Prasad, K. K. - Ray, Hem Shankar - Advances in Rotary Kiln Sponge Iron Plant-New Age International (P) LTD (2009)

Advances in
Rotary Kiln
Sponge Iron Plant
K.K. Prasad
Hem Shankar Ray
Advances in
Rotary Kiln
Sponge Iron Plant
This page
intentionally left
blank
Advances in
Rotary Kiln
Sponge Iron Plant
K.K. Prasad
B.Sc., Hons. (Chem.), B.E., Ph.D. (Mech. Engg.)
Formerly, Deputy General Manager
Research and Development Centre for Iron and Steel
Steel Authority of India Limited, Doranda, Ranchi
Hem Shankar Ray
B.Tech., Hons. (Metallurgical Engg.), M.A.Sc., Ph.D.
Emeritus Scientist
Central Glass and Ceramic Research Institute
Kolkata
Copyright © 2009, New Age International (P) Ltd., Publishers
Published by New Age International (P) Ltd., Publishers
All rights reserved.

No part of this ebook may be reproduced in any form, by photostat, microfilm,
xerography, or any other means, or incorporated into any information retrieval
system, electronic or mechanical, without the written permission of the publisher.
All inquiries should be emailed to [email protected]
ISBN (13) : 978-81-224-2892-6
PUBLISHING FOR ONE WORLD

NEW AGE INTERNATIONAL (P) LIMITED, PUBLISHERS
4835/24, Ansari Road, Daryaganj, New Delhi - 110002
Visit us at www.newagepublishers.com
Preface to the First Edition
A number of books on sponge iron are available and they deal largely with gas based sponge iron
making. There is no comprehensive treatise available on coal based rotary kiln sponge iron making.
This is in keeping with the global popularity of gas based sponge iron making which accounts for
over 90% of the total sponge iron production. But the situation in India is different. Locational
conditions here are such that, except for a limited area in Western coast where regular availability of
natural gas at competitive price can be ensured, the enhancement of sponge iron making capacity
has been through the coal based rotary kiln route.
But all is not rosy for the rotary kiln. Even though rotary kiln is a versatile reactor and is able to
handle raw materials of widely varying properties, but this statement presupposes the prior knowledge
of variation. Apart from this high sensitivity to variation in raw material characteristics, there are
the problems of high-energy consumption and low productivity. Before long the technology must
improve itself to make it more energy efficient and environment friendly.
The present book focusses on the optimised process concept, which has been developed, based
on the extensive R&D work undertaken in the Direct Reduction Process Development (DRPD)
Division, which is under the Research and Development Centre for Iron and Steel (RDCIS) equipped
with a fully functional Rotary Kiln Sponge Iron Pilot Plant and associated laboratories with facilities
for Process Control.
Thus we have highlighted the recent advances in various features of rotary kiln. To make the
text comprehensive a chapter on reaction kinetics has been included as also a section on iron ore and
pellets. Other supporting sections are on aerodynamics, project engineering, auto-ignition, coal
throwing etc.
We hope readers would appreciate this endeavour of ours and help us by offering constructive
criticism.
K.K. Prasad
H.S. Ray
This page
intentionally left
blank
Acknowledgement
The authors would like to thank foremost RDCIS SAIL; the first author compiled a very significant
amount of information presented in this book during his tenure here between 1981 and 2006.
Others include employees of DRPD Division of RDCIS SAIL, other employees who assisted the
project, both within SAIL and from other organisations like HEC and professors and scientists in
IIT’s (Mumbai, Kanpur, Kharagpur, IT-BHU) and RRL’s.
Our special thanks go to Dr. Asim Kumar Ray of RDCIS, SAIL, to provide help and support in
bringing out this book. We would also like to thank Prof. R.C. Gupta of IT, BHU for making
available macro- and micro-graphs of sponge iron as also a few samples and some requisite information.
A number of rotary kiln sponge iron plants of India have provided a good deal of information
regarding their operation, some of which we have incorporated in this write up. We are thankful to
the management of all such plants.
This page
intentionally left
blank
GLOSSARY
(Important Terms, Symbols, Abbreviations and Notes)
IMPORTANT TERMS
Decrepitation = Tendency to break into smaller sizes (degrade) generally on heating
and/or during reduction.
Free-board = Open space in the reactor above the solid charge.
Granular material = Solid material which is not very fine (i.e., not ground) and is in a
relatively close size range (e.g., 10 to 40 mm, 6 to 20 mm, 5 to 18 mm,
etc.).
Mole, Molar = Molecular weight expressed in weight measures (e.g., grams – g-moles;
kilograms – kg-moles).
Reducibility = The ease with which the ore or oxide gets reduced in presence of
(of iron ore) reducing gases – Described in Chapter 9.
Reactivity = The ease with which the coal or char regenerates carbon monoxide by
(of coal char) reaction with either CO2 (Carboxy–reactivity), O2 (Oxy-reactivity) or
H2O (Hydroxy-reactivity) – Described in Chapter 9.
Sponge Iron, Direct Reduced Iron (DRI)
= Iron produced by the reduction of iron oxide (ore) in solid state. Since
the process is accompanied by large reduction in volume (53%), a lot
of pores appear in the metallic product, which looks like sponge under
the microscope.
SYMBOLS
Δ = Denotes changes in properties (e.g. ΔG)
φ = Diameter
α = Angle of Repose
o
= Denotes Standard State (e.g. ΔGo); also denotes degree (temperature,
e.g., °C, °F, °K, °R)
φ = Filling Angle (Semi-vertical angle subtended by charge bed on the
kiln axis – Chapter 6)
ψ = Angle of Inclination (of kiln)
γ = Angle between surface of material and kiln axis (Zero for uniformly
loaded kiln)
G = Gibbs Free Energy
H = Enthalpy
P = Partial Pressure
x Glossary
S = Entropy
T = Temperature (Absolute – Degrees Kelvin or Rankine)
t = Residence Time
ABBREVIATIONS
ABC = After Burning (and Cooling) Chamber
ACAR = Allis Chalmers Agglomerated Reduction
ACCAR = Allis Chalmers Controlled Atmosphere Reduction
BASL = Bihar Alloy Steels Limited
BF = Blast Furnace
BIS = Bureau of Indian Standard
BOD = Biological Oxygen Demand
BOF = Basic Oxygen Furnace
C = Centigrade, Celsius (temperature)
Cal = Calories
Cfix = Fixed Carbon (also FC)
Cp = Heat Capacity at Constant Pressure (Specific Heat of one mole)
CFD = Computational Fluid Dynamics
Co. = Company
COD = Chemical Oxygen Demand
CODIR = Coal Ore Direct Iron Reduction (A sponge iron making process)
CONARC = Converter & Electric Arc composite furnace
CPT = Crossing Point Temperature
Daf = Dry and ash-free
Dia = Diameter
DIN = Deutsches Institute fur Normung e.v. (Standardisation institute for
Germany)
Dmf = Dry and mineral matter free
DNA = Deoxyribo-Nucleic Acid
DPR = Detailed Project Report
DR = Direct Reduction
DRC = M/s Direct Reduction Corporation
DRI = Direct Reduced Iron
DRPD = Direct Reduction Process Development Division (of RDCIS, SAIL)
EAF = Electric Arc Furnace
F = Fahrenheit (temperature)
FC = Fixed Carbon
Glossary xi
Fe(M), Fe(m) = Metallic Iron

Fe(T), Fe(t) = Total Iron
Finmet = Fine Ore Metallisation (A sponge iron making process developed from
(FIOR)
FIOR, Fior = Fine Iron Ore Reduction (A sponge iron making process using fluidised
bed)
FR = Feasibility Report
FY = Financial Year
GCal = Giga Calories (109 Calories)
HBI = Hot Briquetted Iron
HEC = M/s Heavy Engineering Corporation
HEG = M/s Hindustan Electro Graphite
HSL = M/s Hindustan Steel Limited
HyL = Hojalata ya Lamina, or Sheets and Plates (also a sponge iron making
process developed by M/s HyL, Mexico)
IDT = Initial Deformation Temperature
IF = Induction Furnace
IISCO = M/s Indian Iron and Steel Company Limited
IIT = Indian Institute of Technology
IS = Indian Standards (formulated by BIS)
IT, BHU = Institute of Technology, Banaras Hindu University (Varanasi, India)
JIS = Japanese Industrial Standard
JSPL = M/s Jindal Steel and Power Limited
JSW = M/s Jindal South West Limited
JVSL = M/s Jindal Vijayanagar Steel Limited (now JSW)
K = Kelvin (absolute temperature)
KCal, kCal = Kilocalories
KVA = Kilo-Volt Ampere
LOI = Loss on Ignition
m = metre, metres
mm = millimetre, millimetres
Midrex = A sponge iron making process developed by the then Midland Ross
Company
MVA = Mega Volt Ampere
NALCO = National Aluminium Company Limited
NMDC = National Mineral Development Corporation
NML = National Metallurgical Laboratory
xii Glossary
NOx = Oxides of nitrogen

OSIL = M/s Orissa Sponge Iron Limited
QRT = Quick Response Thermocouple
R = Rankine (absolute temperature)
RDCIS = Research and Development Centre for Iron and Steel (of SAIL)
RK = Rotary Kiln
rpm, RPM = Revolutions per minute
RRL = Regional Research Laboratory
R&D = Research and Development
SAF = Submerged Arc Furnace
SAIL = M/s Steel Authority of India Limited
SIIL = M/s Sponge Iron India Limited
SL/RN = Stelco, Lurgi, Republic Steel and National Lead (A sponge iron making
process)
SOx = Oxides of Sulphur
SPM = Suspended Particulate Matter
TDR = TISCO (or Tata) Direct Reduction (A sponge iron making process)
TISCO = M/s Tata Iron and Steel Company Limited (Tata Steel)
tpa/tpy = Tonnes per annum or year
tpd = Tonnes per day
UHP = Ultra High Power
VM = Volatile Matter
NOTES
1. For arithmetic operation *(Star) has been used to denote multiplication to avoid
confusion with alphabet x. Also / (slash) has been used to denote division to avoid
increasing line space.
2. Italics in text generally denote the reference in the Bibliography section.
3. CGS units have been used in the text and not SI units, as is the current trend. One
important conversion factor, which is required for linking SI and CGS systems is
that, one calorie equals 4.186 joules.
List of Figures
Fig. 1.1 Optical (left) and Scanning Electron (right) micrograph of a typical sponge iron sample 1
Fig. 1.2 Accidental smelting of iron in an anthill 2
Fig. 1.3 Some primitive iron making furnaces of India 3
Fig. 1.4 Primitive iron making practice of Orissa 3
Figs. 1.5 A view of the enigmatic Iron Pillar and a probable method of its making 4
& 1.6
Fig. 1.7 Layer-wise reduction of iron oxides in a rotary kiln 5
Fig. 1.8 A possible mechanism of reduction and pore formation during sponge iron
making in rotary kiln 7
Fig. 2.1 Key steps in sponge iron making in rotary kiln 10

Fig. 2.2 A concise schematic representation of a rotary kiln sponge iron plant 10
Fig. 2.3 A schematic of the sponge iron Pilot Plant of RDCIS, SAIL
Fig. 2.4 Essential features of a reduction kiln in a rotary kiln sponge iron plant 11
Fig. 2.5 Material balance in a rotary kiln sponge iron plant 12
Fig. 2.6 Energy balance in a conventional rotary kiln sponge iron plant 13
Fig. 2.7 A comparison of sponge iron production of India and World over the years 13
Fig. 2.8 Another comparison of sponge iron production of India and World indicating
the dominant position of coal based route in India 14
Fig. 2.9 Delicate balance of oxidising and reducing conditions in a sponge iron rotary kiln 16
Fig. 3.1 Ellingham diagram for a few key oxides 20
Fig. 4.1 Gas flow profile when air injection pipe is located along the kiln axis 26
Fig. 4.2 Gas flow profile when air injection pipe is located 50 mm above the kiln axis 26
Fig. 5.1 Schematic representation of the process of mathematical modelling 43
Fig. 6.1 Material movement in a rotary kiln 46

Fig. 6.2 Modes of transverse bed motion in a rotating cylinder 46
Fig. 6.3 Geometrical considerations of material movement in a rotary kiln 48
Fig. 6.4 Particle movement profile in a charge bed due to climbing and rolling 49
Fig. 6.5 Flow of material caused by disturbance 50
Fig. 6.6 Residence time plot for an experiment with fine iron ore as tracer covering 54
both rotary kiln and cooler
Fig. 6.7 Residence time plot for an experiment with fine iron ore as tracer in the rotary cooler 55
Fig. 6.8 Residence time plot obtained while using manganese ore as tracer 56
Fig. 6.9 A typical coal throwing arrangement 57
Fig. 6.10 Calculated distances of throw of coal particles under different conditions 59
xiv List of Figures
Fig. 7.1 Modes of heat transfer in a rotary kiln 62

Fig. 7.2 A simplified method of calculating requirement of air in different kiln segments 64
Fig. 8.1 Schematic diagram representing activation energy barrier and the effect of a catalyst 69
Fig. 8.2 Typical plots of degree of reduction with time at different temperatures 70
Fig. 8.3 Plot of –log (1–α) with time (Straight Lines denote First Order Reaction) 71
Fig. 8.4 Plot of log of velocity constant against reciprocal of absolute temperature 71
(Arrhenius plot)
Fig. 8.5 Reduced time plots for different orders of reaction 72
Fig. 8.6 A possible mechanism of pore formation during reduction of iron oxide 73
Fig. 9.1 A view of the Gorumahisani iron ore mines 77

(Shows high grade rocks associated with lower grade and clayey overburden)
Fig. 9.2 Apparatus for measuring reducibility of iron ore 78
Fig. 9.3 Gas flow diagram of the apparatus for the determination of reactivity of char 80
Fig. 9.4 Determination of Damköhler number 83

Fig. 9.5 Variation of reactivity with fuel ratio 84
(Data from Reuter & Serbent, 1976)
Fig. 9.6 Variation of rotary kiln reduction zone bed temperature with coal char reactivity 85
Fig. 9.7 Chart showing guidelines for coal selection 86
Fig. 10.1 Mechanism of initiation and growth of accretion 90

Fig. 10.2 FeO-SiO2-Al2O3 phase diagram 91
Fig. 10.3 Liquidus line in a typical ash-ore system 92
Fig. 11.1 Sketch of the experimental setup (Perspex Model) at RDCIS, SAIL 98
Fig. 11.2 Photograph of the perspex model 99
Fig. 11.3 Photograph of an experimental setup to study temperature rise during aqueous 99
corrosion of sponge iron
Fig. 11.4 Zones in a sponge iron cargo and stages leading to auto-ignition (Birks) 101
Fig. 11.5 Set up for determination of crossing point temperature 103
Fig. 12.1 Effect of burden metallisation on productivity and coke rate in blast furnace 109
Fig. 12.2 Schematic of a typical mini/midi steel plant of India 110
Fig. 13.1 Schematic of raw material handling and proportioning unit 114
Fig. 13.2 Schematic of a product handling system 115
Fig. 13.3 Laboratory rotary tube furnace 116
Fig. 13.4 Schematic representation of a slip ring gas sealing device 122
Fig. 13.5 Schematic of ACCAR process ported reactor 125
Fig. 13.6 An optimised rotary kiln sponge iron making process 128
Fig. 15.1 Steel consumption in present day economies 141

Fig. 15.2 World crude steel production 141
Fig. 15.3 Long term steel demand projection for India 142
Fig. 15.4 Sponge iron production over the years 144
(Charts depict the phenomenal growth of coal based sponge iron making in India)
Fig. 15.5 Future iron making 145
Fig. 15.6 Future scenario of outputs from coal washeries 147
List of Tables
Table 2.1 DRI Production: India and World 15
Table 3.1 Thermodynamic data of important reactions 21
Table 5.1 Chart for calculating sponge iron composition from iron ore composition 33
Table 5.2 Filled up chart of Table 5.1 34
Table 5.3 Products of heating 100 kg coal 36
Table 5.4 Estimation of iron ore which can be reduced by gases evolved from coal added 37
for generation of only protective char (Basis: 1000 kg or 1 tonne DRI)
Table 5.5 Estimation of iron ore which can be reduced by 100 kg of coal 38
Table 5.6 Estimation of coal heat available from coal for meeting heat requirement 39
(ef. Net calorific value)
Table 5.7 Alternatives considered for calculation along with their boundary conditions 40
Table 5.8 Calculation of heat available from the Combustion of 100 kg coal 40
under rotary kiln conditions for the four alternatives
Table 5.9 Estimation of heat demands of different alternatives 41
Table 6.1 Formulations for residence time in a rotary vessel 50

Table 6.2 Comparison of the experimentally determined residence time
(during iron ore reduction) with the predictions from the model 52
(i.e., formulation no. 6 of Table 6.1)
Table 12.1 Comparison of steel making in arc & induction furnaces 106
Table 13.1 Steps in process design & engineering 118

Table 13.2 Outline of preparation of a project feasibility report 120
Table 13.3 Comparison of the conventional and optimised process 130
Table 15.1 Carbon dioxide emissions in steel making 138

Table 15.2 Carbon dioxide emissions at different stages under Indian conditions 139
Table 15.3 Estimated average specific arisings from Indian coal based sponge iron plants 139
Table 15.4 Comparison of solid arisings from blast furnaces vis-a-vis
rotary kiln sponge iron plants 140
Table 15.5 Comparison of pollutants in exqust gases in blast furnace iron making 140
and coal based rotary kiln sponge iron plants
Table 15.6 Likely Indian steel making scenario in mid twenty first century 143
This page
intentionally left
blank
Contents
Preface to the Second Edition v

Preface to the First Edition vii
Acknowledgement ix
Glossary xi
List of Figures xv
List of Tables xvii
CHAPTER 1
INTRODUCTION 1–8
1.1 What is Sponge Iron and What is DRI? 1
1.2 History of Modern Sponge Iron Making 6
1.3 Iron Ore Reduction at a Glance 6
1.4 Multiple Roles of a Rotary Kiln 8
CHAPTER 2
ROTARY KILN PROCESS OF MAKING SPONGE IRON 9–18
2.1 Historical Background 9
2.2 Important Features 9
2.3 Sponge Iron Pilot Plant of RDCIS SAIL 10
2.4 Features of a Rotary Kiln Sponge Iron Plant 12
2.5 The Indian Scene 13
2.6 Why should we select a Rotary Kiln? 15
2.6.1 Process Strengths 16
2.6.2 Product Strengths 17
2.6.3 Weaknesses of the Process 17
2.6.4 Weaknesses of the Product 17
xviii Contents
CHAPTER 3
THERMODYNAMIC CONSIDERATIONS: FEASIBILITY OF REACTION 19–22
CHAPTER 4
AERODYNAMICS INSIDE A SPONGE IRON ROTARY KILN 23–29
4.1 Thumb Rules Used in Production of Sponge Iron 23
4.2 Sources of Gas 24
4.3 Effects of Air Injection 24
4.4 Limits to Gas Velocity 25
4.5 Gas Flow Pattern 25
4.6 CFD Analysis 25
CHAPTER 5
MATHEMATICAL MODELLING IN ROTARY KILN SPONGE IRON MAKING 31–44
5.1 What is a Model? 31
5.2 What is a Mathematical Model? 32
5.3 How can we make a Useful Mathematical Model? 32
5.4 Example of a Small Mathematical Model 32
5.5 Role of Coal in Rotary Kiln Sponge Iron Making Process 34
5.6 Quantity of Protective Char 35
5.7 Reduction of Iron Oxide 35
5.8 Coal for Combustion 35
5.9 Waste Gas Temperature 42
5.10 Programming Based on Model 42
5.11 Spread-Sheet Programming 42
5.12 Segment-wise Modelling 42
5.13 Prediction from the Model 43
5.14 Summary of the Modelling Process 44
CHAPTER 6
PHYSICAL MOVEMENT OF SOLIDS INSIDE A ROTARY KILN: CHARGE 45–60
MOVEMENT AND COAL THROWING/SLINGING
6.1 Importance of Residence Time 45
6.2 Estimating Gross Residence Time from Input and Output of Solids 45
Contents xix
6.3 Charge Movement: Cascading Flow of Granular Material 46

6.4 Mathematical Treatment of Charge Movement 48
6.4.1 Testing of the Mathematical Expression and its Amendment 50
6.4.2 Bed Disturbance Factor 50
6.5 Importance of Filling Degree 51
6.6 Techniques of Measuring Residence Time 53
6.6.1 Measurement of Residence Time in the Pilot Rotary Kiln of RDCIS SAIL 53
(a) Physical Tracers 54
(b) Chemical Tracers 54
6.7 Importance of Throwing Coal from Discharge End 56
6.8 Coal Throwing Philosophy 59
CHAPTER 7
REQUIREMENT, GENERATION AND TRANSFER OF HEAT IN 61–65
A SPONGE IRON ROTARY KILN
7.1 Material and Energy Balance in a Conventional Rotary Kiln Sponge Iron 61
Making Process
7.2 Generation and Transfer of Heat 61
7.3 Mathematical Treatment of Heat Transfer 62
7.4 Segment-wise Air Requirement 64
CHAPTER 8
REACTION KINETICS 67–73
8.1 Factors Affecting Reaction Rates 67
8.2 Rate Law and Order of Reaction 68
8.3 Birth of the Rate Law or Law of Mass Action 68
8.4 Unimolecular Reaction and First Order Reaction 68
8.5 Decomposition of an Oxide 69
8.5.1 Degree of Reduction 70
8.6 Temperature Dependence of Velocity Constant: The Arrhenius Equation 70
8.7 Experimental Determination of Activation Energy 70
8.8 Variation of Reaction Rate with Temperature 71
8.9 Role of Diffusion 72
8.10 How Sponge Iron Making is a First Order Reaction? 73
xx Contents
CHAPTER 9
RAW MATERIALS FOR SPONGE IRON MAKING 75–87
9.1 Iron Oxide Feed 75
9.2 Iron Ore Lumps 75
9.3 Magnetic and Hematitic Ores 76
9.4 Magnetic Ore Minerals 76
9.5 Laboratory Test on (Method of Determining) Iron Ore Reducibility 77
9.6 Coal for Sponge Iron Making 78
9.7 Peat 78
9.8 Conversion of Peat into Higher Ranks 79
9.9 Indian Coal Deposits 79
9.10 Size of Coal for Sponge Iron Making 79
9.11 Role of Coal in a Sponge Iron Rotary Kiln 79
9.12 Measurement of Reactivity of Coal Char 81
9.13 Gas Analysis Method of Reactivity Measurement 81
9.13.1 Principle 82
9.13.2 Test Conditions 82
9.13.3 Test Apparatus 83
9.13.4 Test Procedure 83
9.14 Indicators to Reactivity 85
9.15 Other Criteria for Coal Selection 87
CHAPTER 10
ACCRETION OR RING FORMATION INSIDE A ROTARY KILN 89–93
10.1 Can Ring Formation be Avoided? 89
10.2 Deposit Formation Through Solid State Sintering 90
10.3 Reasons for Catastrophic Growth of Deposits 91
10.4 Precautions Required 92
10.5 Accretion Removal During Operation 92
CHAPTER 11
SPONGE IRON PROPERTIES: 95–104
RE-OXIDATION AND AUTO-IGNITION OF SPONGE IRON
11.1 Reasons for Re-oxidation Tendency 95
11.2 History of Auto-Ignition 96
Contents xxi
11.3 First Reported Incidence 96

11.4 Eye Witness Report 97
11.5 Investigations Related to Auto-Ignition 98
11.6 Other Significant Results 100
11.7 Clancey’s Hypothesis 100
11.8 Hypothesis of Neil Birks 100
11.9 Ignition Temperature of Sponge Iron: Crossing Point 102
11.10 How Valid is this Hypothesis? 103
11.11 Precautions Needed in Sponge Iron Storage 104
11.12 Conclusions 104
CHAPTER 12
USES OF SPONGE IRON 105–111
12.1 Use of Sponge Iron in Electric Steel Making 106
12.2 Use of Sponge Iron in Submerged Arc Furnace 107
12.3 Use of Sponge Iron in Iron Making 108
12.4 Use of Sponge Iron in Cupola 109
12.5 Use of Sponge Iron in Oxygen Steel Making 109
12.6 Sponge Iron and Hot Metal: Partners in Progress 110
CHAPTER 13
PROCESS DESIGN, ENGINEERING AND OPERATIONAL 113–131
ASPECTS OF AN RK-DR PLANT
13.1 Process Design and Engineering 113
13.1.1 Raw Material Testing 114
13.1.2 Upscaling 116
13.1.3 Sizing of Key Equipment 117
13.1.4 Process and Material Flow Diagrams 117
13.1.5 Preparation of List of Major Equipment 117
13.1.6 Preparation of Layout 117
13.2 Project Engineering 117
13.3 Installation, Commissioning and Project Cost 119
13.4 Process Enhancing Equipment and Concepts 119
13.4.1 Gas Sealing System 119
13.4.2 Underbed Hydrocarbon Injection System—The Ported Kiln Concept 122
13.4.3 Coarse Coal Throwing or Slinging System 123
xxii Contents
13.4.4 Waste Heat Recovery System— Generation of Electrical Power 124

13.4.5 Use of Waste Gas Energy for Preheating 125
13.4.6 An Optimised Rotary Kiln Sponge Iron Making Process 126
13.4.7 Pros and Cons of the Optimised Concept 129
13.5 Kiln Refractories 129
CHAPTER 14
OTHER USES OF ROTARY KILN FOR REDUCTION PURPOSES 133–135
14.1 Magnetising Roasting of Lean Iron Ores 133
14.2 Pre-reduction of Chromite Ore 133
14.3 Pre-reduction of Manganese Ore 134
14.4 Pre-reduction of Titaniferrous Ore 134
14.5 Pre-reduction of Nickeli ferrous Overburden 134
14.6 Use of Rotary Kiln for Gasification of Coal 135
CHAPTER 15
ENVIRONMENTAL ASPECTS OF SPONGE IRON MAKING IN 137–148
ROTARY KILN AND FUTURE PROSPECTS
15.1 Carbon Dioxide Emissions from the Sponge Iron Industry 137
15.2 Other Pollutants and Arisings in Sponge Iron and Steel Industries 138
15.3 Sponge Iron in India 139
15.4 Projections of Steel Consumption 140
15.5 Projections for India 141
15.6 Possible Sources of Additional Iron 141
15.7 Role of Sponge Iron and Newer Processes 142
15.8 Rotary Kiln Sponge Iron Making: Status and Prospects 143
15.9 Use of Sponge Iron in Blast Furnaces 143
15.10 Smelting Reduction in India 144
15.11 Modern Trends in Iron Making 145
15.12 A Peep into the Iron and Steel Making Practice of Twenty Second Century 146
15.13 What Makes Rotary Kiln Sponge Iron Making Click in India? 146
15.14 Another Opportunity for Rotary Kiln Sponge Iron Making? 147
Bibliography 149–155
Index 157–163
1
Introduction CHAPTER
1.1 WHAT IS SPONGE IRON AND form. Since there is no melting, block volume
WHAT IS DRI? remains practically unchanged. As weight is
Sponge iron, also called Direct Reduced Iron reduced and true density increases, true
(DRI), is formed when an oxide of iron is volume is reduced considerably. These
reduced to its metallic form below the melting opposing factors are reconciled by the
points of the metal and the oxide. Naturally, formation of a lot of microscopic pores, and
the shape of oxide is retained in the reduced under the microscope it looks like sponge and
Fig. 1.1 Optical macrograph (left) and Scanning electron micrograph (right) of a
typical sponge iron sample
(Courtsey Prof. R.C. Gupta, IT-BHU)
2 // Advances in Rotary Kiln Sponge Iron Plant
hence its name (Fig. 1.1). Unlike the name, it extraction was probably an accident, but the
is not at all soft to the touch, but a well- art was later refined to make high-grade iron,
prepared sponge iron would be much softer almost pure, followed by solid-state
under the hammer than normal iron or steel. carburisation (cementation – this is also the
The terms ‘Direct Reduced Iron’ and ‘Direct principle behind the case hardening process
Reduction (DR)’ have a historical background. of the presen+t day). The steel produced
In the present context, DR signifies reduction were of finest grades, the artisans were able
without melting, which may appear odd to us to control the chemistry (without knowing
now, but it is not so if we look through the eyes the chemistry) and heat treatment with
of the earlier (primitive to us) iron and steel precision.
makers. A story is being told by the indigenous
Sponge iron was the first extracted iron people of the Bastar region in central India
which man had produced (Fig. 1.2). The first according to which iron was smelted first
Fig. 1.2 Accidental smelting of iron in an anthill

Introduction // 3
accidentally in an anthill. To catch a rat which opening at the base, in a direction where wind
has taken refuse inside the hollow interior of was favourable, to blow in the fire he made.
an anthill with a damaged top (it also housed The suffocating rat tried to come out and was
a tree stub) a hunter covered the top with a caught. But overnight the favourable wind
red rock (!) using dry leaves around to seal intensified the fire and next day an object was
all openings at the top. He then made an found which was much better for their
Fig. 1.3 Some primitive iron making furnaces of India
Fig. 1.4 Primitive iron making practice of Orissa

hunting weapons than anything they had

used before. (Fig. 1.2–On the left is the
original sketch of the authors. The right half
is as it has been reproduced in the book
‘Ferrum Hunters’ by Das & Ray–2005).
When man could find means of attaining
higher temperatures, he could make much
more iron from furnaces of same capacity
through the liquid iron route. In this process
much more carbon was picked up, which had
to be removed during further processing to
make steel, and this obviously was an indirect
process route compared to the traditional
method of making a pure form of iron and
carburising it to make steel. By comparison
the traditional method come to be known as
the Direct Reduction (DR) process and the
iron produced by reduction in solid state as
DRI (Sponge Iron).
Sponge iron was thus originally produced
Fig. 1.5 using coal or charcoal in a mud shaft
Fig. 1.6 A view of the enigmatic Iron Pillar and a probable method of its making
Introduction // 5
(Figs. 1.3 and 1.4) and this turned out to be the contained either no manganese or in
forerunner of the modern day blast furnace traces only.
for producing liquid iron. 3. As there was no control over
India has many iron based ancient phosphorous in the process, high
monuments with remarkable properties. The phosphorous in the iron ore led to
most famous of these is the Iron Pillar of Delhi high phosphorous wrought iron.
(Fig. 1.5). Iron for all these monuments and
Probably such an ore was deliberately
artefacts were extracted through the sponge
chosen for the Delhi Pillar.
iron route – made in mud- or clay-shaft
furnace. Some of the indigenous people in 4. The Iron Pillar of Delhi was probably
Chhattisgarh and Jharkhand States are still made by making high carbon ‘blooms’
practising this technique. Studies of the successively one over the other, and
products made by these people, as well as beating the freshly formed ‘bloom’ on
study of ancient heritage (Delhi Pillar being to the older layer–while keeping a
one), have revealed some points of importance. split hollow cylindrical stone die
1. It was easy to make very low carbon around (Fig. 1.6). This way the bloom
wrought iron by such a process. While layers could be effectively welded to
higher carbon iron could be attempted, each outer, but left the surface
control of carbon was found difficult. extremely rough. The erect iron mass
Large variation of carbon between was then taken down, heated and
subsequent heats was observed. rotated on a manually driven stone
lathe and forged between stone dies –
CO + 1 O2 (from Air) = CO2 (in the Free Board)
2 much like the modern forging of
CO2 + C = 2CO (in
( the Bed)) railroad axles. During this forging
2
CO
CO2 process the surface got heavily decar-
Fe (m)
burised while the interior remained
Partially reduced relatively high in carbon.
5. Studies conducted by Prof.
Balasubramaniam of IIT Kanpur have
indicated that a combination of
Fe2O3
extremely low manganese and high
phosphorous has contributed to the
excellent corrosion resistance of the
Iron Pillar. Too high a carbon level
would have had an adverse effect on
this property.
But, now a days, it is found commercially
FeO + CO = Fe + CO2 feasible to produce sponge iron in a shaft by
Fig. 1.7 Layer-wise reduction of iron oxides in using reducing gases produced in a separate
a rotary kiln reactor. On the other hand, sponge iron
2. All the samples examined by us as production with direct use of coal is effected
well as those reported in literature predominantly in rotary reactor or kiln, which
is our present focus. The reduction process Second World Wars is an example of one such
in a rotary kiln is also used, or is feasible, for attempt which had met with partial success.
a few other oxides or ores which have been Heavy deposit formation could not be
touched upon later. prevented and the process had to be
abandoned on this account.
1.2 HISTORY OF MODERN SPONGE When rotary kiln processes could not
IRON MAKING come up to the expectation to meet the iron
Sponge iron making played a dominant role needs of the industry, attempts were made
in the production of metallic iron before 1800 to reduce iron oxide in shafts using gases
A.D., but could not compete with blast generated externally. In the Wieberg
furnace technology in terms of efficiency and Soderfors process, attempt was made to
economy. However, in recent times, generate reducing gas through the
constraints relating to raw materials and gasification of coke using external energy. But
intermediate inputs to steel making have re- the first truly viable shaft process of sponge
stimulated interest in the development of iron making was developed by M/s Hojalata
ya Lamina (meaning Sheets and Plates) which
technically sound and economically viable
reduced iron ore pellets in stationary bed
high tonnage sponge iron processes. As a
using externally reformed natural gas (HyL
result, various processes have been
process). Later on they converted the set of
developed, tested and adopted throughout
stationary beds to a moving bed reactor, but
the world and presently sponge iron accounts
Midland Ross Company did the same earlier
for over 8% of the total iron production in and got a head start in sponge iron making
the world. Although the share of coal-based market and gained a lead, which they are
variant is less than 2%, the upward trend, maintaining even today (Midrex process). But
especially in India, is very promising. rotary kiln was the only choice in the regions,
Iron metal and its oxides have melting where natural gas availability was restricted.
points close to each other and all are about Consequently, sponge iron making in rotary
1500°C or above. However, there is a tendency kiln flourished first in South Africa, and now
for formation of ‘clusters,’ ‘agglomerates’ and India produces over 70% of coal based sponge
‘accretion’ or ‘ring formation’ during actual iron in the world.
manufacture of sponge iron in the It is very easy to reduce the higher oxides
temperature range of 900 to 1100°C. Ring of iron to FeO stage. One only needs to meet
formation is a phenomenon occurring the heat demand, the reductant requirement
exclusively in a rotary kiln, while clusters and and the temperature level. The key step in
agglomerates are common to both rotary kiln all DR processes is the reduction of FeO to
and shaft processes. In the present context metallic iron stage.
we cannot conceive of a reactor to
manufacture sponge iron, which operates 1.3 IRON ORE REDUCTION AT A
above 1100oC, even though there have been GLANCE
prolonged attempts to produce sponge iron If we expose an iron oxide pellet or lump to
and semi-fused iron at higher temperatures. reducing gases such as CO and H 2 at an
The Krupp Renn process of production of iron adequate temperature, sponge iron is the
“luppen” in Germany between First and product. If we cover iron oxide with coal or
Introduction // 7
charcoal and heat in a muffle furnace, the coal, an inclined rotary furnace was found
result is same. One may be inclined to think handy and convenient.
that physical contact with carbon should cause Figure 1.7 schematically represents how
reduction, but the interior of iron oxide lump iron ore pellet – or for that matter a lump – is
or pellet also gets reduced, which is not in gradually reduced in layers when exposed
contact with carbon. Further, if we try to to rotary kiln conditions. A pellet of
carry out the same exercise under satisfactory metallisation would not have any
progressively reduced pressures, the rate of core of Fe2O3 left in it, while the partially
reduction progressively reduces. It is obvious, reduced layer would be restricted to a very
therefore, that reduction is effected by limited area in the core.
reducing gases, even if we keep iron oxide In Fig. 1.8 we have tried to present
and coal in contact. For facilitating quicker pictorially the mechanism of reduction and
regeneration of these reducing gases from pore formation during sponge iron making.
Fig. 1.8 A possible mechanism of reduction and pore

formation during sponge iron making in rotary kiln
1.4 MULTIPLE ROLE OF A ROTARY of smaller diameter (i.e., of higher L/D ratio)
KILN since heat transfer from flame to the solid
The use of rotary kiln is limited not only to charge is the main requirement. Because of
produce DRI but it is most commonly used its versatile usage in processing materials for
for calcination of lime and refractory material, drying, heating, calcining, agglomerating,
production of cement and also for pre- sintering, reducing, melting, etc., the rotary
treatment (pyro-metallurgical beneficiation) kiln will continue to play an important role
of some non-ferrous ores and waste oxides. in different industries in future too.
These kilns are comparatively longer and
2
Rotary Kiln Process
of Making Sponge Iron CHAPTER
2.1 HISTORICAL BACKGROUND process concepts have been emerging in this

