWt BFBC having 1.1 m inner diameter and 4 m height. Brassica (vegetable oil
residue, bulk density=92 kg/m3), thistle (energy crop, bulk density=220 kg/m3) and
almond shells (agro-industrial biomass, bulk density=400 kg/m3) were fired with
either silica or limestone bed material. Materials were dried before feeding to
combustor. During the tests primary to secondary air ratio was 60/40 and excess air
was 60 %. Carbon conversion efficiency was greater than 98 % in the combustor.
The combustion tests using thistle and almond shell generated agglomerates with
silica sand bed material. Round agglomerates were formed near biomass feeding
point in the thistle combustion test, probably due to high temperature at this point of
silica bed. In the case of almond shells, homogeneous agglomerates formed in a
narrow film on the top of the silica bed, which caused poor fluidization and
shutdown of the combustion test 2 hours after its start-up, as a consequence of higher
residence time of almond shells in this area with respect to other bio-fuels with lower
44
density and higher porosity. Agglomerates were not observed when limestone was
used as the bed material indicating the positive effect of this material to reduce or
even avoid the ash agglomeration. An explanation of the good behavior of limestone
bed material compared with silica bed material was due to the dilution effect which
was caused by adsorption of alkaline salts on the surface of the pores of the material
[75].
As shown in the studies, biomass combustion offers many advantages on reduction
of gaseous emissions. However, it brings many operational problems. To avoid those
operational problems related with high alkali content of biomass, co-firing of these
residues with coal is the most simple and effective alternative. Blending biomass
with coal offers higher combustion efficiencies compared to biomass combustion
only and lower emissions compared to coal combustion only. Moreover, it provides
controllable disposal of residue. Following section reviews the studies on co-firing of
various biomass and coal types.
2.3.2 Biomass and Coal Co-firing Studies
The agricultural residues mostly co-fired with coals in fluidized bed combustors are
rice husk, straw, olive residue, and forestry residues are woodchip, bark and sawdust.
Rice husks can be co-fired with coal to reduce the gaseous emissions from coal
combustion. The influence of co-firing of coal with rice husk (0-15 wt %) and
woodchips (0-25 wt %) on N2O and NOx emissions was investigated in a bubbling
fluidized bed in the study of Shen et al. [96]. The results revealed that co-firing
decreased N2O and NOx emissions. As wood chips and rice husks have high volatile
matter, their addition to the fuel blend resulted in larger release of volatiles in the
lower part of the bed. These volatiles consumed most of the oxygen and form a lean
zone in this area to restrict N2O and NOx formation. Increasing the rice husk share in
the fuel blend from 5 to 15 % reduced NOx and N2O emissions from 180 to 145 ppm
and 150 to 100 ppm, respectively. Reduction rate of N2O emission decreased with
45
increasing biomass share. Also increasing the temperature resulted in higher NOx and
lower N2O emissions.
There are various studies on co-firing of straw with coal. Details of the straw and
coal co-firing in fluidized bed combustion systems and their operating conditions are
summarized in Table 2.7.
Straw has low melting point ash and firing straw alone in fluidized beds results in
unscheduled shutdowns of combustors due to agglomeration of bed material and
fouling of heat transfer surfaces. Co-firing coal and straw reduces the fouling and
agglomeration tendency because of the higher melting point of coal ash and its
ability to react with potassium [97].
Agglomeration is a direct result of stickiness of bed material. Tendency of
agglomeration during co-firing of straw and coal was investigated by Lin and Dam-
Johansen [98] in laboratory scale fluidized bed. Quartz sand was used as bed
material. The main parameter reflecting the agglomeration tendency was
defluidization time. The influence of temperature and coal fraction on defluidization
time was investigated. Defluidization was indicated by a sudden decrease of pressure
drop over the bed to a low level. Defluidization time was extended when straw was
co-fired with coal. Combustion temperature had the most pronounced effect on
defluidization time. The influence of coal fraction was found to be insignificant at
high temperatures, but it became very sensitive at low temperatures. Defluidization
time increased sharply with decreasing temperature when temperature was below a
critical value. This critical temperature was 875 °C at coal fraction of 20-30 wt % in
the fuel blend. It was about 100 ºC higher than the melting temperature of ash from
straw firing only.
To predict the melting behavior of ash components, co-firing experiments were
carried out with coal and straw (50 wt %) in a 30 kWt bubbling fluidized bed in the
study of Sandelin et al. [99]. Mixing the fuels did not lead to a mean value of the fuel
properties or a linear relationship for the behavior of the fuels.
46
Table 2.7: Operating conditions and system properties of straw and coal co-firing
FBCs.
Reference Fuel Blend Operating conditions System Bed material
Lin & Dam-
Johansen [98]
Wheat straw pellets
(70-80 wt %) & Polish Coal
Bed temperature: 870-945 °C Under-bed feeding by screw
feeder
BFBC Lab scale
0.068 m inner diameter &
1.135 m height
Quartz sand
(275-460 μm)
Sandelin et al. [99]
Straw (50 wt %)
& Coal
Bed temperature: 850-880 °C Under-bed feeding by screw
feeder
BFBC 30 kWt
0.108 m inner diameter
& 3 m height
Quartz sand
(300-600 μm)
Lin et al. [97]
Coal (36 %) Woodchips
(40 %) Straw (24 %).
Bed temperature: 825-870 °C MgO was used as additive in
lab scale tests.
CFBC 20 MWt
3.4 m2 cross section & 15 m
height, CFBC
80 MWt BFBC
Lab scale
Quartz sand, fuel
ash & limestone
Glazer et al. [100]
Straw pellets (20, 50 % on energy basis)
& coal
Bed temperature: 850-890 °C Under-bed feeding by screw
feeder
CFBC 25 kWt
0.080 m inner diameter & 5 m
height
Silica sand
(300-600 μm)
Kassman et al. [101]
Straw pellets (10, 20, 30 %
on energy basis) &
bituminous coal
Bed temperature: 850-865 °C Primary/total air ratio : 44 %
U0 : 4.8 m/s
CFBC 12 MW
2.25 m2 cross section &
13.6 m height
Started with silica
sand, then
gradually replaced by coal ash and
lime
Okasha [102]
Rice straw (50, 66% ) &
bitumen - CaO pellets
Over-bed feeding Bed temperature: 750-875 °C
in pilot scale BFBC, Bed temperature: 800 °C &
U0 : 3 m/s in lab scale BFBC
BFBC Lab scale
0.1 m inner diameter &
0.8 m height BFBC
Pilot scale 0.3 m inner diameter &
3.3 m height
Silica sand
(250-500 μm)
47
Significant amount of energy was released by coal through char combustion. In
contrast, high and significant amount of energy was released from straw during
combustion of pyrolysis gases. Consequently, burning behavior of the fuel mixture
influenced the behavior of the char residue, where also the greatest amount of ash
forming elements was found. No agglomeration was observed in the bed. The spent
bed material was completely depleted of chlorine and sulfur. Significant amounts of
potassium, calcium, sodium and aluminum were captured in bed ash.
Agglomeration, deposition, corrosion and emission behavior of co-firing straw (0-60
%, thermal basis) with coal was studied by Lin et al. [97] in full scale 20 MWt
circulating fluidized bed boiler, 80 MWt circulating fluidized bed boiler and a
laboratory scale bubbling fluidized bed combustor.
The 20 MWt boiler was a demonstration plant so called multi-circulating fluidized
bed boiler. The unique feature of the boiler was the pre-separator which was
integrated with the particle cooler embedded as a superheater and an evaporator. The
separated particles prevented the heat transfer surface from direct contact with flue
gas and potentially avoided corrosion. Low fluidization velocity in the coolers
minimized the erosion risk. The tests in this boiler were focused on the local
atmosphere (oxidizing and reducing) and alkali concentration in the gas phase in the
combustor and pre-separator. Fuel blend was composed of 36 % coal, 40 % wood
chips and 24 % straw on thermal basis. The alkaline measurements in gas phase
showed that potassium concentration in the combustor section varied between 0.16
and 2.3 ppm (v/v). In the pre-separator, the potassium vapor concentration was in the
range from 0.06 to 3.7 ppm. The combustor exit had the highest potassium
concentration, while, in the pre-separator part, potassium concentration is the highest
at the top. The high potassium concentration in the top part of the pre-separator was
due to high char concentration in this region, resulting in more release of potassium
during char combustion.
In 80 MWt CFBC [103], co-firing straw with coal resulted in considerably higher
corrosion rates compared to coal firing only. Variation in boiler load had a large
48
impact on operating conditions of the boiler, as well as the chemical reactions in the
combustor, loop-seal and convective pass. When the boiler load decreased,
temperature in all parts of the boiler decreased and the condition in the loop seal
shifted from predominantly reducing to more oxidizing. Decrease in particle re-
circulation rate led to lower elutriation rate of char particles to loop-seal, and more
oxidizing conditions there. Deposition tests in the loop-seal showed that the deposits
were only formed at the full load during co-firing of coal and straw. It appeared that
half of the deposit was potassium chloride. Small amount of char transported to the
loop-seal was believed to be the primary source of potassium chloride. Complete
combustion of the char with a small amount of aeration in the loop-seal led to high
concentration of potassium chloride, forming deposits.
In the same study, laboratory scale experiments were carried out with Polish coal and
wheat straw pellets. A mixture of MgO particles and sand was used as the bed
material (66.7 wt % MgO, 33.4 % sand). Agglomeration tendency of SiO2-MgO was
lower than that of quartz sand, but higher than pure MgO, suggesting silica
containing compounds were responsible for agglomeration tendency of bed material.
Using MgO as an additive reduced the agglomeration tendency, but did not eliminate
it. Fine particles were collected in the cyclone and large agglomerates were found in
the bed material. This meant that agglomeration and attrition/abrasion occurred
simultaneously in the bed material. Significant decrease in SO2 emission was
detected at straw share of 60 % on thermal basis. Potassium in straw was capable of
SO2 capture. Using MgO as additive resulted in higher NO emissions.
Chlorine behaves as a shuttle for potassium transportation to the particle surface
before release as potassium chloride to the gas phase. Influence of operating
conditions and fuel composition on the release of alkali compounds to the gas phase
during combustion and co-combustion of high alkali straws with low sulfur content
coal was investigated by Glazer et al. [100] in a 25 kWt circulating fluidized bed.
The amount of total gas-phase sodium and potassium compounds in the flues gases
were measured with excimer laser induced fluorescence. The results showed that the
release of gaseous species was depended on the fuel composition, K/Cl and K/Si
49
ratios in the fuel. A synergic effect of co-combustion of straw with coal led to a
strong decrease in gaseous alkali concentrations. The part of the alkali metals
released from the straw to the gas phase interacted with clay minerals in the coal to
form alkali–alumina-silicates. Potassium and sodium concentrations were much
lower in co-firing tests compared to straw firing tests. Small additions of straw to
coal led to dramatic increase in gaseous alkali content in the flue gas.
Alkali related problems in straw and coal co-firing was investigated in the study of
Kassman et al. [101] in 12 MW circulating fluidized bed boiler. Straw pellets were
co-fired with Polish bituminous coal. The tests were focused on variation of straw to
coal in combination with different feeding rates of limestone. Chlorine content of
straw and coal were 0.36 % and 0.32 % on dry and ash free basis, respectively. Gas
phase alkali chlorides were measured by means of an in-situ alkali chloride monitor
(IACM). SEM/EDX was used for analyses of collected deposits on the steel deposit
rings. Alkali chlorides existed in the gas phase when the flue gas temperature was
higher than 650 °C. IACM instrument measured the sum of sodium and potassium
chloride concentrations on-line. In the first part of the experiments coal was used as
fuel with increasing lime supply. Potassium chloride level was generally low and
independent of SO2 concentrations. HCl level was high and decreased somewhat by
increasing lime supply. In the second part, straw and coal co-firing experiment was
carried out with increasing lime supply. Alkali supply was equal to a fraction of
straw of 17-19 % on energy basis. Addition of straw pellets increased potassium
chloride concentration with approximately one order of magnitude (from ~2 to ~20
ppm). In the third part, straw fraction in the fuel blend was increased with a constant
lime supply. The lime supply was equal to a Ca/S molar ratio of 6.2-6.9. Potassium
chloride concentration increased with increasing straw share. At straw share of 29 %
on thermal basis, about 50 ppm potassium chloride concentration was measured. This
level was believed to be much too high with respect to potential problems fouling
and super heater corrosion [104]. No clear relationship was seen between increased
lime supply and potassium chloride.
50
Deposits were collected on steel rings during a period of 4 hours using a temperature
controllable probe held at 500 °C. The deposits were analyzed for chlorine, sodium,
potassium, alumina, silica, sulfur and calcium by EDX. In the test series where straw
was co-fired at a constant ratio with increasing lime supply, chlorine was present in
all the deposits. Excess supply of lime led to capture of chlorine due to formation of
calcium chloride in the deposits. In the series carried out with constant lime supply
and increasing straw share, significant amounts of potassium and chlorine were
found in the deposits. Increasing the straw ratio in the fuel blend increased chlorine
concentration in the deposits drastically.
Co-firing reduces SO2 emissions due to low sulfur content of biomass and intrinsic
sorbent capability of biomass ash. In the study of Okasha [102], enhancement of
sulfur retention of rice straw-bitumen pellets by integrating CaO within the pellets
was investigated either by a batch operation in a laboratory scale bubbling fluidized
bed or a continuous operation in a pilot scale bubbling fluidized bed. A series of
batch combustion tests were carried out with feeding a single pellet at different molar
ratios of built-in Ca/S. SO2 concentration was measured throughout the burning time
of entire pellet. SO2 concentration decreased greatly with increasing Ca/S ratio in
the pellet. Sulfur retention was less efficient during devolatilization stage as volatiles
intensively released and stayed a very short time within the pellet. On the other hand,
sulfur retention was very efficient during char combustion due to relatively lower
rate of combustion in this stage. SO2 emission was close to zero at char burning stage
at Ca/S molar ratio of 0.75. Sulfur retention efficiency increased with increasing the
share of straw in the pellet. Increasing the Ca/S ratio within the pellet increased NOx
emissions slightly. This was attributed to the catalytic effect of CaO promoting NOx
formation [105, 106]. Increasing the bed temperature intensified the rate of volatile
release and reduced sulfur retention. In continuous operation sulfur retention was
observed at steady state. At built-in Ca/S ratio of 0.75, SO2 emission was reduced to
65 ppm which corresponded to 96 % retention efficiency. Evidently, the retention
efficiency under continuous operation was considerably higher than that of batch
tests considering the same Ca/S ratio.
51
Olive residues have also been co-fired with coals in FBC units. Details of the olive
residue and coal co-firing FBC systems and their operating conditions are
summarized in Table 2.8.
Table 2.8: Operating conditions and system properties of olive residue
and coal co-firing FBCs.
Reference Fuel Blend Operating conditions System Bed material
Armesto et al. [107]
Alpeorujo (10, 20, 25 wt %)
& Puertollano
lignite (<1, <5 mm)
Bed temperature : 830, 850, 870 °C
Fluidization velocity: 0.7-1 m/s Under/over bed feeding
by screw feeder
N.A.
Armesto et al. [108]
Foot cake (10, 15, 20, 25 wt %, dry basis)
& lignite/ anthracite
Bed temperature : 830, 850, 870 °C
Fluidization velocity: 0.7-1 m/s Under bed feeding by screw
feeder
BFBC 0.1 MWt
0.20 m inner diameter
& 3 m height
N.A.
Cliffe & Patumsawad
[109]
Olive oil waste
(0, 10, 20 wt %) & coal
Bed temperature: 830 to 940 °C Excess air: 50, 90 % Pneumatic feeding
BFBC 10 kWt
0.15 m inner diameter
& 2.3 m height
Sand (850 μm)
Atımtay & Topal [110]
Olive cake (25, 50, 75 wt %) & Tunçbilek
lignite
Bed temperature: 850-865 °C Excess air ratio: 1.1-2.16
u0: 1.76-2.55 m/s
CFBC 0.125 m inner diameter & 1.8
m height
Silica sand
(560 μm)
N.A.: Not available
Alpeorujo is a specific type of residue from olive oil production having high
moisture (50-70 %), volatile matter and alkaline content. Difficulties such as bed
agglomeration, slagging on furnace walls and fouling of heat transfer surfaces arise
during combustion of Alpeorujo due to high potassium and sodium content in its ash.
There were some experiences on burning alpeorujo although continuous failure of
the bed during fluidization had determined severe constraints for commercial
operation. The feasibility of co-combustion of Alpeorujo with lignite was
investigated in the study of Armesto et al. [107] in CIEMAT 0.1 MWt atmospheric
bubbling fluidized bed combustor. Operating parameters in the co-firing tests were
52
furnace temperature, fluidization velocity, biomass share, coal particle size and
position of feeding system. Higher combustion efficiencies were obtained when fuel
is fed from the bottom. Increase in biomass share did not show significant effect on
combustion efficiency when fuel was fed from the bottom. CO2 emissions were
changing within the range of 12.6-13.6 % during feeding from bottom. The
placement of feeding had strong influence on CO emission. CO emissions were
about 3000 mg/Nm3 when fed from the top and reduced to about 200 mg/Nm3 when
fed from the bottom. The effect of alpeorujo share in the fuel mixture on CO
emission was very little. Decrease in particle size led to increase in CO emission.
Contrary to the pervious studies, increasing biomass share resulted in higher SO2
emissions. At fluidizing velocity of 0.7 m/s, increasing biomass share from 10 to 20
% led to increase in SO2 emissions from 1018 to 1211 mg/Nm3 and it stayed constant
through 25 % share. However, at 1 m/s fluidizing air, SO2 emission continued to
increase to 1646 mg/Nm3 with higher share of alpeorujo. This behavior was
explained according to the ash characteristics. The ashes obtained during the tests
with minor SO2 emissions had major content of unburned fuel and the presence of
sulfur (as pyritic sulfur) in these ashes were significant. The NOx emissions
decreased when biomass content in the mixture increased. This was attributed to the
lower nitrogen content of the fuel blend and its higher volatile matter content.
Increasing the biomass share from 10 to 25 wt % reduced NOx emission from 274 to
219 mg/Nm3 at 0.7 m/s fluidizing velocity. At 25 wt % biomass share, increasing the
furnace temperature from 850 to 880 °C resulted in higher NOx emissions. When fuel
blend is fed from top of the boiler, higher NOx emission was detected. During the test
runs cyclone, baghouse and bed material samples were collected for chemical
analysis. The main components were SiO2 and Al2O3 which were the main ash
components in the Puertollano coal ash. Chlorine was found to concentrate in
baghouse ash. Potassium components were concentrated in cyclone ash due to the
low temperature of the combustor at this zone, 350 °C, which was below the
condensation temperature of majority of potassium compounds. The content of
potassium in cyclone ash increased and in the baghouse ash decreased with
increasing alpeorujo share in the fuel mixture.
53
Another investigation on feasibility of co-firing olive residue with coal was carried
out by Armesto et al. [108] with a specific type of olive oil residue so called foot
cake which was rich in moisture (66.4 %) and volatile matter (25 %). Foot cake was
co-fired with both lignite and anthracite. The effect of coal type, biomass share,
operating temperature and fluidizing gas velocity on combustion efficiency and flue
gas concentrations of O2, CO, CO2, SO2, NO and N2O were investigated in the
bubbling fluidized bed pilot plant same as given in the previous study. HCl in the
flue gas was measured discontinuously. The combustion efficiency of foot cake-
lignite blends was obtained to be higher than the efficiency of foot cake-anthracite
blends. This was attributed to the higher volatile matter content of lignite compared
to anthracite. In the case of foot cake-anthracite blends, there was a decline in
freeboard temperature causing high levels of unburned carbon. The combustion of
volatiles was completed in the bed but the particles did not appear to have long
enough residence time for combustion. In the case of foot cake-lignite blends, the
decline in freeboard temperatures was less. Increasing the operating temperature
from 800 to 850 °C resulted in higher conversion of fuel-N to NOx from 30 to 50 %
Higher temperatures led to higher combustion rates and radical concentrations [111].
On the other hand, fuel-N to N2O conversion slightly decreased with increasing
temperature. The effect of biomass share in the mixture did not show significant
influence on combustion efficiency. SO2 emissions from combustion of foot cake-
lignite blends were significantly higher than the SO2 emissions from combustion of
foot cake-anthracite blends due to the difference in sulfur contents of fuel blends.
Increasing the biomass share from 10 to 20 wt % reduced SO2 emissions from 1050
to 850 mg/Nm3 for lignite blend and from 500 to 400 mg/Nm3 for anthracite blend.
The major NOx emissions were obtained during the tests using anthracite-foot cake
blends due to lower volatile matter content of anthracite. On the contrary, the major
N2O emissions were on the tests with lignite-foot cake blends. This result was
opposite to what was expected since N2O conversion was known to increase with
decreasing volatile matter. This could be resulted from reduced flame temperature
caused by high moisture content of foot cake. HCl emissions were close to 1
mg/Nm3. This was attributed to the chlorine retention by calcium in the ash.
54
Similar to foot cake, olive oil waste having high moisture content (60 %) was co-
fired with coal in the study of Cliffe and Patumsawad [109]. The experiments were
carried out in a 10 kWt bubbling fluidized bed. Operating temperature was changed
in the range from 830 to 940 °C. Excess air was changed between 50 and 90 %. Fuel
was fed pneumatically to the bed surface from a sealed hopper through an inclined
feeding pipe and the flow rate was controlled by a screw feeder. Biomass share
greater than 20 wt % caused bed temperature to drop so that combustion could not be
sustained. TG and DTG burning profiles showed that maximum burning rate of olive
oil waste was about 450 °C [112]. The combustion efficiencies of fuel blends having
0, 10 and 20 wt % of olive oil waste were ranging from 91 to 95 %, 85 to 93 %, and
88 to 93 %, respectively. The combustion efficiency of 20 wt % olive oil waste
mixed with coal was slightly higher than that of 10 % olive oil waste mixed with
coal. This was explained by the fact that when the moisture content of the fuel
increased, devolatilization time was extended due to delayed gas evolution from the
fuel. Therefore, fuel mixture had more time in bed to burn. Increased concentration
of olive oil waste in the fuel mixture led to less carryover. Significant combustion
took place in freeboard. Freeboard temperatures increased by addition of 10 wt %
olive oil waste to the coal. CO emissions were changing between 100 and 350 ppm at
6 % O2 in flue gas. CO emissions increased with increasing excess air.
Co-firing of olive residue so called olive cake with lignite was carried out by Atımtay
and Topal [110] in a laboratory scale circulating fluidized bed. Due to high Na2O (50
wt %) content of the olive cake ash, potential operation problems such as fouling of
heating surfaces and agglomeration of the bed material was expected during
operations. The influence of excess air ratio on combustion characteristics and
emissions was investigated in a wide range from 1.1 to 2.16. Biomass share during
the combustion tests were 25, 50 and 75 wt %. Silica sand and ash were used as bed
materials. The densities of olive cake, coal and sand were given as 591, 1374 and
1730 kg/m3, respectively. Superficial gas velocity was changing between 1.76 and
2.55 m/s. Olive cake addition to the coal firing system at all proportions led to an
increase in combustion efficiency up to a certain excess air ratio and further increase
in excess air led to lower combustion efficiencies. Combustion shifted to the upper
55
sections of the combustor. At 25 wt % olive cake co-firing, about 97 % combustion
was obtained. SO2 emission was about 2600 mg/Nm3 and did not change with
increasing excess air ratio. However, at the excess air ratio of 1.50, a sharp decrease
was observed in CO and hydrocarbon emissions from 3100 to 169 mg/Nm3 and from
1960 to 36 mg/Nm3, respectively. NOx emission was almost constant at 220-260
mg/Nm3. During 50 wt % olive cake co-firing, 98 % combustion efficiency was
achieved at the excess ratio of 1.5. SO2 emission was almost constant at around 1600
mg/Nm3. Similarly up to excess air ratio of 1.5, sharp decrease in CO and
hydrocarbon emissions were observed from 4290 to 360 mg/Nm3 and from 2240 to
62 mg/Nm3, respectively. At 75 wt % olive cake co-firing, 98.2 % combustion was
achieved at the excess air ratio of 1.6. SO2 emission was about 1250 mg/Nm3. Until
the excess air ratio of 1.6, sharp decrease in CO and hydrocarbon emissions were
observed from 5800 to 150 mg/Nm3 and from 2750 to 120 mg/Nm3, respectively.
The optimum excess air ratio was given as 1.51-1.60.
Apart from agricultural residues, forestry residues are co-fired in fluidized beds.
Details of the forestry residue and coal co-firing FBC systems and their operating
conditions are summarized in Table 2.9.
To demonstrate the technical feasibility of fluidized bed as a clean technology for the
combustion of low grade fuels and coal/forestry residues Armesto et al. [113]
performed co-combustion experiments in either circulating or bubbling fluidized bed.
A forestry waste, pine chips, high ash content refuse coal and a low grade, high
sulfur content lignite were used in the experiments. Limestone was injected to the
systems for sulfur retention. In circulating fluidized bed, combustion efficiencies
were changing between 90 and 100 %. Refuse coal/biomass blend gave lower
efficiency. As Ca/S ratio was increased, sulfur retention rate increased. At 50 %
biomass share and using refuse coal, increasing the Ca/S ratio from 1.5 to 3,
decreased SO2 emission from 2298 to 898 mg/Nm3 and CO emission from 659 to
380 mg/Nm3. NO emission stayed almost constant at around 300 mg/Nm3.
56
Table 2.9: Operating conditions and system properties of forestry residue
and coal co-firing FBCs.
Reference Fuel Blend Operating conditions System Bed material
Armesto et al. [113]
Pine chips (0-80 %, 0-100 %) & Refuse
coal
Bed temperature: 830-860 °C &
800-850 °C u0: 5-6.2 m/s &
0.6-0.7 m/s
CFBC 0.2 m inner
diameter & 6.5 m height BFBC 1 MWt
1.14 m inner diameter & 4.6 m total height
N.A.
Adanez et al. [114]
Pine bark (0-100 %) &
Lignite/South African coal
Bed temperature: 800-900 °C u0: 4-6 m/s
Excess air: 15-25 % Secondary air: 10-35 %
CFBC 0.3 MWt
0.2 m inner diameter &
6.5 m height
N.A.
Gayan et al. [115],
Pine bark & Sub-
bituminous coal/Spain
Lignite
VTT Bed temperature: 850 °C
Excess air: 18-25 %, Secondary air: 50 %
Biomass share: 50, 60 wt % CIEMAT
Bed temperature: 800-900 °C Excess air: 10-30 %,
Secondary air: 10-35 % Fluidization velocity: 4-6 m/s Biomass share: 0-100 wt %
CFBC VTT: 0.1 MWt
0.2 m inner diameter & 6.5
m height CFBC
CIEMAT: 0.3 MWt
0.17 m inner diameter & 8 m
height
VTT Sand
(100-300 μm)
CIEMAT Sand
(300-500 μm)
Gani & Naruse [116]
Sawdust (50 wt %) & low
grade coal Furnace temperature: 800 ºC
Electrically heated drop-tube furnace
N.A.
Leckner & Karlsson
[117]
Wood chips/ sawdust
(0-100 %) & bituminous
coal
Bed temperature: 850 °C Fluidization velocity: 7.7 m/s
CFBC 12 MW
1.7×1.7 m2 cross section & 13.5 m height
Silica sand & coal ash mixture
Kakaras et al. [118]
Waste wood (0-50 %) &
lignite
Under bed pneumatic feeding Excess air ratio: 1.3-2.1
BFBC 0.095 m inner
diameter & 1.2 m height
Sand (CaCO3)
(500-1000 μm)
Yrjas et al. [119]
Forest residues,
wood chips (0-100 wt %)
& coal
N.A. CFBC 550 MWt
N.A.
Orjala et al. [120]
Bark, sawdust, peat & coal N.A.
CFBC 150 MWt
CFBC 50 kWt
N.A.
N.A.: Not available
57
In bubbling fluidized bed, combustion efficiencies were in the range of 96.9 to 99.5
%. In co-firing refuse coal/biomass blends with 50 % biomass share between Ca/S
ratio of 3.7 and 6.7, no significant change in sulfur retention rate was observed. At 58
% biomass share and using refuse coal, increasing the Ca/S ratio from 3.7 to 6.7
decreased SO2 emissions from 568 to 485 mg/Nm3 and increased CO emission from
199 to 353 mg/Nm3.
Circulating fluidized bed process was found to be a little more efficient than
bubbling fluidized bed process due to higher combustion efficiency and more
efficient utilization of limestone. Increasing the proportion of biomass improved the
efficiency and environmental impact.
Combustion efficiency of pine barks during co-firing with two different kinds of coal
(South African (SA) coal and high sulfur content lignite) was investigated by Adanez
et al. [114] in 0.3 MWt circulating fluidized bed. The effect of operating conditions
such as biomass share (0-100 %), temperature (800-900 °C), excess air (15-25 %),
air velocity (4-6 m/s) and percentage of secondary air (10-35 %) were studied. For
both coals carbon combustion efficiency is increased with increasing biomass share.
Increasing pine bark share from 0 to 100 % increased combustion efficiency of SA-
pine bark and lignite-pine bark blends from 96 to 99.5 % and 98.2 to 99.6 %,
respectively. Higher reactivity of lignite/pine bark blends resulted in higher
combustion efficiencies with respect to South African coal/pine bark blends.
Temperature rise from 800 to 900 ºC resulted in increased combustion efficiencies
for both blends. The solid circulation flow rate increased when gas velocity increased
so that the flow rate of solid losses by the cyclone increased. This acted on the mean
residence time of char particles in bed which decreased the combustion efficiency for
both blends. On the other hand, increasing excess air increased the mean oxygen
concentration in the bed, thus increased the carbon combustion efficiency. Also,
increasing the percentage of secondary air generated a reducing zone in the lower
part of the combustor and therefore reduced the combustion rate.
58
Similar results were obtained in the co-combustion experiments performed with pine
bark and two different coals in 0.1 MWt VTT and 0.3 MWt CIEMAT circulating
fluidized bed combustion pilot plants. Gayan et al. [115] co-fired pine barks with
South African sub-bituminous coal and high sulfur Spanish lignite. In CIEMAT
secondary air was introduced 1.5 m above the distributor plate. Pine bark and coal
were fed to the boiler simultaneously from two hoppers mounted on a balance
through a screw feeder. In VTT pilot plant, secondary air with 50 % share was
introduced at 2 m height. Fuels were fed from separate hoppers and mixed in a screw
feeder. Moisture content of pine barks were decreased from 37 to 11 % during
storage and grinding. The particle size of pine barks were less than 3 mm in VTT
pilot plant, however, in CIEMAT combustor it ranged to 30 mm. The influence of
different operating conditions was investigated in CIEMAT combustor. They were
biomass share, combustor temperature, fluidization velocity, excess air and
secondary air ratio and size distribution of the feed. In VTT pilot plant the influence
of biomass share (0-100 wt %) on combustion efficiency was studied by burning sub-
bituminous coal. In CIEMAT combustor (850 ºC bed temperature, 25 % excess air,
24 % secondary air and 5 m/s gas velocity) increase in biomass share resulted in
increased combustion efficiency of pine bark-lignite and pine bark-sub-bituminous
coal blends from 97.5 to 99 % and from 96 to 99 %, respectively.
In VTT facility (850 ºC bed temperature, 30 % excess air, 40 % secondary air and
2.3 m/s gas velocity), increase in biomass share of pine bark-sub-bituminous coal
blend resulted in increased combustion efficiency from 99.3 to 99.8 %. Increasing
the bed temperature led to increase in combustion efficiency and decreased carbon
concentration in bottom region due higher reaction rates. Also higher efficiencies and
lower char concentrations were obtained for lignite blends due to higher reactivity of
this coal compared to South African coal. Increase in excess air led to higher mean
oxygen concentration in bed which positively affected combustion efficiency. On the
other hand, increase in the percentage of secondary air produced lower oxygen
concentration in bed therefore decreased the combustion efficiency.
59
In general co-firing is expected to lower SO2 and NOx emissions due to lower sulfur
and nitrogen content of the fuel blend; however especially in woody biomass
combustion increasing biomass share results in higher NOx emissions. In the sawdust
and low grade coal co-firing experiments performed by Gani and Naruse [116], NO
and N2O concentrations were found to be higher than coal combustion only even
when the fuel blend had much lower nitrogen content than coal. Such results were
also reported by Leckner and Karlsson [117] who studied NOx emissions from
fluidized bed combustion of high volatile wood chips/sawdust and bituminous coal
mixtures (0-100 wt %) in a 12 MWt CFB boiler. The extreme cases with 100 %
wood or 100 % coal showed that NO emission from wood combustion is higher than
that of coal despite higher nitrogen content of coal relative to wood. A small addition
of coal to wood however, yielded higher emissions of NO than wood combustion
only. Increasing the coal fraction led to higher amount of char in the bed which had
reducing effect on NO emission. This was explained by a much higher release of NO
from coal combustion than wood, caused by high nitrogen content of coal. For
woody biomass, NO rapidly formed in bed section and remained constant along the
combustor, however during coal combustion, formed NO in bed was reduced along
the combustor by reduction reactions of high amount of coal char [121]. When small
fraction of coal was added, reduction in NO was small. Increasing the coal fraction
resulted in higher amount char in the bed which has reducing effect on NO emission.
The NO emissions were measured within the range of 50 to 100 ppm (6 % O2). The
N2O emission mostly originated from the coal. Wood addition had a slight reducing
effect on N2O emission. CO emission was found to increase steadily with increasing
the coal fraction in the blend. SO2 emissions were increased from 20 to 400 ppm
proportional to addition of coal since the sulfur content of wood is negligible.