The production of steel began in ancient area, there were rapid births and deaths of
times; but because of the complexity and slow these processes and process concepts in the
speed of the ancient process, they could not middle of the twentieth century. But those
be carried out on a very large scale. Conse- operating successfully at present have many
quently, they were replaced by the high features in common. Some of the common or
production rate ‘indirect process,’ and the slightly differing features are:
development of modern DR Process did not (i) System of sealing to prevent air ingress
begin until the middle of 19th century. into the reactor,
Perhaps the very first patent in U.K. for (ii) System of throwing or slinging coal
from discharge end of reactor,
sponge iron making was in 1792 presumably
using a rotary kiln. More than 100 DR (iii) System of weigh feeding and
proportioning of raw materials
processes have been invented and operated
since 1920. Most of these have died down. But (iv) System of introducing controlled
amount of air at regular intervals of
some of them have re-emerged in slightly
length in such a way that it does not
different form. oxidise the reduced product in the bed,
As touched upon earlier, sponge iron is (v) System of temperature sensing at
mainly produced from ore by two different regular intervals of length of the
routes – (a) by reducing gases (CO and H2) reactor and recording of the same.
in a shaft furnace, and (b) through direct (vi) System of indirect cooling of sponge
treatment with coal in a rotary kiln. iron-char mixture in a rotary steel
cylindrical shell using water from the
2.2 IMPORTANT FEATURES outside.
The coal based rotary kiln process of making (vii) System of treating waste gases and
sponge iron is the focus of the present write maintaining desired flow profile
up. Although many different processes and through pressure control.
A typical process scheme for making

sponge iron in a rotary kiln is presented in
Fig. 2.1. While Fig. 2.1 shows only the key
steps, a more detailed scheme, as it would
appear for a typical operating plant, is
presented in Fig. 2.2.
2.3 SPONGE IRON PILOT PLANT OF

RDCIS SAIL
The sponge iron Pilot Plant (SIPP) of RDCIS,
SAIL, which would be mentioned a number
of times, was set up in 1980-82 with almost
all the features of a commercial rotary kiln
sponge iron plant. The know-how status
was, however, slightly different at that
Total time of materials in the rotary kiln time. Figure 2.3 represents more appro-
(Residence time)
priately the SIPP of RDCIS SAIL.
Fig. 2.1 Key steps in sponge iron
making in rotary kiln
Dry ESP/
bag filter
ID fan
Ash
Stack
Waste gas
Iron ore
coal
flux
After
Fine Shell-mounted
burning &
coal air fans cooling
n chamber
tion kil
Reduc Inlet hood
Air Water spray

Dust
Dust
blower
Cooler
Product
Dust
Magnetic separator
Fig. 2.2 A concise schematic representation of a rotary kiln sponge iron plant
Rotary Kiln Process of Making Sponge Iron // 11
1 1 1 1 1 Waste gas
5
Air
8 9 11
2 6 6 6 10
][ ][ ][
12
Water 4
3
13 7 7 7 7 7
1. Raw material bins 4. Surge bin 7. Product storage bins 10. Induced draught fan 13. Cooler
2. Belt conveyor 5. Vibratory screen 8. After burning chamber 11. Waste gas stack
3. Bucket elevator 6. Magnetic separator 9. Radial flow scrubber 12. Rotary kiln
Fig. 2.3 A schematic of the sponge iron Pilot Plant of RDCIS, SAIL
This Plant of capacity 5 to 9 tonnes per making. When operated with a good quality
day of sponge iron was commissioned in coal, metallisation level was consistently
March 1982 within the premises of M/s HEC above 90%. On the other hand three of the
at Ranchi, with the objective of adapting and coals tested in Pilot Plant were either not
assimilating coal based sponge iron suitable or were only marginally acceptable.
technology in India. The Pilot Plant was in In such cases obviously metallisation levels
regular operation since its commissioning till were reduced – in extreme cases upto 70%.
1992-93, with 4 to 5 Campaigns each year. Other major results include:
48 campaigns were carried out in the • Development of ore-coal composite
Pilot Plant with various ore and coal pellets technology, which improves
combinations from different deposits in the kiln productivity and reduces energy
country. The two longest campaigns lasted consumption (patented)
62 days each. A total of 26 ore-coal • Pre-heating system of ore (patented)
combinations were processed in the Pilot • Simultaneous injection of under-bed
Plant. Most of the iron ores tested in Pilot hydrocarbon fuel and over-bed air in a
Plant were found suitable for sponge iron rotary reactor
Air
Waste gas
Iron ore
Fine coal tubes Coal

Air
Rotary kiln Flux
Sponge
iron
Char ples
Thermocou
Fig. 2.4 Essential features of a reduction kiln in a rotary kiln sponge iron plant
Fig. 2.5 Material balance in a rotary kiln sponge iron plant
2.4 FEATURES OF A ROTARY KILN gas and conversion cost at similar capacities
SPONGE IRON PLANT are higher. But even then, gas-based
Figure 2.4 indicates the essential features processes have generally found favour due
which are needed in the reduction kiln of a to better and more consistent quality,
sponge iron rotary kiln plant. However, the lower energy consumption and higher
air tubes indicated in this diagram can be module size. It was realised early that
substituted by a ported kiln design, which is rotary kiln processes can be up-scaled only
discussed later in this book. to a limited extent and bigger module size
A typical material balance for the sponge does not mean a higher economy of scale.
iron making process is presented in Fig. 2.5. Modules bigger than 500 tpd were
Here coal is assumed to contain about 20% continuously plagued by problems of
ash, something which is hardly available now accretion formation and were maintenance
a days. Product is shown to be screened into prone. All such modules have now been
three fractions. But due to difficulty in phased out.
screening the -1 mm fraction, it is usual now India has been the largest producer of
a days not to separate out this fraction. Use sponge iron since last few years. Its
of 6 to 20 mm iron ore is indicated. Presently contribution to world DRI production is in
it is more common to use 5 to 18 mm fraction. excess of 25% at present. India has been
A typical energy balance in the form of increasing the gap with the next country
Sankey diagram is presented in Fig. 2.6. Venezuela. Trends indicate that this gap
Rotary kiln processes have had to would continue to increase in the foreseeable
compete with gas-based processes. Gas future. Iran is third on the line and Mexico,
based processes use relatively costlier who was once the world leader, is presently
input such as pellets and reformed natural placed fourth. One thing may be noted though
Input energy
6.0, coal
2.25,
1.7
Waste
Sponge
gases
1.15, Char iron
(Chemical
(Chemical (Chemical
energy +
energy)* energy)
Sensible heat)
0.3, Cooling
losses
(Sensible heat in
solid product)
Unit GCal = 109 calories 0.6, Radiative &

Unaccounted losses
Fig. 2.6 Energy balance in a conventional rotary kiln sponge iron plant
that the entire production by number 2, 3 and Figure 2.8 presents the scene with respect to
4 countries is through gas based route while coal and gas based processes.
in India, more than half the production is by 60 30%
the coal based route.
World DRI
50
2.5 THE INDIAN SCENE production
DRI production million tonnes
Share of India's production

Situation in India is different from the rest of
40 20%
the world. Local conditions here have
favoured coal based rotary kiln units and
presently India has more such modules than 30
India’s
rest of the world put together. Over 300
share, %
modules are presently in production in India.
20 10%
And there is another trend in India of
downscaling. Only in India it has been found DRI
Production
profitable to operate 100 tpd and 50 tpd 20 in India
module (even 25 tpd modules), while else-
where 250 tpd module is considered as the 0 0%
minimum economic size. Table 2.1 and Fig. 2.7 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
present the production figures of sponge iron Year
in India vis-à-vis world, over the years. Fig. 2.7 A comparison of sponge iron
production of India and World over the years
59.80
Sponge iron production in the year 2006,
million tonnes
60.00 –
49.10
50.00 –
40.00 –
India
30.00 – World
20.00 – 15.00 10.70

10.00 –
9.50
5.50 World
0.00 –
Total India
Coal based
Gas based
Fig. 2.8 Another comparison of sponge iron production of India and World indicating
the dominant position of coal based route in India
The Krupp-Renn process, about which in the latter, in spite of the fact that a 2 mtpa
we mentioned in chapter 1, was probably gas based module has been commissioned
the last of the rotary kiln processes, which recently in western India.
attempted to produce iron in fused or As mentioned earlier, nearly 35% of
semi-fused mass. All of the current
India’s sponge iron production is accounted
processes attempt to prevent any type of
for by the three gas based plants located near
fusion during production.
the western coast. M/s Essar Steel Limited
The processes, which are currently in
located in Hazira in the state of Gujarat has
vogue, are Jindal Process (50, 350 and 500 tpd),
five operating modules with a total capacity
SIIL Process (100 tpd), OSIL Process (300 and
of 5.5 mtpy and claim to be the largest sponge
500 tpd), SL/RN (Lurgi) Process (100 and 500
iron plant in the world. Ispat Industries
tpd), Krupp-Codir Process (400 and 500 tpd),
DRC Process (250 to 350 tpd) and TDR Process Limited and Vikram Ispat Limited both
(400 tpd). OSIL Process has evolved from the located in the state of Maharashtra have one
ported kiln concept of ACCAR Process, while module each of capacities 1.4 and 0.9 mtpy.
SIIL Process has been based on the Lurgi or On the coal based front, the plant of M/s
SL/RN Process. Presently, of the total sponge Jindal Steel and Power Limited (JSPL) in
iron produced in the world, coal based rotary Raigarh in the central Indian state of
kiln processes contribute only about 15%, but Chhattisgarh is the largest in India and
considering India alone this percentage is probably in the world. It has ten modules
about 65%. Trends point to substantial increase totalling 1.2 mtpy capacity.
Table 2.1 DRI Production: India and World* Thus we see that the largest coal based
Million Tonnes plant in the world barely stands up to the
smallest gas based module in India. But then
Year India World
the coal based plants make it up in numbers.
1970 0 0.79 As mentioned earlier, over 300 modules of
1975 0 2.81 coal based plants are operating in India. And
1978 0 5.00 situation is changing so fast, almost on daily
1979 0 6.64 basis, that not much point is served by
describing these plants here.
1980 0.01 7.14
Apart from JSPL, some of the plants
1981 0.02 7.92
which operate large size modules (300 to 500
1982 0.03 7.28
tpd) are Bihar Sponge Iron in Jharkhand,
1983 0.04 7.90 Prakash Industries, Nova Iron and Steel,
1984 0.08 9.34 Monnet Ispat, Godavari Ispat and Power all
1985 0.09 11.17 in Chhattisgarh, Sunflag Iron and Steel
1986 0.17 12.53 Company, Lloyds Metals and Engineers, in
Maharashtra, Orissa Sponge Iron, Tata
1987 0.19 13.52
Sponge Iron, in Orissa, GSAL India in Andhra
1988 0.19 14.09
Pradesh, etc. We must make special mention
1989 0.26 15.63 of Sponge Iron India Limited in Paloncha in
1990 0.61 17.68 Andhra Pradesh, which was the first of the
1991 1.15 19.32 commercial plants (originally called a
1992 1.44 20.51 demonstration plant) commissioned in 1980,
1993 2.21 23.65
and which started the race for the installation
of the 100 tpd modules, the total number now
1994 3.12 27.37
being well over one hundred. 50 tpd modules
1995 4.28 30.67
could also have exceeded 100 in number. But
1996 4.84 33.30 another special mention must be made of the
1997 5.26 36.19 25 tpd modules. One is operating in Ramgarh
1998 5.26 36.96 in Jharkhand (Palash Sponge Iron), while at
1999 5.22 38.59
least two modules are operating in Raipur in
Chhattisgarh.
2000 5.44 43.78
2001 5.59 40.51 2.6 WHY SHOULD WE SELECT A
2002 6.59 45.10 ROTARY KILN?
2003 7.67 49.45 The rotary kiln direct reduction (RKDR)
2004 9.37 54.60 processes have been looked upon with
2005 11.10 55.96 apprehension, mainly because there have been
rapid births and deaths of processes in this
2006 15.00 59.80
group. But the fact that it has re-emerged,
*Data taken mainly direct from Midrex
Air (4N2+O2) introduced

CO + CO2 + H2 + H2O + N2 through air tubes
FeO + CO Fe + CO2
↑
CO + ½O = CO2 CO2 +C = CO + CO2
[Partial]
Fig. 2.9 Delicate balance of oxidising and reducing conditions in a

sponge iron rotary kiln
points to certain strengths of this process. Let (ii) As a large freeboard volume is
us examine some of them. available above the solid charge (about
85%), the rotary kiln can tolerate
2.6.1 Process Strengths heavily dust-laden gas. When the kiln
Rotary kiln process has to compete mainly is suitably designed, it would be best
with the shaft process of making sponge iron suited for utilising the Indian high ash
and in some cases with iron making blast non-cooking coals. In shaft reactors,
furnace. As compared to them, the rotary kiln generation of such dust leads to
has some advantages, as also some limitations, choking and channelling which leads
both with respect to the process and the finally to disruption of the process.
product it makes. The major process (iii) Rotary kiln can serve the dual purpose
strengths of rotary kiln are: of a coal gasifier as well as an ore
(i) A rotary kiln can mix the solid charge reducer. Preparation of reducing gas
as it heats and reduces it. Simultaneous from coal is an expensive step, which
mixing helps in the dilution of CO2 is coming in the way of commer-
concentration formed around the iron cialisation of coal gasification based
ore/sponge iron particles – which is DR process. Therefore, rotary kiln DR
necessary for the reduction reaction process has proved commercially
to proceed. viable, even with low productivity per
unit volume, because of this capability 2.6.3 Weaknesses of the Process

to perform two different functions Notwithstanding the above, rotary kiln has
simultaneously. a number of weaknesses. These are coming
Figure 2.9 schematically represents the in the way of its wide acceptability. The main
situation inside a rotary kiln where a delicate process related weaknesses of rotary kiln are:
balance of reducing zone within the
(i) It has very low productivity. Shaft
chargebed and an oxidising zone in the
furnaces, which make sponge iron,
freeboard is always maintained.
give upto five times more output than
(iv) In comparison to blast furnace, the rotary kilns of same inner volume.
temperature of reduction of iron oxide Productivity in rotary kiln is
is much lower in rotary kiln (about
consequently much lower.
1000oC as against 1500 to 2000oC in blast
(ii) The rotating reactor makes it difficult
furnace). This means that much less
to incorporate process control and
energy is required for bringing the
quality control systems. Energy saving
reactants to the temperature of reaction.
measures, such as use of pre-heated
2.6.2 Product Strengths air, are difficult to incorporate. To
Additionally the strengths of the product prevent ingress of atmospheric air an
made by rotary kiln are: elaborate sealing system is required,
which has made the reactor very
(i) It is easy to desulphurise iron ore while “engineering intensive”.
making sponge iron. Consequently the
(iii) The RKDR process has low energy
sponge iron of much lower sulphur
efficiency. The stored energy in
content can be produced as compared
to blast furnace hot metal. For shaft sponge iron is about 1.7 GCal per
process of sponge iron making, prior tonne, while energy usually spent in
and meticulous de-sulphurisation of making it in rotary kiln is about 6
natural gas is necessary to prevent GCal per tonne. Among other things,
poisoning of catalyst used for a lot of energy goes out in waste gases
reforming. (over 2 GCal per tonne).
(ii) Sponge iron produced from rotary kiln (iv) The RKDR process produces some
is obtained in close granular size sponge iron in fine form (-3 mm) which
range. This permits charging in electric is a little difficult to utilise in electric
or other steel making furnaces in a furnaces. While much of the fines are
continuous manner, obviating the need generated due to the nature of ore
for opening and closing of roof. used, the situation is aggravated by
Continuous charging permits partial the tumcbling action within the rotary
refining during melting stage as the kiln, which forces softer particles to
particle passes through the slag layer break down further.
into the mixed layer. If adequate
melting energy is available, refining 2.6.4 Weaknesses of the Product
time, and consequently, operation time In addition the sponge iron made by rotary
can be considerably reduced. kiln has the following weaknesses:
(i) For charging in electric furnaces in rotary kiln sponge iron is much less
substantial quantities, a system of susceptible to re-oxidation as
continuous charging needs to be compared to sponge iron from shaft
installed. This would mean an units using reformed gases.
additional investment for the existing Those who have ventured into sponge
units, which are not having this iron have to endeavour to exploit the
facility. strengths of RKDR to the fullest extent and
(ii) The sponge iron from rotary kiln has would have to try to mitigate the effects of
much lower carbon content (usually its weaknesses suitably. Those who
0.2%) than either the sponge iron from contemplate venturing into sponge iron have
shaft furnace (0.7 to 2%) or the hot to make a thorough analysis as to whether
metal from blast furnace. Carbon in the strengths outweigh the disadvantages or
sponge iron not only helps in adding to not in the scenario they are finding themselves
the opening carbon in molten bath, it in. It becomes the duty of the process
also carries in chemical energy, which developers to put in innovations, which make
helps in reducing the consumption of greater use of the strengths and minimise to
electric power. Too low a carbon content the extent possible the weaknesses of RKDR.
comes in the way of a healthy carbon There are many basic aspects, which need
boil and, therefore, bath carburisers to be considered for making sponge iron in
need to be added. Clean carburisers are rotary kiln, the important ones being:
costly while coke, char or pig iron carries
with it undesirable elements like sulphur (i) Thermodynamics of reduction and
and phosphorous. gasification reactions
(ii) Characteristics of raw materials and
(iii) Sponge iron from rotary kiln carries
their role in the process
with it more gangue and phosphorous
than those from shaft furnace, mainly (iii) Reaction kinetics, roles of reducibility
because shaft furnace uses cleaner of iron ore and reactivity of coal char
inputs. Gangue and phosphorous and thereby the basis of selection of
contents are much higher than they iron ore and coal
are in iron and steel scrap, which means (iv) Movement of solids in the rotary kiln
extra inputs of phosphorous and slag and its residence time
in electric furnaces. (v) Gas evolution and flow rate
(iv) When we compare with scrap and pig (vi) Heat transfer, temperature profile and
iron, all sponge irons are prone to re- process model
oxidation and the product from rotary These would be dealt in the subsequent
kiln is no exception. However, this chapters.
Thermodynamic
3
Considerations:
Feasibility of Reaction CHAPTER
Thermodynamics, and more specifically the forward direction). For isothermal reactions,
free energies of chemical reactions and the free energy (negative of entropy multiplied
equilibrium constants derived therefrom, by absolute temperature) is a more
govern the limits of all chemical and convenient term for expressing entropy. It is
metallurgical processes. in energy units and can be conceived as the
Readers may like to recall that the science total energy less the bound energy. Feasible
of thermodynamics evolved out of the study reactions are indicated by a decrease in free
of those properties of heat which are not energy. And the extent of decrease in free
found in other forms of energy. Heat flows energy is the driving force of the reaction.
much like a liquid seeking its own level (i.e., Thermodynamics and free energy change
stabilisation of temperature). While assume even greater importance in processes
mechanical work can be completely converted involving solids, as in sponge iron making.
into heat through, for example, friction, Sponge iron making gave way to a more
conversion of heat into work can only be indirect process route involving melting and
achieved when accompanied with is now being resurrected only after a greater
simultaneous discard of a certain lower understanding of how to provide a greater
quality (lower temperature) heat. During the driving force so that reaction rates reach such
most efficient operation of such a heat engine, levels, which can compete with indirect
the heat per unit absolute temperature (or processes.
the entropy) remains constant. But as the Figure 3.1 is an approximate represen-
most efficient operation takes infinite time, tation of Ellingham diagram of a few key
such an operation cannot be a practical oxides of our interest. For a more accurate
operation. For actual operation, entropy representation the publications of United
increases. For a proposed operation or States Steel or of the Association of Iron and
reaction, an indicated increase in entropy Steel Engineers may be referred to.
signifies its feasibility (i.e., the process or An Ellingham diagram plots the standard
operation has a tendency to move into the free energy changes of reactions and their
variation with temperature. The term making in rotary kiln. In this Table, the
‘standard’ refers to a state of reference, symbols G and H represents free energy and
generally 1 atmosphere pressure and a heat content (enthalpy) respectively.
temperatue of either 0°C (273°K) or, as in the Superscript ‘°’ indicates standard state and
present case, 25°C (298°K) and sometimes subscripts ‘298’ and ‘1273’ indicate the
1000°C (1273°K). In case of Ellingham temperature in degreees Kelvin (°K). From
diagram for oxides, which is relevant to our the Table we can deduce the following:
subject, the lines lying higher up represent a
(a) The overall reduction process is highly
less stable oxide as compared to the lines
endothermic and additional coal (or
lower down. It can be seen that the line of
any other fuel) is required to be
CO2 formation from CO and O2 lies slightly
combusted to meet this heat demand.
above the line of FeO formation from iron
(b) As stated earlier, under standard
and oxygen. This means that under standard
conditions and 1000oC, reduction of
state the following reaction cannot proceed
FeO to Fe is not feasible (reaction no.
forward.
2 in the Table). But it becomes feasible
FeO + CO ↔ Fe + CO2 if CO/CO2 ratio is maintained above
This is the key reaction in sponge iron 2.52; i.e., the concentration of CO2 in
making process and the reaction is made to the vicinity of iron ore particles is not
move in the forward direction by keeping a allowed to build up above 28%.
large excess of CO over CO2; i.e., by heavily (c) The above is achieved by maintaining
reducing the activity of CO2. a fast rate of the gasification reaction
Table 3.1 presents the thermodynamic (reaction no.6). If the reduction is
parameters that are relevant to sponge iron carried out at a lower temperature,
2C + O2 = 2CO
2CO + O2 = 2CO2
2Fe + O2 = 2FeO
C + O2 = CO2
DG°
kCal/mole O2 2C + O2 = 2CO
4/3 A1 + O2 = 2/3 A12O2
–150
0 Temperature °C 2000
Fig. 3.1 Ellingham diagram for a few key oxides

Table 3.1 Thermodynamic data of important reactions
∆ G and ∆ H values in kCal/kg mole, T in °K
SL. Class of Reaction ∆ GoT ∆ G o298 ∆ G o 1273 ∆Ho298 ∆ H o 1273 Remarks

reaction
1. Fe2O3 + CO = + 2,120 – 10.39 T – 976 –11,106 + 2,270 – 2,532 Always feasible. Easily
Reduction reaction
2FeO + CO2 ccomplished at the start of

reduction zone of rotary kiln.
2. FeO + CO = – 4,190 + 5.13 T – 2,661 + 2,340 – 4,430 – 8,530 Key reaction. CO/CO2 ratio
Fe + CO2 required to be over 2.52 at
1000oC for reaction to
proceed
Combustion reaction
3. C+½ O2 = CO – 26,700–20.95 T – 32,943 – 53,369 – 26,420 – 27,183
Thermodynamic Considerations: Feasibility of Reaction // 21

Gasification and
Combuction (Heat
4. C + O2 = CO2 – 94,200 – 0.2 T – 94,260 – 94,455 – 94,050 – 94,561 generating) reactions
5. C+½ O2 = CO2 – 67,500 +20.75 T – 61,316 – 41,085 – 67,630 – 67,378
6. CO2 + C = 2CO + 40,800 – 41.70 T + 28,373 – 12,284 + 41,210 + 40,195 Gasification of carbon
7. FeO + C = Fe +CO + 36,610 – 36.57 T + 25,712 – 9,944 + 36,780 + 31,665 Over all reaction for the
Overall reaction
reduction zone.
8. Fe2O3 + 3/2 C = + 54,940 – 62.68 T + 36,261 – 24,852 + 55,225 + 40,700 Over all reaction for the
2Fe + 3/2 CO2 RKDR process. The highly
endothermic heat
requirement is met by
burning of additional coal.
Data taken or calculated from Kubaschewski and Alcock – Metallurgical Thermochemistry, Pergamon Press, 1979
higher percentages of CO 2 can be process move forward. This has further been
tolerated. discussed under Reactivity of Coal char in the
While gasification (reaction no. 6) is chapter on Raw Materials (Chapter 9).
feasible at temperature of operation of rotary Kindly note that FeO is a non-
kiln, the rate of gasification is critically stoichiometric compound. Extent of non-
dependent on the nature of surface of char stoichiometry has been ignored and there
(as also the interior, since surface is could be a very marginal error in the data on
continuously consumed) and thereby the this account. Non-stoichiometry means that
nature of coal used. A coal, which yields atom ratio of Fe and O is not exactly 1:1 but
reactive char, which can quickly convert slightly different (roughly about 0.95:1).
traces of CO2 to CO, is necessary to make the
4
Aerodynamics Inside a
Sponge Iron Rotary Kiln CHAPTER
Aerodynamics refers to the science of study have not been able to produce satisfactory
of gas flow and the forces applied by the sponge iron without the use of ‘injection coal’
flowing gas on interacting bodies. The flow i.e., coal thrown or slung from the discharge
pattern of gases inside a sponge iron rotary end of the rotary kiln. A large portion of the
kiln has a strong influence on the satisfactory thrown coal falls on the solid bed close to
performance of the process. One important the discharge end and passes on to cooler
aspect, which the readers would have guessed, before it is fully de-volatilised. Thus de-
is the need to maintain a delicate balance of volatilisation of coal is continued in the early
reducing atmosphere near the surface of solid part of the cooler. As a thumb-rule, for the
charge in the reduction zone and oxidising protection of the sponge iron produced, a
atmosphere at and near the axis of the kiln. ratio of fixed carbon to total iron in the kiln
This and a few other aspects are important in discharge product is maintained at 0.15. This
the study of gas flow characteristics inside a signifies a source of loss of carbon from the
sponge iron rotary kiln. process going unutilised. But, for the sake of
obtaining a satisfactory product at the end,
4.1 THUMB RULES USED IN producers normally do not attempt to reduce
PRODUCTION OF SPONGE IRON this ratio.
Before discussing the aerodynamics or the gas Why so much carbon is required to be
flow characteristics inside a sponge iron rotary kept in the product? In fact if the atmosphere
kiln, we would like to discuss a few thumb- in the cooler is fully reducing i.e., CO2, H2O
rules being followed by rotary kiln sponge iron and O2 are totally absent, there is no need to
manufacturers, which have direct linkage to keep that much carbon in the product. But as
gas flow. These thumb-rules are results of discussed earlier, due to the need to keep a
experience gained in a long history of delicate balance between oxidising and
empirical and industrial practice. reducing conditions within the free-board in
Sponge iron manufacturers using coal the kiln, there is likelihood of some CO2 and
alone as the reducing and heat supply agent, H2O diffusing out from kiln into the cooler
atmosphere. Further, there may be some air There is still a surplus of CO, which keeps
ingress, if the system is not perfectly sealed evolving from the bed. Thus from the point
(or there is otherwise fault in or wearing out at which solid bed has reached about 700oC
of the sealing system). Hence, there is need to till its discharge after full reduction at a
protect the sponge iron from re-oxidation in temperature close to 1000 o C, CO gas
the cooler, until its temperature drops below continuously evolves. Before this point, at first
its ignition point (Crossing Point—kindly refer moisture of the total charge, and then volatile
Chapter 11). matter from feed coal keeps evolving out
But fixed carbon in the char is ill equipped from bed. Even the coal thrown in
to protect the hot sponge iron. It is not the pneumatically from discharge end, which falls
fixed carbon, but the little bit of volatile matter on solid bed in the reduction zone only,
that keeps liberating, which helps in evolves moisture and VM from the solid bed.
protecting the sponge iron. Thus gases evolve almost uniformly
The above discussion was needed before from the solid bed throughout the kiln
we discuss Aerodynamics, in order to length and the sum total comes out from
emphasise the point that at kiln discharge, the the charging end of the kiln. The gas
gas velocity in the free board is not zero and volume is supplemented by the controlled
some gas evolved in cooler joins the rotary kiln amount of air injected through the Central
gases. Burner (a misnomer, since it is not a burner
and only injects air and the Shell Air Fans
4.2 SOURCES OF GAS
(mounted on the kiln shell) via the Air Tubes,
Apart from the above, and the air introduced placed all along the length of the kiln.
in rotary kiln for combustion purposes, almost
all the gas in rotary kiln is generated from 4.3 EFFECTS OF AIR INJECTION
within the solid bed. Main reactions taking Air can be taken to be consisting of one part
place within the bed are oxygen and four parts nitrogen. Nitrogen adds
FeO + CO = Fe + CO2 to the gas volume while oxygen is consumed
CO2 + C = 2CO without any addition to the gas volume.
Overall reaction within the bed becomes 2CO + O2 = 2CO2
(by adding the above two reactions)
2H2 + O2 = 2H2O
FeO + C = Fe + CO
Thus injecting air results in increasing
Thus there is surplus of CO in the bed
gas volume by 80% of the volume of air
which keeps evolving as the iron oxide is being
metallised. Even in the region where Fe2O3 is injected. Air is injected in controlled
being converted to FeO by the reaction, quantities for combusting CO (& other
combustibles) in free board to maintain the
Fe2O3 + CO = 2FeO + CO2
desired temperature profile and meet the
CO2 + C = 2CO
thermal needs of the process.
The overall reaction being,
Thus gas velocity keeps increasing from
Fe2O3 + C = 2FeO + CO nearly nil at kiln discharge end to the
Aerodynamics Inside a Sponge Iron Rotary Kiln // 25
maximum level at charging end of kiln (or reduction zone of the kiln, turbulent flow
at the After Burning and Cooling – ABC pattern is preferred in the preheating zone.
chamber). Based on extended investigations on the flow
of fluids in pipes, Osborne Reynold in 1883
4.4 LIMITS TO GAS VELOCITY introduced the empirical criterion of a
A very high velocity of gas is expected to blow dimensionless number, which later came to
away some of the granular solids from the bed, be known as the Reynold’s Number, in
which is obviously not desired. This is determining whether a particular flow would
expected to happen at fluidising velocity be lamellar or turbulent.
(velocity of gas at which the suspended Reynold’s Number NR is defined by the
granular particles remain in suspension and following expression
do not settle down – the suspended solids thus NR = DVρ/µ
start behaving like a fluid). As a thumb-rule a where D = Diameter of pipe
velocity value of 60m/sec is considered as
Here, diameter of rotary kiln would be
the fluidising (and therefore the limiting)
applicable
velocity. Fortunately, due to the need to
maintain a large free board in the reduction V = Velocity of the fluid
zone of the kiln, designs of conventional Here, gas velocity in the kiln
sponge iron rotary kilns are not constrained ρ = Density of fluid (kiln gas)
by this consideration. However, as and when µ = Viscosity of fluid (kiln gas)
the need arises to evolve more efficient kiln Typically, if the value of NR is below 2,000,
designs, fluidising velocity considerations will the flow would be lamellar, and above 4,000
have to be kept in mind. the flow would have turbulent pattern.
Between the values of 2,000 to 4,000 the flow
4.5 GAS FLOW PATTERN pattern would be of mixed type, in some
In order to keep the oxidising constituents regions lamellar and in other regions
existing at and near the kiln axis in the turbulent. If care is taken to limit the filling
reduction zone, to remain mostly separate degree to within 15% in the reduction zone,
from the surrounding gases, it is important values of Reynold’s Number are then
that the gas flow in this region be lamellar; maintained within 2000 in some part of
i.e., the layer of oxidising gases flow as a reduction zone, close to the discharge point
separate layer and do not mix with the layers of solids, and flow pattern remains lamellar.
of reducing gases surrounding it. In other Beyond this point, towards the charging end
words, in the reduction zone the flow should of solids, and in the entire preheating zone,
not be turbulent, meaning small layers gas volumes increase rapidly and the flow
(eddy’s) of gases do not form and mix laterally pattern then reaches mixed zone and
with the reducing gases in the vicinity. thereafter turbulent zone and thereby heat
On the other hand if turbulence were not transfer operation is facilitated.
created in the preheating zone, heat transfer
would be heavily impaired. So, while a 4.6 CFD ANALYSIS
lamellar flow pattern is desirable in the As mentioned earlier, making sponge iron in
rotary kiln using coal is a delicate balancing
Fig. 4.1
Fig. 4.2
Fig. 4.3
exercise. In the same cross section in it, the output data and the flow direction are
reduction zone we need to maintain fully computed. The flow direction decides which
reducing atmosphere within, and just above, element is the successor to the first element
the solid charge bed, and close to the axis of and the output data of the first element
the kiln we need to oxidise the gases to applies to this second element. The
generate heat and also keep fair percentages computation is thus continued until we reach
of trivalent molecules H2O and CO2 so that the last element.
adequate heat transfer by radiation is Computational steps and the number of
achieved. Readers may note here that input and intermediate data being extremely
bivalent gases like CO, H 2 , N 2 radiate large, and the interactive steps of calculation
negligible amount of heat. And this oxidation being extremely complex, the solution requires
of CO to CO2 and H2 to H2O should be carried a computer capable of storing and handling
out in a manner such that the reducing such a large number of data, and a computer
atmosphere within, and just above, the solid program, which can handle and interpret such
charge is not disturbed. To achieve this, it a complex process.
has been found necessary to maintain a large
In a typical case, RDCIS, SAIL, used
free board. Consequently, it has not been
FLUENT software and a personal computer
found practicable to maintain average filling
with a Pentium-4 processor to compute flow
degree of solids of more than 15% in the
pattern in a typical 100 tpd sponge iron rotary
reduction zone. Sometimes, even this 15%
kiln. The flow pattern generated by the
filling is found to be too large as the bed
introduction of “Combustion Air” at the kiln
atmosphere is found to get disturbed by the
flow pattern in the free board. The result is a discharge end using a 65 mm diameter pipe
lower metallised or a re-oxidised sponge iron. at the rate of about 140 m/s was studied. The
In extreme cases the heat generated in re- kiln is 42 m long and available inside diameter
oxidation makes the sponge iron pieces to fuse is 2600 mm. The kiln is conical at discharge
together and form agglomerates. end with a 45o cone and discharge comes out
It is possible to determine the expected through an opening of 1220 mm diameter or
flow pattern inside a sponge iron rotary kiln 610 mm radius. If the air required (combustion
using Computational Fluid Dynamics or the air) is introduced through an axially placed
CFD technique. Fluid flow is a highly complex pipe of 65 mm dia. at a linear velocity of about
process. Fluid characteristics, specially its 140 m/s, the CFD analysis indicates an airflow
viscosity, keep changing with both pressure profile represented in Fig. 4.1. From this figure
and temperature. The flow pattern can be it can be seen that at 4.2 m or 10% of the length
computed by the finite element analysis, of the kiln, the air injected would tend to strike
where the entire volume is broken into a grid the kiln bottom where the solid bed would be
of very small but finite elemental volumes found during sponge iron making operation.
within which the properties can be assumed This CFD analysis has been carried out in an
to be uniform. Based on the input data to a empty kiln, and can be repeated with the solid
particular elemental volume, and the bed in place.
interactive effect of the elements surrounding The above result predicts the ill effects
of placing the air injection pipe axially, which bed. A rough estimate of the velocity of strike
would disturb the reducing atmosphere would be between 3 to 4 m/s which is much
within the solid bed quite intensely and thus smaller than the general levels of velocities
would come in the way of obtaining a well in the kiln free board. Lowest velocity is
metallised product. indicated when air pipe is placed 320 mm
Figures 4.2, 4.3, 4.4 and 4.5 show the above the axis of the kiln.
results of CFD analysis when the air pipe is The reason why the value of 320 mm was
placed 50 mm, 100 mm, 150 mm and 320 mm chosen is because some of the sponge iron
above the axis. In all these cases it is seen plants, operating 100 tpd kilns, have claimed
that the layers of air flow tend to bend to have found best results when the bulk of
upwards first and, after what may be called
air introduced from discharge end is injected
a recoil from the top surface, tends to strike
using a pipe placed 320 mm above the axis of
the bottom beyond 15 m or 35% of the length
the kiln. Thus, although the optimum location
of the kiln. By this time it would have lost
of air pipe has been an exercise of trial and
most of its momentum, and chemical
reactions would have consumed all O2 in the error in the past but Computational Fluid
stream for this strike on the solid bed to have Dynamics or CFD exercise has brought out the
any deleterious effect. logic behind the suitability of a particular
location.
Additionally, there is a vortex like
formation below these layers of air stream in Very few variations were attempted in the
Fig. 4.2 to 4.5. A portion of kiln gas tends to exercise cited above. A more rigorous
re-circulate over the bottom layer (where application of CFD analysis to the conditions
solid would normally be present in actual of an actually operating rotary kiln should
operation). This layer, which would come in help in further optimisation of process
contact with the solids, would consist of only parameters and thereby further enhancing the
reducing gases. Even then it is not desirable productivity and quality of the rotary kiln
to disturb the gas dynamics within the solid sponge iron making process.
This page
intentionally left
blank
Mathematical Modelling
5
in Rotary Kiln
Sponge Iron Making CHAPTER
Mathematical models are being increasingly approach it with a sense of scepticism,

used in the understanding and control of apprehension and fright. It is, therefore,
industrial processes. Use of Mathematical highly recommended that the potential users
Model ensures that the process operates in should attempt to develop the mathematical
the most optimum regime, where the model themselves. If they are successful in
maximum benefit is realized within the their attempt the rewards are highly
available constraints. Since industrial significant, but even if they are unable to
processes generally operate on a large scale, develop a workable model, the experience
operating it even marginally away from the gained would place them in a far better
optimum results in massive loss of position to interact with a commercial supplier
opportunities. On the other hand a of a mathematical model than he would
mathematical model of an industrial process otherwise have been.
is highly complex and, if the user is not the
maker of the model, one finds it very difficult 5.1 WHAT IS A MODEL?
to understand and implement. Therefore, We use a number of models of various types
those marketing commercial mathematical in our day to day life. A model is something
models, offer a training module along with which represents or simulates or looks like
the model. Further, makers of commercial or even acts like the real thing. We buy for
mathematical models make them for a wider our children a toy car, which is basically
application. Fine tuning the same to make it meant for them to play with, but it also
suitable for the needs of the customer has to simultaneously helps the child to understand
be done after detailed interaction between cars better. Toy car is an example of a physical
the maker and the customer. The actual fine model which is a smaller version of the real
tuning may be done by the maker or the car. Most of the physical models are of this
customer or jointly between them. type. Even prototype working models of a
Mathematical models being highly machine of newer design is a smaller version
complex, the potential users generally of the real thing to come, and is made for the
purpose of the validation of the new design. 5.3 HOW CAN WE MAKE A USEFUL
But physical models need not always be MATHEMATICAL MODEL?
smaller than the real thing. A physical model Formulation of a mathematical model starts
of a DNA molecule is many many times with identifying the basic principles and
bigger just because it is impossible for the fundamental processes involved and
naked eye to physically see objects at converting them into mathematical terms.
molecular scale. This identification process is started with the
analysis of inputs and outputs of material and
5.2 WHAT IS A MATHEMATICAL energy – in other words the Material Balance
MODEL? and the Energy Balance of the process. Once
These are mathematical expressions, which the overall material balance and the energy
theoretically or empirically relate parameters balance are worked out, the material and
of a real world problem to each other. The energy flow within the process is analysed.
definition is best understood through The total process is broken into interlinked
examples. We have consciously or sub-processes and each sub-processes and its
unconsciously been using simple mathematical relationship with other sub-processes are
analysed and converted into mathematical
models since ages. Many times we tell a
terms until a workable mathematical model
friend over the phone “I shall be with you in
is developed. After this first development,
15 minutes.” Let us see what we have done.
the model needs to be validated which is best
We have analysed the activities required to done by inputting actual operating data and
reach the friend. We have estimated time the model output is compared with actual
required for each of the activities and we output. Based on the observed difference, a
have added them up. We have also added a correction or adjustment may need to be
few minutes for contingencies and thus have incorporated in the model. After full
arrived at the final figure. validation, the mathematical model is ready
Ancient Egyptians were experts at making for implementation.
pyramids. They wanted to cut stones at right While converting the actual process steps
angles (90°). For this purpose they made three into mathematical expressions, a number of
sticks of lengths three units, four units and simplifying assumptions may need to be
five units. They joined these sticks at the edges made. It is advisable to keep listing these
and used the largest angle of the triangle assumptions. During review and validation,
these assumptions may need to be examined
formed for matching the corner angles of the
for their validity or may need to be modified.
stones. When we see a lightening flash, we
can count seconds until we hear the thunder, 5.4 EXAMPLE OF A SMALL
divide seconds by five and conclude that the MATHEMATICAL MODEL
cloud that caused this is so many miles away. Let us attempt to make the mathematical
Also if we stand on a quiet sea shore and we model of a small step in sponge iron making
see a boat’s mast setting below the horizon, (actually a sub-process of sponge iron
we can conclude that the boat has gone beyond making), i.e. how to predict the composition
a mile range. and quantity of sponge iron, having been
Mathematical Modelling in Rotary Kiln Sponge Iron Making // 33
given the composition of iron ore (or any It is very easy to fill up the columns next to
other iron oxide feed such as iron ore pellets). the composition of iron ore and appears almost
For this we need the complete analysis of iron a superfluous column. But from here we can
ore. Then we list the analysis in a Table of fill up the next column where oxygen remains
the following type (an Excel or a Lotus 1-2-3 only 5% of the previous column. Moisture and
spreadsheet). Here, items like Fe 2O 3 and LOI are lost in sponge iron making process and
oxygen are not obtained by chemical analysis,
so also Fe2O3. Fe(t), SiO2 + Al2O3 and phos-
but are calculated based on the values of Fe,
phorous remain unchanged. Sulphur and
Fe(metallic), and FeO. Although the
carbon values are difficult to fill at this point
assumption that LOI (Loss on Ignition) is 50%
and we leave these columns blank and add the
CO2 and 50% H2O, will not be needed further
in this part of the model, but would be used column up (we get here the value of 69.35 kg.).
in other parts of mathematical model of By experience we know that a well metallised
sponge iron making. sponge iron is:
Table 5.1 Chart for calculating sponge iron composition from iron ore composition
Sl. No. Element/Item Iron ore Kg of element/ Kg. of element /item % of

composi- item in 100 kg of in sponge iron, 95% element /
tion iron ore reduced (92% item in
metallised) sponge
iron
1. Fe (t) 65.00 65.00
2. Fe (m) 0.00 0.00
3. FeO 0.00 0.00
4. Fe2O3 92.90 92.90
5. Oxygen 27.90 27.90
6. Moisture (H2O) 2.00 2.00
7. Loss on 2.00 2.00
ignition (CO2
+ H2O = 50:50)
8. SiO2 + Al2O3 2.90 2.90
9. S 0.05 0.05
10. P 0.05 0.05
11. C 0.00 0.00
Total 100.00 100.00
Total without sulphur

and carbon
• about 95% reduced (this fact we have 69.50 kg of sponge iron. Therefore, 1500 kg of
already used in calculation) iron ore would give 1042.5 kg of sponge iron.
• contains about 0.02% sulphur and By practical experience we know that during
• also contains about 0.2% carbon rotary kiln sponge iron making, for each tonne
Thus, 69.35 kg of items in sponge iron sponge iron made, about 40 kg. of sponge iron
should account for 99.78% of the items in (or its equivalent iron ore) is lost in various
sponge iron. Thus by multiplying the figures dust streams. Thus, for making 1000 kg of
in the fifth column by a factor of (99.78/69.35) saleable sponge iron we need to put in 1500
we fill up the last column. And by kg of iron ore.
multiplying the carbon and sulphur figures
in the last column by (69.35/99.78) we 5.5 ROLE OF COAL IN ROTARY KILN
complete the fifth column. SPONGE IRON MAKING PROCESS
An important inference obtained from this For taking up the work of preparing the
Table is that from 100 kg iron ore we can obtain mathematical model of the rotary kiln DR
Table 5.2 Filled up chart of Table 5.1

Sl. No. Element/Item Iron ore Kg. of element/ Kg of element/ % of
composition, item in 100 kg of item in sponge element /
% iron ore iron 95% reduced item in
sponge
iron
1. Fe (t) 65.00 65.00 65.00 93.50
2. Fe (m) 0.00 0.00 60.10 86.50
3. FeO 0.00 0.00 6.30 9.10
4. Fe2O3 92.90 92.90 0.00 0.00
5. Oxygen 27.90 27.90 1.40 2.00
6. Moisture (H2O) 2.00 2.00 0.00 0.00
7. Loss on ignition 2.00 2.00 0.00 0.00
(CO2 + H2O
= 50:50)
8. SiO2 + Al2O3 2.90 2.90 2.90 4.10
9. S 0.05 0.05 0.05 0.02
10. P 0.05 0.05 0.05 0.07
11. C 0.00 0.00 0.10 0.20
Total 100.00 100.00 69.50 100
Total without
sulphur and carbon 69.35 99.78
process, we have to understand at first the coal from its ultimate analysis. Then we
role of coal in this process. Coal has got devolatilise the coal to obtain the quantity
multiple roles in this process. It acts as the and composition of the char. Then we calculate
source of heat which is required in the process the quantity of this char in the product to
and this is provided by the controlled obtain FC/VM = 0.15. In practice the char in
combustion of carbon and hydrogen in coal the product discharge will have a much lesser
by air along the length of the kiln. Carbon in fixed carbon than the char, which we have
coal also acts as reductant for removing the assumed here to be with the product. We
oxygen in combination with iron in iron ore. would further be assuming that this char gets
This is a third role of coal which is less mixed with a further quantity of ash and the
understood. Coal in the form of char, which result is the quantity and composition
accompanies the sponge iron produced from obtained in actual practice.
the kiln, helps in protecting the sponge iron
from re-oxidation. 5.7 REDUCTION OF IRON OXIDE
Experience has indicated that a ratio of
We now assume that the VM liberated in the
Fixed Carbon to Total Iron in the product
above step consists of C, H, H2O and CO2.
should be about 0.15 for a smooth operation
Since sulphur largely passes from coal to flux
of the process.
(one solid to another), we consider it non-
While coal provides so many services in volatile and eliminate it from calculations. The
the process, it has some negative roles also available carbon and hydrogen in the VM is
to perform. Coal has the responsibility of then consumed for stoichiometrically
raising the temperature of furnace and the reducing the iron ore.
furnace charge to the processing tempera-
Fe2O3 + 3/2 C = 2 Fe + 3/2 CO2
ture. But in doing so it must heat itself to the
processing temperature first. In doing so it Fe2O3 + 3 H2 = 2 Fe + 3 H2O
liberates the volatile matter and the heat The quantity of iron ore reduced in these
required to liberate the volatile matter has steps is estimated as also the heat deficit
also to be pre-accounted for by coal, before caused in these steps. Next an estimate is
it can liberate heat for other purposes. The made of the coal which is required for
ash within the coal has to be likewise heated reducing the balance of unreduced iron ore.
to the processing temperature. Air needed Also the heat deficit caused by these steps is
for combustion also needs to be brought to estimated.
processing temperatures. Likewise, the waste
gases which leave the reactor at temperatures 5.8 COAL FOR COMBUSTION
somewhat close to processing temperature The accumulated heat deficit has to be made
also carries energy obtained from coal itself. up by burning additional coal. The available
heat from coal is calculated and based on this
5.6 QUANTITY OF PROTECTIVE CHAR the quantity of additional coal required is
It would be convenient to calculate at first estimated. Please note that of the total
the quantity of coal needed to provide the calorific value of coal, some part of the heat
necessary carbon in product discharge to is used up by the coal for its own
bring the level of Fixed Carbon to Total Iron requirement, which is provided for in the
in product to 0.15. First we infer the VM in calculation beforehand.
36 //
Table 5.3 Products of heating 100 kg coal
Assumptions:
Coal composition: Free moisture = 5.6%, Combined moisture =1.8%, Ash = 30.0%, C = 51.4%, H = 3.0%, S = 0.5%, N = 1.3%, CO2 as
Advances in Rotary Kiln Sponge Iron Plant

carbonate = 0.4%, O = 6.0%.
Heat effects on conversion have been neglected Quantity in kilogrammes where not mentioned
Consti- Initial Quantity Quantity Other Gaseous Overall product Volume Volume Char analysis
tuent quan- being remaining consti- product Kg- % of percent-
tity conver- after tuents Quan- consti- Quan- Consti- Quan- Volatile moles consti- age of Qua- Percent
ted conver- consu- tity tuent tity tuents tity or Non of tuents consti- ntity
sion med volatile evolved in VM tuents in
gases the total
gases
evolved
Ash 30.0 0 30.0 - 0 - 0 Ash 30.0 Non 30 43.86%
Volatile
H2O 5.6 0 5.6 - 0 - 0 H2O 5.6 Volatile 0.311
(Free) (Free)
H2O 1.8 0 1.8 - 0 - 0 H2O 1.8 Volatile 0.100 7.81%
Comb- Comb-
ined ined
CO2 0.4 0 0.4 - 0 - 0 CO2 0.4 Volatile 0.009 0.71% 0.57%
N 1.3 1.3 0 - 0 N2 1.3 N2 1.3 Volatile 0.046 3.63% 2.92%
S 0.5 0 0.5 - 0 - 0 S* 0.5 Non 0.5 0.73%
Volatile
O 6 6 0 C 4.5 CO 10.5 CO 10.5 Volatile 0.375 29.28% 23.56%
H 3 3 0 C 9 CH4 12 CH4 12 Volatile 0.750 58.57% 47.12%
4.5 46.9 0 6 H2O 7.4 25.83%
C 51.4 (Total)
9 37.9 H 3 C 37.9 Non 37.9 55.41%
Volatile
Total Kg-moles of VM constituents 1.281 100%
Total Kg-moles of gases evolved 1.592 100%
Total Char 68.4 100.0%
*S passes from one solid (coal and/or iron ore) to flux. Hence it is considered non-volatile
Table 5.4 Estimation of iron ore which can be reduced by gases evolved from coal added for generation of only
protective char (Basis: 1000 kg or 1 tonne DRI)
Assumptions:
Iron ore analysis: Fe (T) = 66.0 % (all as Fe2O3), Gangue=3.7%, Moisture (Free + Combined)=2.0%, Oxygen (with Fe)=28.3%.
Quantity required = 1500 kg
DRI analysis (95% Degree of reduction – Equivalent to 92% Degree of metallisation): Fe(T) = 92.6 %, Fe(M) = 85.7%, FeO = 8.9%,
Gangue = 5.2%, C = 0.2%. Quantity = 1069, comprising 1000 kg saleable, 30 kg fine DRI, and 39 kg DRI lost as dust.
Char analysis: Fixed carbon = 55.41%, Ash = 44.59%. 100 kg coal gives 68.400 kg char on carbonisation
Quantity in kilogrammes
Quan- Quantity of Quantity Quantity Gas Quan- Quantity Quan- Heat Product gas
tity of Carbon in of Char of Coal con- tity of Fe2O3 tity deficit, Cons- Quantity Kg- Percent
Iron in Char needed needed needed stitu- reduced of Iron Kcal tituents moles

DRI to maintain ents by these Ore
Cfix/Fe(T)= 0.15 lib- constitu- reduced
erated ents up
by this to 95%
coal
989.89 148.4841 267.977 391.779 N2 5.093 0 0 N2 5.093 0.1819 1.40%
CO2 1.567 0 0 CO2 1.567
H2 O 28.992 0 0 H 2O 28.992
CO 41.137 82.480 87.48 -5,397 CO2 64.643
CH4 47.013 659.838 699.83 822,422 CO2 129.287
Total 787.31 817,025 H 2O 105.780
H 2O 15.746
from iron
Reduction reactions considered: ore
Fe2O3+3CO=2Fe+3CO2 ∆Ho298 = –11,020 kCal/kg-mole CO2 193.930 4.4075 34.0%
& (Total)
4Fe2O3+3CH4=8Fe+3CO2+6H2O ∆Ho298 = +209,920 kCal/kg-mole H2O 150.518 8.3621 64.6%
(Total)
Total gas 12.952 100%
38 //
Table 5.5 Estimation of iron ore which can be reduced by 100 kg of coal
(Constituents converted as per Table 5.3 has been considered here)
Quantity in kilogramms

Sl. No. Coal Quantity Reaction Amount of Amount of Product gas Heat deficit
constit- with Fe2O3 Fe2O3 that iron ore (kCal)
uents for complete can be red- being redu- Constituent Quantity Kg-
conversion uced up to ced moles
95%
reduction
1 Ash 30.0 - 0 0
2 H2O (Total) 7.4 - 0 0 H2O 45.941 2.5523
3 CO2 0.4 - 0 0 CO2 188.867 4.2924
4 N2 1.3 - 0 0 N2 1.400 0.0500
5 S 0.5 - 0 0
6 CO 10.5 (1) 21.053 22.329 –1,377.50
7 CH4 12 (2) 168.421 178.628 52,480.00
8 C 37.9 (3) 354.620 376.112 116,279.31
(4) 6,745.29
Total: 100 577.069 6.8947 174,127.10
(1) Fe2O3+3CO = 2Fe+3CO2 ∆ Ho298 = –11,020 kCal/kg-mole

(2) 4F2O3+3CH4 = 8Fe+3CO 2+6H2O(g) ∆ Ho298 = +209,920 kCal/kg-mole
(3) Fe2O3+3/2 C = 2Fe+3/2 CO2 ∆ Ho298 = +55,225 kCal/kg-mole
(4) H2O(l) = H 2O(g) ∆ Ho298 = +10,520 kCal/kg-mole
Table 5.6 Estimation of coal heat available from coal for meeting heat requirement (ef. Net Calorific Value)
(Basis: 100 kg coal)

Sl. Coal Oxygen re- N2 coming Product Gas
No. constituents quired for in with the Heat
(Table 5.3) Quantity complete air (kg) Generated Constituents Quantity Kg-moles
combustion (kg) (Kg)
1. Ash 30 0 0 0
2. H2O (Total) 7.4 0 0 –4,324.89 H2O (g) 34.40 1.9111
(converted to
water vapour)
3. CO2 0.4 0 0 0 CO2 188.87 4.2924

4. N2 1.3 0 0 0 N2 544.03 19.4298
5. S 0.5 0 0 0
6. CO 10.5 6 21 25,361.25
7. CH 4 12 48 168 143,812.50
8. C 37.9 101.07 353.73333 297,041.25
Total : 155.07 542.73 461,890.11 767.30 25.6333
Reactions considered: CO + 1/2 O2 = CO2 ∆ fPWV = –67,630 kCal/kg-mole

CH4 + 2O2 = CO2 + 2H2O(g) ∆ fPWV = –191,750 kCal/kg-mole
C + O2 = CO2 ∆ fPWV = –94,050 kCal/kg-mole
H2 O(l ) = H2O(g) ∆ fPWV = +10,520 kCal/kg-mole
40 //
Table 5.7 Alternatives considered for calculation along with their boundary conditions
Ore preheating Temperature Air Mean air Exhaust gas Char recir- Char quan-

of preheating preheating temperature temperature culation tity recirc-
ulated, kg/t
DRI
Alternative-I No 25 oC No 25 oC 1200oC No 0
Alternative-II No 25 oC No 25 oC 1000oC No 0
o o o
Alternative-III No 25 C No 25 C 800 C No 0
Alternative-IV Yes 664 oC Yes 400oC 700 oC Yes 200 (~85%)
(Optimised Process*)
Conventional process regime lies somewhere between Alternatives 1 and 3. * Explained in section 5:13 and Chapter 13
Table 5.8 Calculation of heat available from the Combustion of 100 kg coal under rotary kiln
conditions for the four alternatives
Heat Values in kCal CV = 4618.9 kCal/kg
Products of combustion of coal Heat required to raise the temperature of product constituents
Constituents Quantity Kg- 1200oC Average 1000oC Average 800oC Average 700oC Average
(kg) moles Cp Cp Cp Cp
H2O (g) 34.40 1.91111 21,214 9.447 17,131 9.194 13,244 8.942 11,374 8.817
CO2 188.87 4.29242 61,547 12.203 49,870 11.916 38,579 11.597 33,082 11.418
N2 544.03 19.4298 172,663 7.563 141,341 7.461 110,812 7.359 95,845 7.308
Total 767.30 25.6333 255,424 208,343 162,635 140,301
Alternative1 206,466
Alternative 2 253,547
Available heat
Table 5.9 Estimation of heat demands of different alternatives
Alternative 1 Alternative 2 Alternative 3 Alternative 4
Quan- Heat, Quantity Heat, Quan- Heat, Quan- Heat,
tity, kg kCal Kg kCal tity, Kg kCal tity, Kg kCal
Quantity of iron ore reduced by 699.8 822,422 699.8 822,422 699.8 822,422 177.523 208,621
gases evolved from coal added for
generation of protective char only,
and the heat deficit there from
Coal required for reducing the balance 138.7 241,447 138.7 241,447 138.7 241,447 229.17 399,050
iron ore, and the heat deficit there from
Calcination of flux 75 32,250 75 32,250 75 32,250 75 32,250
Heat required to raise the temperature of 189,530 157,269 125,009 146,789
product gases to the outlet temperature

Radiation and unaccounted losses (assumed) 600,000 600,000 600,000 600,000
Heat put into the system as sensible heat
through pre-heated charge (Negative deficit) 0 0 0 –290,998
Total heat deficit 1,885,649 1,853,388 1,821,128 1,095,711
Coal required for meeting heat deficit 913.296 730.983 608.554 340.718
Coal required for generating 391.78 391.78 391.78 99.38
protective char
Total coal required 1443.7 1261.4 1139.0 669.3
Exit Gas Composition & Quantity
Alternative Alternative 2 Alternative 4

Constituents H2O CO2 N2 Total H2O CO2 N2 Total
Kg-moles 25.9 41.73632 142.279 209.887 14.482 25.58012 66.361 106.423
Percent 12.33% 19.89% 67.79% 100.00% 13.61% 24.04% 62.36% 100.00%
Gas volume, Nm3/t DRI 579.5 934.9 3187.1 4701.5 324.4 573.0 1486.5 2383.9
Air required, Kg-moles/t DRI 177.111 82.55319
Nm3/t DRI 3967.29 1849.19
Assumptions (S=Specific heat, Cp=Molar specific heat): SIron ore = 0.225 kCal/kg, SCoal = S Char = 0.38 kCal/kg; S Flux = 0.33 kCal/kg,
Heat of Calcination of flux = 430 kCal/kg; Average Cp-product gas = 9.0 kCal/kg-mole
Iron Ore Analysis : Fe (T) = 66% (all as Fe2O 3), Gangue=3.7%, Moisture (Free+Combined)=2.0%, Oxygen (with Fe)=28.3%.
DRI Analysis (95% Degree of Reduction – Equivalent to 92% Degree of Metallisation): Fe(T) = 92.6%, Fe(M)=85.7%, FeO=8.9%,
Gangue = 5.2%, C = 0.2%. Quantity = 1069, comprising 1000 kg saleable, 30 kg fine DRI, and 39 kg DRI lost as dust. Coal composition: Free
Moisture = 5.6%, Combined Moisture =1.8%, Ash = 30.0%, C = 51.4%, H = 3.0%, S = 0.5%, N = 1.3%, CO2 as carbonate = 0.4%, O = 6.0%.
5.9 WASTE GAS TEMPERATURE 5.11 SPREAD-SHEET PROGRAMMING