Agglomeration, slagging and fouling problems occur especially during high
potassium and sodium content woody biomass combustion. Biomass ash aggravates
agglomeration by reacting with silica bed material and forming low melting point
compounds which brings about defluidization. Before performing experiments some
theoretical study can be done to investigate ash deposition tendency. To predict
agglomeration tendency of low ash, high combustible content uncontaminated wood,
60
railway sleepers and demolition wood before co-firing with low quality Ptomais
lignite in a laboratory scale bubbling fluidized bed reactor, Kakaras et al. [118] used
base to acid ratio to estimate fouling and slagging tendency of the fuels. If base to
acid ratio is greater than 0.7, the fuel was considered to have low fouling potential.
Base to acid ratio of lignite, uncontaminated wood and railway sleepers were 1.4, 3.9
and 6.3, respectively. However; the alkaline metals sodium and potassium can form
combinations with low fusibility temperatures altering the base to acid influences.
Biomass addition to the system improved the ignition and due to higher combustible
content and heating value, steady state was achieved faster. Increasing waste wood
share in the fuel blend decreased CO emission in the system. In uncontaminated
wood co-firing experiments, increasing biomass share from 0 to 50 % decreased CO
emission from 1000 to 300 ppm in excess air range of 1.3-1.6. Higher excess air
ratios (1.7-2.1) led to higher CO emissions changing from 1300 to 700 ppm by
increasing uncontaminated wood share in the fuel mixture from 0 to 50 %. In all co-
combustion experiments, SO2 emission was reduced due to low sulfur content of the
waste wood and also greater sulfur retention was observed in 20 % thermal input of
demolition wood in the blend due to high CaO content in its ash. Furthermore, low
ash and sulfur content of waste wood contributed to minimization of ash
agglomerates formation during co-combustion of low quality lignite. Slight decrease
in NOx emission during co-combustion was mainly attributed to lower nitrogen
content of waste wood. N2O emission was also reduced from 200 to 125 ppm when
increasing waste wood share from 0 to 50 % in the fuel blend.
The dominating elements in bottom and fly ash samples were calcium, iron and
manganese in the ash samples. Bottom ashes had higher calcium concentration due to
lower volatility of this element [122] and calcite sand used as inert bed material.
High concentration of iron was attributed to presence of iron in lignite. The most
remarkable difference among various fuel blends was the high alkaline metal
concentration in the mixtures with higher wood waste contribution and mainly the
potassium concentration, when uncontaminated wood was added in 50 %. Waste
wood addition in blend up to 30 % thermal input was found to be feasible. Higher
wood share in fuel blend resulted in agglomeration problems.
61
Chlorine deposition was investigated in the study of Yrjas et al. [119] in 550 MWt
circulating fluidized bed boiler co-firing bark, wood chips and forest residue (0-100
wt %) with coal. The short-term deposits were collected on the surface of air cooled
probes with detachable rings from two locations where the flue gas temperatures
were 730 and 530 ºC. Advanced fuel analysis method was carried out to obtain
reactive sodium, potassium and calcium amounts which could react with sulfur to
form sulfates instead of reacting with chlorine to form alkali chlorides. This method
was based on selective leaching by water, ammonium acetate and hydrochloric acid.
Ash forming elements leached out with water and ammonium acetate represented the
more reactive species, while the rest and the leached out with the acid represented the
less reactive species. It was stated that sulfur released from the fuel blend reacted
with the sodium and potassium to form alkali sulfates and this prevented alkali
chloride formation and thus fouling. Chlorine left the system with flue gas in the
form of HCl instead of depositing. Lower amounts sulfur and higher amounts
reactive calcium compounds of biomass fuels resulted in lower SO2 emissions.
Therefore, increasing the biomass share indirectly affected the chlorine deposition
risk. However, all the sodium, potassium and calcium in the fuel blend were not
reactive towards sulfur. The rates of deposit build-up were below 12 g/m2h, indicated
no severe deposit problems.
Similar to the previous study, Orjala et al. [120] stated that highly reactive alkali
metals and chlorine in biomass ash reacted with sulfur and alumina silicates and
chlorine was mostly released as HCl. Even a small amount of chlorine in biomass
could result in potassium and sodium chloride deposition on boiler heat transfer
surfaces. To avoid such problems, co-firing was given as an effective way of
diminishing the corroding tendency of biomass ash since coal brought within itself
protective elements to the combustor. By co-firing, even low sulfur containing coal
could significantly reduce the corrosion risk in the combustor [123].
62
63
CHAPTER 3
EXPERIMENTAL SET-UP, PROCEDURE
AND CONDITIONS
3.1 General Experiments were carried out in the Middle East Technical University 0.3 MWt
Atmospheric Bubbling Fluidized Bed Combustion (ABFBC) Test Rig. Test rig was
originally constructed and operated for the investigation of combustion and in-situ
desulfurization characteristics of low quality Turkish lignites. Therefore, the existing
test rig was modified for co-firing of biomass and coal. Modified test rig is described
in detail in section 3.2 and pre-experimental modifications and work carried out
before the experiments are given in section 3.3. The modifications were carried out
within the scope of a research project, MAG 104M200, financed by the Scientific
and Technical Research Council of Turkey (TÜBİTAK).
3.2 METU 0.3 MWt ABFBC Test Rig The test rig in its present form is displayed in Figure 3.1 whereas its flow sheet is
given in Figure 3.2. As can be seen from the flow sheet, the test rig basically consists
of the following sub-systems:
• air and flue gas system
• modular combustor
• fuel and limestone feeding systems
• ash removal systems
64
Figure 3.1: METU 0.3 MWt ABFBC Test Rig
65
Figu
re 3
.2: F
low
shee
t of M
ETU
0.3
MW
t AB
FBC
Tes
t Rig
66
• cooling water system
• instrumentation and control system
3.2.1 The Combustor
The main body of the test rig is the modular combustor formed by five modules of
equal dimensions. Modular structure of the combustor is intended to provide
flexibility in burning various fuels by addition or removal of heating surfaces. Each
module has an internal cross-section of 0.45 × 0.45 m2 and 1 m height. Inner walls of
each module are refractory lined with firebricks with a thickness of 6 cm. Outer walls
of the refractory bricks are insulated with insulation bricks with thickness of 20 cm.
Further insulation is provided by leaving an air gap of 6 mm between the outer wall
of insulation brick and the inner wall of the steel construction of each module. The
first and fifth modules from the bottom are referred as bed and cooler, respectively,
and the ones in between are referred as freeboard modules.
The bed module provides an expanded bed height of 1 m. It contains 6 water-cooled
U-tubes (25 mm OD, stainless steel) providing 0.35 m2 of cooling surface, 5 ports for
thermocouples, 4 ports for gas sampling probes, one port for LPG distributor and two
ports for feeding fuel/limestone mixtures. One of the feeding ports is 22 cm and the
other is 85 cm above the distributor plate.
In the freeboard and cooler modules, there are 6 ports for gas sampling probes and 9
ports for thermocouples. There exists a water-cooled tube bundle consisting of 11
tubes (26.7 mm OD, carbon steel) with 14 passes installed across the cross-section of
the cooler module providing 4.3 m2 cooling surface in the cooler module for cooling
the stack gases before leaving the combustor.
67
3.2.2 Air and Flue Gas System
Fluidizing air is supplied by a forced draft (FD) fan. Fluidizing air fed by the FD fan
enters the bottom of the windbox through a pipe of 6.5 m long and 7.8 cm ID on
which a manual gate valve, an automatic butterfly valve and a vortex flow meter are
installed. The design of the wind box allows the installation of bed ash removal
system as shown in Figure 3.1. It is a mobile wind box supported by four wheels and
a distributor plate is placed on the top. Air supplied to the wind box by means of the
pipe of 7.8 cm ID diverges to the full cross-section of the combustor at the distributor
plate located 1.4 m above the entrance port. Sieve type distributor plate contains 412
holes, each 4.5 mm in diameter, arranged in a triangular pattern. Within the bed
module air mixes with fuels and limestone for effective combustion and sulfur
capture.
Flue gases and elutriated fines leaving the bed surface enter the freeboard. Sufficient
freeboard height is provided to permit burnout of elutriated lignite fines and
combustible gases.
After leaving the freeboard, flue gases pass through the cooler module to cool the hot
combustion gases. Flue gases leaving the modular combustor enter the cyclone and
then the baghouse filter to leave the elutriated particles before passing through
induced draft (ID) fan to exit from the stack. As the temperature of the flue gases
entering the baghouse filter is limited by the maximum operating temperature of the
bag material which is 260 °C, two alternative systems are provided for the safe
operation of the baghouse filter: A bypass line between the cyclone and the ID fan
and an air dilution system to reduce the flue gas temperature at the inlet to the bag
filter through a slide valve if the temperature exceeds the upper operating limit of the
bag material.
The pipes carrying the flue gases before and after the baghouse filter are 14.0 and 5.3
m long, respectively, and have an ID of 15.3 cm. There is also a bypass pipeline
68
between the cyclone and the ID fan. It has an ID of 12.8 cm and length of 14.5 m.
The outlet of the baghouse filter joins this pipeline 4.2 m before the ID fan.
An orifice plate with a bore diameter of 8.05 cm was installed at the stack gas line
before ID fan to measure the flow rate of the flue gases. The pressure drop across the
orifice plate is measured by means of pressure transmitter. Knowing the temperature
and pressure of the flue gases passing through the orifice plate, the signal from the
transmitter is utilized in the control system to yield molar flow rate. After the orifice
plate, flue gas passes through an automatic butterfly valve and then enters to ID fan.
Exit of the ID fan is connected to the stack having dimensions of 20 cm ID and 16 m
height.
3.2.3 Solids Handling System
Lignite, biomass and limestone are stored in three separate silos and conveyed into
the hoppers of water-cooled screw feeders at controlled flow rates via pre-calibrated
volumetric feeders placed under their respective silos. The lignite/biomass/limestone
mixture can be continuously fed to the bed through water-cooled screw feeders either
22 cm or 85 cm above the distributor plate. Both screw feeders are operated at
controlled speed in such a way that there is no accumulation of feed material in the
hopper. In order to prevent backflow of combustion gases from the combustor water-
cooled screw feeders have gas tight connections.
Bed ash is withdrawn from the bed through 5 cm diameter, 1.1 m long water-cooled
ash removal pipe. Some of the bed ash is disposed and the rest is stored to provide
bed inventory when required. Bed ash drain rate is adjusted from the DCS to obtain
the desired bed pressure drop and hence the expanded bed height. Bed ash particles
are collected in a continuously weighted ash storage bin.
The majority of the elutriable fines produced from solids in the bed and those fed
within the solid streams are captured by the cyclone, having dimensions of 45 cm
OD and 2.12 m height. Cyclone catch particles pass through an air lock (i.e. a rotary
69
valve) and fall onto a diverter. Depending on the position of the diverter, particles are
either discharged from the system to a continuously weighted ash storage bin for
experiments without recycle or flow back to the combustor for re-firing. During co-
firing experiments, the elutriated particles collected by the cyclone were directly
discharged from the system, i.e., recycle was not carried out.
In order to catch fine particles of fly ash (dp ≤ 40 μm) leaving the cyclone, jet-pulse
type baghouse filter with a 100 % collection efficiency for particles greater than 1
μm was utilized. Technical specifications of the baghouse filter are summarized in
Table 3.1.
Table 3.1: Technical specifications of the baghouse filter.
Bag material P84-Polyimide
Weight of the bag material 500 g/m2 (± 5 %)
Thickness of the bag material 1.6 mm (± 0.2 mm)
Maximum operating temperature of the bag material 260 °C
Collectable particle size ≥ 1 μm
Bag diameter 0.16 m
Bag height 2.0 m
Collection area of one bag 1.0 m2
Total number of bags 20
Air to cloth ratio (max.) 0.9 m/min
Maximum pressure drop through the filter 150 mm H2O
Ash collection hopper angle 70°
Pulse interval 17 s
Pulse cycle 68 s
Pulse air pressure 6 barg
During the operation, a filter cake is built up at the outer surface of the bags which in
turn becomes a principal collection medium. As the filter cake gets thicker with time,
70
a pulse of compressed air is directed into the bag from the open top which causes a
shock wave to travel down its length dislodging the filter cake from the outer surface
of the bag. A unique aspect of the pulse jet system is the use of a wire cage in each
bag to keep it from collapsing during normal filtration. The bag hangs from the tube
sheet. A series of parallel pulse jet pipes are located above the bags with each pipe
row having a solenoid valve. This allows the bags to be pulsed clean one row of five
bags at a time. Filter cake cleaned off the surface fall into a hopper and is discharged
to fly ash collecting container. There are two fly ash containers each having a volume
of 0.13 m3. During filtration of flue gases if one container gets full, the maximum
level device located just above the ash discharge opening gives alarm by lighting the
level warning light located on control panel, and the container full of ash is replaced
with the other one after closing the ash discharge opening by leak proof slide valve.
3.2.4 Cooling Water System
Cooling water required for the test rig is passed through a magnetic conditioner and
is then divided into two streams, one for the in-bed tube bundles, and the other for
the tube bundle in the cooler module. Heat transfer areas provided by the bed and
cooler modules are 0.30 m2 and 4.3 m2, respectively. The cooling water in bed enters
lower header and leaves the bed through the upper header. The cooling water for the
cooler module enters the upper header and flows downward to provide counter-
current flow to the up flowing flue gases. Water flow rates are adjusted by means of
either a manual or a pneumatic control valve located at the drain of each stream to
maintain maximum exit temperature of about 60 °C.
3.2.5 Gas Sampling and Analysis System
3.2.5.1 Gas Sampling Probe
In order to acquire spatially resolved gas composition data from the combustor, the
combustion gas is extracted from the symmetry axis of the combustor by using gas
sampling probes which are fabricated for in-situ extractive gas sampling.
71
The details of the probe construction are shown in Figure 3.3. The probe is water-
cooled to ensure structural strength at the high temperatures encountered in the test
rig. The body of the probe consists of 4 coaxially oriented stainless steel tubes with
the outermost one being the cooling-shroud and the innermost one being the suction
tube. The two remaining tubes are for guiding the flow of cooling water in the probe.
Cooling water enters from the shroud from the rear of the probe, travels to the tip in
between shroud and the adjacent tube, turns back over the 2nd innermost tube, and
discharged. The suction tube is heated by means of a variable DC power supply and
is isolated from the adjacent one with teflon pipe. A ceramic filter was located at the
tip of the probe for in-situ filtering of the particulates. The filter itself is held in place
by a stainless steel perforated plate and a fixing nut. Once through the ceramic filter,
the dust-free combustion gas travels through the suction tube which is maintained at
130 °C by the combined effects of electrical heating and the water cooling. Relative
positions of gas probes on the combustor are given in Table 3.2.
Table 3.2: Relative positions of gas sampling probes.
Probe No Distance above the distributor plate, cm
P10 26
P9 56
P8 69
P7 85
P6 123
P5 183
P4 291
P3 344
P2 419
P1 500
72
Figu
re 3
.3: G
as sa
mpl
ing
prob
e
73
3.2.5.2 Gas Sampling and Conditioning System
Once through the probe, the sampled combustion gas is passed through a solenoid
valve and sent to the gas conditioning and analysis system of the test rig by means of
sample line. The sample line itself is maintained at 130 °C by means of a variable
DC power supply so that no water, sulfuric acid or hydrocarbons would condense
along the sampling interface. In addition, all lines and fittings in contact with the gas
sample are made of stainless steel to prevent interferences due to gas adsorption or
heterogeneous reactions.
Figure 3.4: GASS-II Pre-conditioning System
Gas is sampled at a rate which is small enough to cause minimal interference to the
combustion system. After passing through the probe, sample gas is transported
through the heated stainless steel line to a sample pre-conditioning system (GASS-II)
to remove particulate, mist and water vapor from the gas stream. The picture of the
pre-conditioning system is shown in Figure 3.4. The First step in conditioning is
filtering out particulates and aerosols by passing the sample through 1 µm filter
having borosilicate glass filter element with a fluorocarbon binder. Collected liquid
mists are removed from the system periodically by an automatic drain filter.
74
After filtering, sample passes through ammonia scrubber which avoids deposition of
ammonia salts if ammonia is present in the sample stream. Sample stream then
passes through Nafion membrane dryer. As the sample enters the dryer, flow splits
into a number of small diameter Nafion membrane tubes arranged in a parallel
bundle. The membrane selectively removes water vapor by a process of permeation
distillation. Water vapor travels through the tubing walls driven by the difference in
partial water vapor pressure on the opposing sides of the membrane. As the sample
flows from inlet to outlet of the dryer, water is continuously removed. There is a
countercurrent flow of dry purge gas in to the dryer to provide a medium for water
vapor to be carried away. The schematic of the Perma Pure Nafion membrane tubes
are shown in Figure 3.5.
Figure 3.5: Schematic of Perma Pure Nafion Membrane Dryer
After the drier, sample gas is passed through the cooler and particulate filters for
removal of submicron sized particulates then pumped to the analyzers via a teflon-
coated diaphragm-type sample pump for species concentration measurements. After
the measurement of species concentrations, sample gas is vented to the atmosphere.
3.2.5.3 Analytical System
The on-line continuous gas analyzers with which the test rig is equipped are listed in
Table 3.3. Analyzers except Bailey SMA 90 are used for measuring concentrations
of species O2, CO, CO2, NO, N2O and SO2 along the combustor and also at the
75
cyclone exit on dry basis. Bailey SMA 90 measures temporal variation of O2 and CO
on wet basis at the combustor exit.
There is also an alternative route for sampled gas in case of a fault in the analytical
system. Components of this backup system are given in Table 3.4. Details of the gas
conditioning and analysis system are shown in Figure 3.6.
Table 3.3: Gas analyzers.
Instrument Species Sensor type Range
ABB Advanced Optima Magnos 106 O2 Paramagnetic 0-10/0-25 % by vol.
ABB Advanced Optima Uras 14
CO CO2 NO N2O
NDIR
0-5 % by vol. 0-20 % by vol. 0-1000/0-2000 ppm 0-500/0-1000 ppm
Siemens Ultramat 6 SO2 NDIR 0-1 % by vol.
Bailey SMA 90 O2 CO
Zirconium oxide Catalytic RTD
0-25 % by vol. 0-2 % by vol.
Table 3.4: Backup gas analyzers.
Instrument Species Sensor type Range
Leeds & Northrup O2 Paramagnetic 0-15 % by vol.
Anarad AR 600 CO CO2
IR 0-5 % by vol. 0-20 % by vol.
Servomex 1491 NO/NOx Chemiluminescence 0-0.2 % by vol.
76
Figu
re 3
.6: G
as c
ondi
tioni
ng a
nd a
naly
sis s
yste
m
77
3.2.6 Deposit Sampling System
A deposit sampling probe is designed and constructed for the investigation of ash
fouling during biomass co-firing runs. Deposit sampling probe is placed at the top of
the freeboard for simulating the conditions in the superheater region of a
commercial-scale boiler where flue gas temperature is about 850-880 ºC. The surface
temperature of the probe is measured using K-type thermocouple embedded into the
outer surface of the probe near the detachable ring in order to control surface
temperature of the removable ring. A constant probe surface temperature of 500 °C is
maintained during the experiment by adjusting the flow rate of the cooling air.
A removable stainless steel deposit ring having dimensions of 15 mm OD and 20
mm length is attached to deposit probe for sampling the ash deposits in this region.
Deposit samples are collected on the surface of air-cooled deposition probe during
each biomass co-firing run. Schematic description of deposit sampling probe is given
in Figure 3.7.
3.2.7 Instrumentation and Control System
Figure 3.8 shows the process and instrumentation (P&I) diagram of the test rig. As
can be seen from the figure the test rig is extensively instrumented for research
purposes. Instrumentation and analytical systems can be divided into following
categories:
• Data acquisition and control system
• Solid flow control and monitoring
• Air and gas flow control and monitoring
• Cooling-water flow control and monitoring
• On-line continuous gas analyzers
• Pressure sensors
• Temperature sensors
• Solids analyses
78
Figu
re 3
.7: D
epos
it sa
mpl
ing
prob
e
79
Fi
gure
3.8
: P&
I dia
gram
of M
ETU
0.3
MW
t AB
FBC
Tes
t Rig
80
The test rig is equipped with a data acquisition and control system namely Bailey
INFI 90. Real time process data is monitored, manipulated, collected and analyzed
with the aid of control software called Bailey LAN-90 Process Control View
installed on an IBM compatible PC 486 computer running under QNX operating
system. The control system scans the signals coming from all of the instruments
attached to it in a fraction of a second and reports and logs their averages discretely
for 30 seconds of intervals. An uninterruptible power supply is connected to Bailey
INFI 90 and PC in order to enable proper shut-down in case of a electricity cut-off by
preventing corruption of data logged.
Fuel and sorbent feed rates are controlled manually by adjusting the fuel feeder or
sorbent feeder control dial from the computer. The flow rates of fuel and sorbent are
normally set to such values that provide desired excess air and Ca/S molar ratio,
respectively. Cyclone ash and bed ash are collected in respective bins and their flow
rates are followed by load cells placed under respective bins.
The volumetric flow rate of air is measured by a vortex flow meter and adjusted with
an automatic butterfly valve driven by a computer controlled pneumatic actuator. In
order to achieve conversion from volumetric to molar flow, a static pressure tap and
a temperature sensor is placed downstream of the vortex flow meter. The flow rate of
air is normally set to a value to achieve the desired superficial velocity in the
combustor. In order to achieve almost neutral pressure on the bed surface, the flow
rate of exhaust gases is adjusted with an automatic butterfly valve driven by a
computer controlled pneumatic actuator.
In order to measure flow rates of cooling-water flowing through bed and cooler
bundles, two orifice plates are located up streams of their lower and upper headers,
respectively. The pressure drops across the orifice plates are measured by means of
pressure transmitters. The signals from the transmitters are interpreted in the control
system to yield mass flow rate of the cooling-water flowing through in-bed and
cooler bundles. There exist two pneumatic control valves installed on the
downstream of upper and lower headers of bed and cooler bundles, respectively, to
81
adjust the cooling-water flow in each bundle. The flow rates of cooling-water in bed
and cooler bundles are normally set to a value which provides exit water temperature
in the range 40-60 °C.
Pressure sensors are used for measuring differential and gauge pressures at various
positions on the test rig. Measured differential pressures are the pressure drops over
orifice plates, bed and distributor plate pressure drop, and gauge pressures are the
pressure at the bed surface and pressure of air feed at the downstream of the vortex
flow meter.
Table 3.5: Relative positions of thermocouples.
Thermocouple No Distance above the distributor plate, cm
TC1 25
TC2 44
TC3 73
TC4 73
TC5 97
TC6 133
TC7 154
TC8 226
TC9 257
TC10 285
TC11 330
TC12 361
TC13 425
TC14 500
Spatial and temporal variations of gas temperatures along the height of the
combustor are measured by means of thermocouples of K type (Chromel-Alumel)
82
with grounded junction. The tips of the thermocouples are on the symmetry axis of
the combustor. The axial positions of thermocouples are given in Table 3.5. The
temperature of air feed at the downstream of vortex flow meter and temperatures of
cooling water at the exits of bed and cooler bundles are measured by resistance
thermocouples of type Pt-100.
3.3 Pre-Experimental Modifications 3.3.1 Feeding System
Biomass feeding was problematic with the available feeding system which was
originally designed for coal feeding. The feeding system of the test rig utilizes
volumetric feeders, i.e. screw feeders to control the fuel and sorbent flow rates. As
biomass fuels have much lower bulk density compared to that of coal, their flow
rates are much higher compared to those of coals. Moreover, biomass fuels have
higher particle sizes (up to 25 mm) compared to that of coals (up to 10 mm). To
mitigate the feeding problems, new bed screw feeders were designed and constructed
by taking into account the available data. Details of bed screw feeder design
calculations are given in Appendix I.
As test rig was originally designed for coal firing only, there is no additional silo for
biomass storage. Therefore, it was decided to use the existing ash silo for biomass
storage during the experiments. Upon this solution, existing ash silo and its screw
feeder was tested for suitability to biomass feeding. Three biomasses, i.e. olive
residues, hazelnut shell and cotton residue were filled into the ash silo and maximum
and minimum achievable flow rates were measured. It was observed that existing
screw feeder cannot provide the minimum biomass flow rate required during the
experiments for all biomasses and that during hazelnut shell and cotton residue
feeding tests, bridging in the silo occurred causing the stoppage of biomass flow into
the screw feeder. To alleviate the minimum flow rate problem existing electric motor
with its gear drive was replaced to decrease its rotation rate from 140 rpm to 14 rpm.
To sort out the bridging problem of silo, inner surface of the existing silo was
83
covered by Teflon layer and silo was hanged up by chains and a vibrator was added.
This new arrangement provided easy flow of biomasses, especially for hazelnut shell
and cotton residue.
3.3.2 Air and Flue Gas System
Instrumentation air used in the test rig is supplied from the compressor located in the
Unit Operations Laboratory of Chemical Engineering Department. During the
experimental preparations, it was observed that compressed air contains too much
rust particles, moisture and oil which cannot be eliminated by the filters.
Investigations showed that existing compressor requires extensive maintenance and
repairs. Therefore a new air compressor was purchased, installed and connected to
the test rig. Technical specifications of the compressor are given in Table 3.6.
Table 3.6: Technical specifications of new air compressor.
Brand Lupamat LKD 61-555
Maximum pressure 8 barg
Operating pressure 6 barg
Tank volume 500 L
Free air flow 1454 L/min
Number of cylinders 3
Motor Power 7.5 kW
The stack of the existing test rig was too short (~ 3 m) and leading to pollution
around the Unit Operations Laboratory during the start-up as the baghouse filter can
not be taken into operation till the flue gas temperature reaches a value at which no
condensation takes place inside the flue gas lines. Therefore, existing stack of the test
rig was removed and a taller stack was installed. The new stack has dimensions of 20
cm ID and 16 m height.
84
During the experimental preparations, it was observed that ID fan has too much noise
and vibration leading to vibration of whole flue gas line. Upon inspection and
various measurements, the manufacturer found that vibration is caused by the
unbalanced fan blades. Manufacturer made the balance adjustment of fan blades and
the vibration problem was sorted out.
3.3.3 Gas Sampling and Analysis System
Perma Pure drier used in the gas conditioning system of the existing unit was found
to fulfill its service life. Therefore, a new gas conditioning system, namely Perma
Pure GASS-II Pre Conditioning System, was bought directly from the USA and
incorporated in the existing system and commissioned. The new gas conditioning
system is fully automatic and also has its own dry air generator.
A new gas analyzer, ABB Advanced Optima, was bought for on-line measurement of
O2, CO, CO2, NO, N2O on dry basis and incorporated into the existing gas analysis
system. As there is no place in the existing gas analyzer cabinet, a new gas analyzer
cabinet was also purchased and necessary connections to the existing gas analyzer
system were made. Gas analysis system is arranged in such a way that both old and
new analyzers can be used for gas analysis.
3.3.4 Instrumentation and Control System
During the experimental preparations, it was observed that after a certain time no
value is shown on the computer connected to the DCS. At first, cable connections
were checked. However, investigations revealed that one of the cards of the DCS,
i.e., Bailey Infi 90 INBIM02 Bus Interface Module, had a problem. As the
manufacturer of the DCS system, Bailey Controls, is not available now as it was
acquired by the ABB, a used card was supplied from a plant in Turkey. It was
installed and the problem was alleviated.
85
As there was no N2O measurement in the original gas analysis system of the test rig,
a new 4-20 mA signal connection to DCS was made from the new ABB Advanced
Optima gas analyzer. In addition, DCS software was modified to incorporate N2O
measurement.
3.4 Operating Procedures 3.4.1 Procedures before Cold Start-Up
Each experiment requires extensive preparation work which can be summarized as
follows:
1. Preparation of limestone (crushing and sieving)
2. Filling of coal, limestone and ash silos.
3. Calibration of screw feeders of coal, biomass and limestone.
4. Checking the calibration of orifice meters.
5. Checking and cleaning of the gas sampling probes.
6. Checking the operation of the gas sampling and conditioning system.
7. Checking and cleaning of the thermocouples.
8. Checking and cleaning of the initial heating system.
9. Checking the bags of baghouse filter for leakages.
10. Assembly of combustor and windbox.
11. Checking the test rig for air leakages.
12. Calibration of gas analyzers.
13. Checking the operation of DCS system
3.4.2 Cold Start-Up
In order to be able to heat up the combustor during cold start-up, a perforated pipe
LPG burner is provided which is located 15 cm above the distributor plate. LPG is
provided by three LPG tubes with 45 kg capacity each stored outside the laboratory.
The line delivering the LPG to the burner is equipped with a needle valve to adjust
86
the flow rate of LPG and a solenoid valve to activate or deactivate the burner. In
order to speed up the heating of the bed charcoal is also provided at controlled flow
rate.
The normal procedure for start-up from cold conditions is as follows:
1. Power on the airlock under the cyclone.
2. Establish a fluidization air flow of about 0.4 m/s (corresponding to 12
kmol/h or 1.5 m/s at 850 °C) and an induced draft at a minimum level at
cold conditions.
3. Establish cooling water flows to bed and cooler bundles, gas sampling
probes, screw feeders and bed ash removal pipe.
4. Open bypass line for the stack gases not to pass through baghouse filter.
5. Establish the start-up LPG burner.
6. About 20 minutes after LPG burning, charge the combustor with bed
material from a previous run corresponding to about 20 cm H2O pressure
drop in the bed.
7. Establish charcoal flow to help the heating of the bed. Initially, feed
charcoal intermittently and wait till temperature in the bed reaches to 700
°C.
8. Commence feeding coal at a rate of approximately 50 kg/h, meanwhile
follow the stack gas O2/CO analyzer and bed temperatures to check the
start of coal combustion.
9. If temperatures do not increase or O2 concentration at the stack does not
decrease, stop coal feeding and continue with preheating the bed and
observe an evidence for ignition of coal fed.
10. If coal is ignited, start feeding ash and coal, and stop LPG and charcoal.
11. Close bypass valve and take baghouse filter into operation.
12. Increase air flow rate to have superficial velocity of approximately 2 m/s.
13. When the bed depth reaches its expected (or planned) steady-state value,
as indicated by bed pressure drop, stop ash feeding and commence
removing bed material to maintain constant bed depth.
87
14. Adjust the air and coal feed rates to produce the desired steady-state
operating conditions.
3.4.3 Procedure during Runs
Owing to the adequate instrumentation of the test rig, full data collection is achieved
with little effort during the runs. The following parameters are logged by means of
the control and data acquisition system to the PC from the beginning of the start-up
procedure:
The temporal variations of fluidizing air and stack gas flow rates, their temperatures
and static pressures, superficial air velocity at average bed temperature, pressure drop
through the distributor plate and that through the bed, pressure at bed exit, oxygen
and carbon monoxide concentrations at the exit of the combustor, rate of coal and
limestone feed and bed ash drain, weights of the carryover and bed ash bins, bed and
freeboard cooling-water flow rates and their inlet and exit temperatures, and
temporal variation of gas temperatures through the combustor. Spatial variation of
gaseous species concentrations are started to be logged at steady state which is
reached in about 5 h after the start-up. The steady state is deduced by following the
variation of gas temperatures along the combustor and O2/CO concentrations at the
exit of the combustor with respect to time.
The only manually recorded parameter is the biomass flow rate which is adjusted
through a digital motor frequency controller with the aid of a calibration curve.
3.4.4 Shutdown
By the end of gas sampling from the combustor, dry air is passed through sampling
lines and analyzers for a few hours so that no sulfuric acid or hydrocarbons would
condense. Stopping of the fuel feeding shuts down the operation of test rig. Air flow
rate is decreased not to lose part of bed material by elutriation. Water flow rates of
both bed and freeboard coolers are decreased and the bed material is discharged
88
rapidly and stored in a bin. The carryover bin is removed and an empty bin is loaded
over the load cell. Water and air cooling continues for about 2 hours and this is
followed by switching off the fans and letting the system to cool down for another 24
hours.
3.4.5 Post Shutdown
Once the system is cooled completely the wind box is detached and bed material on
the distributor plate is collected and weighted. Representative samples of coal,
biomass, limestone, bottom ash, cyclone ash and baghouse filter ash taken during
runs are analyzed for size distribution and chemical analyses.
3.5 Experimental Conditions
3.5.1 Lignite, Biomass and Limestone Characteristics
Experiments were carried out with Çan lignite. Biomasses utilized in the experiments
were supplied from the local markets and were transported to the laboratory in bags
of 40 kg. Representative samples from coals and biomasses (Figure 3.9) were then
subjected to sieve analyses and proximate and ultimate analyses. Size distributions of
fuels, limestone and bottom ash samples were obtained by sieve analyses method
whereas size distributions of cyclone and baghouse filter ashes were obtained by
Malvern Mastersizer 2000 particle size analyzer. Proximate analyses were performed
by using LECO TGA-701. Ultimate analyses were carried out with LECO CHNS-
932. Calorific values of the fuels are measured by using AC-500 bomb calorimeter.
The results of these analyses together with the calorific values and bulk densities are
summarized in Tables 3.7 and 3.8, respectively. As can be seen from these tables,
lignite is characterized by high ash content (∼25 %) and high total sulfur contents (∼4
%). Ash constituents of lignites and biomasses are shown in Table 3.9. With regard
to ash composition, lignite ash is mainly composed of acidic oxides whereas olive
residue ash is mainly composed of basic oxides.
89
Olive Residue
Hazelnut Shell
Cotton Residue
Figure 3.9: Photographs of biomasses.
90
Cot
ton
Res
idue
6.93
5.38
75.5
7
12.1
4
364
46.7
9
6.48
4.40
36.2
3
0.32
5.78
0.32
17.4
Haz
elnu
t Sh
ell
7.62
1.46
73.0
4
17.8
9
320
49.7
7
5.86
0.56
42.1
5
0.08
1.58
0.11
17.5
Oliv
e R
esid
ue
6.07
4.24
75.6
9
14.0
0
591
50.2
2
6.38
1.72
37.0
3
0.14
4.51
0.14
18.1
Run
10
Lign
ite
17.4
7
24.2
9
31.4
4
26.8
0
905
40.0
4
3.84
0.98
21.4
2
4.29
29.4
3
4.35
13.4
Run
9
Lign
ite
17.0
5
27.0
6
30.9
7
24.9
3
905
39.8
7
3.89
0.88
18.2
8
4.46
32.6
2
4.46
12.5
Run
8
Lign
ite
17.1
4
27.4
6
30.3
6
25.0
4
905
41.9
2
4.01
0.96
15.6
4
4.33
33.1
4
4.33
12.5
Run
7
Lign
ite
16.0
5
24.3
6
32.1
7
27.4
2
905
42.2
2
4.23
1.01
19.3
5
4.17
29.0
2
4.17
13.9
Run
6
Lign
ite
17.1
9
25.2
9
31.2
2
26.3
0
905
46.4
7
4.26
1.15
13.9
5
3.63
30.5
4
3.63
13.4
Run
5
Lign
ite
16.7
5
23.8
9
32.0
4
27.3
3
905
44.8
3
4.00
1.20
17.6
6
3.61
28.7
0
3.66
14.0
Run
4
Lign
ite
16.6
0
22.6
8
32.5
2
28.2
0
905
43.8
1
3.86
1.16
20.6
4
3.34
27.1
9
3.92
14.6
Run
3
Lign
ite
16.9
8
24.5
6
31.5
8
26.8
8
905
45.4
2
4.37
1.11
15.8
5
3.67
29.5
8
3.67
13.6
Run
2
Lign
ite
16.4
8
26.7
4
31.0
5
25.7
4
905
44.9
3
4.09
1.14
13.9
6
3.86
32.0
2
4.07
13.3
Run
1
Lign
ite
16.3
5
28.7
8
29.7
9
25.1
7
905
44.6
0
3.95
1.09
11.9
7
3.98
34.4
1
4.17
12.3
Tab
le3.
7: F
uel a
naly
ses.
Prox
imat
e A
naly
sis
(As r
ecei
ved
basi
s)
Moi
stur
e, %
Ash
, %
Vol
atile
mat
ter,
%
Fixe
d ca
rbon
, %
ρ Bul
k, k
g/m
3
Ulti
mat
e A
naly
sis
(Dry
bas
is)
C, %
H, %
N, %
O, %
(by
diff
eren
ce)
S C
ombu
stib
le, %
Ash
, %
S To
tal,
%
LH
V, M
J/kg
91
Cot
ton
Res
idue
0.00
0
1.66
8
6.35
1
31.6
47
12.5
23
7.59
6
14.5
61
7.89
6
7.90
3
2.98
6
5.49
5
1.37
4
Haz
elnu
t Sh
ell
0.00
0
9.04
0
64.4
54
13.8
74
5.84
1
5.15
7
0.63
3
0.30
7
0.18
9
0.05
5
0.45
2
Oliv
e R
esid
ue
0.00
0
0.12
8
11.9
66
21.5
10
20.8
40
13.5
99
9.31
4
9.32
7
6.17
7
7.13
9
Run
10
Lign
ite
0.00
0
1.09
1
3.78
8
5.15
5
6.15
2
16.3
08
13.3
40
20.9
97
11.4
08
6.18
5
6.44
1
3.71
6
5.42
0
Run
9
Lign
ite
0.00
0
0.88
0
7.18
6
6.39
9
5.93
5
13.7
84
10.6
86
18.0
22
11.0
99
6.58
0
7.28
0
4.63
5
7.51
5
Run
8
Lign
ite
0.00
0
1.26
7
5.05
9
7.10
1
7.81
0
18.7
50
13.2
00
17.7
78
9.06
9
4.92
9
5.17
3
3.49
5
6.36
7
Run
7
Lign
ite
0.00
0
1.29
3
8.13
8
8.86
2
9.68
5
23.4
05
12.9
80
14.5
30
6.16
3
3.16
6
3.25
8
2.93
1
5.58
8
Run
6
Lign
ite
0.00
0
0.95
4
8.14
5
11.7
64
13.4
20
27.0
03
13.1
24
11.4
75
3.94
9
1.93
1
1.91
8
2.12
0
4.19
8
Run
5
Lign
ite
0.00
0
0.45
8
4.45
0
7.98
2
10.3
80
26.0
56
14.0
99
15.2
52
6.14
2
3.16
7
3.29
8
3.95
0
4.76
6
Run
4
Lign
ite
0.00
0
0.87
9
3.89
7
9.83
7
11.4
44
29.5
48
14.2
33
13.3
21
4.83
9
2.25
9
2.49
8
3.43
1
3.81
5
Run
3
Lign
ite
0.00
0
0.50
4
3.98
6
7.81
6
10.1
67
25.3
01
14.4
38
15.4
62
6.24
8
3.24
2
3.54
1
3.96
0
5.33
4
Run
2
Lign
ite
0.00
0
2.03
3
8.43
4
9.23
6
10.4
13
23.3
78
13.2
65
14.1
60
5.55
3
2.68
7
2.55
5
2.68
1
5.60
6
Run
1
Lign
ite
0.00
0
1.07
8
3.04
2
4.89
6
6.10
3
16.8
59
13.5
65
21.4
78
10.8
51
5.82
3
5.55
3
4.09
4
6.65
9
Tab
le3.
8: F
uel s
ize
dist
ribut
ions
.
SIEV
E O
PEN
ING
, mm
19.0
00
16.0
00
12.7
00
8.00
0
6.30
0
4.75
0
3.35
0
2.00
0
1.00
0
0.50
0
0.35
5
0.18
0
0.10
6
0.00
0
92
Cot
ton
Res
idue
0.00
0.81
4.95
10.8
3
14.7
7
0.00
10.2
9
57.5
1
0.85
Haz
elnu
t Sh
ell
2.28
2.59
7.11
38.8
4
6.60
5.50
7.40
27.8
6
1.81
Oliv
e R
esid
ue
31.1
9
5.29
5.17
17.5
2
2.51
2.64
5.21
27.9
5
2.52
Run
10
Lign
ite
50.1
1
22.5
7
11.4
6
7.79
0.55
4.24
1.51
0.18
1.58
Run
9
Lign
ite
55.5
5
21.3
5
11.7
1
5.42
0.53
2.71
1.05
0.20
1.48
Run
8
Lign
ite
51.9
1
21.8
3
12.1
5
7.92
0.58
2.31
1.60
0.33
1.37
Run
7
Lign
ite
51.4
3
22.8
0
12.2
8
7.26
0.52
2.23
1.59
0.26
1.63
Run
6
Lign
ite
50.0
2
23.8
1
12.0
3
8.19
0.63
2.43
1.05
0.24
1.59
Run
5
Lign
ite
52.7
0
19.1
4
11.2
6
7.78
0.52
5.23
1.62
0.21
1.54
Run
4
Lign
ite
55.2
6
20.4
3
10.4
2
7.50
0.59
2.63
1.34
0.19
1.63
Run
3
Lign
ite
51.3
3
21.8
4
10.5
9
9.06
0.63
3.49
1.24
0.27
1.55
Run
2
Lign
ite
56.5
6
17.4
9
10.9
9
9.21
0.57
2.05
1.45
0.31
1.38
Run
1
Lign
ite
57.2
9
19.6
7
12.0
5
4.85
0.82
2.00
1.58
0.21
1.53
Tab
le3.
9: F
uel a
sh c
ompo
sitio
ns.
As o
xide
s, %
Silic
a, S
iO2
Alu
min
um, A
l 2O3
Ferr
ic, F
e 2O
3
Cal
cium
, CaO
Mag
nesi
um, M
gO
Sulfu
r, SO
3
Sodi
um, N
a 2O
Pota
sssi
um, K
2O
Tita
nium
, TiO
2
93
Chlorine content of biomasses analyzed by using ED-XRF (Spectro Xepos) is given
in Table 3.10. Chlorine content of lignite was obtained by EDX (Energy Dispersive
X-Ray Analysis) method in JSM-6400 Electron Microscope (JEOL).
Table 3.10: Chlorine content of fuels, %.
Olive Residue 0.11
Hazelnut Shell 0.02
Cotton Residue 0.05
Lignite 0.06
The morphologies and crystal sizes of the synthesized samples were observed by
JSM-6400 (JEOL) Scanning Electron Microscope. The micrographs of SEM were
taken in the magnification range of 1500 to 5000 times. During the SEM study, EDX
method was applied in order to get elemental composition of the deposit samples.
EDX analyses of the samples were also performed by JSM-6400 Electron
Microscope (JEOL). Ash deposits were identified qualitatively from their X-ray
diffraction patterns taken by a 100 kV Philips twin tube X-ray diffractometer
(PW/1050) using CuKα radiation.
For determination of trace element concentrations in coal, biomass and limestone a
microwave-assisted acid digestion followed by inductively coupled plasma optical
emission spectrometry and mass spectrometry (ICP-OES and ICP-MS, respectively)
were applied. A combination of nitric, hydrochloric and hydrofluoric acid was
employed in microwave oven followed by boric acid addition for removal of the
hydrofluoric acid from the reaction mixture.
The microwave digestion was carried out by using Anton Paar Multiwave 3000 oven.
ICP-OES and ICP-MS measurements were performed using Perkin Elmer Optima
4300 DV and Perkin Elmer DRC II, respectively. Trace element concentrations of
lignite and biomasses used in Runs 2, 5, 8 and 10 are given in Tables 3.11 and 3.12,
respectively.
94
Table 3.11: Trace element concentrations in lignites in Runs 2, 5, 8 and 10.
Element (mg/kg)
Run 2 Lignite
Run 5 Lignite
Run 8 Lignite
Run 10 Lignite
Detection limit
ICP-MS
As 46.1±0.9 41.6±0.5 52.7±0.3 65.7±0.5 0.01
Ba 89.5±2.7 98.0±1.1 94.8±0.8 108.4±0.9 0.008
Cd 0.108±0.002 0.106±0.004 0.114±0.002 0.185±0.008 0.002
Co 5.03±0.08 5.48±0.08 5.57±0.16 5.93±0.08 0.002
Li 23.9±1.0 23.1±0.7 28.8±0.7 27.0±0.9 0.008
Mo 3.70±0.01 4.02±0.08 4.39±0.08 4.09±0.08 0.002
Pb 16.26±0.54 16.59±0.08 12.29±0.16 21.38±0.16 0.017
Sb 0.48±0.04 0.58±0.02 0.414±0.002 0.49±0.02 0.02
Se <0.18 <0.18 <0.18 <0.18 0.18
Sn 2.41±0.04 4.22±0.12 1.98±0.04 3.60±0.32 0.008
Tl 0.777±0.004 0.832±0.004 0.581±0.004 0.604±0.080 0.003
ICP-OES
Cr 6.6±0.1 13.2±0.1 6.4±0.1 6.6±0.3 0.007
Cu 54.0±0.3 57.1±0.2 49.0±1.0 52.7±1.7 0.01
Mn 110.0±1.0 114.0±0.5 96.0±1.0 106.2±0.3 0.0014
Ni <0.015 5.3±0.3 <0.015 <0.015 0.015
P 75.9±2.1 50.3±1.3 47.1±1.5 58.5±3.0 0.08
V 112.1±0.5 121.9±2.0 114.0±1.0 111.4±1.1 0.009
Zn 39.3±1.3 39.9±0.3 34.7±0.1 34.7±0.7 0.006
95
Table 3.12: Trace element concentrations in biomass.
Element (mg/kg)
Olive residue
Hazelnut shell
Cotton residue
Detection limit
ICP-MS
As 0.72±0.02 0.29±0.02 0.12±0.01 0.01
Ba 14.7±0.1 18.2±0.2 2.27±0.02 0.008
Cd 0.017±0.001 0.040±0.001 0.022±0.001 0.002
Co 0.527±0.005 0.380±0.005 0.472±0.004 0.002
Li 0.776±0.005 0.118±0.005 0.136±0.004 0.008
Mo 0.25±0.01 0.13±0.01 1.61±0.01 0.002
Pb 3.24±0.01 3.53±0.04 1.22±0.02 0.017
Sb 0.144±0.001 0.066±0.001 0.028±0.001 0.02
Se <0.18 <0.18 <0.18 0.18
Sn 0.616±0.006 0.110±0.005 1.455±0.014 0.008
Tl 0.016±0.001 <0.003 <0.003 0.003
ICP-OES
Cr 8.98±0.11 1.82 ±0.03 2.23 ±0.01 0.007
Cu 15.3±0.2 7.9±0.2 11.3±0.1 0.01
Mn 26.2±0.2 106.3±0.4 17.1±0.1 0.0014
Ni 5.1±0.1 1.97±0.02 2.8±0.1 0.015
P 907±10 146±3 7361±97 0.08
V 2.91±0.02 1.05±0.03 5.3±0.4 0.009
Zn 14.8±0.1 22.7±0.2 32.2±0.1 0.006
96
Limestone utilized in the firing tests was supplied by Park Thermic, Electric Industry
and Trade, Inc. and originates from Acıbaşı limestone quarry, 10 km away from the
Çayırhan Thermal Power Plant. Limestone delivered to the laboratory had a particle
size below 6 cm. It was subjected to size reduction by crushing it in a jaw-crusher
and a hammer mill consecutively. Crushed limestone was sieved through a 1.18 mm
sieve and top product was crushed again by hammer mill. Particles under the sieve
were utilized in the experiments. A representative sample from limestone was
subjected to sieve and chemical analyses and the results are summarized in Table
3.13. Trace element concentrations in limestone are given in Table 3.14.
Table 3.13: Characteristics of Beypazarı limestone.
Size Distribution Chemical Analysis (wet)
Size (mm) Weight (%) Component Weight (%)
1.000 – 1.180 13.01 Moisture 0.69
0.850 – 1.000 5.09 CaCO3 88.92
0.710 – 0.850 6.01 MgCO3 6.44
0.600 – 0.710 10.85 SiO2 2.91
0.500 – 0.600 3.95 Na2O 0.15
0.425 – 0.500 10.07 K2O 0.08
0.355 – 0.425 6.45 Al2O3 0.39
0.180 – 0.355 16.82 Fe2O3 0.43
0.106 – 0.180 10.13 LOI 42.43
0.000 – 0.106 17.63 d50: 0.41 mm
97
Table 3.14: Trace element concentrations in Beypazarı limestone.
Element (mg/kg) Limestone Detection limit
ICP-MS
As 10.4±0.4 0.01
Ba 97.7±1.6 0.008
Cd 0.062±0.001 0.002
Co 1.08±0.02 0.002
Li 17.1±0.2 0.008
Mo 0.97±0.01 0.002
Pb 5.20±0.05 0.017
Sb 0.302±0.014 0.02
Se <0.18 0.18
Sn 2.22±0.04 0.008
Tl 0.168±0.009 0.003
ICP-OES
Cr 9.2 ±0.1 0.007
Cu 11.1±0.1 0.01
Mn 47.6±0.1 0.0014
Ni 5.9±0.1 0.015
P 59±1 0.08
V 9.3±0.1 0.009
Zn 8.3±0.3 0.006
98
3.5.2 Operating Conditions
In order to investigate the effect of biomass share on emission performance of the
test rig, a total of 10 runs without/with limestone addition were carried out at several
biomass shares. In Run 1, coal is burned without limestone and biomass addition
whereas in Run 2, coal is burned with limestone addition. In Runs 3 to 5, coal is
burned with limestone addition at several olive residue shares, i.e., 15, 30 and 50 wt
%, respectively. In Runs 6 to 8, coal is burned with limestone addition at several
hazelnut shell shares, i.e., 11, 30 and 42 wt %, respectively. In the last two runs,
Runs 9 and 10, coal is burned with limestone addition at several cotton residue
shares, i.e., 30 and 41 wt %, respectively. Runs were carried out consecutively by the
same team. The total duration of operation was about 40 hours. Durations of the runs
are given in Table 3.15. As can be seen from the table the last run was of the shortest
duration due to feeding problems encountered at the beginning of the run. In all the
runs, the lignite was burned in its own ash due to its high ash content. Table 3.16 lists
the operating conditions of the runs. During the runs, parameters other than biomass
share were tried to be maintained constant. Feed point location was 0.22 m above the
distributor plate for all runs. Runs reported in this thesis study refer to the experiment
coded as 070317.
Table 3.15: Durations of the runs.
Run 1 5 hours 52 minutes
Run 2 6 hours 5 minutes
Run 3 3 hours 58 minutes
Run 4 2 hours 13 minutes
Run 5 2 hours 29 minutes
Run 6 3 hours 28 minutes
Run 7 1 hour 56 minutes
Run 8 1 hour 4 minutes
Run 9 1 hour 31 minutes
Run 10 22 minutes
99
Run
10
35.7
25.2
41
48
12.9
2.7
0.0
18.0
0.0
15.0
15.1
21
2.0
857
843
1.15
Run
9
46.0
19.7
30
37
16.7
2.7
5.5
17.3
0.0
14.1
14.5
10
1.9
860
849
1.15
Run
8
32.4
23.3
42
50
13.9
3.3
2.2
12.1
1.9
14.0
14.3
22
1.9
854
835
1.10
Run
7
41.0
17.2
30
35
14.1
2.8
2.5
14.5
1.4
14.0
14.1
21
1.9
853
832
1.16
Run
6
54.3
7.0 11
14
18.6
3.2
5.5
17.1
1.0
14.2
14.4
20
1.9
857
831
1.22
Run
5
30.2
28.8
49
55
11.2
3.3
1.5
11.0
1.7
14.0
14.3
28
1.9
852
849
1.10
Run
4
40.9
18.8
31
36
13.9
2.9
3.6
12.5
1.3
14.0
14.2
23
1.9
846
832
1.14
Run
3
56.6
10.0
15
19
19.1
3.1
8.0
16.6
0.8
14.0
14.5
18
1.9
860
839
1.18
Run
2
68.7
0.0 0 0 22.4
2.7
8.3
19.4
1.2
14.0
14.6
21
1.9
848
817
1.12
Run
1
76.5
0.0 0 0 0.0 0 6.9
14.2
0.4
16.0
16.8
23
2.2
894
866
1.02
Tab
le 3
.16:
Ope
ratin
g co
nditi
ons o
f Run
s 1-1
0.
Coa
l flo
w ra
te, k
g/h
Bio
mas
s flo
w ra
te, k
g/h
Bio
mas
s sha
re, w
t %
Bio
mas
s sha
re (o
n th
erm
al b
asis
), %
Lim
esto
ne
flow
rate
, kg/
h
Ca/
S m
olar
ratio
(bas
ed o
n to
tal S
)
Bot
tom
ash
flow
rate
, kg/
h
Cyc
lone
ash
flow
rate
, kg/
h
Bag
hous
e fil
ter a
sh fl
ow ra
te, k
g/h
Air
flow
rate
, km
ol/h
Flue
gas
flow
rate
, km
ol/h
Exce
ss a
ir, %
Supe
rfic
ial v
eloc
ity, m
/s
Ave
rage
bed
tem
pera
ture
, °C
Ave
rage
free
boar
d te
mpe
ratu
re, °
C
Bed
hei
ght,
m
100
Run
10
63
13
2
4802
3170
13.0
31.2
27.0
Run
9
65
13
2
3301
3281
12.9
39.4
26.9
Run
8
60
12
2
4073
2423
13.0
34.2
30.7
Run
7
63
12
2
3526
2840
12.9
37.2
27.7
Run
6
66
12
2
3094
2634
13.0
40.4
29.2
Run
5
61
12
2
3165
2691
12.9
39.0
29.6
Run
4
63
12
2
3378
2884
12.9
37.0
27.7
Run
3
65
11
2
3435
2917
12.9
37.0
27.7
Run
2
63
11
2
2842
2767
13.0
40.3
28.1
Run
1
54
11
2
3629
1792
12.9
35.5
41.2
Tab
le 3
.16:
Ope
ratin
g co
nditi
ons o
f Run
s 1-1
0 (c
ont’d
).
ΔPB
ed, c
m H
2O
ΔPD
istri
buto
r pla
te, c
m H
2O
Bed
surf
ace
pres
sure
, cm
H2O
Bed
coo
ling
wat
er fl
ow ra
te, k
g/h
Free
boar
d co
olin
g w
ater
flo
w ra
te, k
g/h
Coo
ling
wat
er i
nlet
tem
p., °
C
Bed
coo
ling
wat
er o
utle
t tem
p., °
C
Free
boar
d co
olin
g w
ater
out
let t
emp.
, °C
101
CHAPTER 4
RESULTS & DISCUSSION
4.1 General
This thesis study is based on experimental data collected from a research project for
the investigation of combustion and gaseous emission characteristics of various
biomasses co-fired with typical low quality Turkish lignite having high ash and
sulfur contents. Combustion tests were carried out on the 0.3 MWt ABFBC test rig
located in the Chemical Engineering Department of Middle East Technical
University.
The effect of biomass type and share on combustion and emission performance was
analyzed with respect to the particle size distributions, ash split and partitioning of
major, minor and trace elements in ash streams, combustion efficiency, temperature
and concentration profiles, emissions and deposit formations on heat exchange
surfaces. Lignite was co-fired with olive residues/hazelnut shells/cotton residues, at
olive residue shares of 15, 31 and 49 wt %, hazelnut shell shares of 11, 30, 42 wt %
and cotton residue shares of 30 and 41 wt % in their own ashes. During the tests
parameters other than biomass share were tried to be maintained constant.
4.2 Particle Size Distributions
Particle size distributions of the inlet and outlet streams for runs burning lignite
without and with limestone addition are shown in Figures 4.1.
102
Figu
re 4
.1: S
ize
dist
ribut
ion
of a
ll so
lid st
ream
s in
Run
s 1 a
nd 2
.
103
As can be seen from the figure, particle size decreases in the following order; lignite,
bottom ash, cyclone ash and baghouse filter ash, as expected. Addition of fine
limestone in Run 2 is seen to decrease the particle size of cyclone ash compared to
cyclone ash of Run 1. This can be attributed to the fine particles of limestone
elutriated with the gas. Particle size distributions of inlet and outlet streams in
biomass co-firing runs are illustrated together with those of lignite firing with
limestone run in Figures 4.2-4.4. In the figures, the latter case has no biomass and
hence can be taken as a reference case. Figure 4.2 displays the particle size
distributions of runs with olive residues and the reference case. Introduction of olive
residues results in coarser cyclone ash particles with respect to lignite combustion
with limestone addition (Run 2) despite finer particle size of olive residues compared
to that of lignite. This is mainly due to the introduction of low bulk density olive
residue (591 kg/m3) compared to that of lignite (905 kg/m3) which leads to elutriation
of less dense coarse particles to cyclone. In addition, decrease in limestone flow rate
with biomass addition may reduce the fines fraction in this region. Increasing the
share of olive residue within the fuel feed from 15 to 49 % has almost no influence
on particle size distributions of the ash streams.
Comparison between particle size distributions of ashes without and with hazelnut
shells is shown in Figure 4.3. As can be seen from the figure the effect is only
noticeable in cyclone ash for the largest share of hazelnut in the fuel blend. This may
be considered to be due to the combined effect of lower bulk density (320 kg/m3)
ands coarser particle size of hazelnut shells compare to those of lignite.
Figure 4.4 shows particle size distributions of inlet and outlet solid streams for co-
firing tests carried out with cotton residues. Trends are similar to those of hazelnut
shells.
104
Figu
re 4
.2: S
ize
dist
ribut
ion
of a
ll so
lid st
ream
s in
Run
s 2, 3
, 4 a
nd 5
.
105
Figu
re 4
.3: S
ize
dist
ribut
ion
of a
ll so
lid st
ream
s in
Run
s 2, 6
, 7 a
nd 8
.
106
Figu
re 4
.4: S
ize
dist
ribut
ion
of a
ll so
lid st
ream
s in
Run
s 2, 9
and
10.
107
4.3 Ash Balance, Split and Discharge Compositions
4.3.1 Ash Balance and Ash Split
Ash balance over the combustor for all runs is tabulated in Table 4.1. The closures
and ash splits between bottom and fly ashes are also presented. As can be seen from
the table, ash recovery rates for Runs 1-10 are consistent. Compared to the ash
recovery rates reported in the literature [124, 125] and considering the difficulty in
closing the total solid mass balances over the fluidized bed combustors, the closure
of the total ash balances is acceptable in all runs. About 70 % of the ash is recovered
from fly ash during lignite combustion runs and addition of biomass results in further
increase in ash split to fly ash. In the runs carried out with olive residues, increasing
olive residue share from 15 to 49 wt % leads to increase in ash split to fly ash from
69 to 89 %. Similar trend is also obtained for the co-firing runs performed with
hazelnut shells. Increasing the hazelnut shell share from 11 to 42 wt % results in
increasing ash split to fly ash from 77 to 87 %. Ash split to fly ash in co-firing
cotton residues with lignite at 30 wt % cotton residue share is obtained as 76 %, and
increasing the share of cotton residue to 42 wt % leads to about 100 % ash split to fly
ash. Enhanced shifting of ash split to fly ash by biomass addition is mainly due to the
lower bulk density of the biomass fuels compared to that of lignite leading to more
elutriation of particles from the combustor.
4.3.2 Ash Compositions
4.3.2.1 Major and Minor Elements
Ash compositions of bottom, cyclone and baghouse filter ashes are shown in Figures
4.5-4.7, respectively. Examination of the compositions of ash streams reveals similar
concentrations of compounds in bottom, cyclone and baghouse filter ashes. As can
be seen from figures, addition of biomass reduces SiO2, Al2O3 and Fe2O3
concentrations in ashes irrespective of biomass type.
108
Run
10
8.
68
1.35
7.
44
17
.47
0.00
18
.00
0.00
18.0
0
103
100
Run
9
12
.45
1.06
9.
60
23
.11
5.53
17
.25
0.00
22.7
8 99
76
Run
8
8.
90
0.34
8.
01
17
.25
2.16
12
.09
1.93
16.1
8 94
87
Run
7
10
.00
0.25
8.
12
18
.37
2.50
14
.47
1.41
18.3
8
100
76
Run
6
13
.74
0.10
10
.71
24.5
5 5.
51
17.1
1 1.
01
23
.63 96
77
Run
5
7.
21
1.22
6.
82
15
.25
1.50
10
.95
1.71
14.1
6 93
89
Run
4
9.
26
0.80
8.
03
18
.09
3.55
12
.45
1.31
17.3
1 96
80
Run
3
13
.91
0.42
11
.03
25.3
7 7.
97
16.6
1 0.
75
25
.33
100
69
Run
2
18
.37
0.00
12
.92
31.2
9 8.
28
19.3
6 1.
20
28
.84 92
71
Run
1
22
.01
0.00
0.
00
22
.01
6.91
14
.15
0.39
21.4
5 98
68
Tab
le 4
.1: A
sh b
alan
ce, c
losu
re a
nd sp
lit.
Inpu
t (kg
/h)
C
oal A
sh
B
iom
ass A
sh
So
rben
t
T
otal
Sol
ids I
n O
utpu
t (kg
/h)
Bot
tom
Ash
Cyc
lone
Ash
B
agho
use
Ash
T
otal
Sol
ids O
ut
Clo
sure
, %
Ash
Spl
it, %
109
Figu
re 4
.5: B
otto
m a
sh a
naly
ses o
f Run
s 1-1
0.
110
Figu
re 4
.6: C
yclo
ne a
sh a
naly
ses o
f Run
s 1-1
0.
111
Figu
re 4
.7: B
agho
use
filte
r ash
ana
lyse
s of R
uns 1
-10.
112
As these compounds are mainly originated from lignite, reduction of lignite share in
the fuel blends results in lower concentrations. On the other hand, CaO and SO3
concentrations are increased from Runs 1 to 10 due to limestone addition and sulfur
capture within the bed and increase in inherent CaO fraction by addition of biomass.
Na2O and K2O concentrations in all the ash streams were obtained to be low and
insensitive to biomass addition. This is indicative of absence of problems associated
with basic oxides of biomass during co-firing.
4.3.2.2 Trace Elements
Trace element analyses are carried out for the run with lignite firing with limestone
addition (Run 2) and olive residue and hazelnut co-firing runs with highest biomass
share in the fuel feed (Runs 5 and 8, respectively) by ICP-OES (Cr, Cu, Mn, Ni, P,
V, Zn) and ICP-MS (As, Ba, Cd, Co, Hg, Li, Mo, Pb, Se, Sb, Sn, Tl) in order to
reflect the effect of biomass addition and biomass type on bottom, cyclone and
baghouse filter ash compositions. Trace element compositions of bottom, cyclone
and baghouse filter ashes are given in Tables 4.2, 4.3 and 4.4, respectively.
Concentration of Se (<0.18 mg/kg) could not be measured as its concentration was
below the detection limits of the instrument. Hg concentration could not be
quantified by ICP techniques because it is volatilized during the sample preparation
procedure [126].
Trace element concentrations can also be described in terms of relative enrichment
factors which describe the behavior of elements in bottom and fly ashes. Relative
enrichment factor (RE) is defined by Meij [127] as;
( )( )
( )Element concentration in ash % Ash content of fuelRE = × (4.1)
Element concentration in fuel 100
113
Table 4.2: Trace element concentrations in bottom ash of Runs 2, 5 and 8. Element (mg/kg) Run 2 Run 5 Run 8 Detection
limit
ICP-MS
As 55.1±0.8 50.6±0.5 43.0±0.1 0.01
Ba 549.7±5. 8 407.3±8.0 389.6±1.7 0.008
Cd 0.782±0.040 0.576±0.004 0.599±0.040 0.002
Co 12.74±0.29 11.51±0.25 12.60±0.13 0.002
Li 71.4±0.9 44.8±0.9 39.1±1.0 0.008
Mo 3.76±0.08 5.00±0.12 4.56±0.13 0.002
Pb 53.71±0.79 41.98±0.29 45.00±0.33 0.017
Sb 3.04±0.04 2.08±0.04 1.80±0.08 0.02
Se <0.18 <0.18 <0.18 0.18
Sn 5.86±0.04 4.98±0.12 3.77±0.12 0.008
Tl 1.22±0.04 0.94±0.04 1.073±0.004 0.003
ICP-OES
Cr 37.5±0.2 38.6±0.2 32.9±0.1 0.007
Cu 17.1±0.2 14.4±0.5 16.5±0.2 0.01
Mn 319.0±3.0 275.0±1.0 275.0±1.0 0.0014
Ni 16.6 ± 0.6 14.7 ± 0.5 13.2 ± 0.5 0.015
P 1139 ± 7 832 ± 5 676 ± 5 0.08
V 76.4±0.6 77.0±1.0 83.5±0.7 0.009
Zn 45.3±0.6 37.3±0.5 35.6±0.9 0.006
114
Table 4.3: Trace element concentrations in cyclone ash of Runs 2, 5 and 8. Element (mg/kg) Run 2 Run 5 Run 8 Detection
limit
ICP-MS
As 95.1±1.0 60.6±1.0 77.5±1.4 0.01
Ba 313.6±6.5 269.7±4.3 230.4±4.1 0.008
Cd 0.199±0.004 0.178±0.004 0.166±0.004 0.002
Co 15.68±0.37 14.04±0.04 14.16±0.12 0.002
Li 69.0±1.0 70.1±1.2 52.6±2.0 0.008
Mo 8.78±0.25 6.55±0.08 5.94±0.21 0.002
Pb 27.55±0.42 19.25±0.17 23.01±0.58 0.017
Sb 1.02±0.04 1.04±0.04 0.82±0.04 0.02
Se <0.18 <0.18 <0.18 0.18
Sn 1.62±0.04 1.30±0.04 1.25±0.04 0.008
Tl 1.333±0.040 1.099±0.004 1.293±0.004 0.003
ICP-OES
Cr 22.0±0.2 24.9±0.5 21.2±0.3 0.007
Cu 31.1±0.1 28.7±0.4 22.1±0.2 0.01
Mn 251.0±1.0 203.9±1.8 180.0±0.9 0.0014
Ni 11.3 ± 0.9 11.3 ± 0.8 8.3 ± 0.2 0.015
P 129 ± 8 563 ± 10 132± 5 0.08
V 212.6±0.6 141.3±2.0 130.0±1.0 0.009
Zn 30.1±0.5 28.9±0.4 26.2±0.6 0.006
115
Table 4.4: Trace element concentrations in filter ash of Runs 2, 5 and 8. Element (mg/kg) Run 2 Run 5 Run 8 Detection
limit
ICP-MS
As 565.8 ± 1.4 369.4 ± 3.6 372.8 ± 1.8 0.01
Ba 329.3±3.2 355.5±4.1 348.9±5.7 0.008
Cd 0.389±0.004 0.422±0.004 0.405±0.004 0.002
Co 26.83±0.54 26.31±0.41 27.68±0.54 0.002
Li 151.1±3.4 189.3±1.9 177.6±4.5 0.008
Mo 26.05 ± 0.21 19.04 ± 0.12 19.91 ± 0.25 0.002
Pb 32.70±0.21 25.94±0.25 21.73±0.46 0.017
Sb 2.647±0.004 2.49±0.08 2.16±0.04 0.02
Se <0.18 <0.18 <0.18 0.18
Sn 2.94±0.08 3.95±0.04 2.72±0.004 0.008
Tl 1.26±0.04 1.62±0.04 1.53±0.04 0.003
ICP-OES
Cr 33.0±0.2 37.6±0.4 35.3±0.1 0.007
Cu 119.9±0.9 127.3±2.0 117.6±0.8 0.01
Mn 395.0±1.0 413.0±3.0 543.0±2.0 0.0014
Ni 24.7 ± 0.2 33.2 ± 0.3 28.4 ± 0.4 0.015
P 128±3 1828 ± 17 746 ± 3 0.08
V 461.0±2.0 426.0±5.0 449.0±1.0 0.009
Zn 77.0±1.1 75.1±0.9 78.0±0.5 0.006
116
Trace elements can be classified according to their volatility behavior as Class I,
Class II and Class III elements [127]. Class I elements are defined as elements that
do not vaporize during combustion. Their enrichment factors are about 1 in bottom
and fly ashes and hence, they are equally distributed between bottom and fly ash.