To simplify calculations, the present exercise Tables 5.3 to 5.9 together constitute an overall
includes that portion of the ABC Chamber in model of a typical sponge iron making in
the Sponge Iron Making system where the rotary kiln and the programming and
exit gases have been fully combusted with solutions are based on the ‘spread sheet’
stoichiometric amount of air. The portion of method (‘Lotus’ 1-2-3 or, as in this case,
ABC, where the gas is further diluted with ‘Microsoft Excel’). The spread-sheet method
excess air and is then cooled by water spray, is not the very best way of writing computer
is excluded from calculation. program of this type, but can be an easy and
Since the two functions within the ABC convenient step towards understanding a
chamber are overlapping it is extremely difficult model. Tables 5.3 to 5.9 combine the material
to estimate, let alone measure, the temperature and energy balance of the process with typical
of the gas after complete combustion and before inputs. Many of the desired outputs can be
it is diluted with excess air and cooled with projected or estimated, such as energy
water spray. A reasonable guess is that this consumption, energy in waste gas, waste gas
temperature lies between 800oC to 1200oC. composition, air requirement, quantity of coal
Comparison of results of modelling with those required, etc. Please note that Tables 5.3 to
of actual operating plants indicate that modal 5.9 are based on an iron ore composition
value of this temperature is about 1000oC, i.e., different from that used in Tables 5.1 and 5.2.
a majority of the plants operation is consistent
with the waste gas temperature of about 1000oC 5.12 SEGMENT-WISE MODELLING
after full combustion. Having developed an overall model, we
would like to extend it so that we can estimate
5.10 PROGRAMMING BASED ON what happens at the intermediate stages, such
MODEL as at the middle, at one third or two third
Up to the above we have conceived of a frame- length etc. We would like to know along the
work of calculation of the overall process length of the kiln, the reduction or
parameters. This framework of calculation is metallisation profile, the temperature profiles
often represented in a ‘flowchart’ and has to of the bed and gas, the residence time
be converted to a ‘program’ for a computer required, etc. For this purpose we resort to
to be able to calculate out results from certain “segment-wise modelling” illustraed in
inputs. Earlier with programming language Chapter 7. We divide the kiln into a large
number of segments, and we calculate the
like “Basic” or “Fortran” it was essential first
parameters in each segment and the result
to construct the ‘flowchart’. The ‘flowchart’
becomes an input to the next segment. Within
was (and is) referred to as the ‘model’ as
the segment we assume uniform conditions
separated from the ‘program’ which followed. of temperature of gas, temperature of solid
With the advent of object oriented bed, metallization, gas flow, etc. For this
programming (using computer languages like assumption to be reasonably valid, we have
‘C’, ‘C++’ etc.) making flow chart is no longer to reduce segment size, in successive
an essential part of programming. But calculations, and select such a size of segment
making a flow chart is advisable to easily where size change does not significantly
understand a model. affect the results of calculation.
The kinetics of reduction would be an calculation based on spread sheet

important input to this exercise. The kinetic programming are presented in Tables 5.3 to
parameters like the iron ore reducibility and 5.9. These results and its extensions into
coal char reactivity are widely different for segment wise modelling have indicated a
different ores and coals. These properties few important things:
keep changing with exposure to different 1. When a rotary kiln is operated with
temperatures and for different lengths of time
100% thrown coal, there is about
in the rotary kiln. It is very difficult to
40% saving in coal consumption, as
accurately estimate the reaction rates within
compared to the normal practice of
the rotary kiln segments using the basic
using 40% thrown coal and 60% feed
kinetic properties of the inputs, and this
approach very often does not lead to the coal. However, there would be
development of a practical mathematical some decline in productivity.
model. This result was validated in Pilot Plant
A practical mathematical model would experiments as consumption was
instead use the metallisation profile gene- found to decrease from 938 kg/tonne
rated in laboratory rotary tube furnace using sponge iron to 562 kg/tonne sponge
the particular iron ore and coal combination. iron.
2. Extending the above finding, the
5.13 PREDICTION FROM THE MODEL model has further predicted that if
The results from a typical mathematical model a preheating kiln is incorporated
Idealisation and
approximation
Real world (based on Model
problem experience and
understanding)
Comparison, Mathematical
Correction, representation
Modification,
Implementation
Solution based
Results on Mathematical
from Mathematical model
model experience
Fig. 5.1 Schematic representation of the process of mathematical modelling

upstream of the reduction kiln, development of a practical, workable and

where char is also re-circulated, and useful mathematical model of a process such
the resulting waste gas from the as the process of sponge iron making in a
preheating kiln is used to pre-heat rotary kiln. Such a mathematical process,
air, which is ploughed back into the once developed, helps in understanding of
process, the coal consumption would the process, process simulation and predicting
various outputs for different inputs. These
reduce by 50%. This concept has been
data can be generated without doing costly
referred to as the optimised process
experiments. But the reliability of these
in Table 5.7 and has been discussed in
results depends heavily on how well the
detail in Chapter 13. model has been validated.
3. For a more complete exercise, if we Further, any model is as good as the
want to estimate the rate of revolution assumption made. There is need to keep a
and the kiln inclination, we need continuous check on whether the
further to develop the charge assumptions, which were valid during model
movement and residence time models. validation, have remained valid during
To accurately estimate the gas application or not.
temperature profile required and In the present case of mathematical
consequently the air requirements in modelling of the coal based rotary kiln
different segments, we need to sponge iron making process, the model
develop the heat transfer model. Both developed and its application has helped,
these exercises are also quite involved among other things, in better understanding
and are dealt in the subsequent of the process. This has also pointed towards
sections. some design limitations of the equipment
presently being used. From these results, it
5.14 SUMMARY OF THE MODELLING has been possible to develop an optimized
PROCESS process concept with a preheating kiln, an air
Figure 5.1 adequately depicts the different pre-heater and use of char within the process
interconnected steps involved in the itself.
Physical Movement of Solids
Inside a Rotary Kiln:
6
Charge Movement and
Coal Throwing/Slinging CHAPTER
Sponge iron making in rotary kiln involves quality product, newer kilns are based on
reduction of granular iron ore in solid state much more scientific design data. Residence
by the reducing gases generated by the time requirement is one of the more important
gasification of coal. After the upper layer of design data for such kilns.
iron ore is reduced, the reduction of the Residence time refers to the time spent by
interior requires diffusion of the reducing the solid granular charge in travelling from
gases, mainly CO, to the reduction front and the charging end to the discharge end of the
diffusion of the CO 2 formed, out to the rotary kiln. There is some weight loss and,
surface and freeboard (Fig. 1.7 and 2.9). to a lesser extent volume decrease. In our
For the process to be commercially viable the treatment, volume considerations are more
solids need to be exposed to the reducing important, and bulk density based on mean
gases at as high a temperature as can be volume has been used in further treatment.
tolerated for as long a period as possible. Since One expert had remarked that it is very easy
fusion of constituents causes multiple to conceive that if we put some granular
problems in the kiln, the maximum tempera- material at the higher end of a hollow inclined
ture of solids is controlled to an optimum cylinder, and we rotate it, the material would
level, generally around 1000oC (Fig. 2.1) eventually come out from the lower end. But
if we try to examine the mechanism, we find
6.1 IMPORTANCE OF RESIDENCE TIME it highly complicated.
For designing such a sponge iron kiln, one
needs to know the residence time require- 6.2 ESTIMATING GROSS RESIDENCE
ment and the correlation of the residence TIME FROM INPUT AND OUTPUT
time with the inclination, speed of rotation OF SOLIDS
(revolutions per minute, i.e., rpm) and the In a sponge iron rotary kiln the material is
dimensions of the kiln. While most of the likely to follow a path represented in Fig. 6.1.
early kilns decided on their inclination and About 15% average filling degree of the kiln
rpm based on their experience in getting good is generally maintained in the sponge iron
rotary kilns. Approximately six hours the product and simultaneously looking after
reduction time is required for one of the the safety of the kiln. By using the above
better combinations of ore and coal. The bulk calculation method for an operating kiln, one
density of the charge mix generally lies can estimate the residence time while making
between 1.6 to 1.8 t/m3 while that for dis- assumptions on the average filling degree and
charge material, mostly sponge iron and char, the average bulk density. To check whether
it is between 1.3 and 1.5 t/m3. We may use these assumptions are reasonably correct one
an average figure of 1.4 t/m3 as an example. needs to determine the residence time
About a tonne of sponge iron in discharge is precisely.
accompanied with about 400 kg of char and
ash, making the total discharge per tonne
sponge iron as 1.4 tonnes or 1.0 m3.
...............................
..............................
...............................
..............................
...............................
..............................
...............................
..............................
Let us take the example of a 50-tpd sponge ...............................
..............................
...............................
..............................
...............................
..............................
...............................
..............................
...............................
..............................
...............................
..............................
...............................
..............................
iron kiln. It would have to discharge about 2
m3 per hour of material or about 16 m3 in eight Slipping / Slumping Rolling
Glide conveyance (b) (c)
hours. That means, if a total of eight hours (a)
residence time is envisaged (two hours for
preheating), the solid material at any instant
would occupy 16 m 3 volume in the kiln.
Assuming average filling degree as 15% the
total inner volume of kiln after lining should
be about 110 m3, which is a very important Cascading Cataracting Centrifuging
design data of the kiln. (d) (e) (f)
Fig. 6.2 Modes of transverse bed motion in a

rotating cylinder
6.3 CHARGE MOVEMENT:

CASCADING FLOW OF
GRANULAR MATERIAL
If we keep some granular material in a
horizontal hollow rotary cylinder and we
rotate the cylinder axially, the material would
behave differently for different conditions of
surface friction and speed of rotation.
By the term granular material we mean a
Fig. 6.1 Material movement in a rotary kiln
solid material which is neither too fine in size
(i.e. it is not finely ground) nor is it in as
Here we have assumed the residence time of mined (boulder) form. Rather it is in a close
material in the kiln. In actual practice, the size range such as 10 to 40 mm, 8 to 30 mm, 6
plant operators try to drive the kiln as hard to 20 mm or 5 to 18 mm – the sizes used in a
as possible while maintaining the quality of blast furnace or a sponge iron rotary kiln.
Physical Movement of Solids Inside a Rotary Kiln: Charge Movement and Coal Throwing/Slinging // 47
If, in the above example, the cylinder is obviously has a value slightly lower than the
slightly inclined, we get conditions similar to static angle of repose. When the static angle
that existing in a rotary kiln or a rotary cooler. of repose is exceeded, a small quantity of
The six different behaviours of the granular material slumps from the upper side down
material are represented in Fig. 6.1 – (a) to to the lower side to roughly make the
(f). dynamic angle of repose. After settling in this
Condition of Fig. 6.1(a) occurs when the position the material remains stationary with
friction between the charge material and the respect to the cylinder surface until the static
cylinder surface is very low. The charge angle of repose is reached once again.
material slips on the cylinder surface The above happens when the speed of
(Slipping); the rotation of the cylinder does rotation of the cylinder is slow enough to
not have any effect on the material. If the give the material a resting time between
cylinder is axially inclined the material just slumps. If the rotation is fast enough so that
glides down the inclination (Gliding the static angle of repose of material is reached
Conveyance). Readers may consider this before the slumping is completed, then there
situation as not of practical importance – and is no time for the material to rest. We get a
indeed this would not happen on a refractory movement of charge depicted in Fig. 6.1(c),
surface as that would be sufficiently rough –
which most researchers like to call rolling
but this does happen when refractory is not
movement. Some others have termed it
present, as in a rotary cooler. Initially, rotary
trickling movement. As we would see later,
coolers, which were meant to cool sponge iron
this type of movement is the most desirable
indirectly, out of contact with both air and
in a sponge iron rotary kiln.
water, consisted of just a cylindrical shell of
steel, sprayed from outside with water. With As the rotational speed of cylinder is
continuous abrasion due to movement of progressively increased further, material
sponge iron, the inner surface used to acquire movement changes to cascading (Fig. 6.2 (d)
a mirror finish, leading to gliding conveyance – some additional voids created within the
and thereby insufficient cooling of sponge bed) to cataracting (Fig. 6.2 (e) – also called
iron. Often the upper surface continued to detaching or slinging) to finally centrifuging
remain red hot at discharge. In subsequent (Fig. 6.2 (f)). In the last named condition,
designs, lifters became integral part of cooler centrifugal force makes the material to stick
design which eliminated this problem. to the cylinder surface and any relative
Figure 6.1(b) represents a situation where movement between the material and the inner
the roughness of the cylinder surface carries surface of cylinder completely stops.
the material up with its rotation until a point In the first and the last condition, the
is reached where the static angle of repose is mixing action of material, which is one of the
exceeded. Angle of repose is the highest angle main strengths of rotary kiln, completely
at which a granular material can rest in a stops and therefore are not desirable in a
stockpile. We have used the term static angle rotary kiln. In the condition (b) i.e., cascading,
of repose to distinguish it from the dynamic the mixing action is not continuous, as also
angle of repose which is the angle of repose the surface renewal. The latter would
in a slightly disturbed condition and somewhat impair the heat transfer to the
upper surface of solid material. Henein (1983) therefore, avoided. Some flexibility in rpm
determined conditions which decide the variation is needed to increase or decrease
change in mode of movement from cascading residence time and, therefore, commercial
to rolling. This happens at different rpm’s for kilns are designed to operate between 0.3 to
different diameters of kiln and the nature of 0.6 rpm and generally between 0.4 to 0.5 rpm.
material, but for commercial rotary kiln
sponge iron plants the rpm value hovers 6.4 MATHEMATICAL TREATMENT OF
around 0.3. For small diameter kilns the value CHARGE MOVEMENT
is slightly larger while for large diameter kilns Let us assume that a particle at point A
the value is very marginally lower. So, 0.3 is (Figure 6.3) moves up to the point B with
about the very lowest value of rpm at which rotation of kiln and then slides down. Due
the rolling action is obtained. to the action of gravity, the particle, instead
Cascading and cataracting actions do not give of sliding down to point A, comes down to a
any special advantage over rolling action in point K as shown in Fig. 6.3(a). Let us call
the context of sponge iron making. On the this cycle of bed rotation as one “Cascade”
contrary, overlapping between solid and (not to be confused with cascading action
freeboard starts which is undesirable from mentioned in the previous section).
reduction point of view. Also higher rpm Therefore, due to one sliding movement, or
increases electrical power consumption as also in one Cascade, the net forward movement
wear and tear on the machinery and is, of the particle is AK.
D/2 D/2 B
B O
A.C α
K C
A,K
(a) Longitudinal sectional view (b) Cross-sectional view

(with exaggerated inclination)
Fig. 6.3 Geometrical considerations of material movement in a rotary kiln

φ is the semi-vertical angle subtended by the

charge bed at the centre of the kiln. α should,
in fact, be greater by an amount equal
to 2 π 2 DN 2 sin φ/(φ 2 g cos α ) than the angle of
X
repose, but this is negligible compared to α.
Further, in the rotary kiln condition the
dynamic angle of repose is more relevant and
A
this value is slightly less than α. Both these
differences tend to annul each other and in
in g
any case they are small and can be neglected.
mb
g C
llin Cli
Ro We are, therefore, reasonably justified in
using the static angle of repose in this
B
formulation.
From Fig. 6.3(b),
X'
BC = AB sin α
F = D sin φ sin α (2)
A' A
And forward movement
AK = BC tan ψ
= D sinα sin φ tan ψ (2(a))
Let N be the speed of revolution of the
kiln in revolutions per minute (rpm). Hence,
C
time spent per cascade
= 2φ /(2πN)
= φ /(πN) (minutes) (3)
Forward movement of the particle per minute
B
Profile along X-X' = (D sin φ sin α tan ψ)/(φ /πN)
Fig. 6.4 Particle movement profile in a charge = (πDN sin φ sin α tan ψ)/φ
bed due to climbing and rolling
Let the length of the kiln be L.
From Fig. 6.3(a), Therefore, time required in covering
Forward movement (AK) distance L,
= BC tan ψ (1) T = L/((πDN sin φ sin α tan ψ)/φ)
where ψ = angle of inclination of kiln with = Lφ/(πDN sin φ sin α tan ψ)
horizontal plane. Thus the residence time (T) should
As shown in Fig. 6.3(b), the angle α is the apparently be
static angle of repose of the bed material and = (Lφ/(πN))/(1/(D sin φ sin α tan ψ)) (4)
6.4.1 Testing of the Mathematical factor of half (Fig. 6.4), which would predict
Expression and its Amendment a residence time which is much more than
When attempts were made to calculate double of the realistic estimates.
residence time in the rotary kiln, both It was reasoned that the explanation of
commercial and experimental (cold and hot), this difference in estimates lies probably in
the values obtained were much higher than the movement caused by bed disturbance. A
the realistic estimates based on filling degree correction is therefore required to be applied
and bulk density. for this factor.
Further, the above formulation is based
on the movement of particles on the surface 6.4.2 Bed Disturbance Factor
layer. The forward movement of particles in If we consider a heap of granular material
the sub-surface layers should be (Fig. 6.5(a)) and disturb it by, say, vibrating
proportionately lesser and the average the base, the material would flow on both
forward movement should be reduced by a sides, if the base is horizontal (Fig. 6.5(b)). If
Fig. 6.5 Flow of material caused by disturbance

Table 6.1 Formulations for residence time in a rotary vessel
Sl. No. Author/ Formulation for residence Type of Remarks
organisation time, minutes formulation
1. Sullivan et al., 5.81 L α NLS C/(DN ψ ) Theoretical + C is a factor representing
U.S. Bureau of Mines Empirical constictions.
Generally equal to 2
2. Ginstling et al 0.4698 L/(DN ψ ) Theoretical +
Empirical
3. Saeman L sin α / π DN( ψ + γ cos α ) Theoretical α, ψ & γ are in radians.
4. Vahl and Knigma 0.91 (L/D)2 (D/h0)0.57 Theoretical + For horizontal

(tan α /N)) Empirical hollow cylinders
5. Chaterjee et al k(L3/F)( α / ψ )1.054 (F/NL3)0.981 Statistical Based on Dimensional
Analysis
6. Bandopadhyay et al Lφ/(πDN (sin α sin φ Theoretical
tan ψ + K/N)) + Empirical
the base were inclined, the material would fp = KD/N

flow more towards the inclination. Thus,
where K is an empirical constant.
material flow can occur in an inclined rotary
kiln even without the geometrical effect of Hence,
rotation, and just by providing disturbance Forward movement per cascade
to bed. In a sponge iron rotary kiln, the = D (sin φ sin α tan ψ + K/N)
cascading movement of the material always and
disturbs the bed.
Residence time, T
Thus, it can be concluded that the
movement of charge material from the upper = Lφ/(πDN (sin φ sin α tan ψ + K/N))
to the lower end of the rotary kiln is due to This formulation is a combination of
the following two factors. theoretical and statistical analyses. One may
(i) Forward movement which can be like to compare this formulation with others
calculated by the geometry of available in literature (Table 6.1).
cascading; and This formulation was used to estimate the
residence time of material in a hot pilot
(ii) Additional forward movement due
sponge iron kiln of nominal dimension of 1.3
to constant disturbance of the bed.
m diameter and 12 m length. To know the
Let us denote this additional
actual residence time a few specialised
movement as fp.
techniques were needed which are presented
Thus, we can state that, in subsequent section. The measured values
Forward movement per cascade and those, which can be calculated using the
above model, are compared in Table 6.2.
= D sin φ sin α tan ψ + fp
f p is expected to depend on the various 6.5 IMPORTANCE OF FILLING
operating parameters like φ, N, D, ψ, etc. The DEGREE
formulation of fp may take a number of forms
For commercial production purpose, operators
depending upon the logic that can be applied
would like to load the kiln with as much
for explaining the additional forward
solids as possible. But, as mentioned earlier,
movement. Since these possible logics are
little better than conjectures, it was decided sponge iron making in a rotary kiln is a
to try out a large number of formulations and delicate balancing exercise, particularly in the
then to observe as to which formulation zone where FeO is being reduced to metallic
explains the experimental results better. iron. This has been illustrated in Fig. 2.9.
Highly reducing condition is required within
The experiments towards evaluation of the
the solid bed so that the CO/CO2 ratio never
above were carried out in cold miniature
falls below the limiting value of 2.5 and the
models of rotary kiln. A number of
formulations were tried out for fp and that CO2 formed as a result of reduction is quickly
particular formulation was selected which reconverted to CO. On the other hand, in the
could predict the results of the experimental gas phase, substantial quantities of tri-atomic
observation with the minimum variance of gases (gas molecules containing three atoms,
error. Based on this analysis the best e.g., CO 2, H 2O) are needed so that heat
formulation of fp was found to be transfer by radiation from gas is not
52 //
Table 6.2 Comparison of the experimentally determined residence time (during iron ore reduction) with the
predictions from the model (i.e., formulation no. 6 of Table 6.1)

hampered. Diatomic gases like CO, H2, O2 and particles, exists inside all the time. Markers
N 2 do not radiate heat (or rather the heat can be tracked only when they reach the
radiation is negligible). If the CO2 and H2O discharge end. As there is always a dam at
molecules come near the solid bed, they the discharge end in a sponge iron rotary kiln,
would not only stop the reduction process the marker, which is larger than the normal
but also would re-oxidise the already formed sized material, finds it difficult to cross over
metallic iron. the dam. The results have mostly given
In order to maintain the presence of inflated values and only lower of a series of
substantial amounts of CO 2 and H 2 O results can be relied upon. However, this
molecules in the gas phase without disturbing cannot be considered a scientific method of
the reduction process, a large free board (a
determination of residence time.
place where there is no solid) is maintained
and the process air is introduced at a location (2) Monitoring of Time Gap Between First
which is far removed from the solid bed. It Charging and First Discharging:
has been found in practice that to maintain This time gap also gives only a rough idea of
an effective rate of reduction, solids have to residence time as the filling degree over the
be filled to a level which is not more than length has not stabilised and this data does
15% of the cross-section. This is one major not reflect the situation under steady state
reason why the productivity of sponge iron condition.
rotary kilns is much much lower than that Only these two techniques have been
of, say, blast furnace. But this does not make
employed to our knowledge in actual sponge
rotary kiln unviable since it does perform
iron kilns. More often than not, agencies
multiple functions not possible in many other
transferring the technology have provided
metallurgical reactors.
statistical formulations to their clients and
have asked them to calculate the residence
6.6 TECHNIQUES OF MEASURING
time accordingly. The need for knowing the
RESIDENCE TIME
actual residence time was not felt necessary
A number of techniques have been employed while operating a plant. But this value is
in the past to measure the residence time in
required when one goes to design a new
actual sponge iron rotary kilns but
kiln.
unfortunately, these have not been
documented. Through informal sources, we 6.6.1 Measurement of Residence Time in
have come to know of the following the Pilot Rotary Kiln of RDCIS SAIL
techniques of measurements: The dimension of the kiln, as mentioned in
(1) Use of ‘Markers’ Chapter 2, was 1.3 m dia and 12 m length.
Inert materials of large sizes have been One effective way of measuring the residence
charged and attempt has been made to trace time here could have been by using
their passage through the hot kiln. It is radioactive tracers. But this involves a safety
virtually impossible to see inside an operating hazard and the idea of using it was dropped.
kiln without stopping it, as heavily dusty It was decided instead to use physical and
atmosphere, caused by flying tiny ash chemical tracers in our experiments.
(a) Physical Tracers poor impulse possibly due to easy flowability

Pitfalls of using a physical tracer like the leading to pronounced back mixing.
‘marker’ described above were realised in (b) Chemical Tracers
the beginning. But a different physical tracer
It was realised early that for more accurate
such as iron ore fines is expected to give a
measurement of residence time one must use
value better than that by marker technique.
chemical tracers keeping the physical
To get rough idea of the residence time, iron
properties as far unchanged as possible, as
ore fines were used as the physical tracer in
the presence of even extra fines, for example,
one set of experiments and small sized iron
can change the movement pattern. Towards
ore pellets were used in another set. Two
this end, the following were attempted as
graphs of such determinations, one for the
chemical tracers.
reduction kiln together with the rotary cooler
and the second only for the rotary cooler part (i) Sulphur as chemical tracer
of the Pilot Plant, have been presented in Fig. Iron ore lumps were soaked in calcium
6.6 & 6.7. While experiments with iron ore sulphate solution and dried. They were then
fines yielded significant impulses in the size charged for a specified time instead of the
plot of discharge material, pellets gave very normal iron ore. The product was regularly
Fig. 6.6 Residence time plot for an experiment with fine iron ore as tracer covering both rotary
kiln and cooler
analysed for sulphur content in a Leco C-S (iii) Manganese as tracer

analyser. The time gap between the mean Manganese was used as a chemical tracer in
sulphur impulses from the charging point to two ways. In the first set of experiments, iron
the sampling point gave the time of travel ore was soaked in a strong permanganate
between these two points. In this case, solution, dried and then used. In another set
however, the impulse obtained was feeble. of experiments, Manganese ore itself was
(ii) Carbon as chemical tracer crushed to the same size as iron ore and then
Cold bonded pellets impregnated with coal charged in the kiln.
The composite product was analysed for
were prepared of roughly the same size as
manganese and the concentrations plotted.
iron ore. Iron ore was replaced with these
Figure 6.8 gives the plot of one particular
pellets at a given time and the magnetic
experiment with Manganese ore as tracer.
product was analysed for carbon content,
since the composite pellets were expected to (iv) Use of copper and sulphur simultane-
give sponge iron containing higher carbon. ously as chemical tracers
Residence time was determined as before. Iron ore was soaked in strong copper
Fig. 6.7 Residence time plot for an experiment with fine iron ore as tracer in the rotary cooler
sulphate solution and dried for the purpose (of different kind) were used as raw material.
of residence time measurement. Surprisingly These results are not presented here.
the copper concentration impulse in the
Iron ore fines as tracer could give
product was not significant. But this
significant impulse in the product size plots
technique proved to be a more successful
even at 30 kg or six minutes of feeding. To
variation of the sulphur tracer technique.
have a measurable impulse with chemical
The success of all the experiments
tracers, 12 to 15 minutes of feeding of the
depended on getting a measurable impulse
tracer containing inputs was found to be
of the tracer at sampling point. In some of
necessary. But with iron ore soaked in
the early experiment, no tracer impulse could
calcium sulphate, even higher quantity of
be detected even after careful sampling and
feeding was required for obtaining a
analysis at close intervals. The effect depicted
measurable sulphur impulse. For a
in Fig. 6.5 also caused back mixing of the
commercial kiln, an even higher tracer input
different material charged with the regular
impulse may be required for a detectable and
material of the kiln, making the impulse
reliable output impulse.
flatter and flatter with travelling of larger
distances and passage of time. This technique is simple in concept, and
the results can be utilised for optimising
For the Pilot Plant condition, and while
operations of an existing kiln.
using iron ore as feed material, pellets were
found to be a poor tracer, as even 30 kg of
input could not given a significant impulse 6.7 IMPORTANCE OF THROWING
in the product size plot. However, pellets COAL FROM DISCHARGE END
were found useful as tracer when pellets The final stages of reduction in a sponge iron
Fig. 6.8 Residence time plot obtained while using manganese ore as tracer
rotary kiln are the most important as well as variations of rotary kiln sponge iron making.
the most difficult. As FeO concentration Air is the universally adopted medium for
reduces, the rate of reduction drops. Further, carrying and giving momentum to coal
since the char in the solid bed, formed from particles, as air is anyway needed within the
the feed coal, has spent a few hours at high kiln at this location. Since too much air would
temperature, has lost much of its reactivity, render the atmosphere too oxidising to
i.e., its capacity to convert CO2 quickly back disturb the reduction condition within the
to CO. This also reduces and, beyond a limit, bed, throwing has to be achieved with
stops the reduction process. controlled amount of air.
If fresh coal is introduced during the last From the weigh feeder (or the volumetric
stages of reduction, fresh reactive char can feeder as the case may be) coal would fall in
be made available, which can enhance the irregular spurts. If this is led directly to the
reduction rate and thus greatly enhance the pipe carrying air, there is every chance of
viability of the process. This is done by blocking the air pipe. To avoid this, coal feed
propelling some coal (throwing or slinging) has to be regulated through a feeder, which
to some distance in the reduction zone. would release coal in small quantities at
Since the solid charge is continuously regular intervals. This is done by introducing
moving forward, dropping coal at the kiln a star feeder (having grooves arranged
mouth would be of little use since coal would around a shaft in a star fashion) or some
immediately accompany the kiln discharge variation thereof.
into the cooler. Coal has to be projected or Air in the coal throw pipe comes with a
propelled to some distance in the reduction certain velocity governed by the flow rate
zone. and the cross-sectional area of the pipe. Pipe
Throwing coal from discharge end has is kept roughly parallel to the axis of the kiln.
now become standard practice in almost all Coal particles initially have zero velocity in
Fig. 6.9 A typical coal throwing arrangement

the direction of the kiln axis. The air flow as the particle gains velocity and therefore
drags the coal particles along by its velocity. acceleration rate keeps decreasing, until the
The coal particles gain acceleration within the time when the mixture comes out of the pipe.
throw pipe depending upon the particle Air velocity drops to negligible value and the
surface area, its weight and the difference in velocity of the projected coal starts
the velocities between the particle and air. decreasing, until it falls by gravity on the
The difference in velocity keeps decreasing solid charge bed.
(a) Throw distances with 65 mm air pipe & 7 metre pipe length
7 m Pipe
length, Air
velocity 138
m/s, Air flow
1500 Nm 3/h
7 m Pipe
length, Air
velocity 230
m/s, Air flow
2500 Nm 3/h
7 m Pipe
length, Air
velocity 193
m/s, Air flow
2100 Nm 3/h
Throw distance, m
(b) Throw distances with 65 mm air pipe & 10 metre pipe length
10 m Pipe
length, Air
velocity 138
m/s, Air flow
1500 Nm3/h
10 m Pipe
length, Air
velocity 156
m/s, Air flow
1700 Nm3/h
10 m Pipe
length, Air
velocity 184
m/s, Air flow
2000 Nm3/h
10 m Pipe
length, Air
velocity 166
m/s, Air flow
1800 Nm3/h
10 m Pipe
length, Air
velocity 170
m/s, Air flow
1850 Nm3/h
Throw distance, m
(c) Throw distances with 65 mm air pipe & 15 metre pipe length
15 m Pipe
length, Air
velocity 138 m/s,
Air flow
1500 Nm3/h
15 m Pipe
length, Air
velocity 92 m/s,
Air flow
1000 Nm3/h
15m Pipe
length, Air
velocity 184 m/s,
Air flow
2000 Nm3/h
15m Pipe
length, Air
velocity 156 m/s,
Air flow
1700 Nm3/h
15m Pipe
length, Air
velocity 147 m/s,
Air flow
1600 Nm3/h
Throw distance, m
Fig. 6.10 Calculated distances of throw of coal particles under different conditions
6.8 COAL THROWING PHILOSOPHY On integration between limits when time

Air velocities used in a coal throw pipe are t = 0 also u = 0 and till any time t, we get, for
quite high and always in the turbulent zone. a particular case,
Turbulent flow formulations would apply. As 1/(Va – u) = 1/Va + 0.02394*t
per this,
Here we have assumed the following
FD = 0.44* π *dp2* ρ* (Va – u)/8
values as an example,
where,
dp = 12 mm
FD = Drag force on the coal particle
Va = 120 m/s
dp = Coal particle diameter (Assumed
spherical) ρa = 1.21875 kg/m3
ρa = Density of air ρp = 1400 kg/m3
Va = Velocity of air From these results it can be calculated that

for the coal particle (12 mm spherical) to
u = Velocity attained by coal particles attain 25% of the air velocity only 1.9 m pipe
at any distance x within the pipe length is enough. If 50% of air velocity is to
Acceleration rate of a coal particle is be attained, 12.8 m pipe length is required.
given by For attaining 75% or higher velocity the
fc = FD/m length requirement becomes prohibitively
high, 67.4 metres or more.
where,
Once the coal particle emerges out of the
m = Particle mass throw pipe, the forward drag force stops.
Instead, the particle is dragged back by the presented in Fig. 6.10. It can be observed that
practically stagnant gas in the kiln freeboard. air velocities can be substantially reduced for
Similar considerations in the reverse the same distance of throw if longer pipe
direction apply. If, in this particular case, the lengths are used. This fact has been applied
free fall vertical distance available to coal in the sponge iron rotary kilns based on the
particle is 1.6 m, the calculated distances OSIL Technology.
covered by the particles of various sizes are
Requirement, Generation
7
and Transfer of Heat in
a Sponge Iron Rotary Kiln CHAPTER
7.1 MATERIAL AND ENERGY the process (entirely as chemical energy in

BALANCE IN A CONVENTIONAL coal), more than 2 GCal or over 37% is lost
ROTARY KILN SPONGE IRON in waste gases. Many units having multiple
MAKING PROCESS kilns use the gases to generate electricity and
We know that whatever material or energy sell to local grids. While that is a very
we put into a continuously operating reactor welcome by-product, the new entrepreneurs
who want to make a modest start with, say,
must come out in some or the other form. In
a 25 or 50 tpd single module plant is placed
a continuous and steady process, there cannot
at a great disadvantage. The situation calls
be any accumulation in the process. When one
for evolving means to plough back the waste
attempts to list the inputs and outputs of both
gas energy into the sponge iron making
material and energy, a greater understanding
process itself.
of the process is achieved. The same is true
of the rotary kiln process of sponge iron
making. 7.2 GENERATION AND TRANSFER
A typical material balance of a OF HEAT
commercial rotary kiln sponge iron plant is In a sponge iron rotary kiln, the overall
presented in Fig. 2.5. It can be seen that reduction process is endothermic and tends
while about 2.5 tonnes of solids are charged, to lower the bed temperature.
only about 1.4 tonnes of solids are
discharged. About 1 tonne of solids are Fe2O3 + 3/2 C = 2Fe + 3/2 CO2
gasified in the process and the balance is ΔHo298 = +55,225 kCal/kg-mole
carried away as dust in the waste gas stream. (Reaction no. 6 in Table 3.1 — ‘+’ sign
indicates that heat is absorbed)
A typical energy balance of a commercial
rotary kiln sponge iron plant is presented as To make up for the lost heat the free-
a Sankey Diagram in Fig. 2.6. It is disturbing board gases are maintained at a substantially
to see that out of 6 GCal energy pumped into higher temperature; and the heat transferred
out of the gases to the bed and refractory is H1 = Conductive and convective heat
made up through controlled combustion of transfer from gas to charge bed.
CO in free board gas to CO2. H2 = Radiative heat transfer from gas
2CO + O2 = 2CO2 to charge bed.
∆Ho298 = –135,260 kCal/kg-mole H3 = Conductive and convective heat
(Difference of reactions numbered 3 & 4 transfer from gas to refractory.
in Table 3.1; ‘–’ sign indicates that heat is H4 = Radiative heat transfer from gas
evolved) to refractory.
This reaction is also written as H5 = Conductive and convective heat
transfer from refractory to charge
CO + ½ O2 = CO2 bed.
∆Ho298 = –67,630 kCal/kg-mole H6 = Radiative heat transfer from
Heat from free board gas reaches the refractory to charge bed.
granular solid bed in both direct and indirect H7 = Conductive heat transfer through
manner. Since the gases “see” the charge bed refractory to kiln surface.
only about 25% of the time, the bulk of heat H8 = Convective heat transfer from
transfer takes place through a via-medium i.e., kiln surface to atmosphere.
the refractory surface. The heated refractory H9 = Radiative heat transfer from kiln
surface, in turn, transfers the heat to the solid surface to surroundings.
charge while it goes under the solid bed and
they come in intimate contact during rotation.
There is also transfer of heat through
refractory on to the kiln surface and to the
outside atmosphere. This heat is a loss to the
process and should be kept to a minimum
with suitable insulating layers in the
refractory.
7.3 MATHEMATICAL TREATMENT OF