Class II elements on the other hand, vaporize in the boiler and condense within the
installation. The relative enrichment factor of bottom ash is less than 0.7 because
elements originally present in the vapor phase have no chance to condense on bottom
ash particles. As, Ba, Cd, Co, Cr, Cu, Mn, Mo, Ni, P, Pb, Sb, Sn, Tl, V and Zn
belong to Class II elements. Class III elements such as Hg, Se and Cl are very
volatile and they hardly condense in the boiler. Their enrichments factors are
generally much smaller than 1 [127, 128].
The calculated enrichment factors of trace elements in bottom, cyclone and baghouse
filter ash streams are given in Table 4.5. As can be seen from the table, volatile
elements As, Ba, Co, Cr, Cu, Mn, Mo, Ni, Pb, Sb, Sn, Tl, V and Zn are mostly
enriched in fly ash of Runs 2, 5 and 8. These findings are in accordance with the
previous studies on coal firing [129, 131] and biomass and coal co-firing [132]. The
relative enrichment factors of As, Co, Cu, Mn, Mo, Tl, and V are found to be lower
than 0.7 in all bottom ashes and hence fall into Class II elements which are enriched
in fly ash [127].
4.4 Partitioning of Major, Minor and Trace Elements
Investigation of partitioning behavior of 9 major and minor (Al, Ca, Fe, K, Mg, Na,
S, Si, Ti) and 18 trace elements (As, Ba, Cd, Co, Cr, Cu, Li, Mn, Mo, Ni, P, Pb, Sb,
Se, Sn, Tl, V, Zn) is carried out for the run with lignite firing with limestone addition
(Run 2) and olive residue and hazelnut shell co-firing runs with highest biomass
share in the fuel feed (Runs 5 and 8, respectively) in order to reflect the effect of
biomass addition and biomass type on elements partitioning in bottom, cyclone and
baghouse filter ashes.
117
Filte
r A
sh
2.42
1.11
0.97
1.62
1.57
0.74
2.11
1.08
1.52
6.83
1.67
0.50
1.60
0.45
0.90
1.34
0.52
Cyc
lone
A
sh
0.50
0.73
0.40
0.83
0.94
0.14
0.62
0.36
0.45
2.00
0.30
0.53
0.61
0.21
0.76
0.39
0.18
Run
8
Bot
tom
A
sh
0.28
1.24
1.44
0.74
1.46
0.10
0.46
0.52
0.35
3.17
1.52
1.04
1.34
0.63
0.63
0.25
0.24
Filte
r A
sh
2.88
1.05
1.14
1.45
0.57
0.59
2.62
0.98
1.48
1.08
0.66
0.43
1.14
0.27
0.63
1.13
0.46
Cyc
lone
A
sh
0.47
0.79
0.48
0.78
0.38
0.13
0.97
0.48
0.51
0.37
0.20
0.32
0.48
0.09
0.43
0.37
0.18
Run
5
Bot
tom
A
sh
0.40
1.20
1.56
0.64
0.58
0.07
0.62
0.65
0.39
0.48
0.30
0.70
0.96
0.34
0.37
0.20
0.23
Filte
r A
sh
3.28
0.98
0.96
1.42
1.34
0.59
1.69
0.96
1.88
0.00
0.45
0.54
1.47
0.33
0.43
1.10
0.52
Cyc
lone
A
sh
0.55
0.94
0.49
0.83
0.89
0.15
0.77
0.61
0.64
0.00
0.45
0.45
0.57
0.18
0.46
0.51
0.21
Run
2
Bot
tom
A
sh
0.32
1.64
1.94
0.68
1.52
0.09
0.80
0.78
0.27
0.00
4.01
0.88
1.69
0.65
0.42
0.18
0.31
Tab
le 4
.5: R
elat
ive
enric
hmen
t fac
tors
of t
race
ele
men
ts in
bot
tom
, cyc
lone
and
filte
r ash
es.
As
Ba
Cd
Co
Cr
Cu Li
Mn
Mo
Ni P Pb
Sb
Sn
Tl V
Zn
118
The imbalances are observed in the species mass flow rates. These imbalances could
result from analyzing very small quantity of sample from large quantities of highly
heterogeneous matter, very low concentrations of elements under investigation,
analytical errors and experimental errors in preparing the samples for ICP-OES and
ICP-MS analyses or a combination of all [133].
Figure 4.8 shows the recovery rates of major and minor ash components for Run 2, 5
and 8. As can be seen from the figure, the closure is fairly good for most of the
components. Recovery rates of trace elements are given in Figure 4.9. The
imbalances are observed in the species mass flow rates. These imbalances could
result from analyzing very small quantity of sample from large quantities of highly
heterogeneous matter, very low concentrations of elements under investigation,
analytical errors and experimental errors in preparing the samples for ICP-OES and
ICP-MS analyses or a combination of all [133].
Major and minor ash component partitioning of bottom, cyclone and baghouse filter
ashes produced from Runs 2, 5 and 8 are displayed in Figure 4.10. For each ash
component first, second and third columns refer to Runs 2, 5 and 8, respectively. As
can be depicted from the figure, the partitioning of major and minor ash components
follow the ash split between the bottom ash and fly ash. For the experiments and the
test rig under consideration ash splits to fly ash are found as 71 %, 89 % and 87 %
for Run 2, Run 5 and Run 8, respectively. It can be noted that addition of biomass
shifts the partitioning of major and minor elements from bottom ash to fly ash.
Although, species mass balances over the fluidized bed combustor could not be
closed, some light could be shed to the fate of trace elements in fluidized bed
combustion of high ash content low quality lignites by investigating their partitioning
behavior.
Figure 4.11 illustrates the partitioning of trace elements in bottom, cyclone and filter
ashes for Runs 2, 5 and 8, respectively. For each ash component first, second and
third columns refer to Runs 2, 5 and 8, respectively.
119
Figu
re 4
.8: R
ecov
ery
of m
ajor
and
min
or e
lem
ents
in R
uns 2
, 5 a
nd 8
.
120
Figu
re 4
.9: R
ecov
ery
of tr
ace
elem
ents
in R
uns 2
, 5 a
nd 8
.
121
Figu
re 4
.10:
Maj
or a
nd m
inor
ele
men
ts p
artit
ioni
ng o
f Run
s 2, 5
and
8.
122
Figu
re 4
.11:
Tra
ce e
lem
ents
par
titio
ning
of R
uns 2
, 5 a
nd 8
.
123
The major proportions of As, Cr, Ni, V, Zn were recovered in the cyclone ash for all
runs. This finding complies with the data for coal combustion given in [131, 134]. As
can also be seen from the figure, Ba, Cu, Sn and Mo were mostly recovered in fly
ash. The capture of As in fly ash could be attributed to partial condensation of these
high volatile species due to the low operating temperatures (~350°C) both at the exit
of the combustor due to presence of cooler and in the cyclone [135].
Comparisons between the trace element partitioning of the runs with and without
biomass addition reveal that addition of olive residue and hazelnut shell enhances the
partitioning of As, Ba, Co, Cr, Cu, Li, Mn, Mo, Ni, Pb, Tl, V and Zn to fly ash. Also,
co-firing shifts the partitioning of Cd, P, Sb and Sn from bottom to fly ash. The
reason behind the shift is explained by porous char of biomass [2, 136] leading to
increase in surface area of ash particles over which condensation of vaporized
elements can take place. Li is a non-volatile element; however, it may vaporize with
surrounding alumina silicates and become a part of fly ash. As, Ba, Co, Cr, Cu, Mo,
Ni, Pb, Tl, Sn, V and Zn belong to Class II elements which vaporize and condense in
cooler regions, addition of biomass enhances their partitioning to fly ash.
4.5 Combustion Efficiencies
The effects of operating parameters upon the magnitude of combustibles loss are
investigated by analyzing all solid streams in terms of their carbon contents and CO
emission in the flue gas. The fractional combustibles loss for each run is calculated
as the ratio of the heat loss due to CO emission in the flue gas and unburned
combustibles in bottom, cyclone and baghouse filter ashes to the potential heat of
combustion of coal feed.
Combustion efficiencies calculated from fractional combustibles losses are shown in
Table 4.6. Inspection of the table reveals that combustion efficiencies are very high
(~97 %) for the reactive lignite under consideration despite the absence of cyclone
ash recycle. Combustion efficiency reduces from 97 % (Run 1) to 96 % (Run 2) with
addition of limestone.
124
Run
10
35
.7
3202
11
4310
25.2
41
51
1046
05
2189
15
0.00
1.
87
0 18
.00
3.97
56
00
0.
00
2.41
0
15.1
0 0.
0258
26
3
5863
97
Run
9
46
.0
2989
13
7512
19.7
41
51
8177
5
2192
87
5.53
2.
09
905
17.2
5 4.
30
5814
0.00
2.
26
0 14
.50
0.04
39
431
7150
97
Run
8
32
.4
2985
96
709
23.3
41
85
9751
1
1942
19
2.16
1.
91
323
12.0
9 3.
61
3420
1.93
4.
13
624
14.3
0 0.
0222
21
5
4581
98
Run
7
41
.0
3312
13
5800
17.2
41
85
7198
2
2077
82
5.07
1.
64
653
14.4
7 3.
80
4312
1.41
5.
45
602
14.1
0 0.
0223
21
3
5781
97
Run
6
54
.3
3209
17
4237
7.0
4185
29
295
2035
32
5.51
1.
28
551
17.1
1 3.
50
4699
1.01
4.
79
379
14.4
0 0.
0219
21
3
5843
97
Run
5
30
.2
3343
10
0950
28.8
43
12
1241
86
2251
36
1.50
1.
80
212
10.9
5 3.
50
3001
1.71
9.
34
1251
14.3
0 0.
0516
49
9
4963
98
Run
4
40
.8
3475
14
1797
18.8
43
12
8106
6
2228
62
3.55
1.
67
464
12.4
5 2.
75
2681
1.31
10
.48
1076
14.2
0 0.
0239
23
0
4452
98
Run
3
56
.6
3244
18
3596
10.0
43
12
4312
0
2267
16
7.97
1.
52
946
16.6
1 3.
17
4120
0.75
5.
37
316
14.5
0 0.
0222
21
8
5600
98
Run
2
68
.7
3165
21
7421
0.0
0.0
0.0
2174
21
8.28
1.
62
1049
19.3
6 4.
31
6542
1.20
5.
74
540
14.6
0 0.
0212
20
9
8340
96
Run
1
76
.5
2943
22
5171
0.0
0.0
0.0
2251
71
6.91
1.
31
711
14.1
5 5.
05
5601
0.39
8.
83
270
16.8
0 0.
0347
39
4
6975
97
Tab
le 4
.6: C
ombu
stio
n ef
ficie
ncie
s. In
put:
Coa
l Fee
d R
ate,
kg/
h
Coa
l LH
V, k
cal/k
g
Coa
l Ene
rgy
Inpu
t, kc
al/h
B
iom
ass F
eed
Rat
e, k
g/h
B
iom
ass L
HV
, kca
l/kg
B
iom
ass E
nerg
y In
put,
kcal
/h
T
otal
Ene
r gy
Inpu
t, kc
al/h
Out
put:
Bot
tom
Ash
Dis
char
ge F
low
Rat
e, k
g/h
U
nbur
ned
Car
bon,
%
B
otto
m A
sh E
nerg
y Lo
ss, k
cal/h
C
yclo
ne A
sh D
isch
arge
Flo
w R
ate,
kg/
h
Unb
urne
d C
arbo
n, %
Cyc
lone
Ash
Ene
rgy
Loss
, kca
l/h
Bag
hous
e A
sh D
isch
arge
Flo
w R
ate,
kg/
h
Unb
urne
d C
arbo
n, %
Bag
hous
e A
sh E
nerg
y Lo
ss, k
cal/h
St
ack
Gas
Flo
w R
ate,
km
ol/h
Stac
k ga
s CO
, %
St
ack
Gas
Ene
rgy
Loss
, kca
l/h
T
otal
Ene
r gy
Los
s, kc
al/h
Com
bust
ion
Eff
icie
ncy,
%
125
Reduction in combustion efficiency of about 1 % is in agreement with the findings of
previous study burning similar lignite under similar conditions in the same test rig
[150]. Decrease in efficiency with limestone addition is due to the combined effect of
introduction of a cooler solid which leads to loss of sensible energy from the system
and the net energy loss results from endothermic decomposition reaction of CaCO3
and exothermic formation reaction of CaSO4 at Ca/S molar ratio of 3.
Comparisons of efficiencies of runs with biomass co-firing (Runs 3-10) with that of
base run (Run 2) reveals that combustion efficiency increases with biomass addition.
This increase results from the high volatile matter content of biomass (~75 % on as
received basis) as high volatile matter of biomass rapidly burns and results in highly
porous char accelerating the char combustion as well [2, 136]. Increase in
combustion efficiency with biomass addition is in accordance with previous studies
[108, 114, 115, 118, 155]. During co-firing of olive residue with lignite efficiency
remained constant at 98 % for all runs irrespective of olive residue share.
Insensitivity of combustion efficiency to olive residue share for under-bed feeding
was also reported previously [107].
During co-firing of hazelnut shell, combustion efficiency remains constant at 97 %
up to 30 wt % hazelnut shell in the fuel blend and increases to 98 % with increase of
hazelnut shell share to 42 wt % in the fuel blend. Co-firing of cotton residue with
lignite resulted in 97 % combustion efficiency irrespective of cotton residue share in
the fuel mixture.
126
4.6 Temperature Profiles
Temperature measurements taken during experiments carried out with lignite
combustion with and without limestone addition are displayed in Figure 4.12.
Inspection of the temperature profiles for the experiments with and without limestone
addition shows that temperature decreases considerably in both bed and freeboard
with addition of limestone, as expected. The fall in gas temperature toward the exit is
due to the presence of the cooler in the final module.
In order to see the effect of limestone addition on freeboard temperatures in Run 1
and 2, measured bed and freeboard temperatures are normalized by dividing
freeboard temperatures to the average value of bed temperature of the respective
runs. Figure 4.13 displays normalized temperature profiles of Run 1 and 2. As can be
seen from the figure, addition of limestone shows no significant effect on
temperature difference between bed and freeboard.
The effect of olive residue addition on temperature profiles are illustrated in Figure
4.14. In the figure OR refers to olive residue. Comparisons show that temperatures
slightly increase especially in the freeboard region with increasing olive residue
share in the fuel feed. This is considered to be due to the high volatile content of
olive residue (~76 % on as received basis) and is in accordance with the findings of
previous studies [109, 117]. Normalized temperature profiles of Runs 2, 3, 4 and 5
are shown in Figure 4.15. As can be seen from the figure, increasing olive residue
share leads to reductions in temperature difference between bed and freeboard and
provides more uniform temperature profile across the combustor.
The effect of hazelnut shell addition in the fuel fed on temperature profiles is
illustrated in Figure 4.16. In the figure HS refers to hazelnut shell. As can be seen
from the figure, hazelnut shell addition results in slightly higher temperatures in the
freeboard compared to those of lignite combustion; however, temperatures almost
stay constant along the freeboard irrespective of the hazelnut shell share. Normalized
temperature profiles of Runs 2, 6, 7 and 8 are displayed in Figure 4.17.
127
Figure 4.12: Temperature profiles of Runs 1 and 2.
Figure 4.13: Normalized temperature profiles of Runs 1 and 2.
128
Figure 4.14: Temperature profiles of Runs 2, 3, 4 and 5.
Figure 4.15: Normalized temperature profiles of Runs 2, 3, 4 and 5.
129
Figure 4.16: Temperature profiles of Runs 2, 6, 7 and 8.
Figure 4.17: Normalized temperature profiles of Runs 2, 6, 7 and 8.
130
As can be seen from the figure, addition of hazelnut shell lowers the temperature
difference between bed and freeboard temperatures compared to lignite combustion;
however, increasing hazelnut shell share from 11 to 42 wt % has no significant effect
on the temperature difference between bed and freeboard.
The effect of cotton residue addition on temperature profiles is shown in Figure 4.18.
In the figure CR refers to cotton residue. As can be seen from the figure, cotton
residue addition results in higher freeboard temperatures with respect to lignite
combustion. Freeboard temperatures almost stay constant along the freeboard at
cotton residue shares of 30 and 41 wt %. Normalized temperatures of Runs 2, 9 and
10 are shown in Figure 4.19. Similar to hazelnut shell co-firing runs, increase in
cotton residue share has almost no influence on temperature difference between bed
and freeboard.
Figure 4.18: Temperature profiles of Runs 2, 9 and 10.
131
Figure 4.19: Normalized temperature profiles of Runs 2, 9 and 10.
4.7 Concentration Profiles
4.7.1 O2, CO2 and CO Concentration Profiles
Measured concentrations of O2 and CO2 along the combustor with and without
limestone addition and with increasing biomass shares are illustrated in Figures 4.20
and 4.21. Inspection of the figures shows significant changes in concentrations across
the bed rather than the freeboard. O2 and CO2 concentrations of the first run which
was carried out without limestone addition differs from the ones carried out with
limestone and biomass additions in that higher O2 and lower CO2 concentrations are
measured at the bed exit. These profiles indicate that majority of the combustibles
are burned in bed due to high residence times provided by under-bed feeding and
sufficient bed height.
132
Figu
re 4
.20:
O2 c
once
ntra
tion
prof
iles o
f Run
s 1-6
.
133
Figu
re 4
.21:
CO
2 con
cent
ratio
n pr
ofile
s of R
uns 1
-6.
134
Combustion continues in freeboard rather slowly. Thermogravimetic analysis graphs
of biomasses and lignite given in Appendix B show that devolatilization occurs at
low temperatures in biomass (~350 °C) compared to lignite (~450-500 °C) and
therefore, high amount of volatiles release takes place in bed during co-firing runs.
About 92 and 90 % of combustion takes place across the bed in Run 2 and Runs 3-5,
respectively. Decrease of in-bed combustion by 2 percentage points is considered to
be due to introduction of higher volatile matter with biomass leading to passage and
combustion of volatile matter in freeboard. Increasing the share of olive residue
within the fuel feed from 15 to 49 wt % has almost no effect on O2 and CO2
concentrations. Hazelnut shell addition, in Run 6, also shows no significant influence
on O2 and CO2 concentration profiles.
Figure 4.22 displays CO concentration profiles. CO concentration increases across
the bed and gradually decreases in freeboard. This is due to progressive burning of
CO along the freeboard. Biomass addition to lignite leads to higher CO
concentrations due to introduction of higher volatile matter to the combustor [50,
52]. Increasing biomass share leads to further increase in CO concentrations.
CO concentrations of hazelnut shell co-firing (11 wt % share) are lower than that of
olive residue co-firing (15 wt % share) due to lower volatile matter content of
hazelnut shells compared to that of olive residues (~73 % and ~76 % on as received
basis, respectively).
Discrepancy between CO concentration at the bed exit of Run 4 and that of other
runs is attributed to problems in gas sampling due to bubbles bursting in the splash
zone which influences the measurements and also problems in the gas sampling line
due to high particle concentrations from dense bed which requires frequent purging.
135
Figu
re 4
.22:
CO
con
cent
ratio
n pr
ofile
s of R
uns 1
-6.
136
4.7.2 SO2 Concentration Profiles
SO2 concentration profiles of the Runs 1-6 are shown in Figure 4.23. As can be seen
from the figure, major changes in SO2 concentration takes place across the bed. SO2
concentration is found to be highest in Run 1 where lignite combustion is carried out
without limestone addition due to high sulfur content of lignite under consideration.
Addition of limestone in Run 2 results in lower SO2 concentrations across the bed
compared to Run 1.
SO2 concentrations continue to rise from bed exit to combustor exit. Increase in SO2
concentration along the freeboard is due to progressive release of SO2 by lignite as
well as to insufficient gas residence time for sulfur capture. SO2 profiles of the runs
carried out with biomass addition fall below the SO2 profiles of lignite firing runs.
This is an expected outcome as the sulfur content of biomass under consideration is
an order of magnitude less than that of lignite. This finding is also confirmed by
previous studies [101, 102, 107, 108, 113, 119]. SO2 concentrations of hazelnut shell
co-firing (11 wt % share) is lower than that of olive residue co-firing (15 wt % share)
due to lower sulfur content of hazelnut shells compared to that of olive residues (0.08
wt % and 0.14 wt % on dry basis, respectively).
4.7.3 NO and N2O Concentration Profiles
When a fuel enters the hot fluid bed, volatile nitrogen compounds release and some
nitrogen compounds remains in the solid char. Volatile nitrogen form species such as
NH3 and HCN which are known as the precursors of NO and N2O, respectively
whereas char nitrogen oxidizes directly to NO [137, 138]. As coal has low volatile
content, NO formation is more significant during char combustion. On the other
hand, biomass has high amount of volatiles so that NO and N2O formation is more
significant during volatile combustion rather than char combustion.
137
Figu
re 4
.23:
SO
2 con
cent
ratio
n pr
ofile
s of R
uns 1
-6.
138
Variations in NO concentrations are illustrated in Figure 4.24. As can be seen from
the figure, NO concentrations rise across the bed and remain almost constant along
the freeboard during lignite combustion with and without limestone addition.
Limestone addition in Run 2 results in higher NO concentration at the bed exit. This
is an expected outcome as reactive limestone is known to have catalytic effect on
oxidation of NH3 to NO [105, 106, 139-141].
NO concentrations remain almost constant along the freeboard in co-firing runs.
There are several factors affecting NO concentrations during combustion. Increase in
volatile matter content of the fuels lead to lower NO concentrations [108, 142, 143].
NO concentration can also be reduced by char in the presence of CO [104, 144, 145]
and hence, high CO concentrations along the freeboard may contribute to NO
reduction. Therefore, biomass addition is expected to reduce NO concentrations due
to higher volatile content and higher CO concentrations in the combustor during co-
firing runs. Reductions in NO concentration, on the other hand, are compensated by
introduction of radicals (H and OH) from hydrogen-rich biomass (6.38 wt % and
5.86 wt % on dry basis, for olive residues and hazelnut shells, respectively) during
volatile combustion. These radicals act on HCN to NH3 conversion and lead to
increase in NO concentrations [111, 137]. In addition, higher freeboard temperatures
may enhance NO formation [137, 146]. Heterogeneous catalytic effects of biomass
ash may also be significant in the oxidation of the volatiles. The selectivity in HCN
oxidation towards NO is strongly enhanced by Ca, K and Na in biomass ash [154].
Regarding N2O concentrations displayed in Figure 4.25, it can be noted that fairly
low concentrations prevail along the freeboard even for 100 % lignite combustion.
This is attributed to addition of limestone due to high sulfur content of lignite.
Reduction in N2O concentrations due to limestone addition is also confirmed by
previous studies [105, 106, 139-141]. Increase of olive residue share in fuel feed
decreases N2O concentrations even further. This can be due to the combined effect of
increasing temperature [96, 137, 140, 147] and volatile matter content [148] and also
higher radical concentrations (H and OH) [137, 148] in the fuel feed with increasing
biomass share. Formed N2O can also be destroyed in the presence of these radicals to
form N2 [137, 148].
139
Figu
re 4
.24:
NO
con
cent
ratio
n pr
ofile
s of R
uns 1
-6.
140
Figu
re 4
.25:
N2O
con
cent
ratio
n pr
ofile
s of R
uns 1
-6.
141
In addition to those, calcium, potassium and sodium contents of biomass ash have
catalytic effects on N2O decomposition [148]. Hazelnut shell addition (11 wt %)
leads to lower N2O concentration compared to that of olive residue addition (15v wt
%) due to higher nitrogen content of olive residues (1.72 wt % on dry basis)
compared to that of hazelnut shells (0.56 wt % on dry basis).
4.8 Emissions
Emissions of the Runs 1-10 measured downstream of cyclone are shown in Table
4.7. Inspection of the table reveals that CO, O2 and CO2 emissions of Runs 1 and 2
are similar to each other so that limestone addition has almost no effect on emissions
of these species. Increase in olive residue share from 15 wt % to 49 wt % results in
higher CO emissions compared to lignite firing runs due to high volatile matter
content of olive residues (~76 wt % on as received basis) compared to that of lignite.
O2 and CO2 emissions stay almost constant during olive residue co-firing runs.
Hazelnut shell co-firing shows no significant effect on CO, O2 and CO2 emissions.
Insensitivity of CO emissions to hazelnut shell addition is attributed to its lower
volatile matter content compared to that of olive residues (~73 % and ~76 % on as
received basis, respectively). Cotton residue addition on the other hand, leads to
higher CO emissions compared to those of Runs 1-8. This is attributed to the
combined effect of high volatile matter content of cotton residues (~76 % on as
received basis) and lignite combustion with low amount of limestone addition before
cotton residue co-firing runs due to problems in feeding of the residue. Cotton
residue co-firing shows no significant effect on O2 and CO2 emissions.
CO2 emissions from these tests were also evaluated from the view point of climate
change issue. As can be seen from the CO2 emission values in Table 4.7, CO2
emissions are not affected at all by biomass co-firing. However, as biomass fuels are
CO2 neutral, their contribution to CO2 emission can be considered to be negligible.
Therefore, CO2 emissions can be reduced by the proportion of biomass in fuel feed
for co-firing runs [115, 149].
142
Run
10
5.1
590
558
697
15.3
14
.5
48
6 45
9 13
12
83
285
269
552 60
57
112
Run
9
3.3
903
764
955
17.2
14
.6
92
5 78
3 22
40
76
185
157
321 43
37
72
Run
8
5.1
489
460
575
15.0
15
.0
21
4 20
1 57
6 92
218
205
421 3 3 5
Run
7
5.0
494
465
581
16.0
15
.0
26
9 25
3 72
3 92
228
214
439 6 5 11
Run
6
4.9
503
468
585
16.2
15
.0
29
4 27
3 78
0 92
235
219
449 7 7 13
Run
5
5.1
771
728
910
15.7
14
.8
22
8 21
5 61
6 89
232
219
449 5 4 8
Run
4
5.1
550
519
649
15.6
14
.8
32
1 30
3 86
6 88
253
239
490 11
10
20
Run
3
4.2
532
475
594
16.6
14
.9
44
4 39
7 11
36
88
230
206
422 16
14
28
Run
2
4.8
496
459
574
15.9
14
.8
74
4 68
9 19
70
84
246
228
467 22
20
39
Run
1
5.1
506
477
596
14.8
13
.9
43
46
4097
11
717
8 229
216
443 25
23
45
2 L
imes
tone
Tab
le 4
.7: F
lue
gas e
mis
sion
dat
a.
Com
bust
or O
utle
t
O2,
%
CO
con
tent
, ppm
C
O c
onte
nt1 , p
pm
CO
em
issi
on1 , m
g/N
m3
CO
2 con
tent
, %
CO
2 con
tent
1 , %
SO2 c
onte
nt, p
pm
SO2 c
onte
nt1 , p
pm
SO2 e
mis
sion
1 , mg/
Nm
3 SO
2 ret
entio
n2 , %
NO
con
tent
, ppm
N
O c
onte
nt1 , p
pm
NO
em
issi
on1 , m
g/N
m3
N2O
con
tent
, ppm
N
2O c
onte
nt1 , p
pm
N2O
em
issi
on1 , m
g/N
m3
1 Cor
rect
ed to
6 %
O2
143
SO2 emission is reduced drastically by addition of limestone to lignite in Run 2.
Biomass addition leads to further decrease in SO2 emissions due to negligible sulfur
contents of biomass. Increasing olive residue share from 15 to 49 wt % decreases
SO2 emissions considerably. SO2 emissions from hazelnut shell co-firing runs are
measured to be lower than that of olive residue co-firing runs due to lower sulfur
content of hazelnut shells (0.08 wt %, on dry basis) compared to that of olive
residues (0.14 wt %, on dry basis). Increase in hazelnut share from 11 to 42 wt %
leads to lower SO2 emission due to lower amount sulfur in the fuel feed. SO2
emissions of cotton residues co-firing runs are higher than olive residue and hazelnut
shell co-firing runs due to the same reason explained for high CO emissions.
Sulfur retention efficiencies obtained in all tests are also tabulated in Table 4.7. As
can be seen from the table, as high as 84 % retention efficiency is obtained when
high sulfur content lignite is burned with limestone addition despite the absence of
cyclone ash recycle. This is attributed to increased residence time resulting from
under-bed feeding rather than the reactivity of the limestone utilized as the same
limestone resulted in 69 % retention efficiency when the fuel/limestone mixture was
fed 85 cm above the grid and burned under similar conditions [150].
Co-firing of olive residue results in higher sulfur retention efficiencies compared to
that of lignite (88 % and 84 %, respectively). Increase in olive share in the fuel blend
to 49 wt % leads to further increase in sulfur retention (89 %) due probably to the
additional sulfur capture by the inherent CaO content of biomass [102, 151-153]. Co-
firing of hazelnut shells results in 92 % sulfur retention efficiency irrespective of the
share of hazelnut shell. Increasing the share of cotton residue from 30 to 41 wt %
leads to increase sulfur retention efficiency from 76 to 83 %.
NO emissions stay almost constant during olive residue co-firing runs due to
compensation of high amounts of nitrogen in the fuel blend with coal char and high
volatile matter content. Increasing the hazelnut shell share in the fuel feed leads to
slightly lower NO emissions due to low nitrogen content of hazelnut shell. Co-firing
cotton residue with 30 wt % in the fuel blend leads to lower NO emissions, however,
144
increasing the share to 41 wt % results in higher NO emissions due to increase in
nitrogen content of the fuel blend.
Fuel nitrogen to NO conversion is an important parameter for estimation of NO
emissions. Two different approaches are followed in the literature for determination
of this parameter. The first relates the fuel nitrogen to NO emission to fuel char
combustion [138, 155, 156], the second is just based on NO emission relative to fuel
nitrogen [117]. Both definitions are given below.
2
NO
CFuelCO CO
NFuel
(4.2)CFuel-N to NO conversion, % = 100
N .M × C +CC .M⎛ ⎞ ⎛ ⎞⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎝ ⎠⎝ ⎠
×
where2NO CO CO C ,C and C are the concentrations of NO, CO and CO2 and, CFuel and
NFuel are the carbon and nitrogen contents, respectively. MN and MC are the atomic
masses of nitrogen and carbon, respectively.
(4.3)Nitrogen in the emitted NOFuel-N to NO conversion, % = 100Nitrogen in the fuel ×
Previous studies have shown that fuel nitrogen to NO conversion is a function of
H/N weight ratio [145, 157, 158] as well as the volatile matter content of the fuels
[108, 142, 143] in addition to the operating conditions such as bed temperature,
excess air ratio, etc. In an attempt to see the validity of these correlations for the
results obtained in this thesis study, fuel nitrogen to NO conversions calculated by
using Equations (4.2) and (4.3) together with volatile matter content and H/N ratio
the fuel blends are tabulated in Table 4.8.
145
Table 4.8: Fuel nitrogen to NO conversion.
Volatile matter of fuel feed,
% as received
H/N wt ratio of the fuel feed
Conversion, % (Equation 4.2)
Conversion, % (Equation 4.3)
Run 1 29.8 3.6 7.4 6.8
Run 2 31.1 3.6 7.1 6.8
Run 3 38.2 3.9 5.9 6.0
Run 4 45.9 3.5 5.5 5.8
Run 5 53.4 3.6 4.9 5.3
Run 6 35.8 4.1 7.2 7.4
Run 7 44.4 5.5 7.8 8.9
Run 8 48.3 6.2 8.2 10.1
Run 9 44.3 2.3 2.4 2.9
Run 10 49.5 2.0 3.3 4.0
As can be seen from the table, conversion for olive residue co-firing runs shows no
dependency on H/N ratio but only on the volatile matter content of the fuel blend. On
the other hand, hazelnut shell co-firing runs reveal that conversion increase with H/N
ratio despite increasing volatile matter content of the fuel blend. Cotton residue co-
firing results show no dependency on both parameters. It is worth nothing that
among the three biomasses, cotton residue yields the smallest fuel nitrogen to NO
conversion. These results reveal that predictive accuracy of fuel nitrogen to NO
conversion relationship with volatile matter content and H/N ratio of the fuels is very
poor for systems co-firing fuels with different characteristics.
Inspection of N2O emissions given in Table 4.7 shows that N2O emissions follow a
reducing trend in lignite firing and olive residue co-firing runs. Addition of hazelnut
shells leads to further reduction in N2O emissions; however, cotton residue addition
results in higher emissions due to higher nitrogen content of cotton residue (4.4 wt %
on dry basis).
Although the emissions from a pilot scale test rig are not indicative of the emissions
from commercial size units due to much shorter gas residence times, an attempt was
146
made to find out emission performance of the tests carried out in this study. For co-
firing units emission limits have to be re-defined for different fuel blends. The
emission limits for fuel blends under consideration is calculated by the expression
given in European Directive [159] as;
3
33
Thermal input of biomass (kJ/kg)
× Emission limit of biomass (mg/Nm )
Thermal input of coal (kJ/kg)+
× Emission limit of coal (mg/Nm )Emission limit (mg/Nm ) =
Thermal input
⎧ ⎫⎛ ⎞⎪ ⎪⎜ ⎟⎪ ⎝ ⎠⎪⎨ ⎬
⎛ ⎞⎪ ⎪⎜ ⎟⎪ ⎪⎝ ⎠⎩ ⎭ (4.4)
of biomass (kJ/kg)+Thermal input of coal (kJ/kg)⎧ ⎫⎨ ⎬⎩ ⎭
Gaseous emission limits set by Turkish and European Directives for co-firing units
under consideration are presented in Table 4.9.
Table 4.9: Emission limits set by Turkish and European Union Directives.
Turkish Directive[160] European Union Directive [159]
CO mg/Nm3
SO2 mg/Nm3
NOx mg/Nm3
SO2 mg/Nm3
NOx mg/Nm3
Run 1 200 2000 800 850 400
Run 2 200 2000 800 850 400
Run 3 249 1658 724 726 400
Run 4 294 1347 655 614 400
Run 5 343 1008 579 492 400
Run 6 237 1741 743 757 400
Run 7 290 1376 661 625 400
Run 8 331 1096 599 524 400
Run 9 297 1329 651 608 400
Run 10 324 1140 609 540 400
147
Comparisons between the emission limits given in Table 4.9 and experimental
emissions listed in Table 4.7 shows that CO emissions are above the limits due to
both lack of air staging and short gas residence times in the test rig. SO2 emissions
are found to be lower than the limits set by Turkish Directive except cotton residue
co-firing runs (Runs 9 and 10) due to prolonged lignite firing with insufficient
limestone addition caused by biomass feeding problems encountered before Runs 9
and 10. NO emissions are found to be lower than Turkish emission limits; however,
SO2 and NO emissions are higher than the limits set by European Union Directive.
4.9 Agglomeration and Deposit Formation
Combustion processes with biomass, especially with herbaceous biomasses which
have high alkaline contents are prone to experience agglomeration, fouling and
corrosion [161]. Ash constituents from combustion processes react with flue gas or
with each other to form variety of compounds having low ash melting temperature.
Co-firing biomass with coal on the other hand, reduces risk of operational problem
by introducing the protective compounds such as alumina-silicates in coal which
increases the melting point of the alkaline compounds in the biomass ash [9-11, 42,
45, 47].
Agglomeration tendency is mostly related with melting behavior of the components
in bed material. A bed agglomeration index which expresses the ratio of iron oxides
to sum of potassium and sodium oxides in the fuel ash has been developed, relating
ash composition to agglomerations in fluidized beds [9]. Agglomeration is predicted
to occur when the ratio is less than 0.15. The agglomeration index calculated for fuel
blends in Runs 1-10 are given in Table 4.10. As can be seen from the table, all blends
have low agglomeration tendency.
148
Table 4.10: Bed agglomeration index of fuel blends.
Bed agglomeration index: 2 3
2 2
%(Fe O )BAI =
%(K O+Na O)
Run 1 6.8
Run 2 6.2
Run 3 4.3
Run 4 2.5
Run 5 1.6
Run 6 7.8
Run 7 4.6
Run 8 3.8
Run 9 1.7
Run 10 1.0
Evaluation of agglomeration tendency is also carried out with extracting the melting
behavior data of bottom ash components from ternary phase diagrams. Table 4.11
gives the melting temperatures of bottom ash components of co-firing runs carried
out with highest biomass share (Runs 5, 8 and 10) obtained from different ternary
phase diagrams [162]. As can be seen from the table, melting temperatures of the bed
particles are very high to form agglomerates under fluidized bed combustion
conditions and therefore, during co-firing runs carried out with olive residues,
hazelnut shells and cotton residues, no agglomeration is detected within the bed
material.
149
Table 4.11: Melting temperatures of ternary systems.
Melting Temperature, °C
Ternary System Run 5 (49 wt %
olive residue)
Run 8 (42 wt %
hazelnut shell)
Run 10 (41 wt %
cotton residue)
SiO2-Al2O3-CaO ~1400 Gehlenite
~1440 Gehlenite
~1750 Ca2SiO4
SiO2-Al2O3-K2O ~1700 Mullite
~1700 Mullite
~1700 Mullite
SiO2-Al2O3-Na2O ~1700 Mullite
~1700 Mullite
~1700 Mullite
SiO2-Al2O3-MgO ~1680 Mullite
~1700 Mullite
~1600 Mullite
SiO2-CaO-Fe2O3 ~1470
Pseudo- wollastonite
~1600 Ca2SiO4
~1900 Ca2SiO4
SiO2-CaO-MgO ~1450
Pseudo- wollastonite
~1500 Pseudo-
wollastonite
~1470 Pseudo-
wollastonite
SiO2-CaO-K2O ~1475
Pseudo-wollastonite
~1600 2CaO.SiO2
~1600 2CaO.SiO2
Gehlenite: 2CaO.Al2O3.SiO2,
Mullite: 3Al2O3.2 SiO2,
Pseudo-wollastonite: CaO.SiO2
150
Deposit formation on the heat exchange tubes is another important problem
challenging biomass combustion performance. The feasibility of biomass and coal
co-firing partly depends on its impact on ash deposition since deposits interfere with
operation and eventually lead to corrosion or blockage of gas paths [163].
For predicting the fouling tendency of coals the ratio of sum of potassium and
sodium contents to silica content of the fuel is often used as an index. The ratio
higher than 1, usually indicate fouling problems [11]. Calculated values of base to
acid ratios for fuel blends are given in Table 4.12. As can be seen be seen from the
table, addition of biomass results in no significant increase in the ratio to form
fouling problems.
Table 4.12: Alkali index of fuel blends.
2 2
2
%(K O+Na O)Alkali Index =
%(SiO )
Run 1 0.03
Run 2 0.03
Run 3 0.05
Run 4 0.08
Run 5 0.13
Run 6 0.03
Run 7 0.05
Run 8 0.06
Run 9 0.13
Run 10 0.24
151
As fuel composition and ash melting behaviors are generally insufficient for
prediction of deposit formation and composition [42], an air cooled deposit testing
probe with a detachable ring is designed and constructed for investigation of deposit
formation during co-firing biomass with coal. Deposit probe was placed within port 2
at the combustor height of 4.19 m. The probe was exposed to particle laden flue gas
for about 8 h during olive residue co-firing, 4.5 h during hazelnut shell co-firing and
about 1 h during cotton residue co-firing. Those periods were considered sufficient as
deposit testing probes exposed for 1 to 3 h in full scale boilers can provide reliable
data for prediction of deposit composition [42, 164]. The surface temperature of the
probe is set to 500 °C to represent the surface temperature of the superheater material
at the hottest part of superheater of fluidized bed boilers [164].
Appearance of the deposit rings collected after the co-firing experiments with olive
residue, hazelnut shell and cotton residue are shown in Figures 4.26-4.28,
respectively. As can be seen from the figures no deposit formation took place in the
wind side which implies low risk for fouling for the biomasses under consideration.
Olive residue co-firing runs leads to higher deposit formation compared to hazelnut
shell and cotton residue co-firing runs. Relatively high fouling rate of olive residue
co-firing runs is also confirmed by the calculated values of rate of deposit build-up
(RBU). RBU is defined as the amount of deposit collected per projected surface area
of the probe per unit time (g/m2 h) [24, 165]. RBU values of co-firing runs carried
out with different biomasses are illustrated in Figure 4.29. As can be depicted from
the figure, deposits of olive residue co-firing runs have the higher RBU value than
those of hazelnut shell and cotton residue co-firing runs. Coal is considered to be
non-fouling fuel, therefore, this result may probably due to the higher share of the
olive residue (49 wt %) in the fuel blend compared to shares of hazelnut shell (42 wt
%) and cotton residue (41 wt %) in the fuel blends. RBU value exceeding 20 g/m2h is
usually taken a sign of slagging and fouling problems in the measured area of the
boiler [165, 166]. As all the RBU values are lower than 20 g/m2h, fuel blends can be
considered as non-fouling.
152
Figure 4.26: Appearance of deposit rings after olive residue/lignite co-firing runs.
153
Figure 4.27: Appearance of deposit rings after hazelnut shell/lignite co-firing runs.
154
Figure 4.28: Appearance of deposit rings after cotton residue/lignite
co-firing runs.
155
RB
U (g
/m2 h
)
0
2
4
6
8
10
12
14
Olive residue/lignite Hazelnut shell/lignite Cotton residue/lignite
Figure 4.29: Rate of deposit build-up (RBU, g/m2h)
After each set of the co-firing run, removable ring was separated from the deposit
probe and deposits were scraped off. The deposit powders are coated with carbon
tape and subjected to SEM/EDX analysis. The morphology and size of deposit
crystals observed by SEM at different magnifications are shown in Figures 4.30-
4.38.
As can be seen in the Figures 4.30-4.32, deposits crystals of olive residue co-firing
runs are observed to have irregular shape in the size range of 1 to 3 µm. In Figures
4.33-4.35, deposits from hazelnut shell co-firing runs are observed to have similar
morphology to that of deposits of olive residue co-firing runs. Crystals having
diameter <1 µm and crystals having diameter at around 2-3 µm are observed.
Deposits from cotton residue co-firing runs shown in Figures 4.36-4.38 are observed
to have highly angular structure. The deposit grains are within the size range of 1-5
µm. No agglomerated or molten particles are observed in the deposits.
156
Figure 4.30 SEM micrograph of deposit of olive residue/lignite co-firing runs (a).
Figure 4.31: SEM micrograph of deposit of olive residue/lignite co-firing runs (b).
157
Figure 4.32 SEM micrograph of deposit of olive residue/lignite co-firing runs (c).
Figure 4.33: SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (a).
158
Figure 4.34: SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (b).
Figure 4.35: SEM micrograph of deposit of hazelnut shell/lignite co-firing runs (c).
159
Figure 4.36: SEM micrograph of deposit of cotton residue/lignite co-firing runs (a).
Figure 4.37: SEM micrograph of deposit of cotton residue/lignite co-firing runs (b).
160
Figure 4.38: SEM micrograph of deposit of cotton residue/lignite co-firing runs (c).
The compositions of the deposits analyzed by EDX are shown in Figure 4.39.
Inspection of the figure reveals high concentrations of silicon, calcium, sulfur, iron,
and aluminum in the deposits. Silica and alumina are known as protective coal ash
elements which increase the melting temperatures of the alkalis from biomass ashes
[9-11, 42, 45, 47]. High concentrations of calcium and sulfur suggest formation of
calcium sulfates in the deposits. Potassium concentration in the deposits of olive
residue co-firing runs is found to be higher than potassium content in the other
deposits. This may be attributed to the larger exposure time in olive residue co-firing
test. On the other hand, chlorine concentrations are almost zero in all the deposits
suggesting no formation of alkali chlorides. Therefore, potassium in the deposits
forms potassium sulfates instead of potassium chloride which is a highly fouling
compound leading to corrosion of the superheater tubes of boilers. These findings are
also confirmed with XRD analysis. XRD analysis graph of the deposits of olive
residue and hazelnut shell co-firing are shown in Figure 4.40.
161
Figu
re 4
.39:
Dep
osit
com
posi
tions
.
162
Figu
re 4
.40
: X-r
ay d
iffra
ctio
n pa
ttern
s of t
he d
epos
its.
163
As can be seen from the figure, the deposits are highly crystalline and mainly
composed of calcium sulfate, as estimated from EDX results. Also, in the deposits of
olive residue co-firing runs, potassium sulfate is observed as a minor phase with
calcium sulfate major phase.
As chlorine is not detected in both EDX and XRD analyses of the deposits, the fuel
blends under consideration can be denoted to have low-fouling and corrosion
propensity. Species balance on chlorine carried out for highest share of olive residue
and hazelnut shell co-firing runs (Runs 5 and 8) reveals that 92 % and 86 % of
chlorine is recovered in ash in Runs 5 and 8, respectively. The partitioning of
chlorine in bottom and fly ashes reveal that 100 % and 83 % of chlorine is captured
in fly ashes of Runs 5 and 8, respectively.
164
CHAPTER 5
CONCLUSIONS
5.1 General
In this study, combustion and emission performance of typical Turkish lignite co-
fired with olive residue, hazelnut shell and cotton residue at several shares were
investigated by burning them in their own ashes in 0.3 MWt ABFBC test rig. The
following conclusions were reached under the observations of this study:
• Compared to lignite firing, co-firing of lignite with olive residue, hazelnut
shell and cotton residue results in formation of coarser particles in cyclone
ash.
• SiO2, Al2O3, Fe2O3 concentrations decrease and CaO and SO3 concentrations
increase in bottom, cyclone and baghouse filter ashes during co-firing of
olive residue, hazelnut shell and cotton residue.
• Co-firing shifts major and minor elements partitioning from bottom ash to fly
ash and enhances the partitioning of trace elements to fly ash.
• Co-firing of olive residue and cotton residue increases the combustion
efficiency to 98 % and 97 %, respectively irrespective of biomass share in the
fuel feed. During hazelnut shell co-firing, combustion efficiency is increased
to 98 % with increasing hazelnut shell share to 42 wt %.
• Co-firing of biomasses with lignite leads to slightly higher freeboard
temperatures.
• Co-firing of olive residue results in almost constant O2, CO2 and NO
concentrations, but higher CO and lower N2O and SO2 concentrations across
the freeboard with increasing olive residue share from 15 wt % to 49 wt %.
165
• Co-firing of hazelnut shell results in almost constant O2, CO2 and CO, and
lower SO2 and N2O concentrations across the freeboard at 11 wt % hazelnut
shell share.
• Co-firing of olive residue, hazelnut shell and cotton residue shows no
significant influence on total CO2 emissions, however reduces net CO2
emissions.
• Co-firing of biomasses with lignite leads to higher CO emissions.
• During olive residue co-firing tests, NO emissions stay almost constant while
SO2 and N2O emissions reduce with increasing olive residue share. Hazelnut
shell co-firing results in lower NO, SO2 and N2O emissions with increasing
hazelnut shell share. Cotton residue co-firing, on the other hand, leads to
higher NO and N2O emissions with increasing cotton residue share from 30
wt % to 40 wt %.
• No agglomeration of bed material and fouling problems on heat exchange
surfaces occur during co-firing of lignite with olive residue, hazelnut shell
and cotton residue.
In conclusion, co-firing of high ash and sulfur content, low calorific value Turkish
lignite with olive residue, hazelnut shell and cotton residue is found to be technically
feasible in fluidized bed combustors in terms of combustion and emission
performance, agglomeration and fouling. However, considering the seasonal
availability of these biomasses co-firing application of olive residue, hazelnut shell
and cotton residue with lignite will be more feasible in FBC power plants which are
installed close to the production areas of these residues for reducing pollutant
emissions.
5.2 Suggestions for Future Work
Based on the experience gained in the present study, the following recommendations
for future extension of the work are suggested.
• NO and CO emissions could be lowered by means of staged combustion.
166
• Co-firing experiments could be carried out with recycle of cyclone ash to
provide longer residence time for better utilization of limestone.
• A parametric study with respect to Ca/S ratio and temperature could be
carried out on co-firing of lignite with olive residue, hazelnut shell and cotton
residue for determination of optimum operating conditions.
167
REFERENCES
[1] BP Statistical Review of World Energy June 2007,
http://www.bp.com/statisticalreview, (Last accessed date: 26.08.07). [2] Sami M., Annamalai K., Wooldridge M., “Co-firing of Coal and Biomass
Fuel Blends”, Progress in Energy and Combustion Science, Vol. 27, 171-214, 2001.
[3] Sheinkopf K., “Global Biomass Opportunities”, Independent Energy (USA),
Vol. 23, No. 8, 50-53, 1993. [4] Grassi G., “The Biomass Research and Development Program of the
Commission of the European Communities”, Klass D. L. (Editor), Energy from Biomass and Wastes, Institute of Gas Technology, Chicago, IL, USA, Vol. 16, 15-37, 1993.
[5] International Energy Agency Statistics 2004,
http://www.iea.org/Textbase/stats/index.asp, (Last accessed date: 26.08.07). [6] Advance Techniques for Generation of Energy from Biomass and Waste,
http://www.ecn.nl/fileadmin/ecn/units/bio/Overig/pdf/Biomassa_voordelen.pdf, (Last accessed date: 26.08.07).
[7] International Energy Outlook 2007,
http://www.eia.doe.gov/oiaf/ieo/graphic_data_ electricity.html., (Last accessed date: 26.08.07).
[8] Baxter L., Koppejan J., “Biomass-Coal Co-combustion: Opportunity for
Affordable Renewable Energy”, Euro Heat & Power, Vol. 1, 2004. [9] Bapat D.W, Kulkarni S.V., Bhandarkar V. P., “Design and Operating
Experience on Fluidized Bed Boiler Burning Biomass Fuels with High Alkaline Ash”, in Proceedings of 14th International Conference on Fluidized Bed Combustion (Ed. Preto F.D.S.), Vol. 1, 165-174, ASME, Vancouver, Canada, May 11-14, 1997.
[10] Hughes E.E., Tillman D.A., “Biomass Co-firing: Status and Prospects
1996”, Fuel Processing Technology, Vol. 54, 127-142, 1998. [11] Guanyi C., Mengxiang F., Zhongyang L., Xuantian L., Zhenglun S., Kefa
C., Mingjiang N., Experimental Research on Rice Husk Combustion in
168
CFB Boiler and The Design of a 35 T/H Rice Husk Fired CFB Boiler, in Proceedings of 14th International Conference on Fluidized Bed Combustion (Ed. Preto F.D.S.), Vol. 1, 175-181, ASME, Vancouver, Canada, May 11-14, 1997.
[12] Anthony E.J., “Fluidized Bed Combustion of Alternative Solid Fuels:
Status, Success and Problems of the Technology, Progress in Energy and Combustion Science, Vol. 21, 239–268, 1995.
[13] “International Energy Agency Co-firing Database”, http://www.ieabcc.nl/.
(Last accessed date: 26.08.07).
[14] Statistical Yearbook of Turkey 2006, http://www.turkstat.gov.tr/yillik/yillik_ing.pdf, (Last accessed date: 26.08.07).
[15] http://www.fiskobirlik.org.tr/, (Last accessed date: 26.08.07). [16] Exploitation of Agricultural Residues in Turkey, http://www.agrowaste-
tr.org/ablife/index.php?lang=en/, (Last accessed date: 26.08.07).
[17] Easterly J.L., Burnham M., “Overview of Biomass and Waste Fuel Resources for Power Production”, Biomass and Bioenergy, Vol. 10, 79-92, 1996.
[18] Werther J., Saenger M., Hartge E.U., Ogada T., Siagi Z., “Combustion of
Agricultural Residues”, Progress in Energy and Combustion Science, Vol. 26, 1-27, 2000.
[19] Arvelakis S., Vourliotis P., Kakaras E., Koukios E.G., “Effect of Leaching
on the Ash Behavior of Wheat Straw and Olive Residue During Fluidized Bed Combustion”, Biomass and Bioenergy, Vol. 20, 459-470, 2001.
[20] Tomishige K., Asadullah M., Kunimori K., “Syngas Production by Biomass
Gasification Using Rh/CeO2/SiO2 Catalysts and Fluidized Bed Reactor”, Catalysis Today, Vol. 89, 389-403, 2004.
[21] Kuprianov V. I., Janvijitsakul K., Permchart W., “Co-firing of Sugarcane
Bagasse with Rice Husk in a Conical Fluidized Bed Combustor”, Fuel, Vol. 85, 434-442, 2006.
[22] Özbay N., Pütün A.E., Uzun B. B., Pütün E., “Biocrude from Biomass:
Pyrolysis of Cottonseed Cake, Renewable Energy, Vol. 24, 615-625, 2001. [23] Midilli, A. , Murat Dogru M., Howarth C.R., Ayhan T., “Hydrogen
Production from Hazelnut Shell by Applying Air-blown Downdraft Gasification Technique”, International Journal of Hydrogen Energy, Vol. 26, 29-37, 2001.
169
[24] Theis M., Skrifvars B. J., Hupa M., Tran H., “Fouling Tendency of Ash
Resulting from Burning Mixtures of Bio fuels. Part 1: Deposition Rates”, Fuel, Vol. 85, 1125-1130, 2006.
[25] Senneca O., “Kinetics of Pyrolysis Combustion and Gasification of Three
Biomass Fuels”, Fuel Processing Technology, Vol. 88, No. 1, 87-97, 2007. [26] Kouvo P., Backman R., “Estimation of Trace Element Release and
Accumulation in the Sand Bed during Fluidized Bed Co-combustion of Biomass, Peat and Refused-derived Fuels”, Fuel, Vol. 82, 741-753, 2003.
[27] Scala F., Chirone R., “Fluidized Bed Combustion of Alternative Fuels”,
Experimental Thermal and Fluid Science, Vol. 28, No. 7, 691-699, 2004. [28] Selcuk N., Gogebakan Y., Altindag H., “Co-firing of Steam Coal with High
Sulfur Content Lignite in a Bubbling Fluidized Bed Combustor”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-067 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[29] Altindag H., Gogebakan Y., Selcuk N., “Sulfur Capture for Fluidized-Bed
Combustion of High-Sulfur Content Lignites”, Applied Energy, Vol. 79, 403-424, 2004.
[30] Boavida D., Abelha P., Gulyurtlu I., Cabrita I., “Co-Combustion of Coal
and Non-Recyclable Paper and Plastic Waste in a Fluidized Bed Reactor”, Fuel, Vol. 82, 1931-1938, 2003.
[31] Miccio F., Scala F., Chirone R., “Fluidized Bed Combustion of a Biomass
Fuel: Comparison between Pilot Scale Experiments and Model Simulations”, Journal of Heat Transfer, Vol. 127, 117-122, 2005.
[32] Albina D. O., “Emissions from Multiple-Spouted and Spouted Fluidized
Beds Using Rice Husks as Fuel”, Renewable Energy, Vol. 31, 2152-2163, 2006.
[33] Nielsen H. P., Baxter L. L., Sclippab G., Morey C., Frandsen F. J.,
“Deposition of Potassium Salts on Heat Transfer Surfaces in Straw Fired Boilers: A Pilot Scale Study”, Fuel, Vol. 79, 131-139, 2000.
[34] Bakker R. R., Jenkins B. M., Williams R. B., “Fluidized Bed Combustion
of Leached Rice Straw”, Energy & Fuels, Vol. 16, 356-365, 2002. [35] Miles T. R., Miles T. R. Jr., Baxter L. L., Bryers B. M., Jenkins B. M.,
Oden L. L., “Alkali Deposits Found in Biomass Power Plants: A Preliminary Investigation of Their Extent and Nature”, National Renewable Energy Laboratory, Golden, CO, 1995.
170
[36] Demirbas A., Ilten N., “Fuel Analysis and Thermochemical Processing of
Olive Residues”, Energy Sources, Vol. 26, 731-738 2004. [37] Arvelakis S., Gehrmann H., Beckman M., Koukios E. G., “Studying Ash
Behavior of Agricultural Residues Using Thermal Analysis”, Journal of Thermal Analysis and Calorimetry, Vol. 72, 1019-1030, 2003.
[38] Robinson A. L. , Junker H. , Baxter L. L., “Pilot-scale Investigation of the
Influence of Coal-Biomass Co-firing on Ash Deposition”, Energy & Fuels, Vol. 16, No. 2, 343-355, 2002.
[39] Hulkkonen S., Fabritius M., Enestam S., “Application of BFB Technology
for Biomass Fuels: Technical Discussion and Experiences from Recent Projects”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-132 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[40] Visser H. J.M., Van Lith S.C., Kiel J.H.A., “Biomass Ash–Bed Material
Interactions Leading to Agglomeration in FBC”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-150 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[41] Baxter L.L., Miles T.R., Miles Jr. T.R., Jenkins B. M., Milne T., Dayton D.,
Bryers R.W., Oden L.L., “The Behavior of Inorganic Material in Biomass-Fired Power Boilers: Field and Laboratory Experiences”, Fuel Processing Technology, Vol. 54, 47-78, 1998.
[42] Aho M., “Reduction of Chlorine Deposition in FB Boilers with Aluminum
Containing Additives”, Fuel, Vol. 80, 1943-1951, 2001.
[43] Nielsen H. P., Frandsen F.J., Dam-Johensen K., “Lab-Scale Investigations of High Temperature Corrosion Phenomena in Straw-Fired Boilers”, Energy & Fuels, Vol. 13, 1114-1121, 1999.
[44] Coda B., Aho M., Berger R., Hein K. R. G, “Behavior of Chlorine and
Enrichment of Risky Elements in Bubbling Fluidized Bed Combustion of Biomass and Waste Assisted by Additives”, Energy & Fuels, Vol. 15, 680-690, 2001.
[45] Lisa K., Lu Y., Salmenoja K., “Sulfation of Potassium Chloride at
Combustion Conditions”, Energy & Fuels, Vol. 13, 1184-1190, 1999. [46] Salmenoja K., “Field and Laboratory Studies an Chlorine Induced
Superheater Corrosion in Boilers Fired with Biofuels”, Ph.D. Thesis, Faculty of Chemical Engineering, Åbo Akademi University, Turku, 102, 2000.
171
[47] Hansen L. A., Frandsen F. J., Dam-Johansen K., Sorensen H. S., Skrifvars
B. J., “Characterization of Ashes and Deposits from High-Temperature Coal-Straw Co-firing”, Energy and Fuels, Vol. 13, 803-816, 1999.
[48] Biomass Co-firing: An Efficient Way to Reduce Greenhouse Gas
Emissions,http://ec.europa.eu/energy/res/sectors/doc/bioenergy/cofiring_eu_bionet.pdf, (Accession date: 25/08/07).
[49] Preto F., Antony E.J., Desai D.L., Friedrich F.D., “Combustion Trials of
Rice Hulls in a Pilot Fluidized Bed”, in Proceedings of 9th International Conference on Fluidized Bed Combustion (Ed. Mustonen J.P.), Vol 2, 1123-1128, ASME, Boston, Massachusetts, USA, May 3-7, 1987.
[50] Armesto L., Bahillo A., Veijonen K., Cabanillas A., Otero J., “Combustion
Behavior of Rice Husk in a Bubbling Fluidized Bed”, Biomass and Bioenergy, Vol. 23, 171-179, 2002.
[51] Fang M., Yang L., Chen G., Shi Z., Luo Z., Cen K., “Experimental Study
on Rice Husk Combustion in a Circulating Fluidized Bed”, Fuel Processing and Technology, Vol. 85, 1273-1282, 2004.
[52] Permchart W., Kouprianov V.I., “Emission Performance and Combustion
Efficiency of a Conical Fluidized-Bed Combustor Firing Various Biomass Fuels”, Biosource Technology, Vol. 92, 83-91, 2004.
[53] Kuprianov V. I., Janvijitsakul K., Permchart W., “Co-firing Sugar Cane
Bagasse with Rice Husk in a Conical Fluidized-Bed Combustor”, Fuel, Vol. 85, 434-442, 2006.
[54] Shimizu T., Nemoto T., Tsuboi H., “Rice Husk Combustion in a BFBC
Using Porous Bed Material”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-28 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[55] Natarajan E., Öhman M., Gabra M., Nordin A., Lilidahl T., Rao A.N.,
“Experimental Determination of Bed Agglomeration Tendencies of Some Common Agricultural Residues in Fluidized Bed Combustion and Gasification”, Biomass and Bioenergy, Vol. 15, 163-169, 1998.
[56] Liu H., Lin Z., Liu D., Wu W., Combustion Characteristics of Rice Husk in
Fluidized Beds, in Proceedings of 13th International Conference on Fluidized Bed Combustion (Ed. Heinschel K.J.), Vol. 1, 615-618, ASME, Orlando, Florida, USA, May 7-10, 1995.
[57] Skrifvars B. J., Yrjas P., Kinni J., Siefen P., Hupa M., “The Fouling of Rice
Husk Ash in Fluidized-Bed Combustion. 2. Pilot-Scale and Full-Scale Measurements”, Energy & Fuels, Vol. 19, 1512-1519, 2005.
172
[58] Bakker R.R., Jenkins B. M., Williams R. B., “Fluidized Bed Combustion of
Leached Rice Straw”, Energy & Fuels, Vol. 16, 356-365, 2002.
[59] Tillman D.A., Rossi A.J., Kitto.W.D., “Wood Combustion: Principles”, Processes and Economics, New York: Academic Press , 1981.
[60] Standard Test Method for Fusibility of Coal Coke Ash, Designation: D
1857-87, 1994.
[61] Wall T.F., Creelman R.A., Gupta R.P., Gupta S., Colin C., Lowe A., “Coal Ash Fusion Temperature- New Characterization Technique, and Associates with Phase Equilibria, In Application of Advanced Technology to Ash-Related Problems in Boilers”, Ed. Baxter L., and Desollar R., Waterville Valley, 541-556, 1995.
[62] Coin C.D.A., Kahraman H., Reifenstein A.P., “An Improve Ash Fusion
Test”, in Proceeding of Eng. Found. Ash Conference, Waterville Valley, 1995.
[63] Halder P.K., Basu P., “The Temperature of Burning Carbon Particles in a
Fast Fluidized Bed of Fine Particles”, Chemical Engineering Communication, Vol. 104, 245-255, 1991.
[64] Gerald P.H., Huggins F.E., Dunmyre G.R., “Investigation of High
Temperature Behavior of Coal Ash in Reducing and Oxidizing Atmosphere”, Fuel, Vol. 60, 585-597, 1981.
[65] Hansen L. A., Frandsen F. J., Dam-Johansen K., Sorensen H. S., Skrifvars
B. J., “Characterization of Ashes and Deposits from High-Temperature Coal-Straw Co-firing”, Energy and Fuels, Vol. 13, 803-816, 1999.
[66] Biomass Co-firing: An Efficient Way to Reduce Greenhouse Gas
Emissions,http://ec.europa.eu/energy/res/sectors/doc/bioenergy/cofiring_eu_bionet.pdf. (Last accessed date: 26.08.07).
[67] Lin W., Johansen K.D., Frandsen F., “Agglomeration in Bio-fuel Fired
Fluidized Bed Combustors”, Chemical Engineering Journal, Vol. 96, 171-185, 2003.
[68] Lin W., Krusholm G., Dam-Johansen K., Musahl E., Bank L.,
Agglomeration Phenomena in Fluidized Bed Combustion of Straw, in Proceedings of 14th International Conference on Fluidized Bed Combustion (Ed. Preto F.D.S.), Vol. 2, 831-837, ASME, Vancouver, Canada, May 11-14, 1997.
[69] Skrifvars B. J., Öhman M., Nordin A., Hupa M., “Predicting Bed
Agglomeration Tendencies for Biomass Fuels Fired in FBC Boilers: A
173
Comparison of Three Different Prediction Methods”, Energy & Fuels, Vol. 13, 259-363, 1999.
[70] Öhman M., Nordin A., Skrifvars B. J., Backman R., Hupa M., “Bed
Agglomeration Characteristics during Fluidized Bed Combustion of Biomass Fuels”, Energy & Fuels, Vol. 14, 169-178, 2000.
[71] Öhman M., Nordin A., “The role of Kaolin in Prevention of Bed
Agglomeration during Fluidized Bed Combustion of Biomass Fuels”, Energy & Fuels, Vol. 14, 618-624, 2000.
[72] Brus E., Öhman M., Nordin A., Boström D., “Bed Agglomeration
Characteristics of Biomass Fuels Usinf Blast-Furnace Slag as Bed Material”, Energy & Fuels, Vol. 18, 1187-1193, 2004.
[73] Nielsen H. P., Frandsen F. J., Dam-Johansen K., “Lab-Scale Investigations
of High-Temperature Corrosion Phenomena in Straw-Fired Boilers”, Energy & Fuels, Vol. 13, 1114-1121, 1999.
[74] Senior C.L., Srinivasachar S., “Viscosity of Ash Particles in Combustion
Systems for Prediction of Particle Sticking”, Energy & Fuels, Vol. 9, 277-283, 1995.
[75] Steenari B. M., Lindqvist O., “High-Temperature Reactions of Straw Ash
and the Anti-Sintering Additives Kaolin and Dolomite”, Biomass & Bioenergy, Vol. 14, 67-76, 1998.
[76] Uberoi M., Punjak W. A., Shadman F., “The Kinetics and Mechanism of
Alkali Removal from Flue Gases by Solid Sorbents”, Progress in Energy and Combustion Science, Vol. 16, No. 4, 205-211, 1990.
[77] Koukios E. G., Arvelakis S., NTUA Group, FIBEGAR Final Report,
European Commission, JOULE Program (JOR3-CT95-0021), 1998.
[78] Khraisha Y.H., Hamdan M.A., Qalalweh H.S., “Direct Combustion of Olive Cake Using Fluidized Bed Combustor”, Energy Sources, Vol. 21, 319-327, 1999.
[79] Topal H., Atımtay A.T., Durmaz A., “Olive Cake Combustion in a
Circulating Fluidized Bed”, Fuel, Vol. 82, 1049-1056, 2003. [80] Cammarota A., Chirone R., Scala F., “Bed Agglomeration during Fluidized
Bed Combustion of Olive Husk”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-098 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
174
[81] Scala F., Chirone R., “Characterization and Early Detection of Bed Agglomeration during the Fluidized Bed Combustion of Olive Husk”, Energy & Fuels, Vol. 20, 120-132, 2006.