HEAT TRANSFER
Heat transfer in rotary kiln takes place
through all possible modes viz., conduction,
convection and radiation. While both
conductive and convective heat transfers are
linearly dependent on the temperature
difference, the radiative transfer is
dependent on the fourth power of the
Fig. 7.1 Modes of heat transfer in a rotary kiln
absolute temperature. For mathematical
treatment, nine heat transfer terms can be These nine heat flow terms can be ex-
defined which are pictorially represented in pressed in terms of heat transfer co-efficients
Fig. 7.1. These terms are: as follows:
Requirement, Generation and Transfer of Heat in a Sponge Iron Rotary Kiln // 63
H1 = A1 ( Tg – Tb ), from the freeboard gas to the refractory. This

represents almost two and a half times higher
4 4
H2 = A2 ( Tg – Tb ) rate of heat flux (rate of flow of heat per unit
area) from refractory surface to solid bed as
H3 = A3 ( Tg – Tr ) compared to the heat flux from the freeboard
gas to the refractory surface. The higher heat
4 4 flux is made possible due to intimate
H4 = A4 ( Tg – Tr )
interaction and close contact between
H5 = A5 ( Tr – Tb ) refractory surface and the solid charge bed
as it goes under the latter during kiln
H6 = A6 ( Tr4 – Tb4 ) rotation.
H7 = A7 ( Tr – Ts ) Based on empirical determination of
coefficients by earlier investigators and the
H8 = A8 ( Ts – Ta ) knowledge of refractory properties, the heat
transfer coefficients A1 to A9 can be evaluated
H9 = A9 ( Ts4 – Ta4 ) and the heat flux can be estimated. Such an
estimate would be reliable provided we have
Here T refers to absolute temperature reliable estimates of T terms. Measurement
and the subscripts stand for the following: of Tg is often tricky as radiation from sensor
g = gas tip to the refractory surfaces in the range of
vision (always cooler than gas), tends to
b = charge bed
record a lower than actual value. Only a
r = refractory surface suction pyrometer is likely to record an
accurate gas temperature.
s = kiln surface
Kiln operators, in fact, hardly ever need
a = atmosphere/surroundings to know the actual gas temperature. They
Coefficients A1 to A9 are the relevant heat adjust the various air valves in central burner,
transfer coefficients. Temperatures especially shell mounted air fans etc. to maintain a
in the reduction zone are in such a range that desired level of solid bed temperature. But,
the radiative heat transfer coefficients as and when accretion formation starts,
command a higher weightage than heat operators do want to control the gas
transfer by other modes except for the A5 temperature at the location of accretion to
term, which is the conductive heat transfer prevent rapid growth. It is a good idea to be
term for the heat transferred from refractory guided by the temperature measured by
to the granular solid bed. Schnabel (1977), Quick Response Thermocouple (QRT), not
based on his own heat transfer model, because QRT records gas temperatures
predicted that during the 108o arc of contact accurately, but its indication is closer to
between refractory and the solid charge, refractory surface temperature, and it is the
refractory loses about 80 oC temperature, refractory surface temperature which is more
which it regains during the next 252o exposure important in controlling the growth of
to radiative and convective heat transfer accretion.
7.4 SEGMENT-WISE AIR temperature profile in the kiln. But as and

REQUIREMENT when the process goes out of control, the
It is often a matter of great debate as to what blame very often goes towards not
is the standard amount of air required to be maintaining a standardised air flow pattern.
put in through the central burner and the Figure 7.2, which is nothing but a
various shell air fans. Operators keep on spreadsheet (Excel) chart, represents a
adjusting these flows in order to get a desired simplified method of theoretically estimating
1 2 3 4 5 6 7 8 9 10 11 Segment
Number
Assumptions:
Iron ore rate = 7.5 tph;
Fe = 64%;
Solid Bed Coal rate = 5.6 tph;
Temperature Feed coal = 5.6 tph;
[Measured or 900 1000 1000 1000 1000 1000 950 850 700 400 0 Thrown coal = 2.8 tph,
estimated). °C Fixed carbon in coal =
2.8 tph,
Degree of Thrown coal = 2.8 tph,
reduction 95 95 90 85 80 70 50 30 10 5 0 Fixed carbon in coal =
(Interpolated), % 39% (Wet basis);
VM = 26%; Ash = 30%;
Increase in Degree
Fixed carbon/Total
of reduction over Iron = 0.455;
0 5 5 5 10 20 20 20 5 5 0
previous segment, of the 2.8 tph
% thrown coal, the
following distribution
0.1294
0.1294
0.0000
0.0000
Heat required for
reduction of iron 0.1294 0.1294 0.2589 0.5177 0.5177 0.5177 0.1294 is assumed:
ore. GCal/hr
Heat required for Segment 3 - 7%
0.4256 0.7065
Segment 4 - 22%
heating iron ore and 0 0 0 0 0 0.0000 0.1766 0.2649 0.5249 0 Segment 5 - 34%
flux, GCal/hr Segment 6 - 27%
0.0000
Heat required for Segment 7 - 10%

heating feed coal, 0 0.0000 0.0000 0.0000 0.0367 0.0734 0.1101 0.2202 0 Flux rate 350
GCal/hr kg/hr; Heat
required for full
Heat required for Reduction (100%):
0 0 0.07448 0.2341 0.3618 0.2873 0.1011 0 0 0 0 Fe2 O3 + 3/2C =
heating thrown
3
coal. GCal/hr 2Fe + 2 CO2;
Allocation of ∆H = 55, 225
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
0.2659
radiation and kCal/kg-mole

unaccounted
losses. GCal/hr
0.6294
0.2659
0.3953
0.4698
0.5274
0.1959
0.1348
0.1587
0.8865
0.2659
0.1455
2447.1
3263.1
568.1
568.1
2424.2
2475.4
844.6
1003.7
1344.7
1893.9
2554.9
Air required in the

segment, Nm3/hr
Fig. 7.2 A simplified method of calculating requirement of air in different kiln segments
Requirement, Generation and Transfer of Heat in a Sponge Iron Rotary Kiln // 65
the air requirements in the various segments has been observed that all the air put in by
of the kiln. Portions before and after the kiln the central burner and the coal thrower are
has also been considered as segments, as some not utilised in these segments. As a rough
portion of the process continues there. In this estimate we can say that only 70% of air is
example, the process has been divided into utilised and the remaining amount is utilised
eleven segments. Nine of these are the in the subsequent segments. The same may
sections of the kiln itself, based mostly on be happening in the subsequent segments as
the location of the air tubes. Segment number well in the reduction zone. In the preheating
1 comprises the product discharge transfer zone, the air requirement increases rapidly,
chute and the cooler, while segment 11 the gas volume increases and the flow rate
comprises the feed pipe, the inlet hood and becomes highly turbulent. The utilisation
the ABC chamber. factor should then be higher although an
In chapter 5 we have seen a method of assessment has not been made so far.
determining the total air requirement of the There is possibility of extending this type
process. Figure 7.2 demonstrates how we can of analysis for theoretically estimating the
allocate the requirement to the various average level of gas and refractory surface
segments. In the current example, air temperatures in the different segments.
requirements of segments 1, 2 and 3 have to Refractory temperature in particular is an
be met by the air put in through the central indicative measure of proneness to accretion
burner and the coal thrower. In practice it formation and growth.
This page
intentionally left
blank
8
Reaction Kinetics CHAPTER
The thermodynamic properties, especially 8.1 FACTORS AFFECTING REACTION

the free energy change of a reaction, RATES
establishes whether a particular reaction is As implied earlier, the science of Reaction
feasible or not, but it does not indicate at Kinetics studies rates of chemical reaction.
what rate the reaction would take place. To Chemical reaction rates (meaning mass of
enhance the rates of chemical reaction, reactants transformed per unit time)
increasingly higher temperatures have been generally depend on the following:
employed over the years, as in the case of • Concentrations (or more accurately the
extraction of iron, and it has resulted in thermodynamic activities) of reactants
greater success of the indirect processes over • Temperature of reaction – Increase in
the more direct sponge iron making temperature increases reaction rate
processes. But persistent attempt at • Physical state of reactants – whether
enhancing the rate of solid state reduction in a gaseous phase or in a solution
process has led to re-emergence and (homogeneous reactions) or in
commercial success of the lower temperature separate phases (heterogeneous
and solid state reduction processes, such as reaction as in a sponge iron rotary
the sponge iron making process. Although kiln) – whether in solid blocks (low
the attempts at enhancing the rate of solid surface area) or in powder, colloidal
state reduction has largely been by trial and or emulsion form (high surface area).
error, but the science of the study of Higher the surface area, higher are
Chemical Reaction Kinetics (or simply the reaction sites and therefore higher
Reaction Kinetics) has been a significant are the reaction rates.
contributor. For further enhancement of • Presence of Catalyst or Inhibitor
rates, Reaction Kinetics is expected to play a • Means of excitation – Light, Ultraviolet
key role in future. or Radioactive Radiation, etc.
8.2 RATE LAW AND ORDER OF calculated. This particular relationship is said
REACTION to hold for what is called a “First Order
The most simple reaction is where one Reaction.”
constituent breaks up into a number of
product constituents; e.g., A = B + C + D +.... 8.3 BIRTH OF THE RATE LAW OR LAW
If there is no appreciable reverse reaction OF MASS ACTION
then rate of reaction is represented by In 1799 Proust propounded the Law of
–d[A]/dt, where parentheses ‘[]’ represent Constant Proportions which stated that
concentration of the constituent A. The sign ‘elements combine in definite proportions by
is negative because the concentration of A is weight and the composition of a pure
decreasing. chemical compound thus formed is
‘Rate Law’ states that the rate of reaction independent of the method by which it is
is proportional to the concentration of the prepared.’ Bertholet opposed this law bitterly
reactant(s). Hence, and spent next nearly sixty year trying to
bring evidences against it only to be rebutted
– d[A]/dt = k[A]
one by one. In his quest for gathering such
k is referred to as the Velocity Constant evidences it finally dawned on him and his
of Reaction co-workers that he had been bringing up
or, d[A]/[A] = – kdt evidences of partial combination and the
Integrating, we get extent of combination, in some way,
depended on the concentrations of the
ln [A ] = – kt + C
reactants used. From these results, Guldberg
If, in the beginning (i.e., when t=0), [A] and Waage in 1864 deduced the Law of Mass
= [A] 0 Action, also known as the Rate Law. This law
then, C = ln [A]0 stated that “the rate at which a substance
and ln ([A]/[A]0) = –kt reacts is proportional to its active mass,” and
“the velocity of a chemical reaction is
or [A]/[A]0 = e–kt
proportional to the product of the ‘active
If, in time t, x amount of A has reacted, masses’ of the reactants.” The active mass
then, was assumed to be proportional to the molar
[A] = [A]0 – x concentration, as has been used in derivation
and [A]0 – x = [A]0 e–kt in the last section.
or x = [A]0 (1 – e–kt)
8.4 UNIMOLECULAR REACTION AND
On the other hand, FIRST ORDER REACTION
k = (1/t) ln ( [A]0/[A] ) In the deviation of section 8.2 we have
= – (2.303/t) log {([A]0 – x)[A]0} taken the example of a unimolecular
= – (2.303/t) log (1 – x / [A]0 ) reaction, i.e., a single molecule of A breaks
into a number of molecules. We have also
A plot of – log (1 – x / [A]0 ) against time
said that the final derivation
t would be a straight line with a slope of
k/2.303 from which value of k can be – log(1 – x/[A]0) = (k/2.303)t
Reaction Kinetics // 69
is applicable to a first order reaction. 8.5 DECOMPOSITION OF AN OXIDE

By a similar logic, for a bimolecular Let us consider a chemical reaction where an
reaction of the type A + B = C + D + ....... oxide breaks into its elements.
and starting with equal concentrations of A MO = M + ½O2
and B (i.e., [A]0 = [B]0) we get the relationship Degree of reduction, in such a case, is
1/(1 – x/[A]0) = 1 + akt defined by the ratio of “oxygen removed
from the oxide” to the “total removable
and this applies to a second order reaction. oxygen.” and is represented by the symbol
Although the concept of ‘Order of α . Thus
Reaction’ has evolved from the molecularity α = 1 – [MO]/[MO]0
of reaction, but they need not be the same. and a plot of – log(1 – α ) with t at any given
While molecularity refers to the number of temperatures should give a straight line with
molecules involved in a chemical reaction, the a slope of k/2.303 from which the value of
order of reaction must be derived from the velocity constant of reaction at that
study of the variation of reaction rates with temperature can be calculated. This way the
time, concentration and (as we shall see later) reaction is found to be of the first order. But
temperature. if the plot is not a straight line, the reaction is
not of first order, even though it remains a
Transition state Activation energy barrier
Barrier is
reduced in
Energy
presence
Activation energy
of a
catalyst
A
Free energy change
∆F
reaction
of
B+C+D+
Reaction path
Fig. 8.1 Schematic diagram representing activation energy

barrier and the effect of a catalyst
unimolecular reaction. Very often the increase in temperature would mean that
reactions take place in a number of steps and molecules in general have acquired higher
are much more complex than what they appear energy and there is an increased probability
at first glance. We shall see later that the of greater number of molecules crossing that
reduction of iron oxide to metallic iron in a barrier in a given time.
rotary kiln is also highly complex and the Thus
actual mechanism of reaction can only be
ln k = C´ – E/RT
conjectured at the moment.
or, log k = C” – (E/4.576)(1/T)
8.5.1 Degree of Reduction where R and E are expressed in calories.
We have defined degree of reduction in the Thus, for evaluating E, we need to plot log k
last section. If we take that definition literally against 1/T and measure the slope of the
for different types of iron oxides (such as resulting straight line.
hematite and magnetite) it would become
highly difficult to compare results from
different sets of experiments.
For circumventing this problem, we define
α = 0 for that iron oxide where all iron atoms
are present as Fe2O3 molecules. If we have to
start a study with a sample of say magnetite,
then we say that the sample, to start with, is
already 11.11% reduced and α = 0.1111. It then
becomes easier to compare results of different
studies while using such a basis for defining
degree of reduction.
8.6 TEMPERATURE DEPENDENCE OF

VELOCITY CONSTANT:
THE ARRHENIUS EQUATION
Fig. 8.2 Typical plots of degree of reduction
We have so far dealt with velocity constants with time at different temperatures
of reactions at a particular temperature. In
1887 Arrhenius postulated that the velocity 8.7 EXPERIMENTAL DETERMINATION
constant k of chemical reactions change with OF ACTIVATION ENERGY
temperature as per the relationship
In case of decomposition of an oxide as
k = Ce–E/RT described in section 8.5, plots of degree of
E is referred as the activation energy and reduction against time would be of the type
it can be visualised as a barrier (an energy given in Fig. 8.2. These plots for different
barrier) to the chemical reaction (Fig.8.1). temperatures are experimentally determined.
Reacting molecules must acquire, during Various experimental techniques are followed
random collisions, interactions or excitation, to obtain the desired results, but Thermo-
enough energy so that it is able to cross the gravimetry is a quick and convenient method
dam-like barrier and form the product. An if adequate precautions are taken.
Once these results are obtained then been discussed. Reaction can be of higher
–log(1– α) values are plotted against time t, order (third order and above) as also of
and they should be straight lines if the reaction fractional order. Reactions can also be of zero
is of first order. If reasonable linearity is order where there is no effect of concentration
observed then their slopes are measured on reaction rate. But we hardly ever come
(Fig. 8.3) to obtain k or velocity constant across such a reaction.
values. Thereafter, log k values are plotted
against 1/T (reciprocal of absolute tempera-
T3
tures), and slope of this plot is a measure of
activation energy. In fact, in caloric terms, the
slope is equal to –E/4.576, where E is the T2
activation energy (Fig. 8.4).
If plots in Fig. 8.3 are not straight lines
–log (1–α)
the reaction is not first order. For second
order reaction, instead of – log (1 – α), a T1
different function of α needs to be plotted

along the Y-axis. This function is 1/(1 – α).
This time the plots would be straight lines, Slopes = k/2.303
but not passing through the origin.

Instead of trying to find out the order of Time, t
reaction by trial and error, a short cut is α ) with time (Straight
Fig. 8.3 Plot of –log (1–α
normally applied by using what is called a Lines denote First Order Reaction)
reduced time plot. In this plot α is plotted
against a dimensionless time, which is
obtained by taking ratio of the actual time
with the time required to attain α = 0.5. On
the plot obtained experimentally, theoretical
plots of reaction by different mechanisms are
super-imposed and then the matching with a
particular mechanism can be quickly
observed. Many times in a particular
reduction region, one mechanism may be
followed, which would be giving way to
another mechanism in another region, and
this also can be deduced by comparing with
Fig. 8.4 Plot of log of velocity constant
standard plots (Fig. 8.5). against reciprocal of absolute
temperature (Arrhenius plot)
8.8 VARIATION OF REACTION RATE
WITH TEMPERATURE There is a thumb-rule often quoted for
The treatment presented in the proceeding Metallurgical Reactions, according to which
sections is applicable to the rates of reactions. if we increase temperature by 10ºC, chemical
First order and second order reactions have reaction rates get doubled. It is very unclear
as to how this thumb-rule came into 8.9 ROLE OF DIFFUSION

prominence. On analysing the variation in In case of a gas-solid reaction, as in the
velocity constant at about 1000ºC we find that example cited in the last section on sponge
the above reasoning holds only if the iron making, the reaction product, if solid,
activation energy is of the order of 220 kCal/ tends to form a barrier in the interaction
mole. process. The gaseous species has to diffuse
Let us take the example of iron oxide through the product layer in order that
reduction in solid state by carbon monoxide, further reaction is continued. Alternatively,
as it happens in the sponge iron making the solid reactant may dissolve in the product
process. layer, or in some other way diffuse through
FeO + CO = Fe + CO2 the product layer to interact with the gas
Even though it is a bimolecular reaction, phase. If these diffusion rates are sufficiently
a large majority of researchers who fast, the reactions may still proceed, with the
investigated it experimentally, have reported chemical reaction rate being the controlling
it to be of first order. The activation energy step, and activation energy of the chemical
values though, vary very widely, and are reaction would determine the variation of
much lower than that expected from other reaction rate with temperature. This probably
similar reactions. It is reasoned that the happens in the sponge iron making process
metallic iron which initially forms, catalyses as a majority of researchers have reported
and therefore enhances further reaction. If this process of sponge iron making to be
the activation energy value is in the range of “Chemical Reaction Rate Controlled.”
about 55 to 60 kCal/mole, then the reduction But if diffusion rates are too slow the
rate would be enhanced by only 20% when activation energy of chemical reaction would
temperature is increased by 10ºC. not have any role in the progress of reaction
and we would have a situation where the
progress of reaction is ‘Diffusion Controlled.’
In intervening regions where both chemical
reaction rate and rates of diffusion have
influence on overall progress of reaction we
say that the reaction progress is in the ‘Mixed
Control’ domain.
We would not discuss here the diffusion
mechanisms and diffusion controlled
processes. The treatment is complex and
other comprehensive documents may be
referred for this purpose. We would only
emphasise the need to find means to
circumvent diffusion steps in reactions
involving condensed phases (either solids or
Fig. 8.5 Reduced time plots for different liquids) as diffusion barriers often make
orders of reaction industrial exploitation of process unviable.
FeO Fe2O3 /Fe3O4
[CO]=CO(g)
[C]+[O]=CO(g)
FeO=Fe(m)+[O]
2CO(g)=CO2(g)+[C]
Double arrow depicts mass transport
Fig. 8.6 A possible mechanism of pore formation during reduction of iron oxide
8.10 HOW SPONGE IRON MAKING IS large volume reduction (over 50%) during
A FIRST ORDER REACTION? sponge iron making, a lot of pores form on
As mentioned earlier that although the the product layer, which is formed by the
following reaction reduction on surface of iron ore lump or
pellet. While these pores facilitate easier
FeO + CO = Fe + CO2
access to reducing gases, it is not necessary
is bimolecular, a very large number of for CO molecules to actually meet FeO
researchers have found it to be of first order. molecules to effect further reduction. Carbon
A few reasons may be responsible for this
and oxygen have significant solubilities in
observation.
metallic iron. The dissolved carbon and
• FeO being in condensed phase, its oxygen atoms within the metal layer probably
activity (representing concentration) form an associated molecule (CO dissolved
does not change with time and in metallic iron), which get liberated on the
therefore only the concentration of surface to give CO gas and thereby effecting
CO is affecting the rate of reduction. the reduction of FeO.
• The reaction is, in some way, being
2CO(g) = CO2(g) + [C]
catalysed. Many say that the catalysis
is by the iron metal, which is produced Here, parentheses ‘[]’ mean that the
in the reaction. During catalysis, an constituent is dissolved in Fem layer.
intermediate single molecule is FeO(s) = [Fe] + [O]
formed, which then breaks into the x[C] + x[O]= [CO]x
products. Therefore, the reaction is of [CO]x= x CO(g)
the first order. It is open to future researchers to
Largely following the second reasoning, investigate and prove or disprove this
a hypothesis on the mechanism of reduction hypothesis. At the time of writing it appears
has been presented in Fig. 8.6. Since there is to be a plausible hypothesis.
This page
intentionally left
blank
9
Raw Materials for
Sponge Iron Making CHAPTER
Iron Ore, or more strictly the iron oxide feed, in excess of 1200°C for it to harden. Till
is the source of iron in sponge iron making. recently the higher cost of pellets prevented
Coal is the other major raw material which is its economic use in rotary kiln, but with
used both as a fuel and the reductant in this reduced availability and higher cost of quality
process. The third raw material is the flux, lump ore, coupled with the understanding
either limestone or dolomite, which is used that pellets improve kiln productivity, pellets
to desulphurise iron ore and to prevent are now becoming a viable alternative.
sulphur in coal from entering in sponge iron. Unlike pellets, sinter cannot be used in
Since flux is used in only minor quantities, sponge iron making. Sinter is a very suitable
and its quality has negligible influence on the feed for blast furnace iron making as the
process, it will not be discussed here. emphasis there is on melting rather than solid
state reduction. Since sinter is formed by
9.1 IRON OXIDE FEED surface fusion, the iron oxide structure and
India is fortunate enough to be possessing properties have changed and it is not
large quantities of high grade iron ore amenable to solid state reduction. Iron ore
deposits. These are mined, sized and used briquettes, formed by cold compaction of iron
directly in sponge iron kilns as well as in other ore fines along with a binder, have sometimes
metallurgical reactors like gas based sponge been tried, but without much success. During
iron shaft furnaces and iron making blast reduction, briquettes tend to decrepitate or
furnaces. The other iron oxide feed which is break up resulting in production of large
used in rotary kiln is the iron ore pellets. quantities of fine sponge iron of reduced
Pellets are formed by what is called the value. Also when the kiln carries a large load
“balling” when finely ground iron ore mixed of fines, accretion or ring or deposit formation
with small quantities of moisture and suitable is accelerated.
binders is rotated or rolled over the surface
of a rotating disc or drum generally along 9.2 IRON ORE LUMPS
with slight sprinkling of water. These pellets Almost all high grade iron ore deposits of
are “green” and weak and are fixed in furnace India contain Hematite (Haematite Fe2O3)
mineral. These are associated with undesir- and also the magnetite in BMQ into hematite.
able (gangue) minerals containing Silica SiO2, In this process we get what is called
Alumina Al2O3, etc. Pure hematite has the metamorphosed hematite and in the latter
chemical formula Fe2O3 and therefore contains case, Banded Hematite Quartzite (BHQ). Such
69.94% iron, balance 30.06% being oxygen. hematite deposits have relatively open
Presently, sponge iron plants are able to get structure and are more easily reducible than
iron ore lumps containing about 64% iron. magnetite. However, BHQ remains heavily
From the ore deposits, ore is taken out contaminated with silica. But freaks of nature
by removing the overburden, blasting the has resulted in natural beneficiation of such
iron ore seams, excavating and crushing and deposits. Action of heat, acidic water, air and
sizing. While the sized ore is consumed by sun has resulted in slow dissolution of this
iron making units, ore fines left at mine site iron oxide, its collections in ponds and lakes
are not usable directly and needs to be formed in volcanic craters nearby and slow
beneficiated and/or agglomerated. Further, but massive precipitation through formation
during mining, seams of lower grade material of sediments. These deposits are called
are also excavated, which do not find ready sedimentary type and most of the high grade
consumers and are often kept aside for a hematite deposits are of this type. Since the
possible future treatment by beneficiation and deposit has formed by low temperature
agglomeration (mostly pelletisation). precipitation, it has an open and porous
structure and is therefore highly reducible.
9.3 MAGNETIC AND HEMATITIC ORES At some places, these deposits have been
We have so far talked mainly about hematite enriched further by the action of basic water
(also spelled haematite) ore deposits. But in which leaches away silica. But removal of silica
nature magnetite or Fe3O4 formed first. Hot weakens the bonding of deposit and we get
liquid volcanic eruptions brought out liquid a friable deposit, sometimes even blue dust.
iron silicate at some locations. Silica being The Gorumahisani Iron Ore Mines is an
higher melting, solidified into a silica rich example where both igneous and sedimentary
quartz layer. Solidification of silica resulted iron ore occur in close proximity. Sedimentary
in enrichment of adjoining liquid layer in iron lakes had formed within the crater of the
oxide which latter solidified into magnetite, Gorumahisani Hills.
being the stable compound at that
temperature, i.e., near 1500° C. The process 9.4 MAGNETIC ORE MINERALS
resulted in what is called Banded Iron Magnetite or Fe3O4, derives its name because
Formation (or BIF), having layers of quartz of its magnetic property. Its magnetic
and magnetite of varying thickness. When the intensity is about 40% that of pure iron. Since
layer thicknesses were small, Banded most magnetite deposits are igneous and
Magnetite Quartzite (BMQ) was formed. difficult to reduce, sponge iron producers
Thus, most of the magnetite deposits and avoid using it in rotary kiln. But being
BMQ’s are of igneous nature (i.e., formed magnetic does not always mean the ore is
directly from volcanic eruptions) and have igneous type. Sometimes, hematite marginally
relatively dense structure. Natural action of transforms into its magnetic forms, martite
heat, air and water has caused oxidation and and maghemite. Since these are formed in
re-crystallisation of some magnetite deposits sedimentary deposits, they retain the open
Raw Materials for Sponge Iron Making // 77
Fig. 9.1 A view of the Gorumahisani iron ore mines (Shows high grade rocks
associated with lower grade and clayey overburden)
Courtesy: M/s Ghanshyam Mishra & Sons Pvt. Ltd.
structure and can be as reducible as hematite pores which form on the surface allow gases
in rotary kiln. to access the unreduced regions of the lump
or pellet and thereby makes the reduction
9.5 LABORATORY TEST ON (METHOD commercially viable. As reduction is effected
OF DETERMINING) IRON ORE mostly by the regenerated carbon monoxide
REDUCIBILITY in the rotary kiln, what is measured in
We have talked of more and less reducible laboratory for assessing its suitability is the
iron ore in general terms. But for the purpose reducibility toward carbon monoxide.
of quantitative assessment, whether a As hematite ores can be reduced upto 33%
particular source of iron oxide is more or less without effecting any metallisation,
reducible than the other, a more precise measuring reduction rates up to 33%
definition is required. reduction region would not serve any
Reducibility refers to the ease of removal purpose as the key step in the process is the
of oxygen, which is combined with iron in reduction of FeO to metallic iron.
the iron oxide feed. Generally, reducibility is measured as the
As the reaction process is gas solid rate of reduction under standardised
interaction process, reduction in the interior condition after reducing gases have removed
of lumps and pellets should have been very 40% of the oxygen – what would have
difficult, had it not been accompanied by a originally been with iron in Fe2O3 form (i.e.,
large volume reduction (52%). The resulting at atomic ratio of O/Fe = 0.9). Results are
expressed as “percent reduction per minute.” reduced material (50%) and generally higher
In this connection, one may refer to the latest levels of gangue associated with magnetite,
revisions of ISO/TC 102/SC3, ISO/DP 4695, and also, as mentioned earlier, due to
DIN 1770, JIS (M6495) and IS 8167 of BIS relatively denser structure of magnetite as
standards on Reducibility of iron ore, sinter most magnetite deposits are formed from the
and pellets. melts of volcanic eruptions.
9.6 COAL FOR SPONGE IRON

MAKING
Coal is a combustible mineral substance (rock)
generally black or blackish in colour. Coal is
the main source of meeting industrial energy
requirement. It is still far ahead of its cleaner
alternatives such as petroleum, natural gas,
electrical and atomic energy. Coal is called a
fossil fuel as it is formed by fossilisation of
wood and plant matter over centuries and
millennia.
Coal is formed from ancient plant or
woody matter initially by biological action
(peat) followed by physico-chemical
compaction over millions of years. Most of
the coal deposits were formed during, what
is called, the Carboniferous Period ranging
from about 360 to 300 million years ago. This
period was highly conducive to growth of
plants and trees of gigantic sizes and numbers.
But the conditions of rainfall, river movement
and shifting of earth’s crust resulted in
periodic destruction of the rain forests, their
Fig. 9.2 Apparatus for measuring reducibility submergence or carry over and accumulation
of iron ore
of woody matter into large lakes and their
Almost all high-grade hematite ores have subsequent submergence by plate shifting.
been found to have adequate reducibility for
sponge iron making. With higher gangue 9.7 PEAT
content, the reducibility generally decreases, By micro-biological degradation the lowest
probably due to hindrance caused by gangue rank of coal i.e., peat, was formed, in which
to the diffusion path of gases. Ores containing the woody structure is retained. It contains
large percentage of magnetite are found to more than 50% free moisture as mined. But
have much lower reducibility as compared on suitable storage and exposure to sun, free
to hematite ore, possibly due to marginally moisture can be brought down to below 20
smaller pore volume formation in the % when it can be used as a slow burning fuel.
9.8 CONVERSION OF PEAT INTO 9.10 SIZE OF COAL FOR SPONGE

HIGHER RANKS IRON MAKING
Whenever, natural forces caused sub- For sponge iron making, coal from feed end
mergence of peat deposits and it was covered is required to be roughly of 5 to 20 mm size
with an overburden it was subjected to and the coal from discharge end (thrown coal)
pressure and temperature (e.g., geothermal can be a judicious mixture of fine coal
heat). It slowly transformed progressively (0 to 5 mm) and lumpy coal (up to 20 mm).
into higher ranks viz., lignite or brown coal, Use of ground coal is presently not
high volatile bituminous, medium volatile considered practical in rotary kiln.
bituminous, low volatile bituminous, semi-
anthracite and anthracite coals. Progressively 9.11 ROLE OF COAL IN A SPONGE
the deposits gained in carbon percentage IRON ROTARY KILN
while it lost hydrogen, oxygen, etc. Carbon Coal is required to perform two vital
chains become longer and the coal compounds functions inside the sponge iron rotary kiln.
became like a resin (at medium to low volatile Firstly, it provides the requisite heat for
bituminous range) which further broke up bringing the reactants to the reduction
when the rank increased. temperature. Secondly, carbon in the char
The above may appear like too much of a generated from the de-volatilisation of coal
detail, but it is at this point (resin like acts as a reductant for converting the iron
formation) that the coal started becoming oxide into metallic iron. Of the iron oxides,
unsuitable or less suitable for sponge iron FeO or Wüstite is the most difficult to reduce.
making. If coal is able to reduce FeO into metallic iron
under rotary kiln conditions, then the coal is
9.9 INDIAN COAL DEPOSITS suitable for rotary kiln sponge iron making
Fortunately for sponge iron makers (and less process. We would see later that once the
fortunately for blast furnace iron makers), coal is found capable of reducing FeO to
most of the Indian Coal deposits are of lower metallic iron, it is automatically able to
rank type. On the other hand, rather provide the necessary heat for the process.
unfortunately, most of these lower rank In sponge iron rotary kiln, FeO is reduced
deposits have mineral matters well dispersed as per the following steps:
within their seams. Therefore, they yield FeO + CO Fe + CO2
large amounts of ash on burning and their
value is correspondingly reduced. Mineral Free energy considerations indicate that
matter being well dispersed, the coals are not for this reaction to proceed in the forward
easily amenable to beneficiation. Mineral direction under rotary kiln conditions, the
matters are liberated only at high levels of concentration (percentage) of carbon
grinding. Grinding makes the coal less monoxide will have to be higher than 2.5 times
suitable or even unsuitable for most the concentration (percentage) of carbon
applications and need to be agglomerated dioxide. Thus any carbon dioxide either
before use. The process of beneficiation of formed by this reaction, or any other source,
coal thus becomes uneconomical for most has to be removed quickly from the site of
applications. reaction. Carbon in char can convert CO2 to
NiCr-Ni double thermocouple,

controller, limit temperature
cut-out indicator
Measurement
NiCr-Ni thermocouple (a)
Plug-in coupling
Flexible tube Scavenging
for gas supply
Quartz Reactor (B)
Support
NiCr-Ni thermocouple (b)
Plug-in Coupling
40-90 mm
Tube furnace (A)
Plug-in coupling II
for flexible tube Shut-off valve
III
IV Conical ground joint
Precision
manometer
Three-way cock, X
Flow meter
Metering
Calibration
valve Three-way cock, Y
Flow meter
Gas
dryer
Filter
Gas analyser (C) Two colour printer for

% CO & °C (D)
Shut off valve

To exhaust
I Plug-in coupling
for flexible tube
Flexible
connecting Nozzle for flexible
tube tube of 6mm
Pressure
reducing
valve
CO2 Fig. 9.3 Gas flow diagram of the apparatus for the
(From CO2 cylinder) determination of reactivity of char
CO thus providing additional CO for relative reactivity in terms of percentage of

reduction. carbon reacted under specified conditions.
CO2 + C = 2 CO Both apparatus and procedures are complex.
To further complicate the problems, the
The more rapidly carbon in char is able
results differ with different char preparation
to convert CO2 to CO, the more suitable the
methods. The reactivity of char is adversely
coal is for rotary kiln sponge iron making.
affected, if temperatures used in its
We say that the char from coal has adequate
preparation are higher, if it is exposed to
“reactivity”.
maximum temperature for longer period of
We thus define reactivity as the ability of time, and the results are vitiated if a small
carbon in char to quickly react with CO2 to amount of VM is retained in the char. The
form CO. More strictly we are defining latter often happens if attempt is made to
“Carboxy-Reactivity” which is relevant for make larger quantity of char in one batch. In
rotary kiln sponge iron making. Other forms future revisions BIS will have to address these
of reactivities are “Oxy-Reactivity” – ability aspects in the standard.
of carbon in char to quickly combine with
oxygen to form CO – and “Hydroxy- 9.13 GAS ANALYSIS METHOD OF
Reactivity” – ability of carbon in char to quickly REACTIVITY MEASUREMENT
combine with steam to form CO and H2.
We would like to present here briefly the gas
By combining the reduction and the analysis method of measuring the reactivity
gasification reactions, we get the overall of coal char. IS: 12381 of BIS details two such
reduction reaction in rotary kiln sponge iron methods out of which the method involving
making. continuous recording of percent CO in the
FeO + C = Fe + CO exit gas using an infra red CO analyser has
There is thus a surplus of carbon monoxide been selected. Other than the gas flow
gas which keeps coming out of the bed in the scheme, the apparatus is quite simple. The gas
reduction zone. Partial combustions of this flow diagram along with the apparatus is
CO supplements at least part of the heat represented in Fig. 9.3.
required in the process. Thus we see that once In this method, both instantaneous
coal is found suitable for the role of reduction, reactivity and average reactivity over a
it can automatically fulfil the role of heat period can be measured. In all the alternative
provider at least partially. methods, only average reactivity over a
period can be measured. We feel that the
9.12 MEASUREMENT OF REACTIVITY results obtained are more reliable as it can
OF COAL CHAR easily eliminate the error due to any residual
For the methods of measurement of the volatile matter in the sample. Other methods
(carboxy-) reactivity of coal char, the latest can reduce the error on this account by
revision of IS: 12381 may be referred. The extending measurement period but cannot
apparatus and the procedures mentioned here completely eliminate it.
either measure reactivity in terms of cubic This method is also free from errors
centimetre of CO2 converted to CO per g. arising out of dropping or blowing off of ash
carbon in the bed per second, or measure a particles.
9.13.1 Principle After plotting the graph, the area under