[82] Rasul M.G., Rudolph V., “Fluidized Bed Combustion of Australian
Bagasse”, Fuel, Vol. 79, 123-130, 2000. [83] Kuprianov V.I., Permchart W., Janvijitsakul K., “Fluidized Bed
Combustion of Pre-Dried Thai Bagasse”, Fuel Processing Technology, Vol. 86, 849-860, 2005.
[84] Skrifvars B.J., Backman R., Hupa M., Sfiris G., Åbyhammar T., Lyngfelt
A., “Ash Behavior in a CFB Boiler during Combustion of Coal, Peat or Wood”, Fuel, Vol. 77, 65-70, 1998.
[85] Lyngfelt A., Lecknere B., “Combustion of Wood-Chips in Circulating
Fluidized Bed Boilers- NO and CO Emissions as Functions of Temperature and Air-Staging, Fuel, Vol. 78, 1068-1072, 1999.
[86] Preto F., “Combustion of Wood Processing Residues in a Circulating
Fluidized Bed”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-171 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[87] Miccio F., Scala F., Chirone R., “FB Combustion of a Biomass Fuel:
Comparison between Pilot Scale Experiments and Model Simulations”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-033 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[88] Chirone R., Miccio F., Scala F., “Mechanism and Prediction of Bed
Agglomeration during Fluidized Bed Combustion of a Biomass Fuel: Effect of the Reactor Size”, Chemical Engineering Journal, Vol. 123, 71-80, 2006.
[89] Tranvik A. C., Öhman M., Sanati M., “Bed Material Deposition in
Cyclones of Wood Fuel Fired Circulating Fluidized Beds (CFBCs)”, Energy & Fuels, Vol. 21, 104-109, 2007.
[90] Wòjtowicz M. A., Pels J. R., Moulijn J. A., “Combustion of Coal as a
Source of N2O Emission”, Fuel Processing Technology, Vol. 34, 1-71, 1993.
[91] Leckner B., Karlsson M., “Gaseous Emissions from Circulating Fluidized
Bed Combustion of Wood”, Biomass & Bioenergy, Vol. 4, 379-389, 1993.
175
[92] Lyngfelt A., Åmand L. E., Leckner B., “Reversed Air Staging - A Method for Reduction of N2O Emissions from Fluidized Bed Combustion of Coal”, Fuel, Vol. 77, 953-959, 1998.
[93] Environment Canada 1989, Reference Method for Source Testing:
Measurement of Releases of Semi-Volatile Organic Compounds from Stationary Sources, EPS1/RM/2, 1989.
[94] Fiorentino M., Marzocchella A., Salatino P., “Segregation of Fuel Particles
and Volatile Matter during Devolatilization in a Fluidized Bed Reactor - II. Experimental”, Chemical Engineering Science, Vol. 52, 1909-1922, 1997.
[95] Llorente M. J. F., Cuadrado R. E., Laplaza J. M. M., García J. E. C.,
“Combustion in a Bubbling Fluidized Bed with Bed Material of Limestone to Reduce Biomass Ash Agglomeration and Sintering”, Fuel, Vol. 85, 2081-2092, 2006.
[96] Shen B.X., Mi T., Liu D.C., Feng B., Yao Q., Winter F., “N2O Emission
under Fluidized Bed Combustion Condition”, Fuel Processing Technology, Vol. 84, 13-21, 2003.
[97] Lin W., Johnsson J.E., Jensen A.D., Dam-Johansen K., Combustion of
Biomass in Fluidized Beds- Problems and Some Solutions Based on Danish Experiences, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-124 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[98] Lin W., Johansen K.D., “Agglomeration in Fluidized Bed Combustion of
Biomass-Mechanisms and Co-Firing with Coal”, in Proceedings of 15th International Conference on Fluidized Bed Combustion (Ed. Reuther R.B.), FBC99-0120 in CD-ROM, ASME, Savannah, Georgia, May 16-19, 1999.
[99] Sandelin K., Coda B., Backman R., Berger R., Hein K.R.G., “Melting of
Ash Components When Co-Firing Coal, Straw and Paper Sludge under Bubbling Fluidized Bed Conditions”, in Proceedings of 16th International Conference on Fluidized Bed Combustion (Ed. Geiling, D. W.), FBC01-129 in CD-ROM, ASME, Reno, Nevada, USA, May 13-16, 2001.
[100] Glazer M.P., Schürmann H., Monkhouse P., Jong W., Spliethoff H., “Co-
Combustion of Coal with High Alkali Straw, Measuring of Gaseous Alkali Metals and Sulfur Emissions Monitoring”, in Proceedings of 8th International Conference on Circulating Fluidized Beds”, (Ed. Cen, K.), in CD-ROM, Hangzhou, China, May 10-13, 2005.
[101] Kassman H., Anderson C., Högberg J., Ǻmand L. E., Davidsson K., “Gas
Phase Alkali Chlorides and Deposits during Co-Combustion of Coal and Biomass”, in Proceedings of 19th International Conference on Fluidized
176
Bed Combustion (Ed. Winter F.), FBC2006-99.F in CD-ROM, Vienna, Austria, May 21-24, 2006.
[102] Okasha F., “Enhancing Sulphur Self-Retention by Building-in CaO in
Straw-Bitumen Pellets”, Fuel Processing Technology, Vol. 88, 401-408, 2007.
[103] Hansen P.F.B., Lin W., Dam-Johansen K., “Chemical Reaction Conditions
in a Danish 80 MWth CFB-Boiler Co-Firing Straw and Coal”, in Proceedings of 14th International Conference on Fluidized Bed Combustion (Ed. Preto F.D.S.), Vol. 2, 831-837, ASME, Vancouver, Canada, May 11-14, 1997.
[104] Henderson P., Kassman H., Andersson C., “The Use of Ammonium
Sulphate in Waste-Fired Boilers to Reduce Superheater Fouling and Corrosion and NOx emissions, Proc Swedish Finnish Flame Days, Borås, Sweden, 2005.
[105] Johnsson J. E., “Formation and Reduction of Nitrogen Oxides in Fluidized-
Bed Combustion”, Fuel, Vol. 9, 1398-1415, 1994. [106] Leckner B., Amand L. E., “Emissions from a Circulating and Stationary
Fluidized Boiler: A Comparison”, in Proceedings of 9th International Conference on Fluidized Bed Combustion (Ed. Mustonen J. P.), 891-897, ASME, Boston, Massachusetts, USA, May 3-7, 1987.
[107] Armesto L., Veijonen K., Bahillo A. Cabanillas A., Plumed A., Salvador
L., “Co-Combustion of Coal and Biomass Wastes in Fluidized Bed”, in Proceedings of 16th International Conference on Fluidized Bed Combustion (Ed. Geiling D.W.), FBC01-0044 in CD-ROM, ASME, Reno, Nevada, USA, May 13-16, 2001.
[108] Armesto L., Bahillo A., Cabanillas A., Veijonen K., Otero J., Plumed A.,
Salvador L., “Co-Combustion of Coal and Olive Oil Industry Residues in Fluidised Bed”, Fuel, Vol. 82, 993-1000, 2003.
[109] Cliffe K.R., Patumsawad S., “Co-Combustion of Waste from Olive Oil
Production with Coal in a Fluidised Bed”, Waste Management, Vol. 21, 49-53, 2001.
[110] Atimtay A.T., Topal H., “Co-Combustion of Olive Cake with Lignite Coal
in a Circulating Fluidized Bed”, Fuel, Vol. 83, 859-867, 2004. [111] Hamalainen J.P., Aho M.J., “Formation of Nitrogen Oxides from Fuel-N
through HCN and NH3: A Model-Compound Study”, Fuel, Vol. 73, 1894-1898, 1994.
177
[112] Vitolo S., Petarca L., Bresci B., “Treatment of Olive Oil Industry Wastes”, Biosource Technology, Vol. 67, 129-137, 1999.
[113] Armesto L., Cabanillas A., Bahillo A., Segovia J.J, Escalada R., Martinez
J.M., Carrasco J.E., “Coal and Biomass Co-Combustion on Fluidized Bed: Comparison of Circulating and Bubbling Fluidized Bed Technologies”, in Proceedings of 14th International Conference on Fluidized Bed Combustion (Ed. Preto F.D.S.), Vol. 1, 301-311, ASME, Vancouver, Canada, May 11-14, 1997.
[114] Adànez J., Diego L.F., Gayàn P., Garcio-Labiano F., Cabanillas A., Bahillo
A., “Co-Combustion of Biomass and Coal in Circulating Fluidized Bed. Modeling and Validation”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-064 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[115] Gayàn P., Adànez J., Diego L.F., Garcia-Labino F., Cabanillas A., Bahillo
A., Aho M., Veijonen K., “Circulating Fluidized Bed Co-Combustion of Coal and Biomass”, Fuel, Vol. 83, 277-286, 2004.
[116] Gani A., Naruse I., “NO and N2O Formation/Decomposition Characteristics
during Co-Combustion of Coal with Biomass”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-024 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[117] Leckner B., Karlsson M., “Emissions from Circulating Fluidized Bed
Combustion of Mixtures of Wood and Coal”, in Proceedings of 12th International Conference on Fluidized Bed Combustion, (Ed. Rubow L.), Vol. 1, 109-115, ASME, San Diego, California, USA, May 9-13, 1993.
[118] Kakaras E., Vourliotis P., Grammelis P., Co-combustion of Lignite with
Waste Wood in a Lab-scale Fluidized Bed, in Proceedings of 16th International Conference on Fluidized Bed Combustion (Ed. Geiling D. W.), FBC2001-004 in CD-ROM, ASME, Reno, Nevada, USA, May 13-26, 2001.
[119] Yrjas P., Skrifvars B.J., Hupa M., Roppo J., Nylund M., Vainikka P.,
“Chlorine in Deposits during Co-firing Biomass, Peat and Coal in a Full Scale CFBC Boiler”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-097 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[120] Orjala M., Kärki J., Lind T., Oksa M., Vainikka P., “Deposit Formation
and Hot Corrosion in Biomass Firing- Challenges and Problem Solving in Pilot and Full Scale CFB”, in Proceedings of 8th International Conference on Circulating Fluidized Bed, (Ed. Cen, K.), in CD-ROM, Hangzhou, China, May 10-13, 2005.
178
[121] Leckner B., Karlsson M., “Gaseous Emissions from CFB Combustion of
Wood”, Biomass and Bioenergy, Vol. 4, 379-389, 1993.
[122] Manninen H., Perkio A., Palonen J., Peltola K., Ruuskanen J., “Trace Metal Emissions from Co-Combustion of Refuse Derived and Packaging Derived Fuels in a Circulating Fluidized Bed Boiler”, Chemosphere, Vol. 32, 2457-2469, 1996.
[123] Rupar K., Sanati M., “The Release of Terpenes during Storage of
Biomass”, Biomass and Bioenergy, Vol. 28, 29-34, 2005. [124] Åmand L. E., Westberg H. M., Karlsson M., Leckner B., Coda B., Hocquel
M., Berger R., Hein K. R. G., Feng X., Abul Milh M., “Co-Combustion of Dried Sewage Sludge and Coal - The Fate of Heavy Metals”, in Proceedings of 16th International Conference on Fluidized Bed Combustion (Ed. Geiling, D. W.), FBC01-108 in CD-ROM, ASME, Reno, Nevada, USA, May 13-16, 2001.
[125] Åmand L. E., Leckner B., “Metal Emissions from Co-Combustion of
Sewage Sludge and Coal/Wood in Fluidized Bed”, Fuel, Vol. 83, 1803-1821, 2004.
[126] Brendon B. M., Dugwell D. R., Kandiyoti R., “Partitioning of Trace
Elements during Combustion of Coal and Biomass in a Suspension-Firing Reactor”, Fuel, Vol. 81, 2002.
[127] Meij R., “Trace Element Behavior in Coal-Fired Power Plants, Fuel
Processing Technology, Vol. 39, 199-217, 1994.
[128] Xu M., Yan R., Zheng C., Qiaro Y., Han J., Sheng C., “Status of Trace Element Emissions in a Coal Combustion Process: A Review”, Fuel Processing Technology, Vol. 85, 215-237, 2003.
[129] Demir I., Hughes R. E., Maris P. J. D., “Formation and Use of Coal
Combustion Residues from Three Types of Power Plants Burning Illinois Coals”, Fuel, Vol. 80, 659-1673, 2001.
[130] Klika Z., Bartonova L., Spears D. A.,” Effect of Boiler Output on Trace
Element Partitioning during Coal Combustion in Two Fluidised-Bed Power Stations”, Fuel, Vol. 80, 907 -917, 2001.
[131] Åmand L. E., Leckner B., “Metal Emissions from Co-Combustion of
Sewage Sludge and Coal/Wood in Fluidized Bed”, Fuel, Vol. 83, 1803-1821, 2004.
179
[132] Grammelis P., Skodras G., Kakaras E., “Effects of Biomass Co-firing with Coal on Ash Properties. Part 1: Characterization and PSD”, Fuel, Vol. 85, 2310-2315, 2006.
[133] Brown J. A. R., “Overview of Trace Element Partitioning in Flames and
Furnaces of Utility Coal-Fired Boilers”, Fuel Processing Technology, Vol. 39, 139-157, 1994.
[134] Selçuk N., Gogebakan Y., Gogebakan Z., “Partitioning Behavior of Trace
Elements during Pilot-Scale Fluidized Bed Combustion of High Ash Content Lignite”, Journal of Hazardous Materials, Vol. B137, 1698-1703, 2006.
[135] Gulyurtlu I., Lopes M. H. Abelha P., Cabrita I., Oliveira J. F. S., “The
Study of Partitioning of Heavy Metals during Fluidized Bed Combustion of Sewage Sludge and Coal, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-079 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[136] Senneca O., “Kinetics of Pyrolysis, Combustion and Gasification of Three
Biomass Fuels”, Fuel processing Technology, Vol. 88, 87-97, 2007.
[137] Winter F., Wartha C., Hofbauer H., “NO and N2O Formation during the Combustion of Wood, Straw, Malt Waste and Peat”, Bioresource Technology, Vol. 70, No. 1, 39-49, 1999.
[138] Konttinen J., Hupa M., Kallio S., Winter F., Samuelsson J., “NO Formation
Tendency Characterization for Biomass Fuels”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-78025 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[139] Liu H., Gibbs B. M., “The influence of Calcined Limestone on NOx and
N2O Emissions from Char Combustion in Fluidized Bed Combustors”, Fuel, Vol. 80, 1211-1215, 2001.
[140] Li Z, Lu Q., Na Y., “N2O and NO Emissions from Co-firing MSW with
Coals in a Pilot Scale CFBC”, Fuel Processing Technology, Vol. 85, 1539-1549, 2004.
[141] Zijlma G. J., Gerritsen.G. W., Van Den Bleek C. M, “NOx Formation and
Reuction with NH3 in Fluidized Bed Combustion - The Influence of the O2 Concentration on the Kinetics”, in Proceedings of 15th International Conference on Fluidized Bed Combustion (Ed. Reuther R.B.), FBC99-0168 in CD-ROM, ASME, Savannah, Georgia, May 16-19, 1999.
[142] Shimizu T., Tachiyama Y., Souma M., Inagaki M., “Emission Control of
NOx and N2O of Bubbling Fluidized Bed Combustion”, in Proceedings of
180
11th International Conference on Fluidized Bed Combustion, ASME, 695-700, 1991.
[143] Selçuk N., Batu A., Oymak O., “NOx Emissions from Combustion of High
Sulfur Lignite in an ABFBC Test Rig”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-109 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[144] Molina A., Eddings E. G., Pershing D.W., Sarofim A.F., “Char Nitrogen
Conversion: Implications to Emissions from Coal-Fired Utility Boilers”, Progress in Energy and Combustion Science, Vol. 26, 507-531, 2000.
[145] Chyang C. S., Wu K. T., Lin C. S., “Emissions of Nitrogen Oxides in a
Vortexing Fluidized Bed Combustor”, Fuel, Vol. 86, 234-243, 2007.
[146] Valk M, “Atmospheric Fluidized Bed Coal Combustion Research Development and Application”, Elsevier, 1995.
[147] Boemer A., Braun A., Renz U., “Emission of N2O from Four Different
large Scale Circulating Fluidized Bed Char Combustion”, in Proceedings of 12th International Conference on Fluidized Bed Combustion, ASME, San Diego, California, USA, 585-597, 1993.
[148] Liu D., Wang J. H., Chen H. P. Zhang S. H., Huang L., Lu J. D., “Emission
Control of N2O by Co-combustion of Coal and Biomass and Narrow Pulse Corona Discharge”, in Proceedings of 15th International Conference on Fluidized Bed Combustion (Ed. Reuther R.B.), FBC99-0064 in CD-ROM, ASME, Savannah, Georgia, May 16-19, 1999.
[149] Baxter L., “Biomass-Coal Co-Combustion: Opportunity for Affordable
Renewable Energy”, Fuel, Vol. 84, 1295-1302, 2005.
[150] Selcuk N., Gogebakan Y., Harmandar H., Altindag H., “Effect of Recycle on Fluidized-Bed Combustion and Emission Characteristics of High-Sulfur Lignite” Combustion Science and Technology, Vol. 176, 959-975, 2004.
[151] Spliethoff H., Hein K. R. G, “Effect of co-combustion of biomass on
emissions in pulverized fuel furnaces”, Fuel Processing Technology, Vol. 54, No. 1-3, 189-205, 1998.
[152] Nordin A., “Optimization of Sulfur Retention in Ash when Co-Combusting
High Sulfur Fuels and Biomass Fuels in a Small Pilot Scale Fluidized Bed”, Fuel, Vol. 74, 615-622, 1995.
[153] Kokko A., Nylung M., “Biomass and Coal Co-Combustion in Utility Scale
- Operating Experience of Alholmens Kraft”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.),
181
FBC2005-035 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
[154] Löffler G., Wargadalam V. J., Winter F., “Catalytic Effect of Biomass Ash
on CO, CH4 and HCN Oxidation under Fluidized Bed Combustor Conditions”, Fuel, Vol. 81, 711-717, 2002.
[155] Liu D. C., Wu Z. S., Shen B. X., Lin Z. J., “The Relative Importance of
Char and Volatile Nitrogen on Formation of Nitrous Oxides and Nitric Oxides”, Energy & Fuels, Vol. 13, 1252-1254, 1999.
[156] Goel S. K., Morihara A., Tullin C. J., Sarofim A. F., “Effect of NO and O2
Concentration on N2O Formation during Coal Combustion in a Fluidized Bed”, in Proceedings of 25th International Symposium on Combustion, 1051-1059, 1994.
[157] Hämäläinen J. P., Aho M. J., Tummavuori J. L., “Formation of Nitrogen
Oxides from Fuel-N through HCN and NH3: A Model-Compound Study”, Fuel, Vol. 73, 1894-1898, 1994.
[158] Aho M. J., Rantanen J. T., “Emissions of Nitrogen Oxides in Pulverized
Peat Combustion between 730 and 900 °C”, Fuel, Vol. 68, 586-590, 1989.
[159] Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants, Official Journal of the European Communities, L 309, 27/11/2001.
[160] Endüstriyel Kaynaklı Hava Kirliliğinin Kontrolü Yönetmeliği, 7 Ekim
2004, Resmi Gazete Sayı 25606, Çevre ve Orman Bakanlığı.
[161] Llorente M. J. F., Cuadrado R. E., Laplaza J. M. M., Garcìa J. E. C., “Combustion in Bubbling Fluidised Bed with Bed Material of Limestone to Reduce Biomass Ash Agglomeration and Sintering”, Fuel, Vol. 85, 2081-2092, 2006.
[162] Osborn E. F., Muan E.F., “Phase Diagrams of Oxide Systems”, American
Ceramic Society and the Edward Orton Jr., Ceramic Foundation, 1960.
[163] Ǻmand L. E., Leckner B., Eskilsson D., Tullin C., “Deposits on Heat Transfer Tubes during Co-combustion of Biofuels & Sewage Sludge”, Fuel, Vol. 85, 1313-1322, 2006.
[164] Enestam S., “Prediction of Ash Behavior & Deposit Formation in FBC of
Biofuel Mixtures”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-78137 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
182
[165] Skrifvars B. J., Yrjas P., Hupa M., Aho M., Silvennoinen J., Eteläaho R.,
Kouki J., Saari K., “Fireside Deposit Formation in Biomass Fired FBC: A Comparison between Tests performed in Three Significantly Different Sized Combustors”, in Proceedings of 17th International Conference on Fluidized Bed Combustion (Ed. Pisupati, S.), FBC2003-074 in CD-ROM, ASME, Jacksonville, Florida, USA, May 18-21, 2003.
[166] Skrifvars B. J., Yrjas P., Laurén T., Hupa M., Dittrich M., “The ǺBo
Akademi Database- Ash Behavior Measurements in Full Scale Boilers”, in Proceedings of 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), FBC2005-78140 in CD-ROM, ASME, Toronto, Ontario, Canada, May 22-25, 2005.
183
APPENDIX A
WORLDWIDE CO-FIRING POWER PLANTS
Tables A.1, A.2 and A.3 refer to coal-fired pulverized fuel, bubbling and circulating
fluidized bed power plants, respectively. Power plant names written in bold refer to
the plants having commercial co-firing operation. Tests have been performed with
every commercially significant fuel type (lignite, sub-bituminous coal, bituminous
coal, petroleum coke) with every major category biomass (herbaceous and woody
fuel types generated as residues and energy crops) [8]. As can be seen from Table
A.1, the majority of co-firing applications are carried out in pulverized fuel power
plants. Table A.2 and A.3 show that mostly woody type biomass is co-fired with
coals in bubbling and circulating fluidized bed boilers.
184
Co-
fired
Fue
l
Plan
tatio
n fo
rest
w
aste
and
gre
en
was
te
Woo
d w
aste
Woo
d w
aste
Woo
d w
aste
(s
awdu
st, s
havi
ngs)
Woo
d w
aste
(fre
sh
saw
dust
)
Woo
d w
aste
(p
lant
atio
n sa
wm
ill
resi
due
and
cons
truct
ion
and
dem
oliti
on w
aste
tim
ber)
Woo
d w
aste
(s
awdu
st, s
havi
ngs)
Woo
d w
aste
Woo
d ch
ips
Prim
ary
Fuel
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Hea
t/w
t %
-/5
N.A
.
-/5
-/5
-/5
-/5
N.A
.
N.A
.
3/-
Out
put
MW
e
1040
4×12
5
2×66
0
4×50
0
2×66
0
2×50
0
4×35
0
2×35
0
124
Bur
ner
Type
N.A
.
N.A
.
T-fir
ed
T-fir
ed
Wal
l- fir
ed
T-fir
ed
N.A
.
N.A
.
T-fir
ed
Ow
ner
Wes
tern
Pow
er
CS
Ener
gy
Del
ta E
lect
ricity
Mac
quar
ie
Gen
erat
ion
Del
ta E
lect
ricity
Del
ta E
lect
ricity
Stan
wel
l C
orpo
ratio
n
Taro
ng E
nerg
y
Ver
bund
A
ustri
an
Hyd
ropo
wer
AG
Plan
t Nam
e
Muj
a
Swan
bank
B
Val
es P
oint
Lid
dell
Mt P
iper
Wal
lera
wan
g
Stan
wel
l
Taro
ng
St. A
ndrä
Loca
tion
Col
lie
Ipsw
ich,
SE
Que
ensl
and
Lake
Mac
quar
ie,
New
cast
le, N
ew
Sout
h W
ales
Lidd
ell,
New
So
uth
Wal
es
Lith
gow
, New
So
uth
Wal
es
Lith
gow
, N
ew S
outh
W
ales
Roc
kham
pton
Taro
ng
St. A
ndrä
Cou
ntry
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tria
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13].
Con
tinen
t
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Aus
tralia
Euro
pe
N.A
. : N
ot a
vaila
ble.
185
Co-
fired
Fue
l
Bar
k, sa
wdu
st, w
ood
chip
s
Woo
d ch
ips
Stra
w
Stra
w
Stra
w/m
isca
nthu
s
N.A
.
Stra
w
Stra
w
Woo
d, st
raw
Woo
d
Sew
age
slud
ge
Prim
ary
Fuel
Pulv
eriz
ed
Polis
h ha
rd c
oal
Pulv
eriz
ed
coal
Pulv
eriz
ed
coal
Pulv
eriz
ed
coal
Pulv
eriz
ed
coal
Pulv
eriz
ed
coal
Pulv
eriz
ed
coal
Pu
lver
ized
co
al
Lign
ite
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Hea
t/w
t %
3/-
N.A
.
20/-
20/-
N.A
.
N.A
.
N.A
.
N.A
.
-/7
N.A
.
N.A
.
Out
put
MW
e
137
540
150
350
250
150
108
100
350
75
Bur
ner
Type
T-fir
ed
T-fir
ed
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Wal
l-fir
ed
N.A
.
N.A
.
Ow
ner
Ver
bund
A
ustri
an
Hyd
ropo
wer
A
G
Elec
trabe
l
Mid
kraf
t Po
wer
Co
Mid
kraf
t Po
wer
Co
Elsa
m
E2
Mid
kraf
t
N.A
.
N.A
.
N.A
.
N.A
.
Plan
t Nam
e
Bio
coco
mb
Rui
en
Stud
stru
pvae
rket
#1
Stud
stru
pvae
rket
#
4
N.A
.
Ave
døre
N.A
.
Bay
ernw
erke
A
G
VEA
G
VEA
G
Saar
berg
wer
ke A
G
Loca
tion
Zeltw
eg
Rui
en
Aar
hus
Aar
hus
Am
ager
Cop
enha
gen
Esjb
erg
Bav
aria
Lübb
enau
Mag
debu
rg
Saar
berg
Cou
ntry
Aus
tria
Bel
gium
Den
mar
k
Den
mar
k
Den
mar
k
Den
mar
k
Den
mar
k
Ger
man
y
Ger
man
y
Ger
man
y
Ger
man
y
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
N.A
. : N
ot a
vaila
ble.
186
Co-
fired
Fue
l
Woo
d
Woo
d pe
llets
, oliv
e w
aste
Peat
, woo
d
Sew
age
slud
ge
Ker
nels
, pap
er
slud
ge, s
hells
, fib
ers
Was
te w
ood
Pape
r slu
dge
Bio
mas
s pel
lets
(p
aper
slud
ge,
com
post
resi
due)
Pulv
eriz
ed w
ood
Prim
ary
Fuel
Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Hea
t/w
t %
N.A
.
100/
-
N.A
.
-/4
N.A
.
-/8
N.A
.
N.A
.
N.A
.
Out
put
MW
e
180
279
320
600
403
600
600
2×51
8
602
Bur
ner
Type
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Ow
ner
Fortu
m
Pow
er &
H
eat A
B
Upp
sala
En
ergi
e A
B
NU
ON
EPZ
Esse
nt
Esse
nt
EON
Elec
trabe
l
Plan
t Nam
e
Vas
tham
nsve
rt C
HP
Has
selb
yvae
rket
Upp
sala
Ene
rgie
Hem
weg
cent
rale
8
Bor
ssel
e 12
Am
erce
ntra
le 9
Am
erce
ntra
le 8
Maa
svla
ktec
entra
le
1+2
Gel
derl
and
Loca
tion
Hal
sing
bour
gi
Stoc
khol
m
Upp
sala
Am
ster
dam
Bor
ssel
e
Gee
rtrui
denb
erg
Gee
rtrui
denb
erg
Maa
svla
kte,
R
otte
rdam
Nijm
egen
Cou
ntry
Swed
en
Swed
en
Swed
en
The
Net
herla
nds
The
Net
herla
nds
The
Net
herla
nds
The
Net
herla
nds
The
Net
herla
nds
The
Net
herla
nds
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
N.A
. : N
ot a
vaila
ble.
187
Co-
fired
Fue
l
- Cer
eal r
esid
ues
Sew
age
slud
ge
Var
ious
Var
ious
Var
ious
Var
ious
Var
ious
Var
ious
Woo
d
Woo
d
Woo
d
Woo
d
Prim
ary
Fuel
Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Hea
t/w
t %
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
Out
put
MW
e
1980
2034
2400
4000
1960
2035
1995
2010
1000
1455
1200
2000
2100
Bur
ner
Type
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
Ow
ner
Edf
Pow
erge
n
Scot
tish
Pow
er
Dra
x Po
wer
Brit
ish
Ener
gy
AEP
AEP
Pow
erge
n
Inte
rnat
iona
l Po
wer
Inno
gy
Scot
tish
Pow
e r
EdF
Inno
gy
Plan
t Nam
e
Wes
t Bur
ton
Kin
gsno
rth
Lon
gann
et
Dra
x
Eggb
orou
gh
Ferr
ybri
dge
Fidd
lers
Fer
ry
Rat
cliff
e
Rug
eley
Abe
rthaw
Coc
kenz
ie
Cot
tam
Did
cot
Loca
tion
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
Cou
ntry
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
N.A
. : N
ot a
vaila
ble.
188
Co-
fired
Fue
l
Woo
d
Woo
d
RD
F
Urb
an w
ood
was
te, p
etro
leum
co
ke
Saw
dust
, tre
e tri
m
Woo
d ch
ips
Woo
d re
sidu
e,
will
ow
Shre
dded
pal
let
woo
d w
aste
Switc
hgra
ss
Was
te w
ood
Prim
ary
Fuel
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Pulv
eriz
ed C
oal
Hea
t/w
t %
N.A
.
N.A
.
N.A
.
-/10
-/13
-/30
20/-
-/12
-/12
N.A
.
Out
put
MW
e
970
1085
75
160
100
108
90
120
60
100
Bur
ner
Type
N.A
.
N.A
.
N.A
.
Cyc
lone
Fron
t Fi
red
T-fir
ed
T-fir
ed
Fron
t-w
all-
fired
cy
clon
e
T-fir
ed
N.A
.
Ow
ner
Pow
erge
n
Inno
gy
N.A
.
NIP
SCO
Sout
hern
C
ompa
ny/
Geo
rgia
Pow
er
Com
pany
New
Yor
k St
ate
Elec
tric
& G
as
Nia
gara
M
ohaw
k Po
wer
C
orp.
Nor
ther
n St
ates
Po
wer
Com
pany
Sout
hern
C
ompa
ny/
Ala
bam
a Po
wer
C
ompa
ny
N.A
.
Plan
t Nam
e
Iron
brid
ge
Tilb
ury
Am
es
Mun
icip
al
Bai
ley
Gen
erat
ing
Stat
ion
# 7
Ham
mon
d G
ener
atin
g St
atio
n #
1
Gre
enid
ge
Gen
erat
ing
Stat
ion
# 6
Dun
kirk
St
eam
St
atio
n
BL
Stat
ion
# 1
Gad
sden
St
eam
pla
nt
#2
Geo
rgia
Po
wer
Loca
tion
N.A
N.A
.
Am
es, I
owa
Che
ster
ton,
In
Coo
sa,
Geo
rgia
Dre
sden
, N
ew Y
ork
Dre
sden
, N
ew Y
ork
Engl
and
Gad
sden
, A
laba
ma
Ham
mon
d
Cou
ntry
UK
UK
USA
USA
USA
USA
USA
USA
USA
USA
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Euro
pe
Euro
pe
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
189
Co-
fired
Fue
l
Var
ious
gro
und
woo
d
Var
ious
gro
und
woo
d
Chi
pped
railr
oad
ties/
PR
B
Urb
an w
ood
was
te/S
hosh
one
coal
/PR
B b
lend
RD
F
Switc
hgra
ss
Switc
hgra
ss
Saw
dust
Seed
cor
n, so
y be
ans
Prim
ary
Fuel
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Hea
t/w
t %
-/3
-/3
-/5
-/20
2/-
-/15
2.5/
-
-/20
1/-
Out
put
MW
e
190
138
840
469
350
2×50
650
272
450
Bur
ner
Type
T-fir
ed
Wal
l-fire
d
Supe
r-C
ritic
al
Cyc
lone
Cyc
lone
Wal
l-fire
d
Wal
l-fire
d
T-fir
ed
Cyc
lone
Cyc
lone
Ow
ner
Rel
iant
Ene
rgy
Rel
iant
Ene
rgy
Kan
sas c
ity
(MO
) Pow
er &
Li
ght
Nor
ther
n In
dian
a Pu
blic
Ser
vice
C
ompa
ny
Lake
land
El
ectri
c
Mad
ison
G
as &
Ele
ctric
C
ompa
ny
IES
Util
ities
Inc.
TVA
Otte
r Tai
l Pow
er
Co.
Plan
t Nam
e
Shaw
ville
G
ener
atin
g St
atio
n #
3
Shaw
ville
G
ener
atin
g St
atio
n #
2
La C
ygne
G
ener
atin
g St
atio
n #
1
Mic
higa
n C
ity
Gen
erat
ing
Stat
ion
# 12
Lake
land
El
ectri
c #
3
Blo
unt S
treet
Ottu
mw
a G
ener
atin
g St
atio
n #
1
Alle
n (T
.H)
Foss
il Pl
ant
Big
Sto
ne
Plan
t # 1
Loca
tion
John
stow
n,
Penn
sylv
ania
John
stow
n,
Penn
sylv
ania
Kan
sas C
ity,
Kan
sas
Lake
M
ichi
gan,
In
dian
a
Lake
land
, Fl
orid
a
Mad
ison
, W
isco
nsin
Mar
shal
ltow
n,
Iow
a
Mem
phis
, Te
nnes
see
Milb
ank,
So
uth
Dak
ota
Cou
ntry
USA
USA
USA
USA
USA
USA
USA
USA
USA
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
190
Co-
fired
Fue
l
Sand
er d
ust
Woo
d ch
ips
Har
dwoo
d sa
wdu
st
Rai
lroad
ties
Shre
dded
ra
ilroa
d tie
s
Saw
dust
Saw
dust
from
pa
llets
Prim
ary
Fuel
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l Pu
lver
ized
C
oal
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Hea
t/w
t %
1/-
-/20
-/5
25/-
N.A
.
-/12
36/-
Out
put
MW
e 25
0,
319,
48
0,
490
165
180
75
170
32
46
Bur
ner
Type
Opp
osed
fir
ed
Wal
l fire
d
T-fir
ed
T-fir
ed
T-fir
ed
Wal
l fire
d
T-fir
ed
Ow
ner
Geo
rgia
Po
wer
C
ompa
ny
Sant
ee C
oope
r
TVA
Illin
ois P
ower
C
ompa
ny
Duk
e Po
wer
C
ompa
ny
Rel
iant
ene
rgy
Sout
hern
C
ompa
ny/
Sava
nnah
El
ectri
c &
Po
wer
C
ompa
ny
Plan
t Nam
e
Har
lee
Bra
nch
Gen
erat
ing
Stat
ion
Jeff
erie
s G
ener
atin
g St
atio
n #
3 &
# 4
Kin
gsto
n Fo
ssil
Plan
t # 5
Ver
mili
on P
ower
St
atio
n #
1 Le
e (W
.S) S
team
St
atio
n #
3
Sew
ard
Gen
erat
ing
Stat
ion
# 12
Kra
ft/
Riv
ersi
de
Plan
ts #
2
Loca
tion
Mill
edge
ville
, A
tlant
a,
Gea
orgi
a M
onck
s C
orne
r,
Sout
h C
arol
ina
Oak
ridge
, Te
nnes
see
Oak
woo
d,
Illin
ois
Pelz
er, S
outh
C
arol
ina
Pitts
burg
h,
Penn
sylv
ania
Port
Wen
twor
th,
Geo
rgia
Cou
ntry
USA
USA
USA
USA
USA
USA
USA
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a N
orth
A
mer
ica
Nor
th
Am
eric
a
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
191
Co-
fired
Fue
l
Was
te p
aper
sl
udge
Was
te w
ood
Kiln
drie
d w
ood/
pe
trole
um c
oke/
PR
B b
lend
Pape
r pel
lets
Rai
lroad
ties
Saw
dust
Prim
ary
Fuel
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Pulv
eriz
ed
Coa
l
Hea
t/w
t %
1/-
N.A
.
5/-
5/-
7/-
5/-
Out
put
MW
e
250
54
560
165
175
182
Bur
ner
Type
T-fir
ed
Cyc
lone
Cyc
lone
Cyc
lone
Fron
t wal
l fir
ed
Ow
ner
Tri-S
tate
G
ener
atin
g &
Tr
ansm
issi
on
Ass
ocia
tion,
Inc.
Nor
ther
n St
ates
Po
wer
Tam
pa E
lect
ric
Com
pany
Ass
ocia
ted
Elec
tric
Coo
pera
tive,
In
c.
TVA
Plan
t Nam
e
Esca
lant
e G
ener
atin
g St
atio
n #
1
SEPC
O
Kin
g (A
llen
S.)
Gen
erat
ing
stat
ion
# 1
Gan
non
(F.J.
) G
ener
atin
g St
atio
n #
3
Thom
as H
ill
Ener
gy C
ente
r
# 2
Col
bert
Foss
il Pl
ant #
1
Loca
tion
Prew
itt, N
ew
Mex
ico
Sava
nnah
Still
wat
er,
Min
neso
ta
Tam
pa,
Flor
ida
Thom
as H
ill
Res
ervo
ir,
Col
umbi
a,
MO
Tusc
umbi
a,
AL
Cou
ntry
USA
USA
USA
USA
USA
USA
Tab
le A
.1: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
pul
veriz
ed fu
el p
ower
pla
nts [
13] (
cont
’d).
Con
tinen
t
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
192
Co-
fired
Fue
l
Peat
, bar
k, o
il
Peat
, woo
d ch
ips,
bark
, oil
Woo
d an
d pa
per
was
te
Peat
, woo
d w
aste
Peat
, woo
d w
aste
, H
FO
Bar
k, sl
udge
, fib
er w
aste
Pe
at, w
ood
was
te,
HFO
Woo
d w
aste
, pea
t, oi
l
Peat
20
%,
pelle
ts 8
0%
Woo
d w
aste
, pea
t, oi
l
Woo
d, re
fuse
der
ived
fu
el (R
DF)
Prim
ary
Fuel
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l (f
orm
erly
)
Coa
l
Coa
l
Hea
t/w
t %
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
100/
-
N.A
.
80/-
Out
put
MW
t
155
218
36
17.5
, 24
20
60
20
100
25
120
N.A
.
Out
put
MW
e
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
35
N.A
.
N.A
.
18
Ow
ner
Oce
an S
ky C
o
PT In
had
Kia
t Pu
lp &
Pap
er
UPM
Kym
men
e C
ompa
ny
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Skel
lefte
a K
raft
N.A
.
Taco
ma
Publ
ic
Util
ities
Plan
t Nam
e
N.A
.
N.A
.
Lohj
a Pa
per M
ill
Out
okum
pu O
y
N.A
.
Rau
ma
Pape
r Mill
Sein
ajok
i Ene
rgy
Nyk
oppi
ng E
nerg
y
Hed
esby
n
Sode
rene
rgi A
B
City
Of T
acom
a St
eam
Pla
nt N
o. 2
Loca
tion
N.A
.
N.A
.
Lohj
a
Out
okum
pu
Piek
sam
aki
Dis
trict
Hea
ting
Rau
ma
Sein
ajok
i
Nyk
oppi
ng
Skel
lefte
a
Söde
rtälje
Taco
ma,
W
ashi
ngto
n
Cou
ntry
Indo
nesi
a
Indo
nesi
a
Finl
and
Finl
and
Finl
and
Finl
and
Finl
and
Swed
en
Swed
en
Swed
en
USA
Tab
le A
.2: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
bub
blin
g flu
idiz
ed b
ed p
ower
pla
nts [
13].
Con
tinen
t
Asi
a
Asi
a
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
193
Co-
fired
Fue
l
Slud
ge
RD
F
Lign
ite, g
as, o
il,
woo
d
Lign
ite, o
il, w
ood
Lign
ite, w
ood,
se
wag
e sl
udge
Stra
w
Peat
, woo
d, sl
udge
Peat
, RD
f, w
ood
Peat
, woo
d w
aste
Peat
, bar
k, sa
wdu
st
Lign
ite, w
ood
was
te,
oil,
gas
Prim
ary
Fuel
Coa
l
Lign
ite
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Hea
t/w
t %
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Out
put
MW
t
N.A
.
N.A
.
38
55
94
60
240
98
18
22
84
Out
put
MW
e
N.A
.
20
N.A
.
N.A
.
N.A
.
17
85
N.A
.
N.A
.
8 N.A
.
Ow
ner
Ba
Yu
Pape
r
N.A
.
N.A
.
N.A
.
N.A
.
Mid
kraf
t Pow
er
Co
Kai
nuun
Voi
ma
Oy
IVO
Kuh
mon
Lam
po
Oy
Lies
ka
Etel
a-Sa
von
Ener
gia
Plan
t Nam
e
N.A
.
Prov
inci
al E
lect
ricity
A
utho
rity
of T
haila
nd
Solv
ay O
ster
reic
h
Patri
a Pa
pier
&
Zells
toff
Lenz
ing
AG
.
Gre
naa
Co-
Gen
erat
ion
Plan
t
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Loca
tion
Peik
ang
Chi
ang
Mai
Eben
see
Fran
tsch
ach
Lenz
ing
Gre
na
Kaj
aani
Kok
kola
Kuh
mo
Lies
ka
Mik
keli
Cou
ntry
Taiw
an
Thai
land
Aus
tria
Aus
tria
Aus
tria
Den
mar
k
Finl
and
Finl
and
Finl
and
Finl
and
Finl
and
Tab
le A
.3: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
circ
ulat
ing
fluid
ized
bed
pow
er p
lant
s [13
].
Con
tinen
t
Asi
a
Asi
a
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
N.A
. : N
ot a
vaila
ble.
194
Co-
fired
Fue
l
Peat
, bar
k, w
ood
Peat
, woo
d w
aste
Peat
, slu
dge,
bar
k
Woo
d, R
DF
Coa
l was
te, w
ood
was
te
Peat
, woo
d
Woo
d, b
ark
Var
ious
was
tes
Woo
d
Woo
d, p
eat
Woo
d, p
eat,
bark
, w
ood
was
te, o
il
Woo
d, o
il
Prim
ary
Fuel
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Hea
t/w
t %
90/-
100/
-
N.A
.
N.A
.
N.A
.
N.A
.
90/-
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Out
put
MW
t
550
160
26
50
15
55
80
125
80
25
43
Out
put
MW
e
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
9.6
40
N.A
.
N.A
.
N.A
.
N.A
.
Ow
ner
Alh
olm
ens K
raft
N.A
.
Rau
ma
Mill
Sand
e Pa
per
Mill
A/S
N.A
.
N.A
.
Stor
a En
so L
td
Bris
ta K
raft
AB
N
orrk
opin
g K
raft
Nuk
opin
g En
ergi
verk
O
ster
sund
s Fj
arrv
arm
e
Cal
edon
ian
Pape
r plc
Plan
t Nam
e
Piet
arsa
ari
N.A
.
N.A
.
N.A
.
Hun
osa
pow
er
stat
ion
Ave
sta
Ener
give
rk
Stor
a En
so F
ors
Mill
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Loca
tion
Rau
hala
hti m
unic
ipal
C
HP
plan
t
Rau
ma
Sand
e
La P
ered
a
Alv
esta
Fors
Mär
sta
Nor
rkop
ing
Nuk
opoi
ng
Öst
ersu
nd
Scot
land
Cou
ntry
Finl
and
Finl
and
Finl
and
Nor
way
Spai
n
Swed
en
Swed
en
Swed
en
Swed
en
Swed
en
Swed
en
UK
Tab
le A
.3: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
circ
ulat
ing
fluid
ized
bed
pow
er p
lant
s [13
] (co
nt’d
).
Con
tinen
t
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
Euro
pe
N.A
. : N
ot a
vaila
ble.
195
Co-
fired
Fue
l
RD
F fr
om w
aste
pa
per &
pla
stic
s
Slud
ge
Ant
hrac
ite, w
ood
Tire
s
Woo
d, o
il
Ant
hrac
ite, w
ood,
oil
Prim
ary
Fuel
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Coa
l
Hea
t/w
t %
40/-
N.A
.
N.A
.
N.A
.
N.A
.
N.A
.
Out
put
MW
th
125
168
149
260
132
Out
put
MW
e
35
65
N.A
.
N.A
.
76
N.A
.
Ow
ner
Slou
gh E
stat
es
Sout
heas
t Pap
er
Bla
ck R
iver
Pa
rtner
s U
DG
Nia
gara
G
oody
ear
Rum
ford
Cog
en
Co.
P.H
.Gla
tfelte
r C
o
Plan
t Nam
e
Slou
gh H
eat a
nd
Pow
er L
td.
N.A
.
N.A
.
N.A
.
Rum
ford
Cog
en
Co.
Sprin
g G
rove
Pa
per M
ill
Loca
tion
Slou
gh
Dub
lin, G
A.
Fort
Dru
m
Nia
gara
Fal
ls
Rum
ford
, Mai
ne
Sprin
g G
rove
, Pe
nnsy
lvan
ia
Cou
ntry
UK
USA
USA
USA
USA
USA
Tab
le A
.3: W
orld
wid
e sa
mpl
es o
f co-
firin
g ex
perie
nced
circ
ulat
ing
fluid
ized
bed
pow
er p
lant
s [13
] (co
nt’d
).
Con
tinen
t
Euro
pe
Nor
th
Am
eric
a N
orth
A
mer
ica
Nor
th
Am
eric
a N
orth
A
mer
ica
Nor
th
Am
eric
a
N.A
. : N
ot a
vaila
ble.
196
APPENDIX B
TGA GRAPHS OF FUELS
Time, min
0 10 20 30 40 50 60 70 80 90 100
Wei
ght,
%
0
10
20
30
40
50
60
70
80
90
100
Tem
pera
ture
, °C
0
100
200
300
400
500
600
700
800
900
1000
WeightTemperature
Figure B.1: TGA graph of lignite.
197
Time, min
0 10 20 30 40 50 60 70 80 90 100
Wei
ght,
%
0
10
20
30
40
50
60
70
80
90
100
Tem
pera
ture
, °C
0
100
200
300
400
500
600
700
800
900
1000
WeightTemperature
Figure B.2: TGA graph of olive residue.
Time, min
0 10 20 30 40 50 60 70 80 90 100
Wei
ght,
%
0
10
20
30
40
50
60
70
80
90
100
Tem
pera
ture
, °C
0
100
200
300
400
500
600
700
800
900
1000
WeightTemperature
Figure B.3: TGA graph of hazelnut shell.
198
Time, min
0 10 20 30 40 50 60 70 80 90 100
Wei
ght,
%
0
10
20
30
40
50
60
70
80
90
100
Tem
pera
ture
, °C
0
100
200
300
400
500
600
700
800
900
1000
WeightTemperature
Figure B.4: TGA graph of cotton residue.
199
APPENDIX C
POINT VALUES OF MEASUREMENTS
Table C.1: Sampling probe readings of O2 concentrations of Runs 1-3, dry mole %. No Height Run 1 Run 2 Run 3
1 5.00 4.76 4.85 4.36
2 4.19 - - -
3 3.44 - 4.69 4.32
4 2.91 - 4.64 4.28
5 1.83 - 5.24 4.86
6 1.23 7.50 6.11 6.01
Table C.2: Sampling probe readings of O2 concentrations of Runs 4-6, dry mole %. No Height Run 4 Run 5 Run 6
1 5.00 5.02 5.12 5.62
2 4.19 - - -
3 3.44 5.10 5.27 4.69
4 2.91 5.20 4.85 4.90
5 1.83 5.40 5.63 5.52
6 1.23 5.36 6.68 5.87
200
Run
6
15.5
0
-
16.3
8
15.9
7
15.3
4
14.6
0
Run
5
15.6
4
-
15.4
7
15.7
6
14.8
9
13.5
1
Run
4
15.7
1
-
15.6
2
15.5
4
15.4
0
15.2
7
Run
3
16.5
3
-
16.4
7
16.3
6
15.7
7
14.1
5
Run
2
16.0
7
-
16.3
0
16.3
5
15.3
7
14.2
2
Run
1
15.1
5
- - - -
11.9
3
Hei
ght,
m
5.00
4.19
3.44
2.91
1.83
1.23
Tab
le C
.3: S
ampl
ing
prob
e re
adin
gs o
f CO
2 con
cent
ratio
ns o
f Run
s 1-6
, dry
mol
e %
.
No 1 2 3 4 5 6
201
Run
6
500 - 517
630
1516
3214
Run
5
838 - 896
1900
3696
5478
Run
4
568 - 567
591
891
948
Run
3
570 - 607
831
2083
4814
Run
2
500 - 497
541
1136
3228
Run
1
525 - - - -
2913
Hei
ght,
m
5.00
4.19
3.44
2.91
1.83
1.23
Tab
le C
.4: S
ampl
ing
prob
e re
adin
gs o
f CO
con
cent
ratio
ns o
f Run
s 1-6
, dry
mol
e pp
m.
No 1 2 3 4 5 6
202
Run
6
201 - 273
321
178
72
Run
5
224 - 193
256
95
12
Run
4
315 - 272
239
231
252
Run
3
331 - 380
457
271
147
Run
2
630 - 499
553
427
79
Run
1
4507
- - - -
3211
Hei
ght,
m
5.00
4.19
3.44
2.91
1.83
1.23
Tab
le C
.5: S
ampl
ing
prob
e re
adin
gs o
f SO
2 con
cent
ratio
ns o
f Run
s 1-6
, dry
mol
e pp
m.
No 1 2 3 4 5 6
203
Run
6
224 - 249
245
266
277
Run
5
220 - 252
254
293
325
Run
4
248 - 253
250
257
261
Run
3
236 - 251
260
288
320
Run
2
246 - 267
269
290
297
Run
1
227 - - - - 271
Hei
ght,
m
5.00
4.19
3.44
2.91
1.83
1.23
Tab
le C
.6: S
ampl
ing
prob
e re
adin
gs o
f NO
con
cent
ratio
ns o
f Run
s 1-6
, dry
mol
e pp
m.
No 1 2 3 4 5 6
204
Run
6
10
- 8 9 14
13
Run
5
4 - 5 7 10
7
Run
4
11
- 11
13
16
15
Run
3
16
- 16
17
21
11
Run
2
23
- 22
24
24
21
Run
1
26
- - - - 27
Hei
ght,
m
5.00
4.19
3.44
2.91
1.83
1.23
Tab
le C
.7: S
ampl
ing
prob
e re
adin
gs o
f N2O
con
cent
ratio
ns o
f Run
s 1-6
, dry
mol
e pp
m.
No 1 2 3 4 5 6
205
Run
10
855
857
857
855
861
864
858
851
853
845
822
808
700
322
Run
9
859
860
860
859
863
868
862
856
861
853
830
816
705
320
Run
8
852
854
854
853
858
861
855
845
846
836
810
795
706
312
Run
7
851
852
853
851
856
859
853
840
841
832
806
790
702
313
Run
6
855
857
857
856
859
861
855
839
839
829
804
788
700
313
Run
5
849
851
853
851
857
864
859
857
861
853
830
816
722
316
Run
4
844
846
847
845
850
855
850
842
843
833
808
794
696
310
Run
3
859
860
861
859
863
867
861
848
848
838
812
796
693
316
Run
2
847
848
848
847
850
851
845
826
824
814
788
772
679
312
Run
1
892
893
894
893
900
903
896
879
874
862
834
816
719
339
Hei
ght,
m
0.25
0.44
0.73
0.73
0.97
1.33
1.54
2.26
2.57
2.85
3.30
3.61
4.25
5.00
Tab
le C
.8: T
herm
ocou
ple
read
ings
of R
uns 1
-10,
°C
No 1 2 3 4 5 6 7 8 9 10
11
12
13
14
206
Run
10
1.00
1.00
1.00
1.00
1.00
1.01
1.00
0.99
1.00
0.99
0.96
0.94
0.82
0.38
Run
9
1.00
1.00
1.00
1.00
1.00
1.01
1.00
1.00
1.00
0.99
0.96
0.95
0.82
0.37
Run
8
1.00
1.00
1.00
1.00
1.00
1.01
1.00
0.99
0.99
0.98
0.95
0.93
0.83
0.37
Run
7
1.00
1.00
1.00
1.00
1.00
1.01
1.00
0.99
0.99
0.98
0.95
0.93
0.82
0.37
Run
6
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.98
0.98
0.97
0.94
0.92
0.82
0.37
Run
5
1.00
1.00
1.00
1.00
1.01
1.01
1.01
1.01
1.01
1.00
0.97
0.96
0.85
0.37
Run
4
1.00
1.00
1.00
1.00
1.00
1.01
1.00
0.99
1.00
0.98
0.95
0.94
0.82
0.37
Run
3
1.00
1.00
1.00
1.00
1.00
1.01
1.00
0.99
0.99
0.97
0.94
0.93
0.81
0.37
Run
2
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.97
0.97
0.96
0.93
0.91
0.80
0.37
Run
1
1.00
1.00
1.00
1.00
1.01
1.01
1.00
0.98
0.98
0.96
0.93
0.91
0.80
0.38
Hei
ght,
m
0.25
0.44
0.73
0.73
0.97
1.33
1.54
2.26
2.57
2.85
3.30
3.61
4.25
5.00
Tab
le C
.9: N
orm
aliz
ed te
mpe
ratu
res o
f Run
s 1-1
0, °C
/°C
.
No 1 2 3 4 5 6 7 8 9 10
11
12
13
14
207
APPENDIX D
CHEMICAL ANALYSES OF ASH STREAMS
Table D.1: Chemical analyses of bottom ashes of Runs 1-4.
Run 1 Run 2 Run 3 Run 4
Loss on ignition, % 1.31 1.62 1.52 1.67 Components as oxides, wt %
Silica, SiO2 47.88 32.74 24.68 19.81 Aluminum, Al2O3 23.69 16.09 14.53 15.39 Ferric, Fe2O3 8.85 5.68 5.56 6.39 Calcium, CaO 8.97 28.30 34.67 36.14 Magnesium, MgO 1.17 1.60 1.99 2.05 Sulfur, SO3 4.83 11.99 14.88 15.47 Sodium, Na2O 1.83 1.23 1.25 1.97 Potasssium, K2O 1.24 1.17 1.19 1.23 Titanium, TiO2 1.54 1.20 1.23 1.56
208
Table D.2: Chemical analyses of bottom ashes of Runs 5-7.
Run 5 Run 6 Run 7
Loss on ignition, % 1.80 1.28 1.64 Components as oxides, wt %
Silica, SiO2 28.75 14.65 26.14 Aluminum, Al2O3 13.28 7.59 9.17 Ferric, Fe2O3 5.49 3.83 4.45 Calcium, CaO 31.25 46.14 37.88 Magnesium, MgO 1.80 2.76 1.94 Sulfur, SO3 15.61 21.71 17.09 Sodium, Na2O 1.19 1.70 1.20 Potasssium, K2O 1.27 0.76 1.08 Titanium, TiO2 1.37 0.87 1.06
Table D.3: Chemical analyses of bottom ashes of Runs 8-10.
Run 8 Run 9 Run 10
Loss on ignition, % 1.91 2.09 1.87 Components as oxides, wt %
Silica, SiO2 27.52 21.23 24.62 Aluminum, Al2O3 12.42 8.01 7.94 Ferric, Fe2O3 4.88 3.69 4.30 Calcium, CaO 37.93 45.60 39.95 Magnesium, MgO 1.34 1.35 1.93 Sulfur, SO3 12.58 17.53 18.08 Sodium, Na2O 1.01 0.94 1.15 Potasssium, K2O 1.10 0.78 0.89 Titanium, TiO2 1.22 0.86 1.13
209
Run
10
3.97
17.5
2
2.45
6.89
53.1
0
1.89
15.9
3
0.36
1.10
0.76
Run
9
4.30
22.9
3
6.28
7.54
43.4
7
2.01
14.8
1
0.61
1.37
0.98
Run
8
3.61
31.5
1
7.96
6.98
31.4
8
2.06
15.9
4
1.46
1.31
1.30
Run
7
3.80
21.9
1
9.40
8.01
38.7
1
2.12
15.6
4
1.82
1.17
1.22
Run
6
3.50
24.8
3
6.77
7.64
40.6
0
2.92
13.6
6
1.72
1.00
0.86
Run
5
3.50
19.5
8
7.51
6.67
45.1
0
4.63
12.6
5
1.60
1.43
0.83
Run
4
2.75
25.9
2
8.76
5.67
38.9
3
5.05
12.4
6
1.35
0.99
0.88
Run
3
3.17
29.7
1
7.75
6.18
33.0
1
4.51
15.5
6
1.18
0.81
1.29
Run
2
4.31
23.2
7
8.04
7.40
39.9
6
4.61
13.4
7
1.36
0.97
0.91
Run
1
5.05
51.6
1
20.9
2
10.4
8
7.41
0.69
4.05
2.34
0.77
1.73
Tab
le D
.4: C
hem
ical
ana
lyes
of c
yclo
ne a
shes
of R
uns 1
-10.
Loss
on
igni
tion,
%
Com
pone
nts a
s oxi
des,
wt %
Silic
a, S
iO2
Alu
min
um, A
l 2O3
Ferr
ic, F
e 2O
3
Cal
cium
, CaO
Mag
nesi
um, M
gO
Sulfu
r, SO
3
Sodi
um, N
a 2O
Pota
sssi
um, K
2O
Tita
nium
, TiO
2
210
Run
10
2.41
32.1
6
5.98
17.5
2
22.9
1
1.54
15.8
1
1.97
1.07
1.05
Run
9
2.26
31.1
5
6.72
14.6
1
28.3
5
1.40
13.5
2
1.86
0.97
1.41
Run
8
4.13
27.6
0
8.42
14.6
8
27.4
6
1.95
15.2
7
1.55
1.02
2.04
Run
7
5.45
29.5
5
8.78
14.8
0
25.2
4
2.54
14.7
5
1.67
1.12
1.56
Run
6
4.79
22.3
4
10.2
9
15.9
5
27.5
8
3.24
15.6
9
1.98
1.26
1.66
Run
5
9.34
22.2
1
8.49
14.2
9
27.8
0
3.11
18.7
9
2.37
1.19
1.76
Run
4
10.4
8
26.5
3
9.22
13.1
8
28.2
5
2.76
15.9
4
1.57
0.86
1.68
Run
3
5.37
25.7
4
6.60
12.5
9
32.4
6
2.01
16.6
7
1.72
1.00
1.20
Run
2
5.74
32.1
9
9.69
12.6
1
22.9
5
1.40
17.8
0
1.14
0.51
1.72
Run
1
8.83
45.8
2
14.6
8
17.3
6
9.49
0.80
7.93
1.57
0.47
1.88
Tab
le D
.5: C
hem
ical
ana
lyes
of b
agho
use
filte
r as
hes o
f Run
s 1-1
0.
Loss
on
igni
tion,
%
Com
pone
nts a
s oxi
des,
wt %
Silic
a, S
iO2
Alu
min
um, A
l 2O3
Ferr
ic, F
e 2O
3
Cal
cium
, CaO
Mag
nesi
um, M
gO
Sulfu
r, SO
3
Sodi
um, N
a 2O
Pota
sssi
um, K
2O
Tita
nium
, TiO
2
211
APPENDIX E
TABULATED SIZE DISTRIBUTIONS
Table E.1: Particle size distribution of bottom ash of Run 1.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.003 0.003
4 4.750 0.123 0.127 6 3.350 4.140 4.266 10 2.000 6.572 10.838 18 1.000 47.066 57.904 35 0.500 37.191 95.095 45 0.355 4.578 99.673 80 0.180 0.102 99.776 140 0.106 0.048 99.824 PAN 0.000 0.176 100.000
212
Table E.2: Particle size distribution of bottom ash of Run 2.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.580 0.580
4 4.750 1.458 2.038 6 3.350 9.269 11.307 10 2.000 11.480 22.788 18 1.000 36.110 58.898 35 0.500 31.478 90.376 45 0.355 8.795 99.172 80 0.180 0.550 99.722 140 0.106 0.089 99.811 PAN 0.000 0.189 100.000
Table E.3: Particle size distribution of bottom ash of Run 3.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/8 16.000 0.000 0.000 1/2 12.700 0.287 0.287
5/16 8.000 0.000 0.287 1/4 6.300 0.510 0.797
4 4.750 0.929 1.726 6 3.350 8.052 9.778 10 2.000 11.454 21.232 18 1.000 37.571 58.803 35 0.500 33.463 92.266 45 0.355 7.567 99.833 80 0.180 0.154 99.988 140 0.106 0.012 100.000 PAN 0.000 0.000 100.000
213
Table E.4: Particle size distribution of bottom ash of Run 4.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.693 0.693
4 4.750 1.472 2.166 6 3.350 10.001 12.167 10 2.000 11.805 23.971 18 1.000 30.786 54.758 35 0.500 32.934 87.692 45 0.355 11.265 98.957 80 0.180 0.866 99.823 140 0.106 0.177 100.000 PAN 0.000 0.000 100.000
Table E.5: Particle size distribution of bottom ash of Run 5.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.628 0.628
4 4.750 1.755 2.383 6 3.350 10.238 12.621 10 2.000 11.295 23.915 18 1.000 28.815 52.730 35 0.500 33.934 86.664 45 0.355 12.307 98.972 80 0.180 0.848 99.820 140 0.106 0.180 100.000 PAN 0.000 0.000 100.000
214
Table E.6: Particle size distribution of bottom ash of Run 6.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.533 0.533
4 4.750 0.670 1.202 6 3.350 2.930 4.132 10 2.000 3.134 7.266 18 1.000 19.696 26.962 35 0.500 51.820 78.782 45 0.355 20.240 99.022 80 0.180 0.533 99.554 140 0.106 0.171 99.725 PAN 0.000 0.275 100.000
Table E.7: Particle size distribution of bottom ash of Run 7.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.883 0.883
4 4.750 2.297 3.180 6 3.350 10.811 13.991 10 2.000 10.225 24.215 18 1.000 25.919 50.134 35 0.500 35.159 85.293 45 0.355 13.165 98.458 80 0.180 0.900 99.358 140 0.106 0.313 99.671 PAN 0.000 0.329 100.000
215
Table E.8: Particle size distribution of bottom ash of Run 8.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 1.210 1.210
4 4.750 2.283 3.494 6 3.350 10.481 13.974 10 2.000 10.390 24.364 18 1.000 26.559 50.923 35 0.500 36.107 87.031 45 0.355 12.243 99.273 80 0.180 0.339 99.613 140 0.106 0.047 99.660 PAN 0.000 0.340 100.000
Table E.9: Particle size distribution of bottom ash of Run 9.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 1.013 1.013
4 4.750 2.340 3.353 6 3.350 10.147 13.500 10 2.000 9.530 23.030 18 1.000 25.887 48.917 35 0.500 37.141 86.057 45 0.355 13.340 99.398 80 0.180 0.371 99.768 140 0.106 0.036 99.804 PAN 0.000 0.196 100.000
216
Table E.10: Particle size distribution of bottom ash of Run 10.