For intimate contact carbon dioxide should the curve is counted to locate the instant (i.e.,
be passed through a column of char. Since the time) when the integrated area equals
carbon monoxide is much lighter than carbon 0.5 g. of carbon removed.
dioxide, channelling in the column is expected The reactivity may be defined as the
if gases pass up through the column. When instantaneous rate of evolution of carbon
gases are passed down the column all the monoxide per unit carbon present. The
problems mentioned above are overcome, but carbon remaining at that instant (when 0.5 g
lower layers of char encounter gas which is C is removed) is 0.5 g. less than that present
getting progressively enriched by carbon in the beginning of the experiment and is
monoxide. To account for this phenomenon therefore equal to (W*Cfix – 0.5) g.
the concept of Damköhler Number has been It can be shown that
incorporated. Damköhler Number represents
Reactivity
the ratio of residence time of the gas to the
mean life of a reacting species in the gas. = (V*(t + 273)/(t0 + 273)/3.6)*Da/
Use of this approach enables us to (W*Cfix– 0.5)
measure reactivity values instantaneously and Where Da is the Damköhler Number,
the need to average out over a period is which is given by
obviated. Da = 2*loge((200 – Vol.%CO)/
It is customary to measure reactivity at a (200 – 2*Vol.%CO)) – Vol.%CO/
point when 0.5g. of carbon has been gasified (200 – Vol.%CO)
from a sample of 5g. To determine the instant In the present case
(i.e., the time) when gasification of 0.5g. of Reactivity = (V*1000/36000)*(1273/
carbon is completed, a graph is drawn at the 298)*Da/(5*Cfix – 0.5)
end of the experiment, between instantaneous
= 10.68*Da/(5*Cfix – 0.5)
gasification rate of carbon (g.C/h) and time.
The instantaneous gasification rate of carbon Room temperature has been assumed to
is obtained from the %CO in the exit gas and be 25°C. For matching results, when
the gas flow rate. A sample plot is given in conducted at a different room temperature,
Fig. 9.4. The instantaneous gasification rate it is recommended that gas flow may be
can be shown to be related to the percent adjusted using Gas Law to obtain the same
carbon monoxide in the exit gas by the expression for Reactivity.
following relationship.
9.13.2 Test Conditions
Instantaneous gasification rate
A sample of 5 g. of char is taken having a size
= V*(273/(t0 + 273))*(12/22.4)/ range of 0.5 to 1 mm. The following test
(200/Vol.%CO – l) g. carbon/hr. conditions are maintained:
= Vol.%CO/(45.27 – 0.228* Vol.%CO) a) Temperature : 1000 ± 5°C
g.C/hr. b) Test duration : 30 minutes approx.
where V = Volume flow rate of CO2 in c) Gas composition : Pure and
dry CO2 gas
litres/hour,
t0 = room temperature in degrees Celsius d) Gas flow rate : 9 litres/hr.
9.13.3 Test Apparatus 5 g. of char sample having a grain size of

The set up for reactivity determination is 0.5 to 1.0 mm. is filled in the reaction tube
represented in Fig. 9.3. This essentially and the upper glass tube with the thermo-
consists of an electrically heated tubular couple placed in position. The tip of the
furnace (A), a quartz reactor tube of thermocouple should be well within the
approximately 15 mm internal diameter (B), sample and care should be taken that it is not
an infra red detector for continuous in contact with the side of the reaction tube.
measurement of carbon monoxide in the exit Gentle tapping should stabilise the filling.
gas (C), and a recorder for recording The depth of the bed should be 40 to 90 mm.
temperature and %CO in the exit gas (D). The depending on the bulk density of the char.
furnace is controlled preferably by a PID The reaction tube is then mounted on the
controller with an accuracy of ± 5°C. The support outside the furnace and is connected
furnace tube should be 200 mm long and to carbon dioxide cylinder as indicated in
about 25 mm in diameter.
Fig. 9.3. Gas is allowed to pass for about 15
9.13.4 Test Procedure minutes (Scavenging) for flushing out all
The reaction tube is constricted in the lower traces of air. Simultaneously, furnace is
end as shown in Fig. 9.3. A plug of quartz brought to the set temperature of 1000 ± 5°C.
wool is placed above the constriction and is Once scavenging is complete, the reaction
carefully pressed in taking care not to block tube is removed from the conical ground
gas passage. glass joint III,, inserted into the furnace and
For measurement of sample temperature, is connected to a similar ground glass joint
thermocouple tip is placed in the uniform IV as indicated in Fig. 9.3. The thermocouple
temperature zone of the sample. (a) is connected to the printer for continuous
Thermocouple should be adjusted using a sample temperature record.
locking ring.
Fig. 9.4 Determination of Damköhler number

Fig. 9.5 Variation of reactivity with fuel ratio

Printer is started and carbon dioxide gas for each measurement point. This is a measure
is passed at the rate of 9 litres per hour (at of grammes of carbon being gasified per hour
25°C) with both three-way cocks ‘X’ and ‘Y’ (g.C/hr), and this is also plotted against time
having been set to ‘measurement’. After as shown in Fig. 9.4. By measuring area under
measurements are complete, gas flow is this plot, that particular time can be located
stopped, reaction tube is taken out of the at which point the gasification of 0.5 g. of
furnace and placed back on the support. carbon from the sample is complete. The
The % CO in the exit gas is plotted against corresponding per cent CO in the gas is
time in minutes (Fig. 9.4). From the values of located and this indicates the Damköhler
% CO, initial flow rate of CO2 (9 litres per Number.
minute at 25oC in this case), standard litres Reactivity is reported as follows:
of CO evolving per minute can be calculated Reactivity = 10.68*Da/(5*Cfix – 0.5) c.c./g. – sec.
Canada
1100
Bed temperature in rotary kiln reduction zone
Brazil
Ne w Zealand
Anthracite
Germany
U.S.A
1000 Canada
Bituminous coal
900
Lignites
0.4 0.5 0.8 1.0 2.0 3.0 4.0 5.0 8.0 10 15 20

Reactivity, cc/g.s
Fig. 9.6 Variation of rotary kiln reduction zone bed temperature with coal char reactivity
9.14 INDICATORS TO REACTIVITY

Since the results of reactivity tests are of • Lower rank coals have higher reactivity.
doubtful reliability, sponge iron makers rarely All high volatile bituminous coals are
go for these tests unless and until they suitable for sponge iron making
encounter some problems. Although M/s • Even lower rank coals give more reactive
Lurgi, Germany, have made elaborate char (e.g., Lignite). But these are associated
investigations establishing coal char reactivity with two other problems:
to be the most important property for
• Very high moisture
selection of coal for sponge iron making,
actual reactivity tests are rarely used in India • Tendency to become powdery and get
air borne on heating.
for this purpose. Some of the important
findings of Lurgi are being reproduced here After partial drying these can be used as
in the form of graphs (Figs. 9.5 and 9.6). Instead thrown coal in sponge iron making.
operators prefer to use indicators in locating • A fuel ratio value of lower than 1.5 (i.e.,
reactive coals keeping the following in mind. FC/VM < 1.5) ensures adequate reactivity
May not be suitable for reasons Suitable for Unsuitable for

of handlability, need for pre- rotary kiln rotary kiln
treatment, etc. sponge iron sponge iron
making making
Char reactivity
(Qualitative)
Reactivity
0
Caking property
(Qualitative)
Caking
property
60 VM 6
Mean fuel ratio (FC/VM)

50
5
40
4
VM (daf)
30
3
20
2
FC/VM
10
1
0
0
Semi- anthracite
Meta- anthracite
Medium volatile
High volatile
Low v olatile
Anthracite
Sub-
Peat Lignite bituminous
Bituminous Anthracite
Rank
Fig. 9.7 Chart showing guidelines for coal selection

• Coals with fuel ratios up to 1.8 generally by 2%. But this figure is much less than the
have acceptable reactivity damage ash causes in other metallurgical
• Caking and swelling coals yield low reactors. Where ash needs to heated to much
reactive chars and are not suitable for higher temperatures, melted, flux is provided
sponge iron making. for slagging of ash, flux needs to be heated
For the same reason, any type of coke is and calcined for which additional coal/coke
not suitable for sponge iron making. (with addition ash) has to be added. In
• Pre-reacted, re-circulated and higher contrast, in rotary kiln, ash only occupies kiln
density chars have poor reactivities, since volume and needs to be heated to the
these chars have been exposed to high processing temperature. Ash, thus, does least
temperature for a long time. harm in sponge iron rotary kiln as compared
Figure 9.7 gives a qualitative diagram for to other metallurgical reactors.
selection of coal based on reactivity. A low fusing ash manifested by a low
Initial Deformation Temperature, tends to
9.15 OTHER CRITERIA FOR COAL form agglomerates when used in the thrown
SELECTION coal. It also increases tendency of accretion
Ash amount and initial deformation formation, although this aspect has not been
temperature of ash under mildly reducing conclusively proved. Using such coals as feed
condition are the other criteria to be coal has been found not to give any problem
considered for coal selection. An increase in in operation, as the char from feed coal rarely
ash content by 1% reduces kiln productivity sees temperatures in excess of 1000oC.
This page
intentionally left
blank
10
Accretion or Ring Formation
Inside a Rotary Kiln CHAPTER
Accretion or Ring Formation is a perennial a shut down these ‘Rings’ or blockages have
problem, which all the rotary kiln sponge iron to be removed using pneumatic concrete
makers have to face and live with. When a breaking device, or any other suitable
freshly lined kiln goes on production, the means.
surface colour of refractory slowly darkens
as it absorbs iron oxide. After some duration 10.1 CAN RING FORMATION BE
of operation, and especially if the process goes AVOIDED?
temporarily out of control, deposits are seen While the problem cannot be eliminated,
building up at various places on refractory there are means of control of accretion and
surface. Deposits may be localised or, more early detection and management of any large
commonly, may be forming around the deposits, which might be forming on the
circumference (like a ring, hence called ring refractory surface.
formation). Sometimes the deposit formation It is assumed that a lower ash fusion
is also accompanied with formation of temperature of coal means a higher
agglomerates or partially fused association propensity to accretion formation, but this
of solid particles. Agglomerates come out thumb rule has not been properly
into the product discharge circuit but, substantiated yet. Ash fusion temperature
sometimes get attached to the deposits on (the Initial Deformation Temperature or IDT)
the refractory surface and help it to grow in is almost always above 1150ºC. The
size rapidly. refractory surface of a properly operating
Ring formation makes the inner diameter kiln would never reach this temperature.
of kiln progressively smaller at certain Those ash compositions with IDT close to
locations. While this renders uniform 1150ºC are high in FeO, while those with IDT
movement of charge extremely difficult, in 1250ºC and above are low in FeO. In rotary
extreme case of ring formation the kiln kiln condition where FeO is in plenty the
diameter may get fully blocked forcing a above difference in FeO content would not
stoppage of the production campaign. During matter.
Fig. 10.1 Mechanism of initiation and growth of accretion
10.2 DEPOSIT FORMATION THROUGH

SOLID STATE SINTERING rather complicated, let us ignore for a moment
A particle of ash and of iron ore caught the minor constituent i.e., CaO (about 5%) in
together in refractory crevices would slowly the quaternary. In the ternary phase diagram
sinter over a length of time to reach a low FeO-SiO2-Al2O3, the lowest melting consti-
melting composition (Fig. 10.1). But even the tuent melts at 1083ºC (Fig. 10.2 & 10.3). We
lowest melting composition has softening can make a simplifying assumption that
point of 1070ºC and above (in the quaternary accretion mainly comprises of FeO + SiO2 +
phase diagram FeO-SiO2-Al2O3-CaO). Since Al2O3 and the presence of flux may depress
discussing a quaternary phase diagram is the fusion point by 13 to 15 oC.
Accretion or Ring Formation Inside a Rotary Kiln // 91
Thus, ash and ore particles can sinter and

are likely to make a composition with a fusion
point of 1083oC, which can be lowered by 13
to 15 oC in presence of flux containing CaO
and MgO, but even this composition would
not normally melt in an efficiently controlled
kiln. While deposits formed by solid state
sintering are inevitable, it would not cause
any hindrance to the production process as
long as it remains solid. Only when it melts,
it starts gathering other solids around it
(mostly semi reduced sponge iron and ash)
and re-solidifies into a larger mass. This
generally happens when for some reason,
the kiln temperature, which is generally
maintained near 1000ºC (actual
temperature used depends on the
characteristics of iron ore and coal)
goes out of control. In majority of
cases this happens because the
thermocouple tips, which
measure continuously the
kiln temperature, get
covered with deposits.
Thus when the operator
tried to control the kiln
temperature to about
1000ºC, the tempera-
ture of kiln is actually
significantly higher
than 1000ºC.
Fig. 10.2 FeO-SiO2-Al2O3 phase diagram
10.3 REASONS FOR CATASTROPHIC

GROWTH OF DEPOSITS therefore acquires a higher temperature than
Since freeboard gas, and the refractory rest of the refractory. Agglomerates, which
around, are always at higher temperatures are already partially fused, tend to get glued
than the solid material in any cross section of to the top of the deposit and form a bigger
the kiln, a protrusion on a refractory surface protrusion. It thus becomes a vicious circle.
receives higher heat radiation than rest of the Once a protrusion is formed it becomes
refractory surface. The protrusion or deposit difficult to arrest its rapid growth.
(Ore)
Fig. 10.3 Liquidus line in a typical ash-ore system
10.4 PRECAUTIONS REQUIRED reflect the gas temperature at all. As the

To prevent such happenings it is a general thermocouple tip ‘sees’ the refractory for
practice to measure the kiln temperature using more than 70% of time and the solid bed for
quick response thermocouple (QRT which are the rest of the time, whatever heat radiation
made of thin mineral insulated thermo- it receives from gas is being continuously
couples). These QRT’s are inserted tempo- radiated to the refractory and solid bed.
rarily through ports on the kiln specially Consequently, the recorded temperature is
made for them or normally meant for much less than the gas temperature. Since it
collecting samples. A daily comparison of the ‘sees’ refractory surface for the longest time,
records of QRT’s and the corresponding the recorded temperature is closer to the
permanent process control thermocouples temperature of refractory surface. Since the
helps a long way in circumventing the refractory surface temperature controls the
problem of Accretion Formation. fact that any molten particle falling on it either
Kiln operators can use the QRT records immediately solidifies and falls off or
in another way. When this thermocouple is otherwise fuses to the surface and sticks and
inserted inside the running kiln, during thereby starts formation of ring, control of
rotation the thermocouple is for some time this temperature is extremely important. The
within the solid charge, while at other times QRT-gas temperature at any point can,
it is sensing the free board gas. Consequently therefore, be taken to be a measure of “the
operators record the minimum temperature accretion proneness” at that point.
as the QRT-bed temperature and the highest
temperature as the QRT-gas temperature. 10.5 ACCRETION REMOVAL DURING
While the QRT-bed temperature is a OPERATION
relatively accurate estimate of the average bed This is something which has been attempted
temperature, QRT-gas temperature does not since long but has met with very limited and
Accretion or Ring Formation Inside a Rotary Kiln // 93
uncertain success. Some accretions fall off on What has been adopted with somewhat
their own, particularly which are fragile. But, limited success is quenching the site of
in sponge iron making we hardly come across accretion with water during the operation
such accretions. The accretions are usually itself. Although the results are not always
quite hard and some of it are so adherent to positive, but the good thing is that no adverse
refractory that on removal during shut down effect of water quenching has been reported,
it takes away a layer of the refractory. which is rather surprising. Those who use this
When accretion is in an accessible region, technique regularly say that there is no
such as near the kiln mouth at discharge end, discernible adverse effect on the product
a long concrete breaker can be inserted to quality. To date, this appears to be the only
break it off. But very rarely accretions are so practicable method of removing accretion
conveniently located. Gunning it down with during operation even though the outcome
regular bullets or heavier projectiles have been is erratic. Some apparently have thrown
considered but has never been attempted to common salt over the deposit site but have
our knowledge as their use is highly not reported positive result. Theoretically
dangerous both to the personnel and speaking, this measure would rather enhance
equipment. Same consideration applies to the growth of deposit instead of controlling
throwing of explosives on the accretion site. it.
This page
intentionally left
blank
Sponge Iron Properties:
11
Re-oxidation and Auto-
ignition of Sponge Iron CHAPTER
11.1 REASONS FOR RE-OXIDATION iron burns without any flame. In a burning
TENDENCY sponge iron heap one may see a short light
As mentioned earlier, the main rotary kiln blue flame at isolated locations, which is
product, the Sponge Iron or DRI is highly caused by burning of carbon monoxide,
porous; the porosity could go as high as 52%. which is generated within the heap. We had
To this Figure of 52% we should add the once witnessed the spectacle of burning of
porosity of the parent iron oxide. The shredded scrap bundles kept along with
porosity exposes a lot of surface area to the sponge iron, which had caught fire.
atmospheric air for oxidation and the sponge Coal based DRI is much more stable and
iron has to be cooled out of contact from any no such accident has come to our knowledge
air, oxygen, moisture or even carbon dioxide. so far. However, it does not mean that coal
Once at room temperature the propensity to based DRI cannot catch fire. We had seen it
oxidise, also called pyrophoresity, is less acute catching fire when sparks from gas cutting
but it still demands careful storage and torch were falling directly on it for a long
transportation. Under normal atmospheric time.
conditions, sponge iron re-oxidises and loses
Modern commercial sponge iron are
metallisation rather too slowly to be of major
prepared under conditions which make it
concern in a production facility, as long as
wet conditions are avoided. Once exposed even more prone to re-oxidation, so much so
to rain, or even moist air, the loss of that stocks and piles are now known to catch
metallisation is rapid. fire under apparently normal conditions of
The rotary kiln DRI has got a bad name storage, usage and transport.
mostly because of its gas-based counterpart, The phenomenon of auto-ignition of
which is highly pyrophoric and has caused sponge iron is relatively newly discovered,
many a fire with catastrophic effect. even though the sponge iron has been known
Shiploads have been burnt with massive through ages and its susceptibility to
damage to ships a number of times. Sponge oxidation is also well known since long.
11.2 HISTORY OF AUTO-IGNITION processes, using reformed natural gas as

Sponge Iron was first extracted by man about reductant. Units based on HyL, Midrex,
4,500 years ago. This sponge iron was made Armco and Purofer Processes came into
in semi-fused condition and was immediately commercial production, which made large
beaten to shape after re-heating. This quantities of gas based sponge iron available
immediate conversion practice was a ritual. for shipment. From this point the problem of
This ritual was probably started because of auto-ignition raised its head.
the tendency of oxidation of sponge iron on
exposure to atmosphere, which led to poor 11.3 FIRST REPORTED INCIDENCE
results, if the conversion to final tool was not A shipload of sponge iron was sent from
immediate. Thus the practice of guarding it Germany to Portugal by the sea route. The
against re-oxidation was started even when ship by name ‘Cape Crest’ arrived at Lisbon
the propensity to re-oxidation was not port in 1977 and the cargo was found to be
consciously realised. on fire. It is not clear how much of the damage
With the advent of blast furnace, the was caused to the ship and whether any part
practice of sponge iron making was relegated of the cargo could be saved. In May 1979 a
to insignificance. The primitive practice shipload of sponge iron from Europe, after
however, continued in many remote places crossing Atlantic Ocean arrived at Houston,
including remote places of India. Texas in USA and the cargo once again was
During the World Wars, severe constraint found to be on fire. In December of same year
was imposed on iron and steel resources, as a sponge iron shipment from Canada was
the consumption in wars was very high. found to have caught fire on arrival at
Germany started supplementing its iron Barcelona, Spain.
needs during the wars using Krupp-Renn There could have been other incidence of
process which made sponge iron in a rotary fire, which were not reported, but after these
kiln using coal. In this process the sponge iron
disasters stringent shipping norms were
was partially fused in the final stages of
introduced. Ships for sponge iron were
reduction to make what they called ‘Luppen.’
required to use McGregor Hatch (an airtight
Since Luppen was fused and consolidated in
hatch). Apart from this, at the time of loading
the last stages it was not prone to re-
of sponge iron in a ship hold, the hold was
oxidation. But the process was plagued with
heavy deposit formation and fell into disuse required to be flushed with carbon dioxide
when strain on iron resources eased after the gas. As it turned out this precaution did not
World Wars. The Krupp-Renn process was give much reprieve.
to latter re-emerge as the Krupp-Codir In June 1981 two shiploads of sponge iron
process of sponge iron making. Till this time (probably from Indonesia) arrived at
there was no recognition of the possibility of Vishakhapatnam port. During inspection both
auto-ignition of sponge iron. the cargoes were found to be OK. But during
In 1950s Germany, Mexico and United waiting period and by the time unloading
States independently attempted to develop could start, fire was detected in both the
commercially viable sponge iron making consignments.
Sponge Iron Properties: Re-oxidation and Auto-ignition of Sponge Iron // 97
To overcome the fire problem Midrex But in 1985 an identical incidence took
started hot briquetting the sponge iron into place at BASL itself and this time the fire was
pillow shape. The resulting HBI was much a massive one and lasted over 3 days. Late in
more consolidated and stable for shipping the second day we had reached site of fire.
purpose. HyL also followed suite. Midrex and A part of the pile, which had taken a conical
HyL tried to produce as much HBI as possible shape, because of angle of repose
but it was more costly than normal sponge requirement, was looking normal with just a
iron. HBI was considered very safe and no small bluish flame at the top of the cone. On
fire hazard was envisaged. But there was a approaching the pile we could feel the heat
newspaper report that an HBI cargo on the radiation from some distance itself. On
way from Venezuela to Spain caught fire. This removing a part of the apparently normal
happened in March 1982. Midrex denied that outer surface, a white-hot interior could be
it was HBI and claimed that it was a cargo of observed. The appearance from a distance
briquetted sponge iron from Fior process. It was definitely misleading. The sponge iron
may be noted here that Venezuela operates pile on one side was overlapping with a pile
all the 3 types of units: Midrex, HyL and Fior of shredded scrap. The bales of shredded
(now Finmet). Midrex stand was probably scrap in contact were also white hot, and the
correct, as, at that time, the Venezuelan plant fire was spreading into the scrap pile as well.
was briquetting only Fior sponge iron.
With the help of a front-end loader we
11.4 EYE WITNESS REPORT brought a small quantity of burning sponge
iron into a relatively open flat space and we
We had only once the privilege of seeing a
tried short experiments on this small burning
pile of sponge iron on fire. But there have been
pile. First we tried blowing nitrogen through
a number of incidences when our colleagues
the pile with the help of a perforated steel
in the organisation had seen the incidences of
pipe. Then we tried carbon-di-oxide,
fire.
followed by argon gas. All these only
In 1981 a shipment of sponge iron received
resulted in enhancing the fire, since the
from Indonesia was lying in stock in Alloy
convection current associated with the gas
Steels Plant of Steel Authority of India Limited
flow helped to suck in surrounding air into
at Durgapur, in the eastern Indian state of
the burning pile.
West Bengal. When our colleagues had the
opportunity to see the pile, the fire in the pile Since it was not possible to cover the large
had been extinguished apparently by a very burning pile and keep it out of contact with
copious spray of water. The pile had turned air, such an experiment on the small burning
black and was emitting evaporational steam. pile was not tried. We tried spraying water.
In the same year the sponge iron obtained Soft spray did not have much effect. Evolving
apparently from the same source was lying in steam brought in air from surroundings just
stock in the Bihar Alloy Steels Limited (BASL), like the experiments with blowing in of inert
an electric arc furnace based steel plant at gases. But heavier sprays were effective. It
Patratu in the eastern Indian state of Bihar was therefore decided that with the help of
(now Jharkhand). The pile had caught fire and front-end loaders and grab buckets, the
was extinguished by heavy water spray before unaffected sponge iron would be separated
it could spread over a large area. out from the burning portion and thereafter
the burning part would be flooded with during extinguishing of fire apparently because
water. of the high heat evolution, which helps in
This was done in the night and we could removing surplus water by vaporisation.
see that fire was extinguished successfully next
day. Thereafter, we took a few samples and 11.5 INVESTIGATIONS RELATED TO
physically and chemically studied the burnt AUTO-IGNITION
pieces. To our surprise even in the sample After the bitter incidences of auto-ignition
taken from regions, which were red hot, the described above, detailed investigations were
particles showed only superficial oxidation. taken up almost parallelly by M/s Lurgi
Substantial portion of the interior was Germany, HyL Mexico and Midrex USA.
unaffected and was in metallised form. This Details of the internal studies are not readily
indicates that, if detected somewhat early, available but Mr. V.J. Clancey, UK
much of the value in burning sponge iron pile subsequently studied, summarised and
can be saved. Contrary the popular belief that interpreted the findings to which we will
water converts sponge iron back to Fe2O3, it return a little later.
does not happen to any significant extent Subsequently, National Science Foundation
To pH meter 100
To H2 Analyser
To oxygen Thermometer aluminium tubing
analyser Detachable top
Rubber
stopper
15mm 50
Opening
1000
5φ 4 Holes
20
10
Perspex vessel
3-way cock, Skirt plate

one on each
side (All around)
25
Air/O2
Air/O2
100
Base 100
plate Water
(MS) outlet
200
800 φ Gate 15 to 20 mm
valve
Fig. 11.1 Sketch of the experimental setup (Perspex Model) at RDCIS, SAIL
of USA sponsored a project to study the The temperature rise during the different
problem of Auto-ignition. Joint collaborators experiments was also measured. While
to this project were University of Pittsburgh, temperature increase was observed and there
USA and the Research and Development was some trend of higher temperature rise
Centre for Iron and Steel (RDCIS), Steel with higher salinity, but the extent of
Authority of India Limited (SAIL), Ranchi, temperature rise was marginal as well as
India. inconsistent. These investigations did not lead
In the investigations of RDCIS SAIL, to any significant insight towards the reasons
which is the corporate R&D of SAIL, and is for auto-ignition, but were otherwise useful
located at Ranchi, a Perspex model was made in providing aqueous corrosion rate data of
which simulated the hull of a ship. Coal based coal based sponge iron and in confirming the
sponge iron samples were kept within this mechanism of aqueous corrosion of sponge
physical model and were made to react with iron.
water of different salinity in different
RDCIS, SAIL further carried out EDAX
prolonged experiments. Each experiment
investigation of the corrosion product
lasted for at least 3 to 4 days. The gases
obtained in the above experiments. The
evolved could be collected within this model
results supplied some missing links to Prof.
and these gas samples were analysed. The
gases were found to be very high in Neil Birks at the University of Pittsburgh USA
hydrogen level and higher salinity of the and helped him in formulating a plausible
water used resulted in higher volume of hypothesis on auto-ignition of sponge iron.
liberation of hydrogen.
Fig. 11.2 Photograph of the Fig. 11.3 Photograph of an experimental

perspex model setup to study temperature rise
during aqueous corrosion of
sponge iron
11.6 OTHER SIGNIFICANT RESULTS (3000 to 9000 cm2/gm). This makes sponge
In an investigation carried out somewhere in iron capable of adsorbing on its surface a
Europe, a fairly large gas based sponge iron number of gases from atmosphere. Sponge
heap was kept in open atmosphere for nine iron has a stronger affinity for adsorbing
months including a full three months of rainy moisture from atmosphere even in relatively
season. After nine months it could be seen dry weather. It is not uncommon for sponge
that the surface had turned reddish brown iron under ambient condition to adsorb about
and surface samples showed almost zero one percent moisture. However, the amount
metallisation. The surface had also become adsorbed is highly dependent on humidity
virtually impervious obviously because of and temperature.
increase in volume associated with corrosion.
Higher temperatures tend to decrease the
On the other hand when samples were
quantity of adsorbed moisture. When the
taken about 10 cm below the surface, the
sponge iron is charged in melting bath almost
samples were black. The samples had lost very
little of its metallisation and there was no entire adsorbed moisture is released without
evidence of substantial temperature rise inside loss of metallisation. Since this aspect does
the pile. not cause any problem during melting, such
users tend to completely ignore this aspect.
In another similar set of experiments, fresh
gas based sponge iron was piled over an But things are different when sponge iron
already corroded pile. After sometime, is transported in the hull of a ship. In a typical
measurements inside the pile showed shipment there would be about 6000 tonnes
evidences of temperature increase. Although of sponge iron, which would carry with it
several variations of this were tried, and all about 60 tonnes of adsorbed moisture. Due
pointed to increase in temperature, but in to temperature fluctuation during
controlled experiments the increase in transportation as much as 30 tonnes of this
temperature was hardly of the magnitude, adsorbed moisture may sublime and
which could justify and explain auto-ignition. condense on the walls of ship’s hull. After
sometime this water may drip down and
11.7 CLANCEY’S HYPOTHESIS collect at the bottom. Thus a part of cargo
As mentioned earlier Mr. V.J. Clancey of U.K. may become wet even if there has not been
studied the actual occurrences of fire in any leakage in the ship’s hull. Since it is very
cargoes of sponge iron and also the related common for the walls of ship’s hull to have
reports made by Lurgi Germany, HyL Mexico salt deposits, lower layers of sponge iron
and Midrex USA. He derived some important would thus get wetted by saline water
conclusions and presented his hypothesis on leading to enhanced corrosion.
the reasons for enhanced oxidation of sponge Clancey thus pointed out reasons for
iron during shipment but could not pinpoint enhanced danger during shipment of sponge
the exact reason for triggering of auto- iron but could not explain any further.
ignition.
Clancey observed that sponge iron is 11.8 HYPOTHESIS OF NEIL BIRKS
highly surface active material because of its Neil Birks of the University of Pittsburgh
high porosity (over 50%) and high surface area USA accepted Clancey’s observation and
Top of cargo Top of pile
Primarily zone
Cool high
gases Vapour Ignition temperature zone
zone Zone oxidation
zone
Water Water
Reflux wetting vapour drains
condenses
Corrosion zone
Capillary wetting
reactions
evolve heat zone
zone
Ship's hull
Formation of Zones in a Sponge Iron Cargo Stage I - Rust Formation
Top of pile Top of pile

zone zone
zone zone
zone
zone
zone zone
zone zone
zone zone
Stage II – Reduction STAGE III – Oxidation and Ignition