ASTM MESH #
SIEVE OPENING, mm
DIFFERENTIAL WEIGHT, %
CUMULATIVE WEIGHT, %
5/16 8.000 0.000 0.000 1/4 6.300 0.866 0.866
4 4.750 2.671 3.537 6 3.350 10.960 14.497 10 2.000 9.775 24.272 18 1.000 26.182 50.454 35 0.500 42.894 93.348 45 0.355 5.964 99.311 80 0.180 0.244 99.555 140 0.106 0.127 99.682 PAN 0.000 0.318 100.000
217
Table E.11: Particle size distribution of cyclone ash of Run 1.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 1261.915 0.000 28.251 86.920 1124.683 0.110 25.179 87.730 1002.374 0.410 22.440 88.490 893.367 1.130 20.000 89.210 796.214 2.390 17.825 89.900 709.627 4.510 15.887 90.560 632.456 7.510 14.159 91.200 563.667 11.310 12.619 91.830 502.377 15.770 11.247 92.440 447.744 20.720 10.024 93.030 399.052 25.980 8.934 93.610 355.656 31.370 7.962 94.180 316.979 36.710 7.096 94.730 282.508 41.860 6.325 95.270 251.785 46.710 5.637 95.780 224.404 51.200 5.024 96.270 200.000 55.290 4.477 96.730 178.250 58.960 3.991 97.150 158.866 62.240 3.557 97.520 141.589 65.160 3.170 97.850 126.191 67.760 2.825 98.100 112.468 70.090 2.518 98.310 100.237 72.210 2.244 98.530 89.337 74.130 2.000 98.750 79.621 75.910 1.783 98.970 70.963 77.550 1.589 99.170 63.246 79.070 1.416 99.370 56.368 80.480 1.262 99.550 50.238 81.780 1.125 99.710 44.774 82.980 1.002 99.830 39.905 84.090 0.893 99.930 35.566 85.110 0.796 99.990 31.698 86.050 0.710 100.000
218
Table E.12: Particle size distribution of cyclone ash of Run 2.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 1002.374 0.000 22.440 78.370 893.367 0.030 20.000 80.030 796.214 0.150 17.825 81.640 709.627 0.380 15.887 83.190 632.456 0.860 14.159 84.700 563.667 1.780 12.619 86.150 502.377 3.200 11.247 87.540 447.744 5.050 10.024 88.840 399.052 7.280 8.934 90.060 355.656 9.870 7.962 91.190 316.979 12.770 7.096 92.210 282.508 15.930 6.325 93.120 251.785 19.310 5.637 93.920 224.404 22.860 5.024 94.630 200.000 26.520 4.477 95.250 178.250 30.230 3.991 95.790 158.866 33.940 3.557 96.270 141.589 37.610 3.170 96.700 126.191 41.190 2.825 97.100 112.468 44.650 2.518 97.470 100.237 47.960 2.244 97.820 89.337 51.130 2.000 98.160 79.621 54.130 1.783 98.480 70.963 56.980 1.589 98.780 63.246 59.670 1.416 99.070 56.368 62.200 1.262 99.330 50.238 64.600 1.125 99.560 44.774 66.860 1.002 99.750 39.905 69.010 0.893 99.890 35.566 71.050 0.796 99.980 31.698 72.990 0.710 100.000 28.251 74.860 25.179 76.640
219
Table E.13: Particle size distribution of cyclone ash of Run 3.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 893.367 0.000 20.000 86.900 796.214 0.010 17.825 87.850 709.627 0.450 15.887 88.780 632.456 1.700 14.159 89.690 563.667 3.760 12.619 90.580 502.377 6.620 11.247 91.440 447.744 10.230 10.024 92.280 399.052 14.520 8.934 93.070 355.656 19.350 7.962 93.810 316.979 24.560 7.096 94.510 282.508 29.970 6.325 95.150 251.785 35.400 5.637 95.740 224.404 40.680 5.024 96.270 200.000 45.680 4.477 96.750 178.250 50.300 3.991 97.190 158.866 54.480 3.557 97.580 141.589 58.210 3.170 97.930 126.191 61.500 2.825 98.250 112.468 64.400 2.518 98.540 100.237 66.950 2.244 98.800 89.337 69.220 2.000 99.030 79.621 71.240 1.783 99.240 70.963 73.080 1.589 99.420 63.246 74.770 1.416 99.590 56.368 76.320 1.262 99.720 50.238 77.770 1.125 99.840 44.774 79.130 1.002 99.930 39.905 80.420 0.893 100.000 35.566 81.630 31.698 82.770 28.251 83.870 25.179 84.920 22.440 85.920
220
Table E.14: Particle size distribution of cyclone ash of Run 4.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 893.367 0.000 20.000 87.470 796.214 0.020 17.825 88.380 709.627 0.630 15.887 89.270 632.456 2.050 14.159 90.140 563.667 4.230 12.619 90.980 502.377 7.210 11.247 91.790 447.744 10.950 10.024 92.580 399.052 15.380 8.934 93.320 355.656 20.380 7.962 94.020 316.979 25.760 7.096 94.680 282.508 31.330 6.325 95.280 251.785 36.890 5.637 95.840 224.404 42.270 5.024 96.350 200.000 47.310 4.477 96.820 178.250 51.920 3.991 97.240 158.866 56.040 3.557 97.630 141.589 59.670 3.170 97.970 126.191 62.830 2.825 98.290 112.468 65.590 2.518 98.570 100.237 68.010 2.244 98.830 89.337 70.160 2.000 99.060 79.621 72.090 1.783 99.260 70.963 73.860 1.589 99.440 63.246 75.500 1.416 99.600 56.368 77.020 1.262 99.740 50.238 78.460 1.125 99.850 44.774 79.810 1.002 99.930 39.905 81.080 0.893 100.000 35.566 82.290 31.698 83.430 28.251 84.510 25.179 85.540 22.440 86.520
221
Table E.15: Particle size distribution of cyclone ash of Run 5.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 893.367 0.000 20.000 87.380 796.214 0.020 17.825 88.290 709.627 0.760 15.887 89.180 632.456 2.290 14.159 90.030 563.667 4.540 12.619 90.870 502.377 7.590 11.247 91.680 447.744 11.390 10.024 92.460 399.052 15.890 8.934 93.200 355.656 20.940 7.962 93.910 316.979 26.370 7.096 94.570 282.508 31.970 6.325 95.190 251.785 37.530 5.637 95.760 224.404 42.870 5.024 96.290 200.000 47.850 4.477 96.770 178.250 52.370 3.991 97.210 158.866 56.390 3.557 97.600 141.589 59.920 3.170 97.960 126.191 62.980 2.825 98.290 112.468 65.650 2.518 98.580 100.237 67.980 2.244 98.840 89.337 70.060 2.000 99.070 79.621 71.940 1.783 99.280 70.963 73.670 1.589 99.460 63.246 75.280 1.416 99.610 56.368 76.800 1.262 99.740 50.238 78.240 1.125 99.850 44.774 79.600 1.002 99.940 39.905 80.900 0.893 100.000 35.566 82.130 31.698 83.290 28.251 84.390 25.179 85.440 22.440 86.430
222
Table E.16: Particle size distribution of cyclone ash of Run 6.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 2000.000 0.000 35.566 68.770 1782.502 0.050 31.698 70.760 1588.656 0.130 28.251 72.650 1415.892 0.230 25.179 74.450 1261.915 0.350 22.440 76.180 1124.683 0.480 20.000 77.860 1002.374 0.590 17.825 79.490 893.367 0.700 15.887 81.080 796.214 0.810 14.159 82.640 709.627 0.930 12.619 84.160 632.456 1.080 11.247 85.640 563.667 1.450 10.024 87.050 502.377 2.150 8.934 88.390 447.744 3.180 7.962 89.640 399.052 4.590 7.096 90.790 355.656 6.430 6.325 91.830 316.979 8.690 5.637 92.750 282.508 11.370 5.024 93.580 251.785 14.410 4.477 94.310 224.404 17.770 3.991 94.950 200.000 21.360 3.557 95.530 178.250 25.120 3.170 96.050 158.866 28.970 2.825 96.530 141.589 32.850 2.518 96.980 126.191 36.690 2.244 97.400 112.468 40.440 2.000 97.800 100.237 44.060 1.783 98.190 89.337 47.520 1.589 98.550 79.621 50.810 1.416 98.890 70.963 53.910 1.262 99.200 63.246 56.820 1.125 99.470 56.368 59.530 1.002 99.700 50.238 62.070 0.893 99.870 44.774 64.450 0.796 99.980 39.905 66.680 0.710 100.000
223
Table E.17: Particle size distribution of cyclone ash of Run 7.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 2000.000 0.000 35.566 64.820 1782.502 0.180 31.698 67.020 1588.656 0.470 28.251 69.170 1415.892 0.850 25.179 71.270 1261.915 1.250 22.440 73.330 1124.683 1.640 20.000 75.340 1002.374 1.990 17.825 77.320 893.367 2.300 15.887 79.240 796.214 2.590 14.159 81.120 709.627 2.940 12.619 82.940 632.456 3.400 11.247 84.680 563.667 4.060 10.024 86.330 502.377 4.990 8.934 87.870 447.744 6.240 7.962 89.280 399.052 7.830 7.096 90.560 355.656 9.770 6.325 91.710 316.979 12.030 5.637 92.720 282.508 14.580 5.024 93.610 251.785 17.340 4.477 94.380 224.404 20.270 3.991 95.050 200.000 23.310 3.557 95.630 178.250 26.420 3.170 96.150 158.866 29.550 2.825 96.630 141.589 32.670 2.518 97.060 126.191 35.760 2.244 97.470 112.468 38.800 2.000 97.860 100.237 41.780 1.783 98.230 89.337 44.680 1.589 98.580 79.621 47.490 1.416 98.910 70.963 50.220 1.262 99.210 63.246 52.860 1.125 99.480 56.368 55.400 1.002 99.700 50.238 57.860 0.893 99.870 44.774 60.250 0.796 99.980 39.905 62.560 0.710 100.000
224
Table E.18: Particle size distribution of cyclone ash of Run 8.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 1124.683 0.000 20.000 89.670 1002.374 0.060 17.825 90.500 893.367 0.490 15.887 91.290 796.214 1.490 14.159 92.040 709.627 3.030 12.619 92.760 632.456 5.270 11.247 93.450 563.667 8.260 10.024 94.110 502.377 12.010 8.934 94.740 447.744 16.440 7.962 95.340 399.052 21.420 7.096 95.930 355.656 26.790 6.325 96.490 316.979 32.310 5.637 97.040 282.508 37.790 5.024 97.570 251.785 43.040 4.477 98.070 224.404 47.910 3.991 98.540 200.000 52.330 3.557 98.960 178.250 56.240 3.170 99.330 158.866 59.670 2.825 99.640 141.589 62.670 2.518 99.850 126.191 65.280 2.244 99.990 112.468 67.610 2.000 100.000 100.237 69.730 89.337 71.690 79.621 73.550 70.963 75.340 63.246 77.080 56.368 78.740 50.238 80.340 44.774 81.840 39.905 83.250 35.566 84.550 31.698 85.740 28.251 86.840 25.179 87.850 22.440 88.790
225
Table E.19: Particle size distribution of cyclone ash of Run 9.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 2000.000 0.000 35.566 69.990 1782.502 0.020 31.698 71.860 1588.656 0.040 28.251 73.650 1415.892 0.070 25.179 75.370 1261.915 0.110 22.440 77.050 1124.683 0.150 20.000 78.690 1002.374 0.200 17.825 80.310 893.367 0.240 15.887 81.920 796.214 0.350 14.159 83.510 709.627 0.680 12.619 85.070 632.456 1.310 11.247 86.590 563.667 2.280 10.024 88.040 502.377 3.630 8.934 89.400 447.744 5.400 7.962 90.660 399.052 7.600 7.096 91.810 355.656 10.210 6.325 92.830 316.979 13.180 5.637 93.730 282.508 16.460 5.024 94.520 251.785 19.980 4.477 95.200 224.404 23.660 3.991 95.790 200.000 27.410 3.557 96.300 178.250 31.150 3.170 96.750 158.866 34.830 2.825 97.160 141.589 38.380 2.518 97.530 126.191 41.790 2.244 97.880 112.468 45.030 2.000 98.210 100.237 48.100 1.783 98.520 89.337 51.020 1.589 98.810 79.621 53.800 1.416 99.090 70.963 56.460 1.262 99.350 63.246 59.000 1.125 99.570 56.368 61.420 1.002 99.750 50.238 63.730 0.893 99.890 44.774 65.930 0.796 99.980 39.905 68.010 0.710 100.000
226
Table E.20: Particle size distribution of cyclone ash of Run 10.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 893.367 0.000 20.000 89.640 796.214 0.030 17.825 90.440 709.627 0.950 15.887 91.220 632.456 2.770 14.159 91.970 563.667 5.360 12.619 92.700 502.377 8.840 11.247 93.400 447.744 13.160 10.024 94.080 399.052 18.210 8.934 94.740 355.656 23.820 7.962 95.370 316.979 29.760 7.096 95.970 282.508 35.790 6.325 96.560 251.785 41.660 5.637 97.110 224.404 47.180 5.024 97.640 200.000 52.220 4.477 98.140 178.250 56.690 3.991 98.600 158.866 60.570 3.557 99.010 141.589 63.890 3.170 99.370 126.191 66.730 2.825 99.660 112.468 69.160 2.518 99.860 100.237 71.280 2.244 99.990 89.337 73.180 2.000 100.000 79.621 74.930 70.963 76.570 63.246 78.140 56.368 79.640 50.238 81.070 44.774 82.430 39.905 83.690 35.566 84.870 31.698 85.970 28.251 86.980 25.179 87.920 22.440 88.800
227
Table E.21: Particle size distribution of baghouse filter ash of Run 1.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 84.720 12.619 0.050 0.893 86.990 11.247 0.490 0.796 89.020 10.024 1.580 0.710 90.860 8.934 3.270 0.632 92.510 7.962 5.590 0.564 94.000 7.096 8.560 0.502 95.320 6.325 12.170 0.448 96.480 5.637 16.380 0.399 97.460 5.024 21.130 0.356 98.250 4.477 26.320 0.317 98.860 3.991 31.820 0.283 99.320 3.557 37.500 0.252 99.650 3.170 43.220 0.224 99.880 2.825 48.860 0.200 100.000 2.518 54.290 2.244 59.440 2.000 64.220 1.783 68.610 1.589 72.580 1.416 76.150 1.262 79.350 1.125 82.190
228
Table E.22: Particle size distribution of baghouse filter ash of Run 2.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.470 12.619 0.070 0.893 88.420 11.247 0.600 0.796 90.170 10.024 1.810 0.710 91.750 8.934 3.670 0.632 93.180 7.962 6.210 0.564 94.490 7.096 9.440 0.502 95.670 6.325 13.360 0.448 96.710 5.637 17.930 0.399 97.610 5.024 23.060 0.356 98.350 4.477 28.620 0.317 98.910 3.991 34.480 0.283 99.350 3.557 40.480 0.252 99.660 3.170 46.460 0.224 99.880 2.825 52.270 0.200 100.000 2.518 57.800 2.244 62.940 2.000 67.640 1.783 71.860 1.589 75.600 1.416 78.890 1.262 81.760 1.125 84.270
229
Table E.23: Particle size distribution of baghouse filter ash of Run 3.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.950 12.619 0.050 0.893 88.910 11.247 0.320 0.796 90.650 10.024 1.440 0.710 92.210 8.934 3.220 0.632 93.600 7.962 5.650 0.564 94.860 7.096 8.800 0.502 95.980 6.325 12.650 0.448 96.960 5.637 17.170 0.399 97.800 5.024 22.270 0.356 98.490 4.477 27.840 0.317 99.010 3.991 33.730 0.283 99.410 3.557 39.800 0.252 99.700 3.170 45.870 0.224 99.890 2.825 51.800 0.200 100.000 2.518 57.460 2.244 62.740 2.000 67.580 1.783 71.930 1.589 75.800 1.416 79.180 1.262 82.140 1.125 84.710
230
Table E.24: Particle size distribution of baghouse filter ash of Run 4.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 15.887 0.000 1.125 85.550 14.159 0.380 1.002 87.560 12.619 1.370 0.893 89.350 11.247 2.880 0.796 90.960 10.024 4.970 0.710 92.410 8.934 7.670 0.632 93.740 7.962 10.950 0.564 94.940 7.096 14.810 0.502 96.030 6.325 19.190 0.448 96.990 5.637 24.030 0.399 97.820 5.024 29.240 0.356 98.490 4.477 34.710 0.317 99.010 3.991 40.320 0.283 99.410 3.557 45.940 0.252 99.690 3.170 51.450 0.224 99.890 2.825 56.750 0.200 100.000 2.518 61.750 2.244 66.370 2.000 70.580 1.783 74.360 1.589 77.720 1.416 80.680 1.262 83.280
231
Table E.25: Particle size distribution of baghouse filter ash of Run 5.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.930 12.619 0.210 0.893 88.840 11.247 1.120 0.796 90.550 10.024 2.590 0.710 92.090 8.934 4.690 0.632 93.480 7.962 7.430 0.564 94.740 7.096 10.830 0.502 95.870 6.325 14.870 0.448 96.870 5.637 19.480 0.399 97.730 5.024 24.600 0.356 98.430 4.477 30.120 0.317 98.970 3.991 35.890 0.283 99.380 3.557 41.770 0.252 99.680 3.170 47.630 0.224 99.880 2.825 53.320 0.200 100.000 2.518 58.730 2.244 63.770 2.000 68.370 1.783 72.520 1.589 76.210 1.416 79.450 1.262 82.290 1.125 84.760
232
Table E.26: Particle size distribution of baghouse filter ash of Run 6.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 12.619 0.000 0.893 88.030 11.247 0.070 0.796 89.900 10.024 0.900 0.710 91.570 8.934 2.350 0.632 93.070 7.962 4.430 0.564 94.420 7.096 7.230 0.502 95.630 6.325 10.730 0.448 96.690 5.637 14.930 0.399 97.600 5.024 19.760 0.356 98.340 4.477 25.120 0.317 98.910 3.991 30.890 0.283 99.350 3.557 36.900 0.252 99.660 3.170 42.990 0.224 99.880 2.825 49.010 0.200 100.000 2.518 54.820 2.244 60.280 2.000 65.330 1.783 69.900 1.589 73.990 1.416 77.600 1.262 80.760 1.125 83.510 1.002 85.910
233
Table E.27: Particle size distribution of baghouse filter ash of Run 7.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.160 12.619 0.080 0.893 88.210 11.247 0.080 0.796 90.030 10.024 0.970 0.710 91.660 8.934 2.470 0.632 93.120 7.962 4.600 0.564 94.450 7.096 7.460 0.502 95.640 6.325 11.020 0.448 96.700 5.637 15.280 0.399 97.600 5.024 20.170 0.356 98.340 4.477 25.600 0.317 98.910 3.991 31.420 0.283 99.340 3.557 37.480 0.252 99.660 3.170 43.610 0.224 99.880 2.825 49.650 0.200 100.000 2.518 55.450 2.244 60.910 2.000 65.920 1.783 70.460 1.589 74.490 1.416 78.030 1.262 81.130 1.125 83.810
234
Table E.28: Particle size distribution of baghouse filter ash of Run 8.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.860 12.619 0.050 0.893 88.800 11.247 0.320 0.796 90.520 10.024 1.410 0.710 92.060 8.934 3.140 0.632 93.450 7.962 5.520 0.564 94.710 7.096 8.620 0.502 95.840 6.325 12.420 0.448 96.840 5.637 16.900 0.399 97.710 5.024 21.970 0.356 98.410 4.477 27.530 0.317 98.960 3.991 33.440 0.283 99.370 3.557 39.530 0.252 99.670 3.170 45.640 0.224 99.880 2.825 51.620 0.200 100.000 2.518 57.320 2.244 62.640 2.000 67.510 1.783 71.890 1.589 75.760 1.416 79.150 1.262 82.090 1.125 84.640
235
Table E.29: Particle size distribution of baghouse filter ash of Run 9.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.950 12.619 0.140 0.893 88.820 11.247 0.910 0.796 90.500 10.024 2.370 0.710 92.030 8.934 4.520 0.632 93.420 7.962 7.370 0.564 94.690 7.096 10.950 0.502 95.830 6.325 15.190 0.448 96.850 5.637 20.050 0.399 97.710 5.024 25.400 0.356 98.420 4.477 31.100 0.317 98.970 3.991 37.020 0.283 99.380 3.557 43.000 0.252 99.680 3.170 48.870 0.224 99.890 2.825 54.510 0.200 100.000 2.518 59.820 2.244 64.730 2.000 69.170 1.783 73.150 1.589 76.670 1.416 79.760 1.262 82.470 1.125 84.850
236
Table E.30: Particle size distribution of baghouse filter ash of Run 10.
Size, µm CUMULATIVE WEIGHT, % Size, µm CUMULATIVE
WEIGHT, % 14.159 0.000 1.002 86.350 12.619 0.060 0.893 88.380 11.247 0.580 0.796 90.190 10.024 1.760 0.710 91.820 8.934 3.590 0.632 93.280 7.962 6.080 0.564 94.600 7.096 9.270 0.502 95.780 6.325 13.120 0.448 96.810 5.637 17.610 0.399 97.690 5.024 22.650 0.356 98.410 4.477 28.130 0.317 98.960 3.991 33.900 0.283 99.380 3.557 39.820 0.252 99.680 3.170 45.740 0.224 99.890 2.825 51.520 0.200 100.000 2.518 57.040 2.244 62.200 2.000 66.940 1.783 71.230 1.589 75.060 1.416 78.450 1.262 81.430 1.125 84.050
237
APPENDIX F
SIZE DISTRIBUTION GRAPHS
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 1 Lignite
Figure F.1: Particle size distribution of lignite fed in Run 1.
238
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 2 Lignite
Figure F.2: Particle size distribution of lignite fed in Run 2.
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 3 Lignite
Figure F.3: Particle size distribution of lignite fed in Run 3.
239
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 4 Lignite
Figure F.4: Particle size distribution of lignite fed in Run 4.
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 5 Lignite
Figure F.5: Particle size distribution of lignite fed in Run 5.
240
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 6 Lignite
Figure F.6: Particle size distribution of lignite fed in Run 6.
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 7 Lignite
Figure F.7: Particle size distribution of lignite fed in Run 7.
241
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 8 Lignite
Figure F.8: Particle size distribution of lignite fed in Run 8.
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 9 Lignite
Figure F.9: Particle size distribution of lignite fed in Run 9.
242
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 10 Lignite
Figure F.10: Particle size distribution of lignite fed in Run 10.
Sieve opening, mm
0 1 2 3 4 5 6 7
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Olive residue
Figure F.11: Particle size distribution of olive residue fed in Runs 3, 4 and 5.
243
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Hazelnut shell
Figure F.12: Particle size distribution of hazelnut shell fed in Runs 6, 7 and 8.
Sieve opening, mm
0 2 4 6 8 10 12 14 16 18 20
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Cotton residue
Figure F.13: Particle size distribution of cotton residue fed in Runs 9 and 10.
244
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 2 Limestone
Figure F.14: Particle size distribution of limestone fed in Runs 2, 3, 4 and 5.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 6 Limestone
Figure F.15: Particle size distribution of limestone fed in Runs 6, 7 and 8.
245
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 10 Limestone
Figure F.16: Particle size distribution of limestone fed in Runs 9 and 10.
246
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 1 Bed drain ash
Figure F.17: Particle size distribution of bottom ash of Run 1.
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 2 Bed drain ash
Figure F.18: Particle size distribution of bottom ash of Run 2.
247
Sieve opening, mm
0 2 4 6 8 10 12 14 16
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 3 Bed drain ash
Figure F.19: Particle size distribution of bottom ash of Run 3.
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 4 Bed drain ash
Figure F.20: Particle size distribution of bottom ash of Run 4.
248
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 5 Bed drain ash
Figure F.21: Particle size distribution of bottom ash of Run 5.
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 6 Bed drain ash
Figure F.22: Particle size distribution of bottom ash of Run 6.
249
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 7 Bed drain ash
Figure F.23: Particle size distribution of bottom ash of Run 7.
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 8 Bed drain ash
Figure F.24: Particle size distribution of bottom ash of Run 8.
250
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 9 Bed drain ash
Figure F.25: Particle size distribution of bottom ash of Run 9.
Sieve opening, mm
0 1 2 3 4 5 6 7 8
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 10 Bed drain ash
Figure F.26: Particle size distribution of bottom ash of Run 10.
251
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 1 Cyclone ash
Figure F.27: Particle size distribution of cyclone ash of Run 1.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 2 Cyclone ash
Figure F.28: Particle size distribution of cyclone ash of Run 2.
252
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 3 Cyclone ash
Figure F.29: Particle size distribution of cyclone ash of Run 3.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 4 Cyclone ash
Figure F.30: Particle size distribution of cyclone ash of Run 4.
253
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 5 Cyclone ash
Figure F.31: Particle size distribution of cyclone ash of Run 5.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 6 Cyclone ash
Figure F.32: Particle size distribution of cyclone ash of Run 6.
254
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 7 Cyclone ash
Figure F.33: Particle size distribution of cyclone ash of Run 7.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 8 Carryover ash
Figure F.34: Particle size distribution of cyclone ash of Run 8.
255
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 9 Cyclone ash
Figure F.35: Particle size distribution of cyclone ash of Run 9.
Sieve opening, mm
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 10 Cyclone ash
Figure F.36: Particle size distribution of cyclone ash of Run 10.
256
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 1 Baghouse filter ash
Figure F.37: Particle size distribution of baghouse filter ash of Run 1.
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 2 Baghouse filter ash
Figure F.38: Particle size distribution of baghouse filter ash of Run 2.
257
Sieve opening mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 3 Baghouse filter ash
Figure F.39: Particle size distribution of baghouse filter ash of Run 3.
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 4 Baghouse filter ash
Figure F.40: Particle size distribution of baghouse filter ash of Run 4.
258
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 5 Baghouse filter ash
Figure F.41: Particle size distribution of baghouse filter ash of Run 5.
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 6 Baghouse filter ash
Figure F.42: Particle size distribution of baghouse filter ash of Run 6.
259
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 7 Baghouse filter ash
Figure F.43: Particle size distribution of baghouse filter ash of Run 7.
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 8 Baghouse filter ash
Figure F.44: Particle size distribution of baghouse filter ash of Run 8.
260
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 9 Baghouse filter ash
Figure F.45: Particle size distribution of baghouse filter ash of Run 9.
Sieve opening, mm
0.000 0.005 0.010 0.015
Res
idue
, %
0
10
20
30
40
50
60
70
80
90
100
Run 10 Baghouse filter ash
Figure F.46: Particle size distribution of baghouse filter ash of Run 10.
261
APPENDIX G
CALIBRATION CURVES
Frequency, Hz
0 5 10 15 20 25 30 35
Lign
ite fl
ow ra
te, k
g/h
0102030405060708090
100110120130140
y = 4.304 xR2 = 1.000
Figure G.1: Calibration curve for lignite flow rate.
262
Frequency, Hz
0 10 20 30 40 50 60 70 80 90
Oliv
e re
sidu
e flo
w ra
te, k
g/h
0
10
20
30
40
50
60
70
80
90
100
110
y = 1.250 xR2 = 1.000
Figure G.2: Calibration curve for olive residue flow rate.
Frequency, Hz
0 10 20 30 40 50 60 70 80 90 100 110
Haz
elnu
t she
ll flo
w ra
te, k
g/h
0
10
20
30
40
50
y = 0.466 xR2 = 1.000
Figure G.3: Calibration curve for hazelnut shell flow rate.
263
Frequency, Hz
0 10 20 30 40 50 60
Cot
ton
resi
due
flow
rate
, kg/
h
0
10
20
30
40
50
y = 0.787 xR2 = 0.999
Figure G.4: Calibration curve for cotton residue flow rate.
Frequency, Hz
0 10 20 30 40
Lim
esto
ne fl
ow ra
te, k
g/h
0
10
20
30
40
50
60
70
80
y = 2.377 xR2 = 0.999
Figure G.5: Calibration curve for limestone flow rate.
264
APPENDIX H
DEPOSIT COMPOSITIONS
Table H.1: Deposit compositions.
Element, wt %
Olive residue/ lignite co-firing
deposit
Hazelnut shell/ lignite co-firing
deposit
Cotton residue/ lignite co-firing
deposit
Na 3.12 2.00 1.72
Mg 3.70 2.13 2.51
Al 5.05 7.62 5.68
Si 16.79 27.93 23.95
P 1.28 0.00 4.54
S 20.21 19.00 15.10
Cl 0.27 0.00 0.00
K 11.59 1.62 4.12
Ca 27.76 25.01 25.66
Ti 0.45 0.97 0.94
Mn 0.78 0.00 0.00
Fe 9.00 13.72 15.78
265
Figure H.1: EDX analysis graph of olive residue and lignite co-firing deposit.
Figure H.2: EDX analysis graph of hazelnut shell and lignite co-firing deposit.
266
Figure H.3: EDX analysis graph of cotton residue and lignite co-firing deposit.
APPENDIX I
SCREW FEEDER DESIGN CALCULATIONS
Biomass feeding was problematic with the available feeding system which was
originally designed for coal feeding. As biomass fuels have much lower bulk density
compared to coal bulk density, their flow rate was much lower compared to that of
coals. To mitigate the feeding problems, a new bed screw feeder is designed by
taking account of the available data. To check the design approach existing ash
feeder is utilized. The results of the calculations and comparisons are as follows.
Ash Feeder Design Check
The flow rate of available spiral ash feeder is checked with the theoretical and design
values. The calculation procedure for volumetric and mass flow rates are given
below.
L = 0.046 m
R2 = 0.029 m R1 = 0.010 m
0.019 m
Figure I.1: Schematic description of ash feeder spiral.
267
( )
( )
2 22 1
3
1
2
2 2
Q = Π R -R L N 60
where
Q : volumetric flow rate (m /h)R : inside radius (m)R : spiral radius (m)L : pitch lenght (m)N : revoltion per minute (rpm)
Q = Π 0.029 -0.010 0.046 N 60 = 0.006425 N
m = ρ
⋅ ⋅ ⋅ ⋅
⋅ ⋅ ⋅ ⋅
i
i
i
i
Bulk BulkQ = 0.006425 Nρ⋅ ⋅ ⋅i
⋅
Bulk densities for hazelnut shell and olive residue are measured as 260 and 600
kg/m3, respectively. By utilizing the above expression, mass flow rates of both
biomasses are calculated and then compared with the measured values of hazelnut
shell and olive residue flow rate. Results are given in Tables I.1 and I.2 and Figures
I.2 and I.3.
Table I.1: Comparison of measured and calculated hazelnut shell flow rate.
Frequency, Hz Measured Flow Rate, kg/h
Calculated Flow Rate, kg/h Difference, %
0.0 0.0 0.0 0.0
30.0 13.2 14.0 6.3
40.0 18.0 18.7 3.9
50.0 22.0 23.4 6.3
60.0 27.0 28.1 3.9
75.0 33.0 35.1 6.3
85.0 36.0 39.8 10.4
90.0 38.0 42.1 10.8
268
0
5
10
15
20
25
30
35
40
45
0 20 40 60 80 100Frequency, Hz
Flow
rate
, kg/
h
MeasuredCalculated
Figure I.2: Comparison of measured and calculated hazelnut shell flow rate.
Table I.2: Comparison of measured and calculated olive oil residue flow rate.
Frequency, Hz Measured Flow Rate, kg/h
Calculated Flow Rate, kg/h Difference, %
0.0 0.0 0.0 0.0
10.0 12.3 10.8 12.2
20.0 24.5 21.6 11.9
30.0 37.0 32.4 12.5
40.0 49.2 43.2 12.2
50.0 61.0 54.0 11.5
60.0 74.0 64.8 12.5
75.0 96.0 81.0 15.7
269
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80
Frequency, Hz
Flow
rate
, kg/
hMeasuredCalculated
Figure I.3: Comparison of measured and calculated olive residue flow rate.
As can be seen from the above tables and figures, theoretical and measured values
are in good agreement. The differences between the theoretical and measured values
are acceptable considering the motor frequency controller accuracy which is around
10 %. Following the verification of design approach, over bed and under bed screw
feeders design have been performed.
Bed Screw Feeder Design
In the figure given below the dimensions of new bed screw feeders are shown. The
main constraints in the design of new feeder are the available space for installation,
cooling jacket and availability of the materials. To maximize the diameter of the
feeder spiral for feeding of large biomass particles, wall thicknesses are reduced. To
provide easy flow of materials and also increased durability, stainless steel (SS-304)
was used in the manufacturing of new feeders. New feeder design is then checked
with 45 % hazelnut shell and 55 % lignite blend having 500 kg/m3 bulk density.
270
L = 0.05 m
R2 = 0.025 m R1 = 0.010 m
0.015 m
Figure I.4: Schematic description of new bed feeder spiral.
( )( )
2 22 1
2 2
3Bulk blend
Bulk blend
Q = Π R -R L N 60
Q = Π 0.025 -0.010 0.050 N 60 = 0.004948 N
N = 50 rpmρ = 500 kg/m
m Q = 0.004948 500 50 123.7 kg/h
m 97 kg/h (from stoichiometric combustion calculat
ρ
⋅ ⋅ ⋅ ⋅
⋅ ⋅ ⋅ ⋅ ⋅
= ⋅ ⋅ ⋅
=
=
i
i
i i
iion)
As 123.7 > 97 kg/h, which is obtained from stoichiometric combustion calculations,
the new feeder is suitable for co-firing processes. Figure I.5 shows drawing of the
new bed feeder.
271
Figu
re I.
5: S
chem
atic
dra
win
g of
the
new
bed
feed
er.
272
273
CURRICULUM VITAE
PERSONAL INFORMATION Surname, Name : Göğebakan, Zuhal Nationality : Turkish (TC) Date and Place of Birth : 21 March 1979, Denizli Marital Status : Married Phone : + 90 312 2104387 Fax : + 90 312 2102600 E-mail : [email protected]
EDUCATION Degree Institution Year of Graduation PhD METU, Chemical Engineering 2007 MS METU, Chemical Engineering 2003 BS METU, Chemical Engineering 2001 High School Anatolian High School, Denizli 1997
WORK EXPERIENCE Year Place Enrollment 2001-2007 METU, Chemical Engineering Research Assistant 2000 July Deniz Textile, Denizli Intern Engineering Student
ACADEMIC EXPERIENCE Assistantship to some undergraduate courses:
Chemical Engineering Design
Mathematical Modeling in Chemical Engineering
274
Fundamentals of Heat and Mass Transfer
Novel Topics in Separation Processes
Chemical Engineering Laboratories I, II, III
Chemical Engineering Economics
Industrial Organization and Management
Introduction to Chemical Engineering
Assistantship to some graduate courses:
Combustion Technology
Advanced Heat Transfer
Fluidization
• Attended to “2nd Chemical Engineering Conference for Collaborative Research in Eastern Mediterranean” in Ankara, Turkey in 20-24 May 2001.
• Attended to “3rd Chemical Engineering Conference for Collaborative Research in Eastern Mediterranean” in Thessaloniki, Greece in 13-15 May 2003.
• Attended to “18th International Conference on Fluidized Bed Combustion” in Toronto, Ontario, Canada in 22-25 May 2005.
AREAS OF EXPERTISE
• Fluidized bed combustion
• Co-firing of biomass and coal
• Deposit formation in FBC FOREIGN LANGUAGES Advanced English, Basic German COMPUTER SKILLS
• Software packages: Microsoft Office (Word, Excel, PowerPoint, FrontPage,Visio), AutoCAD, Sigma Plot, MathCAD.
• Programming Languages: Visual Basic.
275
REFEREED PAPERS IN CONFERENCE PROCEEDINGS
[1] Coskun Z., Yucel H., Culfaz A., “Synthesis and Characterization of ZSM-35”, 3rd Chemical Engineering Conference for Collaborative Research in Eastern Mediterranean, Thessaloniki, Greece, in CD-ROM (2003).
[2] Selcuk, N., Gogebakan, Y., and Gogebakan, Z., “Partitioning of Trace Elements during Fluidized Bed Combustion of High Ash Content Lignite”, Proceedings of the 18th International Conference on Fluidized Bed Combustion (Ed. Jia L.), ASME, Toronto, Ontario, Canada, Paper No: 58 in CD-ROM (2005).
[3] Gogebakan Z., Gogebakan Y., Selcuk N., “Co-firing of Olive Residue with Lignite in Bubbling FBC”, accepted to be published in Proceedings of 5th Mediterranean Combustion Symposium, Monastir, Tunisia, September 9-13, (2007).
REFEREED PAPERS IN JOURNALS
[1] Selcuk, N., Gogebakan, Y., and Gogebakan, Z., “Partitioning Behavior of Trace Elements during Pilot Scale Fluidized Bed Combustion of High Ash Content Lignite”, Journal of Hazardous Materials B137 (2006) 1698-1703.
[2] Gogebakan Z., Yucel H. and Culfaz A., “Crystallization Field and Rate Study for the Synthesis of Ferrierite”, Journal of Industrial and Engineering Chemistry Research 46 (2007) 2006-2012.
RESEARCH PROJECTS
[1] Selçuk N., Göğebakan Y., Göğebakan Z., Uygur A. B., Moralı M., Co-firing of Biomass with Coal in Bubbling Fluidized Bed Combustors, TÜBİTAK MAG-104M200 (Continuing).
HOBBIES Traveling, Shopping, Movies, Formula 1, Painting.