Fig. 11.4 Zones in a sponge iron cargo and stages leading to auto-ignition (Birks)
realised that the situation in the hull of a ship Prof. Birks thus evolved a hypothesis that
becomes similar to the cases where fresh auto-ignition of sponge iron takes place in the
sponge iron is piled over wetted and corroded following manner:
sponge iron. He carried his reasoning from (i) When sponge iron gets wetted by
that point and also studied reports related to water it undergoes an electro-chemical
burnt cargoes. He realised that fine powdery corrosion and gets converted into
and highly pyrophoric sponge iron was found oxides and hydroxides. In the process
in most cases of burnt cargo. Since such a there is liberation of heat and also
sponge iron was never stacked to start with, there is liberation of hydrogen.
this must have been generated in-situ. Prof.
Birks carried out a few controlled experiments Fe + (x + 1) H2O = FeO.xH2O + H2
simulating the ship’s hull and could find ∆ H = –ve
evidence of in-situ formation of such fine (ii) Since sponge iron is highly porous it
powdery sponge iron. acts as a heat insulator in spite of its
high metallic content. Therefore point is that temperature at which under

corrosion generally results in local normal ambient conditions, the temperature
over-heating. rise in sponge iron become uncontrollable.
(iii) In a ship’s hull the lower layers of A quick method of determining this
sponge iron are likely to get wetted temperature is to determine the “Crossing
by water. This wetting may be Point Temperature.” This determination can
through leakage or other reasons as be very easily made in the laboratory by
explained by Clancey. The H2 gas thus keeping a sample of about 500 gm of say 3
liberated by wetting and corrosion to 10 mm sponge iron in a half litre Pyrex/
may get entrapped in different Corning glass beaker. A thermocouple is
portions of sponge iron pile. embedded in the middle portion of the
(iv) Since corrosion also raises the sample in the beaker. The beaker containing
temperature this may help in reduction the sample is placed in an air oven or a muffle
of iron oxide available in the vicinity furnace, which has already been preheated
by the entrapped H 2 . Since the and stabilised at about 150 o C. The
reduction takes place at a relatively temperature of the furnace or oven is slowly
lower temperature the surface area of
and steadily raised at a rate of 2 to 5 oC/
the sponge iron produced would be
minute. Both the temperatures, those of the
much higher than normal sponge iron
furnace and of the sample, are monitored
and is therefore highly pyrophoric.
and plotted against time. To start with, the
(v) Over a period of time such highly
furnace temperature would be keeping
pyrophoric sponge iron may be
ahead of the sample temperature. But after
generated at a number of places.
reaching a certain temperature level the
Once there is a leakage of air into
sample temperature would be crossing and
such an area, oxidation may start
leading to elevation of temperature. going beyond the furnace temperature. The
It may happen that in a certain temperature at this crossing point of the two
volume, oxygen leakage into the curves is recorded as a crossing point
central portion takes place well after temperature of sponge iron and can be
the surrounding areas are oxidised viewed as the ignition temperature of
and thereby the temperature rises. sponge iron.
This temperature may rise above the The crossing point temperature of
‘ignition point’ of the freshly commercial sponge iron normally lies
generated sponge iron at the centre between 200 to 350 oC. Coal based sponge
and under these conditions, when air iron, made in rotary kiln, has a crossing point
and oxygen leaks into this area, the temperature close to the higher limit. The
sponge iron automatically ignites. crossing point temperature of gas based
sponge iron lie in the middle to lower regions
11.9 IGNITION TEMPERATURE OF of this range. The sponge iron primarily
SPONGE IRON: CROSSING reduced by H2, as in the HyL process, has
POINT the lowest crossing point temperature among
It may be noted that what we call ignition the commercial sponge irons.
Fig. 11.5 Set up for determination of crossing point temperature
H2 has got a very high diffusivity and One may argue that there have been
small molecular volume. At relatively lower reports of auto-ignition in the stockyard of a
temperatures the reaction kinetics may few units using sponge iron where there was
become unfavourable for the reduction of certainly no wetting of the pile with water.
FeO to Fe by CO; but reduction by H2 would But such reports of ‘auto-ignition’ may be
still be possible. Sponge iron produced by H2 viewed with suspicion. In a sponge iron using
at lower temperatures in say, controlled unit there are lots of source of ignition
conditions of a laboratory would have supporting objects (flying sparks, even
smaller pores in much larger numbers; and scrapped artillery shells). Therefore these
would therefore have much higher surface reports are unlikely to be true incidences of
area than that of the sponge irons which we auto-ignition. However, these incidences do
are used to. The crossing point of such a underline the fact that sponge iron is highly
sponge iron may be even below 200 o C. pyrophoric, hazardous and needs to be
Formation of such a sponge iron somewhere handled with care.
inside a large pile would naturally make it There has so far been no report of auto-
very vulnerable to auto-ignition. ignition of sponge iron made by coal based
route in rotary kiln. A study of their crossing
11.10 HOW VALID IS THIS point indicates that they have a higher
HYPOTHESIS? ignition temperature. Since temperatures
Confidence in the hypothesis of Prof. Birks used in rotary kiln are somewhat higher than
would have increased if it were validated in that used in sponge iron shaft, they (coal
control experiments. Unfortunately based sponge iron) have gone through the
simulation condition of auto-ignition is step of surface sintering, which reduces
extremely difficult and certainly not possible number of pores at the surface of the sponge
to reproduce in laboratory. It requires very iron. But question arises whether coal based
large stack of sponge iron where a number sponge iron can be prone to auto-ignition
of pockets of accumulation of H2 could form. even though the crossing point temperature
is higher? If hypothesis of Prof. Birks were carbide only enhances the fire, some people
valid, even this sponge iron would be think that water spray enhances sponge iron
amenable to auto-ignition; only it should take fire. While that is not correct, it should be
somewhat longer time. Whether or not this kept in mind that light water spray, as also
happens can be confirmed only after large blowing in nitrogen or argon helps in
shipments of coal based sponge iron become enhancing convection currents which results
commonplace. in more air being made available to sponge
iron and thereby the existing fire is enhanced.
11.11 PRECAUTIONS NEEDED IN
SPONGE IRON STORAGE 11.12 CONCLUSIONS
In light of the above, we can identify a few 1. Auto-ignition of sponge iron is the
essential precautionary steps related to outcome of electrochemical corrosion
handling and storage of sponge iron. Since of sponge iron. Moisture required for
the large surface area makes it prone to corrosion need not come from external
atmospheric re-oxidation, which is enhanced sources and may remain loosely
in presence of moisture, sponge iron should bonded to the surface of sponge iron.
be handled and stored in such a way that it is However, moistening of the surface
kept away from water or moisture as far as alone does not lead to auto-ignition.
possible. Therefore it should be kept in 2. High porosity of sponge iron makes
covered bay with hard floor having such an it highly surface active as well as a
elevation so that water does not collect at the heat insulator.
bottom. 3. Corrosion within the pile of sponge
But when large tonnages are handled, it iron leads also to loss of permeability
is often not possible to find covered yard for owing to volume expansion. Hydrogen
all the sponge iron stocks. If forced to store gas liberated as a result of corrosion
in open, the pile should be kept undisturbed may get entrapped in some locations.
until use. Whenever reclaiming is started the This may reduce iron oxide available
pile should be quickly consumed. in the vicinity and thus may result in
Some ‘Don’ts’ should be closely followed. the formation of fresh and finely
One should never store fresh sponge iron over divided metallic iron of low ignition
a corroded pile. Such a process would temperature. On gaining access to air
contribute heavily towards proneness to auto- auto ignition may start.
ignition. As and when fire actually starts in a 4. In control experiments auto ignition
sponge iron pile the only sure way to has not been observed in ambient
extinguish it is to use copious water spray or conditions (only beyond 220oC).
flooding with water. 5. A plausible hypothesis of auto ignition
Apprehension exists about the use of has been evolved, but needs to be
water. Like spraying water on calcium validated.
12
Uses of Sponge Iron CHAPTER
The production of sponge iron and use of regard is not so distant future.
sponge iron in steel making are both growing The major reason for the wide acceptance
in India by leaps and bounds. The yearly of sponge iron is its closely sized granular
production of sponge iron has already nature. The close size range makes it
crossed 16 million tonnes. The coal based amenable for continuous charging into
rotary kiln and the gas based routes are reaction vessels. Continuous charging means
contributing about 60:40 in this production. that processing and charging can go parallel
The demand for any item is directly and there is no separate charging time.
dependent on its usefulness. And the market Further, during continuous charging in
demand or success in the market is solely electric arc furnace (EAF), heat losses due to
dependent on the faith of the consumers on removal of roof for charging through bucket
its potential use. This was something lacking is minimized and sometimes completely
when tonnage sponge iron first appeared in eliminated. The sponge iron particles fall
the Indian markets in the 1970’s (through through refining slag and in the process melts
imports) and the 80’s (through indigenous and gets refined, further saving on refining
production). time.
At that time, sponge iron was projected Other favourable aspects of sponge iron
as a substitute for iron and steel scrap. As are its low sulphur content (generally around
metallic content of sponge iron was much 0.02%) and absence of tramp elements (Cu,
lower than scrap, sponge iron played second Sn, As, Sb, Pb, Cr, etc.). Significant levels of
fiddle to scrap for a long time. But with the dephosphorisation have been reported while
experience gained in its usage, many charging sponge iron, possibly due to the
favourable properties of sponge iron were presence of oxides like FeO in the sponge iron.
discovered, and sponge iron then turned out Further, FeO in sponge iron induces “boil”
to be, at least in India, the second major in metallic melt necessary for effective heat
metallic feed for steelmaking. It may possibly transfer and for aiding in the refining process.
overtake the blast furnace hot metal in this Separate boil inducing constituents are not
Table 12.1 Comparison of steel making in arc and induction furnaces
Electric
Electric arc
Item induction
Criteria furnace (EAF) Remarks
no. furnace (IF)
steel making
steel making
1. Refining Possible Not Possible Removal of P, S & C is possible only in
Operations EAF
2. Addition of Possible Possible Due to stirring by eddy currens,
Minor homogenisation of additives is more
Constituents easily possible in IF
3. Specific Lower (increasese Higher (about 750 For mini steel plants, the difference is
Power at lower kwh/tonne fr 70% small
Consumption capacities from DRI charge)
about 450 to 700
kwh/tonne)
4. Electrical High (higher peak Very Low (lower High electrical noise in EAF operation
Fluctuations demand) peak demand) also leads to high acoustical noise.
IF operation is relatively very quiet
5. Tap to Tap Larger Smaller Ultra High power (UHP) EAF's can
Time have tap to tap time close to IF's
6. Heating Localised heating All round heating Slag gets heated only by contact with
Configuration near the arc in the meallic metal in IF
portion
7. Heat Size Higher (25 to Lower (up to 30
80/100 tonnes) tonnes)
8. Melting Rate Lower High UHP EAF's have melting rates
comparable to IF's
9. Specific Lower About 50% higher
Refractory
Cost
10. Specific Higher Lower (power EAF starts operation with short circuit
Capital (complicated) supply unit is condition. Current fluctuations are high;
Investment power supply unit) simpler) also heavy impedance is required in the
circuit
11. Skill Level Higher Lower
Required
12. Versatility Higher Lower Possible to make much larger number
of grades in EAF's
required. However, some additional the major portion of gas based sponge iron is
carburiser is needed. consumed in-house by the producers either
in their UHP-EAF or CONARC furnaces.
12.1 USE OF SPONGE IRON IN M/s Sunflag at Bhandara, Maharashtra are
ELECTRIC STEEL MAKING also consuming their sponge iron in-house
Due to problems associated with gas based although they are a coal based sponge iron
sponge iron (auto-ignition tendency) and producing unit. Many other coal based
relatively high cost of hot briquetting the DRI, sponge iron producing units have now
Uses of Sponge Iron // 107
integrated steel making units in their is required all the time to allow sponge iron
operation. But most of them use induction charge to meet the liquid metal without hind-
furnaces for melting sponge iron. rance. Sponge iron would ignite and burn
The small merchant producers of sponge with atmospheric oxygen if it is held even
iron are mainly feeding induction furnace steel momentarily above the surface of the melt.
producers. Normally these steel producers Sponge iron requires higher energy for
are using 70% sponge iron in the charge with melting as compared to scrap. But by
10% cast iron and 20% steel scrap. The liquid increasing the rate of heat input along with
steel produced which is about 70% of the total continuous charging, this drawback has been
metallic charge, is cast either in pencil ingots largely overcome. Being granular, it starts
or are continuously cast into billets. melting from all sides (i.e., all faces of a
In the 1970’s and 80’s most of the scrap in sponge iron particle starts melting
India were melted in electric arc furnaces simultaneously) and “melts like butter” and
(EAF’s) and in a smaller number of induction this is not possible with heavy melting scrap.
furnaces (IF’s). While use of Sponge iron in Faster melting ensures lesser radiation loss
EAF’s were attempted immediately, IF and therefore, the higher energy requirement
is largely compensated.
operators were wary of its usage since its
induced a relatively violent reaction and
12.2 USE OF SPONGE IRON IN
necessitated changes in operation practice as
well as design of operating platform. SUBMERGED ARC FURNACE
But, with the experience gained, charging In a Submerged Arc Furnace (SAF), the
of sponge iron in induction furnace has really electrical arc struck between the two
electrodes remains covered with the solid
picked up. Even though electric power
charge material. Consequently this furnace is
consumption is higher per tonne steel, at small
more energy efficient as compared to EAF.
capacity levels the difference is hardly
But since there is no access available for
significant, while electrode consumption is
operators to reach and work the melt, this
totally absent. Since there is no arcing there
unit is not conducive to quality steel making.
are no sudden power fluctuations (i.e., there
Instead the melt can easily be allowed to get
is absence of electrical ‘noise’).
saturated with carbon and then we get hot
In Table 12.1 a comparison has been metal.
presented of the steelmaking in EAF and IF This route of making hot metal cannot
using sponge iron. No wonder, mini steel presently compete with the blast furnace iron
producers using sponge iron prefer adopting making process. But there is a possibility of
induction furnace route and it would remain making quite low sulphur and low
their preference as long as the market phosphorous hot metal which would be
demand is consistent with the inputs. suitable for conversion to ductile iron or
Needless to say, there is very limited room spheroidal graphite cast iron. In 1990-91 the
to diversify with such an installation. National Metallurgical Laboratory (NML)
In Table 12.1 low skill level requirement carried out extensive trials in its pilot 500 KVA
is mentioned for IF operation. But skilled submerged arc furnace, with the assistance
removal of the slag (or rather the oxide scum) of SAIL. The trials proved the versatile nature
of SAF in that very inferior DRI could also The effect of charging sponge iron in small
be melted without any problem. During the blast furnaces have been studied extensively
trials, hot metal of quality suitable for in the past by U.S. Bureau of Mines, U.S. Steel
conversion to ductile iron was always Corporation, Steel Company of Canada and
obtained. Ahmsa (Altos Hornos de Mexico S.A.). The
Although this concept is being used in results are summarised in Fig. 12.1. This plot
New Zealand and probably also in Myanmar indicates that, for 10% metallisation of
and South Africa for making hot metal, but burden, both productivity and coke rate
these are examples of commercialization with improve by about 8%. But at higher rates of
a special type of raw material or for specific charging the benefits change. Decrease in coke
site conditions. We cannot say that, at the rate drops to 5.5%. Productivity improvement
moment, the practice of melting of sponge to 9.5% is indicated, but this increase may
iron in SAF to make hot metal has been not be statistically significant. The recent
commercially adopted. experiences of Tata Steel and Steel Authority
of India Limited have been on similar lines.
12.3 USE OF SPONGE IRON IN IRON Because of decreasing benefits, charging of
MAKING sponge iron by more than 10% may not be
Since sponge iron composition is closer to contemplated anytime in future. But even up
steel than pig iron, its use as raw material to to 10% the technical benefits are immense. It
making pig iron in blast furnace appears at is estimated that sponge iron charging would
first, a wasteful exercise. It is indeed not be economically viable if the price of sponge
favoured at present since the input cost of iron is up to 40% of the price of coke. There
sponge iron does not make it economically was a period in the year 2005 when Durgapur
viable for pig iron making. Steel Plant experienced coke shortage and was
But at times of hot metal shortage, the purchasing additional coke at about Rs.
overall plant economics often justifies use of 22,000/- per tonne. Sponge iron was available
sponge iron in enhancing production from at Rs. 9000/- per tonne and the coke shortage
blast furnaces as well as save on expensive was mitigated by the use of a small percentage
coke usage. Some technical disadvantages of (about 5%) of sponge iron. The situation has
sponge iron in blast furnaces have been now eased, but relative to sponge iron the
pointed out, but these can all be overcome average price of coke has to go up only.
by obtaining from the producers sponge iron The Middletown plant of M/s A.K. Steel
of slightly higher size (say + 8 mm) while in the United States has been operating their
compromising slightly on the metallic Fe blast furnaces consistently at very high
content (say 77% as against the usual 80%). productivities since 1996. Productivity of 4.37
The other disadvantages of higher CO in the t/m3-day was reported in the year 2004 but
top gas and higher gas temperature would the measures taken were not explained. They
hardly be observable at the charging levels were known to be charging HBI (hot
being contemplated (i.e., 10% metallisation briquetted iron, i.e. compacted sponge iron)
of burden, approximating to about 9% of iron and high rates of the same appear to be the
oxide feed). major reason behind increased productivity.
80 80
×
70 70
60 60
Productivity increase, %
Coke rate reduction, %

×
50 50
40 40
×
×
30 × 30
×
20 × 20
× Productivity Increase, %
× Coke rate reduction, %
10 10
0 0
0 20 40 60 80 100
Burden metallisation, %
Fig. 12.1 Effect of burden metallisation on productivity and coke rate in blast furnace
12.4 USE OF SPONGE IRON IN 12.5 USE OF SPONGE IRON IN

CUPOLA OXYGEN STEEL MAKING
During 1980’s the Kulti unit of India Iron and The maximum amount of steel is presently
Steel Co. Ltd., were operating cupolas, in one produced through hot metal and basic oxygen
of which sponge iron was charged to the furnace (BOF) route. BOF steelmaking
extent of 20% of metallic charge. No problem requires coolant during oxygen lancing so that
was observed during sponge iron usage. Even the finish temperature does not go higher
though increased resistance to blast and than desired. The best coolant for BOF is the
higher coke rate were anticipated, these were revert steel scrap which simultaneously
found negligible. Flux requirement increased increases the yield by 18 to 30%. Sponge iron
and it was felt that the sponge iron probably can very well be used to replace steel scrap.
travels faster down the shaft as it is smaller With wide adoption of continuous casting,
in size. Extra coke charging before start of quantity of revert scrap is going on
charging sponge iron helped in smooth decreasing, and presently we see sponge iron
operations. as the only alternative. Cooling capacity of
If the above precautions are taken there sponge iron being marginally higher, it is
should not be any problem in use of sponge required in lesser quantities so that the final
iron as cupola feed. Only lack of economic yield of steel is 2 to 3% lower. Sponge iron
viability is preventing use of sponge iron in carries with it a small quantity of FeO which
cupolas. may increase the tendency of slopping of
Fig. 12.2 Schematic of a typical mini/midi steel plant of India
BOF, but this aspect has not been conclusively 12.6 SPONGE IRON AND HOT METAL:
established in the trials conducted in India. PARTNERS IN PROGRESS
Sponge iron is poised to become a standard Most of the entrepreneurs contemplating
coolant in BOF operation. newer steel making units have realised the
importance of marrying together the use of during use in electrical furnaces, the heat
blast furnace hot metal and sponge iron in liberated in the oxidation of carbon helps in
steel making stage. This is done either in an saving a part of the costly electrical energy.
oxygen converter where sponge iron is It is little wonder that most of the sponge
relatively a minor input or in an electric arc iron producers are contemplating expansion
or induction unit where hot metal would be
with a mini blast furnace and vice versa (i.e.
a minor input among the two.
mini blast furnace based units plan to go for
By using them together, the disadvan-
sponge iron production). It is expected that
tages of higher than desirable carbon in hot
metal and undesirable iron oxide in sponge mini-and midi-steel plants of India may
iron are compensated by each other. Further, follow schemes depicted in Fig. 12.2.
This page
intentionally left
blank
Process Design, Engineering
13
and Operational Aspects of
an RK-DR Plant CHAPTER
When an entrepreneur thinks of utilising the Process Design (also referred to as Basic
natural resources of iron ore and non-coking Design) and Process Engineering.
coal deposits to make sponge iron to meet
the market demand of the region, which we 13.1 PROCESS DESIGN AND
are presuming he has already assessed, then ENGINEERING
he looks for a process supplier who can give Process Design refers to the clear under-
a suitable technology. Technology suppliers standing and determination of all process
in sponge iron area have now proliferated. control variables and their optimisation and
Even the technologies, which started with actual design of the equipment and facilities
widely different concepts, have now necessary for carrying out the process.
converged to a nearly common process flow Process design aspect has been mentioned
typically indicated in Fig. 2.2. Fig. 13.1 in chapter 6 (section 6.2) where the volume
represents a typical raw material preparation, requirement of a particular capacity kiln has
handling and proportioning unit while the been discussed, while discussing the impor-
product handling system is represented in tance of determination of residence time in a
Fig. 13.2. rotary kiln.
The technology supplier develops his When an entrepreneur gives his
know-how based on past information, requirement to the technology or process
experience and scientific principles and supplier, he gives the following information:
interpolations. Once his technology has
1. The size of plant required;
proved successful he has only to update his
old document to cater to the next customer. 2. Sources of Raw Materials, Utilities and
But each customer has his unique need. The Power
process of converting the customer’s need 3. The plant site.
into a project document is referred to as The process supplier arranges to test the
Process Design and Engineering – raw materials for determining the suitability
incorporating two overlapping steps of for the process, the productivity expected and
Fig. 13.1 Schematic of raw material handling and proportioning unit
process consumptions based on the properties 13.1.1 Raw Material Testing

of the raw materials. For interpreting test For a DR plant, normally a Rotary Tube
results and other input information to project (Salvis) furnace is used as a bench-scale testing
the data on productivity, consumption device. This test can predict the desirable
pattern, operating parameters, etc. the process process parameters, viz., residence time,
supplier needs to have nomograms or temperature, degree of fill, etc., needed for
software or model of the process. He also obtaining good metallisation.
needs to detail out processing techniques and
While ore reducibility and coal char reactivity
prepare an operation manual. It is also
required that he draws the overall schematic can be measured individually (please refer
of the process, prepares specifications of main chapter 9) and based on the results obtained
technological units and of critical equipment a ruling can be given that a particular
and components and prepares basic combination would be suitable for sponge
engineering drawings thereof. In short, he is iron making, but confidence in test result
now doing the process engineering, and this would certainly be better if tests are carried
also encompasses the steps of project out in conditions closely simulating the actual
definition, preparation of feasibility report conditions in rotary kiln. One such test is the
and project engineering and implementation Rotary Tube test, very commonly referred
which follow. Table 13.1 lists the steps to as Salvis test (IS:11284-1985).
involved in process design and engineering. In fact, this test is relatively simpler than
Process Design, Engineering and Operational Aspects of an RK-DR Plant // 115
both the Ore Reducibility and Coal Char process but it further increases the propensity
Reactivity tests. So usually, the ore-coal to accretion or ring formation.
combination is first tested in Rotary Tube test Figure 13.3 gives a diagram of the
(after chemical and proximate analysis). If the equipment for Rotary Tube Test. It consists
combination does not pass the test then only essentially of an electrically heated tubular
ore and coal are individually characterised reactor of 150 mm internal diameter and 650
to decide which one is the real, or more of, mm reaction zone length. The tubular reactor
the culprit. Apart from the compatibility test, is rotated at a predetermined rpm while
the Rotary Tube also gives indication of the nitrogen gas is passed continuously
agglomeration tendency of coal (caking throughout the experiment. After the tube is
propensity) as also the decrepitation tendency heated to a predetermined temperature
of iron ore (tendency to generate fines). Both (generally 400oC, 700oC or 1000oC), iron ore
these are very important indicators for rotary (5 to 20 mm) and coal (1 to 15 mm) in
kiln operation. An agglomerating coal is predetermined ratio is mixed and charged
bound to give a char with low reactivity, into the hot zone of the tube. The temperature
while decrepitated ore is not only loss to the of tube is then raised to the maximum
Surge bin
+20 mm
3-20 mm
Screen
Magnetic
separator
–3 mm
Rotary
cooler Magnetic Magnetic
separator separator Belt weighers
Belt weighers Belt weighers

Cooler discharge
Conveyor Sponge Char Sponge Char Sponge Char

iron iron iron
–3 –3 3-20 3-20 +20 +20
Product day bins mm mm mm mm mm mm
Despatch/Disposal/Re-use/Storage
Fig. 13.2 Schematic of a product handling system

temperature (1000 to 1020oC) and maintained or operating conditions can be exhibited

there for one and a half to three hours. more quickly and economically than that
Thereafter, the mixture is cooled within the would be possible on a full sized prototype.
Rotary Tube furnace under continued flow
of nitrogen to room temperature. The 13.1.2 Upscaling
product is magnetically separated and both Let us look at the Process Design problem
fractions are size analysed and thereafter from the process or technology developer
chemically analysed. The percentage of +15 point of view that ultimately has to supply
mm of char is a measure of agglomeration the Process Technology either directly to the
property of coal while –3 and –1mm of entrepreneur or through an intermediate
magnetics give an indication of the technology supplier. Normally his design of
decrepitating tendency of iron ore or its the process is based on experience of
friable nature. Friable iron ores are not operation of similar sized plants and units.
suitable for sponge iron making in rotary kiln But sometimes, the requirement of the
without pelletisation. entrepreneur is of a different size plant, in
The overall magnetisation (Fe(M)/Fe(T)) which case the process developer has to
of the magnetic fraction should be above 90% resort to upscaling (or downscaling as the case
for the combination to be commercially may be). Help is then taken from the
viable. mathematical model of the process (Chp. 5),
A successive phase, i.e., Pilot Plant testing, which has been prepared based on data
may then be taken up to generate data for generated from earlier experiments and
the ultimate full scale plant and also produce operating experience of Pilot Plants and other
small quantities of sponge iron for research units. The model is validated and modified,
and trial purposes. As the flexibilities in a Pilot if necessary, using data of some operating
Plant are more, the effects of changing design plants. Thereafter the mathematical model of
Fig. 13.3 Laboratory rotary tube furnace

process is proportionately up-scaled. is a chance that we may end up with

Corresponding process flow diagrams are unaccounted accumulation and unexpected
then prepared and then the individual shortages which may require additional
equipment are designed. investment to correct for in future.
We have talked about proportionate up-
13.1.5 Preparation of List of Major
scaling. This ‘proportionate up-scaling’ may
Equipment
not always be linear and the designer has to
At this point the entire process design has
draw upon his experience, and the experience
been conceived although it may not be frozen.
of others in the field, to decide upon the
The designer now prepares the “System
method of this ‘proportionate up-scaling.’
Specification of Major Units” in the plant
Sometimes a subjective decision would have
which details the equipment requirement of
to be taken.
the unit and their interrelationship.
13.1.3 Sizing of Key Equipment Simultaneously the “Sizing Specification of
Based on the data generated and the size of Major Units” is also prepared which meets
plant required, the reactor volume can be the needs of the project. This exercise also
worked out based on the example in chapter 6 helps in estimating the requirement of utilities
(section 6.2). Based on experience with similar and power.
sized plant and also on scientific reasoning The next step is listing out the major
and scientific interpolation, the reactor sizing equipment and item and indicating their major
can be done. For example, a typical 50 tpd design parameters. Some typical examples of
(15,000 tpa) rotary kiln sponge iron plant has these major equipment can be found in section
a reactor size of 2.1 m diameter and 30 m 13.4.
length; the corresponding cooler size is 1.5
m Φ and 15 m length. 13.1.6 Preparation of Layout
From prior experience the process developer
13.1.4 Process and Material Flow lays out the equipment in his mind at various
Diagrams levels and then makes on paper a conceptual
Since the successful process flow sheets block diagram of the layout at different levels
roughly correspond to the concepts presented of importance. For some of the key regions
in Fig. 2.2, 13.1 and 13.2, to prepare a fairly he makes the elevation block diagram also.
detailed process flow diagram is not a difficult If he does not find any inherent conflict he
task. Thereafter a material flow diagram freezes his concept. However, it is easier said
needs to drawn separately or may simply be than done. Conflicts keep on surfacing again
superimposed. The material flow diagram and again and that makes the task of making
emanates from the Material Balance as also a layout a long drawn out process. Roughly
from the Energy Balance (may refer to Fig. at this point, i.e., on finalisation of layout,
2.5 & 2.6) the process design activity ends.
The importance of Material and energy
balancing has been stated in section 7.1. In 13.2 PROJECT ENGINEERING
addition to that if one does not make prior Once the process and technology concepts
material and energy balance calculation there have been frozen and an agreement or an
Table 13.1 Steps in process design and engineering
Sl.
Steps Details Remarks
No.
1. Evolution of conceptual Inputs required: Conceptual design is generally represented

design based on input ● Labouratory investigation in the form of a schematic diagram
information ● Modelling supported by a short technical write-up
● Bench and pilot scale validation
● Data available in literature on
comparable processes
● Fundamental aspects
2. Evolution of basic design, ● Preparation of plant and Layout is prepared either

involving: Equipment layout ● Mentally, or
i) Specification of inputs ● Evolving specificaions of raw ● To-the-scale skethces, considering
and outputs materials, poducts and other levels of various equipment and
ii) Flow schemes inputs and outputs facilities
iii) Plant layout ● Preparing process / material ● Area requirement is worked out
flow diagram ● Layout is checked for possible
interferences, maintenance and
safety aspects
3. Basic engineering, ● Volume of work is to be ● Most of the inputs to project feasibility
involving estimated under the heads report would be estimated or evolved
i) Preparation of list of a) Equipment/plant & Mechinery at this stage.
equipment b) Refractory ● Availability of infrastructure support is a
ii) Estimation of tonnage/ c) Concreting/Civil work key factor in selection of site. Site
volume of work ● Capital cost is to be estimated accessibility by rail and road,
iii) Estimation of capital using relevant data, e.g. availability of water, other utilities and
cost a) Land requirement services, electrical power, etc. would help
iv) Estimate production b) Soil type in reducing the cost of development of
cost at optimum level c) Land development needs infrastrucutre
v) Prepare activity d) Extra facilities required ● Activity schedule should be presented
schedule & Bar chart e) Cost of civil work, structures, in the form of a bar chart
vi) Estimate cash equipment, installation, ● Cash flow in an industry is like the flow
requirement commissioning, etc. of blood in a human body and meticulous
vii) Prepare estimated ● Prepare cash flow statement for estimate is needed to preclude chances
profit and loss (P&L) a) Construction period of cash shortage
statement for each of b) First ten years of operation ● IRR and NPV are considered the most
the years of ● Estimate separately for each of important financial parameters for the
construction and the 10 years of operation evaluation of a project
operation a) Percentage capacity utilisation
viii) Estimate internal rate b) Maintenance expenses
of return (IRR) and the c) Addition, modification and
net present value construction (AMC) expenses
(NPV) (if required)
4. Preparation of project Detailed in the next Table Project execution and implementation follow,
feasibility report ending with commissioning and guarantee
tests
understanding has been reached with the 13.3 INSTALLATION, COMMISSIONING

entrepreneur, a Project Feasibility Report is AND PROJECT COST
prepared. The entrepreneur may prepare it The gestation period of a DR plant and cost
himself, may ask the process or technology of the overall project depends upon the
supplier to do so or may hire an independent infrastructural facilities already available,
Design Consultant. The report may possibly proximity to rail/road/sea connections, cost-
be prepared in two stages. The preliminary cum-availability of equipment and facilities,
(Pre-feasibility) report may focus only on the technology type, expertise, know-how,
technical feasibility and the economic viability availability of raw materials, process
of the project, while the detailed report requirements, etc.
(Detailed Project Report or DPR) would During commissioning and testing
additionally have plant and project details period, trials are carried out both under load
with process design figures. These reports and no-load conditions. Standardised coal
often help the entrepreneur or promoter to (within a fixed range of sizes, reactivity, ash
raise capital from other investors and lending softening temperature and composition) and
institutions. Table 13.2 gives an outline of iron ore (having specified decrepitability,
items required in the preparation of a composition and reducibility) are chosen.
feasibility report. Close monitoring of all equipment and systems
Thereafter Technical Specification (TS) is need to be done during commissioning and
drawn for tendering purpose. A major post commissioning stabilisation.
portion of the tender specification is the
preparation of Detailed Engineering 13.4 PROCESS ENHANCING
Drawings of equipment, system and other EQUIPMENT AND CONCEPTS
facilities of the proposed plant. It has to be
With the present emphasis on improved
decided beforehand whether the work order
energy efficiency, higher productivity, better
for execution of the project would be given
product quality and better environment
in parts or as turnkey contract. This decision
compatibility, a number of optional facilities
would be based on the available expertise –
have been developed by various institutions
both in open market and in-house – and an
to augment the existing system for ‘high
assessment of their relative merit. Depending
performance DR plants.’ Some of the
on whether it is part-wise contract or turnkey,
innovations are gas sealing system,
the manufacturing drawings may be prepared preheating system, under-bed hydrocarbon
by the equipment supplier selected, or by the injection system, coarse coal throwing or
Design Consultant, or by the Technology slinging system, external coal devolati-
Supplier. Thereafter the steps to follow are lisation, etc.
civil and structural construction, equipment
construction, installation, cold testing and 13.4.1 Gas Sealing System
complete trial run with all interlocking and It is the system of preventing air-ingress into
sequencing. The process finally culminates in the rotating kiln, which has made the RK-DR
hot testing, commissioning, guarantee tests, process of sponge iron making technically
post commissioning stabilisation, etc. successful. Earlier attempts were concentrated
Table 13.2 Outline of Preparation of a Project Feasibility Report
Sl. No. Contents Details Remarks
1. Title and Introductory Names of Name should give an idea of contents

section ● Project
● Report preparing agency Reference should be given to the relevant
● Sponsoring organisation interaction with the sponsoring organisation
Statement of purpose
2. Executive summary A synoptic view of all the essential Balance to be maintained between brevity
findings of each chapter and content while maintaining readability
3. Project background ● Details about project sponsores; Briefly state

their strengths and weaknesses; ● Conclusions arrived at and decisions taken
relevant experience from these former studies;
● Project history ● Pre-investment studies, e.g.,
● Introduction to the process Opportunity studies, Land surveys, etc.
● Studies and investigations
already performed
4. Market and plant ● Study of demand and market of Describe:

capacity the product ● Factors taken into consideration for going
● Sales forecast and marketing into this project
● Production programme ● Existing size and capacities of the industry
● Determination of plant capacity ● Its part growth and the estimated future
growth
● Major problems and prospects
● Sales forecast and marketing
● Anticipated competition
● Quantitative relationship between sales,
plant capacity and material inputs
5. Material inputs Describe: Inputs to be listed separately under the

● Input requirements following heads:
● Their present and ● Raw materials:
potential supply positions ● Processed industrial materials;
● An estimate of annual ● Components;
costs of local and ● Auxiliary materials;
imported materials ● Factory supplies;
● Utilities, especially power
6. Process/Technology Analyse: Credentials of prospective technology

selection ● Alternative routes and processes suppliers, their background and experience,
available are also to be analysed
● Criticial study of alternatives
● Process selection with key
criteria for selection
7. Site selection Pros and cons of alternative sites Land requirement and its cost also to be
and reasons for selecting a stated
particular site to be stated
(Contd...)
Sl. No. Contents Details Remarks
8. Project engineering ● Statement of scope

● Plant Layout, Design and
Description
● Equipment selection
● Raw material sourcing
● Utilities and Auxiliaries
● Instrumentation and control
equipment
● Civil engineering works
● Construction Schedule
9. Plant organization Estimate: Overheads based on:

and overhead costs ● Plant operation schedule ● Factory overheads
● Organisation structure ● Administrative overheads;
● Manning and Training ● Financial overheads.
● Overhead costs
10. Manpower ● Estimate requirements based Managerial, executive, supervisory, skilled

on organisation structure and unskilled manpower should be
● Estimate manpower cost based separately identified
on salaries
11. Implementation Should include: ● Bar Chart is necessary

scheduling ● Implementation time schedule ● PERT Chart to be included for larger
● Implementation costs projects
12. Financial and Based on: Should include:

economic evaluation; ● Investment costs ● Pay-off period
comprising: ● Working capital requirement ● Simple rate of return
● Investment costs ● Capital structure ● Break-even point
● Project financing ● Fixed and variable costs ● Internal rate of return
● Production cost ● Financial evaluation ● Sensitivity analysis
● Economic
evaluation
on the draught control and maintaining slight slides over the stationary hood. The sliding
positive pressure in the reactor to achieve this surface with the hood is sealed by means of
end. Subsequently, mechanical seals were asbestos pad pressed with mechanical springs
tried which followed pumping of temperature while the sliding slip ring plate is pressed by
resistant grease (High Temperature Grease) means of pneumatic cylinders against the slip
into the surface of contact. The present widely ring plate fixed to the kiln shell. High
accepted system of sealing is schematically temperature grease is pumped continuously
represented in Fig. 13.4 and comprises two in between the two plates of slip rings for
mechanical slip ring plates in contact, one the dual purpose of lubricating the contact
welded to the rotating shell of the kiln while surface and sealing any small crevices that
the other, with suitable mounting device, would otherwise cause gas leakage.
13.4.2 Underbed Hydrocarbon Injection injected alternately air and hydrocarbon

System—The Ported Kiln Concept depending on whether it was passing through
The system of gas sealing described above the freeboard (gaseous zone) during rotation
was developed by M/s Lurgi, Germany as of the kiln or was under the solid charge bed
part of the SL/RN process, (developed jointly (Fig. 13.5). This was the overbed air and
by Stelco (Steel Co. of Canada), Lurgi, underbed oil injection system and was made
Republic Steel and National Lead. In the 70s possible by the rotary valve - striker arm
M/s Allis Chalmers announced the concept. In larger kiln the valve-striker system
development, first of ACAR Process (Allis has to be supported by a pneumatic triggering
Chalmers Agglomerated Reduction Process), device and solenoid valve system.
and subsequently of ACCAR Process (Allis The overbed air injection is a necessary
Chalmers Controlled Atmosphere Reduction part of the process (Chapter 2) so that
Process), which claimed many additional controlled oxidation of the CO in freeboard
advantages. They used a ported kiln concept helps in maintaining a near constant
with a large number of tuyeres in the temperature profile over the entire reduction
reduction zone of the kiln. The same tuyeres zone of the kiln. In the Lurgi concept this
Inlet for high

temperature
grease
Fig. 13.4 Schematic representation of a slip ring gas sealing device

requirement was met by provision of air- ACCAR design and has supplied it as OSIL
tubes, which are temperature resistant technology for overbed air injection in newer
permanent metallic tubes, which release the units.
air from the shell mounted air fans along the In the Sponge Iron Pilot Plant Complex
axis of the kiln, so that the air released does of RDCIS, SAIL a project was undertaken to
not oxidise the reduced iron in the charge parallely develop an underbed oil injection
bed. The Lurgi concept does not require the system. A cold prototype was successfully
complicated rotary valve-striker arm, developed which was converted to a hot
pneumatic trigger and solenoid valve system. prototype and thereafter was transplanted
On the other hand, the air tubes use up a part with minor modifications on to the Pilot Plant.
of the inner volume of kiln and makes the Instead of transplanting the system on the
process ever so slightly more accretion prone. kiln it was installed in the beginning of the
Experiences of operating kilns have shown rotary cooler, where the temperature of
that the perennial problem of accretion or ring sponge iron was still very high. Up to the
formation generally starts from the base of point of oil injection, the cooler was refractory
such air tubes. And in converting existing lined to retain the temperature in the solid
cement or calcination kilns to sponge iron bed.
kilns, diameter is generally a constraint, and The outcome of the development, even
by utilising the ported kiln concept and valve– though was technically satisfactory, the
striker arm device rather than air tubes, one benefit obtained could not justify the
can save on the inner volume available to the expenditure on the quantity of oil injected.
process.
Even though the underbed oil injection
But Allis Chalmers originally planned the system had been a new concept evolved in
system to inject hydrocarbons while the the 70s, the tuyere design existed earlier and
tuyere passed under the solid charge bed was used for underbed steam injection for
during rotation of the kiln. They claimed, selectively reducing Fe2O3 to Fe3O4 in lean iron
probably rightly, a higher productivity, a ores, so that the product could be beneficiated
lower temperature of reduction (i.e., lesser by magnetic separation (the magnetising
propensity to accretion formation) and a roasting method of beneficiating lean iron
better product having higher carbon content ores).
and a generally higher level of metallisation.
The plant of M/s Orissa Sponge Iron 13.4.3 Coarse Coal Throwing or Slinging
Limited (OSIL) set up at Palaspanga in System
Keonjhar District in eastern India, was the At the charging end of the rotary kiln,
first and only commercial plant set up on the granular iron ore (generally 5 to 25 mm size)
ACCAR process concept. The commissioning is charged along with coarse coal and flux.
trials were started in 1982 but the plant could Fine coal (0 to 3 mm) is generally not charged
be fully commissioned only in 1985. By that mainly because the ultra-fines therein tend
time the increased oil prices made the oil to get carried away along with the waste gas.
injection process unviable. Apart from the energy lost this way, the fine
Even though OSIL has abandoned oil coal suspended in the hot gas presents a safety
injection system, they have modified the hazard as well.
Towards the latter half of the kiln, where metres or more before being ‘slung’ into the
FeO is being reduced to metallic iron, the char kiln freeboard. This way they have been
in the bed does not remain as reactive as it successful in throwing from the discharge end
was when freshly charred. But it is in this 60 to 70% of the total requirement of coal.
part of the kiln that more reactive char is The more we use coal from the discharge end
needed since the final stages of reduction are the more energy efficient the process would
the most critical ones. be as the volatile matter in the coal would be
To supply this reactive char, almost all utilised within the kiln itself. From the feed
variants of the process throw (or sling) a small coal, the volatile matter joins the waste gases
(or even larger) percentage of coal as fine coal and are not utilised within the kiln.
from discharge end. This throwing or slinging In our zeal to throw more coal from
is done with the help of air as medium. Even discharge end we may be sending in more
though the quantity of air used is only a small and more of air (as carrier) as beyond a limit,
fraction of what is needed for complete the air may disturb the delicate balance that
combustion of this coal, the transportation or is maintained between oxidising and reducing
throwing is achieved in ‘Dilute Phase’ (as layers in the rotary kiln (Fig. 2.9). Too much
against Dense Phase Transportation of air with the coal may lead to oxidation of the
Pulverised Coal and other finely ground already reduced sponge iron resulting not
material in many applications). only in loss of metallisation, but also rise in
While the principle of throwing appears temperature and chance of accretion
quite simple, there are a number of formation. From this one can conclude easily
precautions to be taken while undertaking that in a larger diameter kiln one should be
throwing of coal. Further, there are a number able to throw a higher quantity and per-
of limitations with coal throwing as well. centage of coal without problem.
These have been detailed in chapter 6 and 13.4.4 Waste Heat Recovery System—
we summarise below some of the important Generation of Electrical Power
points.
In Fig. 2.6 and Chapter 7, it has been
Fine coal particles cannot gather enough emphasised that a large part of the energy,
momentum to reach sufficient length of kiln put in as coal, goes out with waste gas – both
by themselves; it is the current of freeboard as sensible heat and chemical energy. The
gas moving towards the charging end, which plant of M/s ISCOR at Vanderbijl Park in
“shoulders” the fine particles to reach South Africa was the first plant to recover
appreciable distances. this energy to generate electrical power.
The coarser particles can gather sufficient In India M/s Jindal Steel and Power
momentum to reach larger distances. But for Limited (JSPL) located at Raigarh in the
this the particles should be given sufficient central Indian state of Chhattisgarh has also
time and distance to attain full momentum installed a large thermal power plant (150
before being released into the kiln MW) which is based on energy from waste
environment. OSIL plant utilises an optically gases from ten modules of rotary kiln. M/s
aligned long pipe, fully straight, where the 6 HEG Limited in the same state, located nearby
to 25 mm size coal is carried by air for full 35 at Durg, also generates thermal power using
heat from waste gases from just two kilns, The Tata Steel’s development, the Tisco
each of capacity 30,000 tpa. The quantity of Direct Reduction (TDR) process originally
electrical power generated by both these envisaged in-situ pelletisation of iron ore and
plants are well in excess of their internal induration of green pellets by the burnt waste
requirement, and the surplus power is being gases. A prototype was installed but there
sold to the state electricity grid. The viability were difficulties with initial trial runs. After
of this concept has been possible because of it was realised that pellets are not going to
remunerative repurchase arrangement with be viable for the rotary kiln process, the
the state electricity board. This has also concept was abandoned.
helped the state in overcoming its electrical The waste gases are highly dust laden,
power shortage. mostly with fine ash from coal as also with
13.4.5 Use of Waste Gas Energy for fine iron ore and fine coal particles, generated
Preheating during tumbling within the kiln. A shaft
reactor containing granular iron ore could
Even though power generation from the
have been used to recover the heat from
energy in waste gas is one feasible option for
waste gases, but for the suspended dust,
utilising waste energy, it is always preferable
which would choke the bed voids, thus
to plough as much of this energy back into
preventing further passage of gas. Partial
the process as possible. Some plants do use
cleaning of gas beforehand would require
the fully burnt off gas for drying of raw
cooling of the gas and would therefore defeat
materials with significant benefits, but this
process can use only a fraction of the energy the purpose of the system.
available. And since drying anyway takes Heating of iron ore in another rotary kiln
place in the beginning of the kiln, the drying is an attractive possibility. Since preheating
device as mentioned above, amounts only to of coal would lead to loss of its volatiles, coal
some extension of the kiln volume. should not be preheated beyond the drying
range. At RDCIS SAIL a heat transfer model
Fig. 13.5 Schematic of ACCAR process ported reactor

was developed based on the principles length. The preheating kiln, where a large
presented in Chapter 7. The model was volume reduction of gas is expected due to
validated with the data collected in its pilot temperature drop, may take a conical shape
rotary kiln when it was run in the preheating for greater effectiveness, in which case the
mode. Based on the model the following charging end diameter would be 2.1 m. The
observations were made: diameters given above are all unlined
1. For effecting the same level of heat diameters. Refractory lining and insulating
transfer the length requirement of kiln layers together are expected to occupy a
increases only marginally with decreasing thickness of 250 mm and the lined inside
diameter. As a thumb rule the length diameters would be lesser by 0.5 m.
requirement may be assumed constant. The above analysis has led to the concept
2. With increasing filling degree of the of an optimised rotary kiln sponge iron
kiln, the efficiency of heat transfer increases. making process, which was mentioned in
Since there are practical difficulties in Chapter 5 and is described in the next section.
operating the kiln beyond 45 to 50% filling
13.4.6 An Optimised Rotary Kiln Sponge
degree, and also because substantial
Iron Making Process
freeboard is required for easy passage of gas,
a filling degree of about 40% is considered Figure 13.6 schematically represents the
optimum for a rotary kiln to be an effective optimised combination of equipment for
recuperator. running a low energy consuming rotary kiln
sponge iron plant. As mentioned earlier, the
In contrast to the above, for a reduction
preheating kiln is so designed as to heat the
kiln, the filling degree must be much lower.
iron ore to about 700°C or more using the
Practice has shown that higher than 15% filling
energy in the waste gas. The discharge end
degree impairs the process of reduction. This
of this preheating kiln, where the waste gases
may be because for reduction purpose a large
enter in counter current fashion, and the
free board is needed to maintain the delicate
combustibles are burnt with preheated air,
balance between oxidising and reducing
effectively works as the After Burning
zones.
Chamber (ABC – also called After Burning
Thus the design of the sponge iron rotary and Cooling Chamber) and no separate ABC
kiln is a compromise between the is required. The rest of the preheating kiln
requirements of reduction and heat transfer. effectively acts as a cooler for the gases and
If the two functions are separated out in two its temperature should come down to about
different kilns then there need not be any 500°C.
design compromise, and the entire function Along with iron ore, the flux and
can be carried out in an overall lower kiln recirculated char can also be heated. This
volume. way flux would be calcined and the
For a 500 tpd plant using coal with 6.4% recirculated char would get ground since iron
moisture and 24% ash, the dimensions ore would act as the grinding medium. With
expected for the reduction kiln are 5 m Φ fresh surface of char being exposed, gasi-
and 40 m length, and for the preheating kiln fication would be easier. The low reactivity
these are 3 m Φ at discharge end and 30 m of char deters its recirculation in the normal
process, but this way not only would the char the temperatures in actual ABC’s are much
be rendered reactive and useful, there may lower than what is indicated here. This does
be enhanced heat transfer in the preheating not affect the results of this calculation.
kiln due to partial reduction of Fe2O3 to FeO. However, it is difficult to ascertain with
In Chapter 5 we have presented a method which variant an actual process needs to be
of theoretically calculating the coal compared. This is the reason why three
requirement in sponge iron making process similar variants, differing only in the
in rotary kiln. Based on this calculation, the theoretical maximum temperature of ABC,
expected coal consumption in the normal as have been presented.
well as optimised process can be compared.
The processes presently commercialised
In the calculations presented in Chapter 5, in India would generally correspond
four variants are considered. In all variants, somewhere between variants (i) and (iii).
waste gases are assumed to be completely
In variant (iv), inputs have been taken
combusted i.e. no volatile matter is allowed
to go out free or partially reacted. 200 kg/t from a heat transfer calculation carried out
of char is assumed to be recirculated in the based on a model developed and tested in
fourth variant. If char addition is not desired the pilot rotary kiln, operating the kiln as
it can be replaced by 151 kg of coal. though it were a preheating kiln only. These
To recall, the variants considered in calculations indicate that, for a module
Chapter 5 are: producing 500 tpd sponge iron, a preheating
kiln of 3 m dia (2.5 m with refractory lining)
(i) Conventional process, waste gas
going out at 1200oC and 30 m length, and preheating charge to
664oC would mean waste gases going out at
(ii) Conventional process, waste gas
700oC. This is without any drying of charge,
going out at 1000oC
reduction of iron ore or calcination of flux
(iii) Conventional process, waste gas
taking place in the preheating kiln. If a partial
going out at 800oC
reduction and/or calcination takes place, heat
(iv) Process with preheating of iron ore, transfer would be further enhanced and
char and flux to approx. 700oC (664oC)
energy saving would be still better. Even
and air preheating to an average
traces of CO and/or H 2 are adequate to
temperature of 400oC.
reduce Fe 2O 3 to FeO, and this is highly
In variants (i), (ii) and (iii) the system has
feasible in the preheating kiln. Outlet gas
been considered up to a point where the
temperature should come down to about
combustibles in the waste gases have been
500oC.
fully burnt with stoichiometric amount of air,
but the heat so developed has been assumed A 500 tpd kiln of conventional type is
to be still retained by the gas in the After likely to have dimensions 5 m Φ * 70 m length.
Burning and Cooling (ABC) Chamber. In In the preheating zone, length required for
actual practice, burning and cooling zones the same extent of heat transfer increases very
(cooling by water spray) cannot be sharply marginally with decreasing diameter. Thus,
demarcated. Rapid transfer of heat starts when preheating kiln is used, a possible
before completion of combustion. As a result combination of dimension would be
Fig. 13.6 An optimised rotary kiln sponge iron making process
(a) Preheating kiln 3 m Φ * 30 m length and calculation with the actual Cp values of the
(b) Reduction kiln 5 m Φ * 40 m length constituents, which are functions of tempera-
Tables 5.3 to 5.9 present the method of ture, and integrate over the relevant tempera-
calculation of coal requirement of sponge iron ture range to obtain a more accurate result.
rotary kiln for the four different variants. In Figure 13.6 represents a possible plant
Table 5.9 we can find the heat deficits for based on the optimised concept. The diagram
these alternatives and thereby the total coal has been made keeping 500 tpd plant in mind
requirement. although it could well be applicable for a 50
One simplifying assumption which has tpd or even a 25 tpd plant. Essential equipment
been made is that the average molar specific have been presented in the diagram and it can
heat of waste gases has been assumed to be form the basis for preparing detailed flow
9.0 kcal/kg mole - oK. One can repeat the diagram and the equipment list.
13.4.7 Pros and Cons of the Optimised 13.5 KILN REFRACTORIES

Concept In direct reduction kilns, selection of proper
As mentioned earlier, for utilising the energy refractory material does not pose a serious
in waste gases, many of the existing plants problem mainly because of relatively low
have resorted to electric power generation. operating temperatures. Dense-fired alumino-
While this is a viable option for large plants silicate refractories in the range of 45-85%
(of capacity) 60,000 tpa and above, but for alumina should meet the requirements,
smaller plants, such power generating units provided they are resistant to abrasion and
are not viable. The result is that many units to highly reducing conditions. Normally
of capacity 15,000 tpa (50 tpd) have come up Andalusite based firebricks with low thermal
and are operating off and on as they are conductivity (1.0 w/mk) and high cold
finding it profitable to run only during the crushing strength (100 N/mm2) are preferred.
periods of good market demand. Over the However, the experiences of the operating
years, they have not been able upgrade the plants are more in favour of using high
process to make it viable during the alumina castables. Even though it calls for
depressed market conditions or are unable higher investment initially, performance has
to generate enough funds for installing so far been found to be much more reliable.
additional units to a capacity level where The reason for the higher reliability of
installation of an electric power-generating castable refractory may be due to the better
unit would be viable. resistance it offers to thermal fatigue. Schnabel
With the optimised process concept, (1977) estimated that refractory surface gains
maximum possible energy would be utilised a temperature of about 80°C during the time
within the process with marginal extra it is exposed to free-board; and loses it when
installation cost. Operating cost would reduce it is in contact with the solid charge. Since
drastically since coal consumption would be this happens with each revolution of the kiln,
halved. The After Burning and Cooling one can imagine the extent of heating and
function would be conducted inside the cooling cycles to which the refractory surface
preheating kiln, which would also carry out is exposed.
the function of grinding of char for complete For using castables, the inner shell of the
utilisation of its carbon content. Only the Air- kiln has to be welded with large number of
preheater would be an extra installation. But evenly spaced stainless steel anchors of ‘U,’
this would also be in lieu of the Air Cooler ‘V’ or ‘Y’ shape. Silica wool or similar
or Gas Conditioner which are presently insulating blanket of 25 to 50 mm thickness is
needed to make the waste gas suitable for thereafter laid out against the shell followed
being treated in the dust removal unit by 50 to 75 mm layer of insulating blocks.
(electrostatic precipitator or bag filters). Finally, 125 to 150 mm layer of castables are
Table 13.3 gives a comparison of the laid out and then dried and cured as per
optimised process with the conventional instructions specified for the castables.
process. The optimised process can be seen While the three layer refractory lining is
to be much more environment friendly, the ideal refractory design using castables,
besides other benefits mentioned earlier. most of the plants in India have not adopted
this design. Instead they have a single layer grinding action on the insulating layers, lying
of castables with no effort at heat conser- in between, with each revolution. Under such
vation. Consequently the surfaces of kiln shell a condition, before a year is completed, the
reach temperatures of 400oC or more while insulating layer converts into powder and
the three layer design of same total thickness stops giving back-up support to the main face
could give shell temperatures as low as 100oC layer. The main face layer thus becomes
or even less. vulnerable to its own weight and the weight
Reason behind the non-adoption of the of the material in the kiln.
three layer design has been the refractory But the increased radiation loss is an
failures encountered in the early indigenised avoidable loss to the process and it is
units in India. The cause of these failures could surprising that serious efforts are not being
be a combination of poor quality of insulating made to make the insulating layer concept a
layers, poor workmanship in laying the success. One reason could be that the
individual layers and a faulty anchoring additional radiation loss represents only
design which allowed a very slight relative about 0.2 GCal/tonne of energy loss, which
movement between the main face castable (i.e. represents only about 3% increase in coal
the outer layer exposed to reacting materials) consumption. Doing maintenance work
and the kiln shell during the rotation of the around a kiln whose outside temperature is
kiln. Even the slightest relative movement about 400oC is not easy, but apparently the
between these two layers causes the slight operators have got used to these conditions.
Table 13.3 Comparison of the conventional and optimised process

Process →
Conventional Optimsed*
Items ↓
Process Parameters
5.5 to 6.0 2.7 to 3.0
Energy consumption
GCal/t-DRI GCal/t-DRI
Coal consumption
1100-1200 540-580
[Coal of 10% Moisture, 25%
Kg/t-DRI Kg./t.DRI
Ash; Wet basis]
2.0 to 2.5 0.35 to 0.40
Loss of energy in waste gases
G Cal/t-DRI GCal/t-DRI
Sensible energy loss in solid 0.28 to 0.32 0.2 to 0.22
product GCal/t-DRI GCal/t-DRI
2.1 to 2.3 1.05 to 1.15
CO2 emission
t/t-DRI t/t-DRI
Design Aspects
Preheating kiln Not present Present

Shape of preheating kiln Not applicable Conical
Air preheater
Air preheater Not present Present
Electrostatic precipitator Generally present Present
* This is an improved version of Indian patent No. 170432 (SAIL-DR Process—January 21, 1987).
There is one benefit the operators gain in this development of higher strength and abrasion
situation. If the surface temperature at any resistant insulating castable. Once such a
time falls sharply below 400oC (measured material is available at reasonable price, the
either by an infra-red camera or hot chalk or entire lining could just be of this single layer.
a sliding contact thermocouple), they conclude This would not only mean reduced weight
that accretion has formed at this point. If the (insulating refractories are highly porous and
temperature goes sharply up, they conclude thus have very low block density) but lower
that refractory layer has fallen off. thickness requirement resulting in further
But a 3% continuous and extra coal reduction in weight.
consumption is not something which should High strength and abrasion resistant
be ignored. After overcoming defects in insulating blocks are already available and
quality, workmanship and design, the may possibly be considered for use. But
insulating layer may well be re-introduced seeing the history, entrepreneurs may have
in future plants. more faith in castables and would, therefore,
There is a parallel development taking wait for the development of high strength
place in refractory area, and that is the and abrasion resistant insulating castables.
This page
intentionally left
blank
14
Other Uses of Rotary Kiln
for Reduction Purposes CHAPTER
Rotary kilns are used for a variety of purposes magnetic separation. The magnetite concen-
like cement making, and roasting and trate is collected in the ferromagnetic fraction.
calcination of lime, dolomite and other In India, high grade iron ore is in great
refractory materials. While these are demand and the deposits are being depleted
examples of oxidising application there are a at a very fast rate. To find iron ore lumps
few examples of reduction in rotary kiln other containing 64% or more of iron, suitable for
than sponge iron making. sponge iron making, has become difficult. On
the other hand, there is accumulation of lower
14.1 MAGNETISING ROASTING OF grade overburden at mining sites and also
LEAN IRON ORES lower grade fines at production as well as
sizing facilities. A good way of removing the
In Chapter 13 the magnetising roasting of lean
accumulation would be to beneficiate them.
iron ores have been mentioned. India is
fortunate to be endowed with very high- Even though the beneficiation plants
presently operating and those planned do not
grade hematite deposits. But many other
have or contemplate having facility of
countries of Asia and Europe are not so
magnetising roasting, but there is a distinct
fortunate. The lean iron ore deposits need to
possibility that as the available grades
be beneficiated to 65% plus iron. After decrease further, this particular concept may
sufficient size reduction to effect a reasonable re-emerge and be contemplated for
liberation of iron oxide and gangue, the ore installation.
is treated in a rotary kiln where a mildly
reducing condition is maintained. To prevent 14.2 PRE-REDUCTION OF CHROMITE
reduction beyond the desirable level, ORE
superheated steam is injected under the bed, In the natural chromite ore, chromium occurs
using the ported kiln concept as depicted in in various states of oxidation, starting with
Fig. 13.6, so that formation of Fe3O4 is ensured. Cr3+ to Cr6+. While using this ore in electric
After cooling, the product is subjected to arc furnaces for the production of ferrochrome
and other chromium based alloys, costly economising on electric power, we are not
electric power is needed for meeting the aware of actual adoption of this system.
endothermic heat required during reduction
of the various oxides of chromium. A lot of 14.4 PRE-REDUCTION OF
this electric power can be saved if the ore is TITANIFERROUS ORE
pre-reduced using coal and other reductants The metal titanium has very unique
in a rotary kiln. Ore would get preheated in properties, unmatched by any other metal or
this process and if charged hot, the sensible alloy. The lack of existence of an economic
heat would help in further saving of electric method of production has prevented the
power. widespread use of this metal. While its
In conditions similar to sponge iron rotary reactive nature at high temperature has forced
kiln, chromite ore cannot be reduced to the adoption of very special extraction
metallic stage, but all chromium is expected technique, the fact that the ores available on
to be brought to trivalent Cr3+ or lower state. land contain very small percentage of
Even this is of great help in the electric titanium has made matters worse. It is only
smelting process of making chromium- the beach sands of Kerala and Orissa in the
containing alloys. southern and eastern shores of India, which
are presently being exploited or considered
M/s Outokumpu, Finland has developed
for the extraction of titanium.
a technology of using the smelter off gas,
Tonnage wise a large quantity of titanium
which is rich in reducing gases – largely carbon dioxide occurs in or near iron ore deposits as
monoxide - for preheating and pre-reduction Titaniferrous Iron Ore. These deposits are
of chromite ore before charging in the electric rarely exploited for using their titanium
smelting furnace. The ferro-chrome plant of content (except for a very special application
Tata Steel at Bamnipal in the eastern Indian of charging titanium in iron blast furnaces for
state of Orissa is a plant based on this protecting the hearth refractory). A concept
technology. is however being tried, where this ore would
be smelted to produce pig iron (by-product)
14.3 PRE-REDUCTION OF and titanium as TiO would be contained in
MANGANESE ORE the slag. The slag would be further processed
Just like chromite, manganese occurs in for titanium extraction. If this scheme is
manganese ore in various states of oxidation, proved successful, pre reduction of the ore –
from Mn2+ to Mn4+. And like chromite, the possibly in rotary kiln – has to be an important
sponge iron making conditions can, at best, step for economising on electric power in the
convert all manganese to bivalent stage – no process.
metallisation of manganese is possible. Before
smelting in electric ferromanganese furnace, 14.5 PRE-REDUCTION OF NICKELI-
if the ore is pre-reduced and preheated in a FERROUS OVERBURDEN
rotary kiln, considerable saving in electric Nickel is extracted throughout the world
power is expected. from sulphide-based ores. India does not
While many ferromanganese plants have have such deposits and therefore India has
been considering adoption of this concept for to import its entire requirement of nickel. It
Other Uses of Rotary Kiln for Reduction Purposes // 135
is in this context that the nickeliferrous feasible at the moment and that is coming in
overburden of chromite ore deposit in the way of its commercialization.
Sukhinda in eastern Indian state of Orissa However, coal gasification processes exist
assumes importance. This overburden which can convert coal to a mixture of CO
contains significant percentages of nickel and and H2 through controlled combustion with
cobalt and a concept is being pursued where oxygen. In most of the faster processes the
Ni and Co would be reduced in a controlled ash is melted and needs to be treated with a
manner to NiO and CoO and then extracted flux to make the resultant slag to flow out
through suitable leachant. In a sponge iron easily from the gasification reactor. For high
rotary kiln, nickel and cobalt would be fully ash coals this step turns the process unviable.
metallised. Therefore, to effect a controlled Jindals are in the process of installing a coal
reduction to bivalent stage would require gasification based sponge iron plant at Angul
considerable control and R&D efforts. But in the eastern Indian state of Orissa where
reduction in rotary kiln is a distinct possibility sponge iron would be made in a shaft furnace
and appears to be worth pursuing. A using gas from coal made in a Sasol-Lurgi
technique very similar to that of magnetising gasifier. The latter would be producing the
roasting of iron ore should be a feasible gas at lower temperatures so that ash remains
proposition. solid and is discharged as such, i.e., without
fluxing. This plant may prove to be the mirror
14.6 USE OF ROTARY KILN FOR of the future large sponge iron plants to come.
GASIFICATION OF COAL However, projections indicate that for smaller
As ash in coal adversely affects the operation merchant DR plants, rotary kiln would remain
of all reactors using coal or coke, many the more viable option.
attempts have been made to minimize or One more option is available for coal
eliminate this drawback. Coal beneficiation gasification and that is to carry out the
by gravity and similar methods has been process in a rotary kiln. In the present sponge
extensively adopted. But, in Indian coal iron rotary kilns both the reduction process
deposits, the mineral matter (source of ash) and the gasification of coal are being carried
is very finely dispersed in the coal seams. out simultaneously. Then, without the iron
Consequently, significant ash elimination is ore, why can the rotary kiln not produce a
accomplished only with substantial loss of reducing gas from coal? We feel that this is
yield. Hence, successful adoption has been definitely feasible and future researchers
only at places where the rejects are gainfully should investigate this possibility which has
utilized. immense potential for India. Unlike the
One apparently very attractive alternative sponge iron rotary kiln, there would be a
is to convert the value in coal (C & H) into a need to enhance gas and bed interaction
liquid or gas which would leave behind the through the use of higher filling (hence
mineral matter. Solvent extraction of coal has smaller kiln), use of lifters and under bed
been found technically feasible although it is injection of air or oxygen using the ported
not economically viable. Here the value in coal kiln concept. Sponge iron would then be made
is dissolved into an organic solvent and in a shaft, but rotary kiln will remain an
filtered. Full recovery of solvent has not been important contributor.
This page
intentionally left
blank
Environmental Aspects of
15
Sponge Iron Making in Rotary
Kiln and Future Prospects CHAPTER
The subjects of effect on environment and the 15.1 CARBON DIOXIDE EMISSIONS
future of sponge iron making have been linked FROM THE SPONGE IRON
together with a reason. Even though the INDUSTRY
immediate future of sponge iron making is The average CO2 level in atmosphere was
dependent on the usefulness of sponge iron 0.0295% at the end of nineteenth century,
vis-à-vis competing items and the value it which increased to 0.035% in 100 years. The
offers to society, in the long term sponge iron greenhouse effect it causes, has led to global
can survive and flourish only if it offers these warming. The ice cap at North Pole has
values along with minimum damage to punctured at many places. The larger ice cap
environment. Any iron and steelmaking in the Antarctic is also receding. Glaciers
operation causes incremental damage to feeding perennial rivers are reducing in size
environment, through the release of noise and we are already experiencing much more
pollution, release of waste heat, solid wastes extreme weather conditions.
(slags, ashes, char, etc.), polluting liquid Unfortunately, we cannot conceive of an
constituents (Phenol, tar, aniline etc.), economic iron and steel making route which
particulate matter (dust) and harmful gaseous does not release CO2. The cleaner fossil fuel –
constituents like CO, SOx, NOx and also CO2. petroleum and natural gas – are in limited
Extensive attempts have been made to control supply. Sponge iron making using such a fuel
all the above and keep them within limits and reductant liberates much lesser amount
which the environment can tolerate and of CO2 but this route cannot be expected to
absorb, except for the liberation of CO2. meet fully the iron and steel needed, nor can
Normally CO 2 is not considered a it be expected to replace the blast furnace route
pollutant. But with the rapid industrialization of iron making. But a variation of this route is
of human society, the rate of net release of expected to be the most environment-friendly
CO2 to atmosphere has increased very rapidly and hence the acceptable route of iron and steel
and we are already seeing glimpses of the extraction of the next century, about which we
catastrophic change it can cause. would discuss at the end of this chapter.
Table 15.1 gives us an idea of the Table 15.1 Carbon dioxide emissions in steel making g
quantity of CO2 being released in the Sl. No. Process routes Global practice Indian practice
atmosphere and also compares the
Indian Scenario with the global practice. 1. BF–BOF 2.4 3.5
Table 15.2 tries to split the CO 2 2. Coal based DR–EAF 1.8 2.0
emissions between iron making and
3. Gas Based DR–EAF 0.8 1.0
steelmaking stages and demonstrates
that most of the emissions take place in 4. Corex–BOF 2.5 2.5
iron making stage only. Scrap based 5. Scrap Based EAF 0.05 0.05
steel making has to remain the favoured
option for environment protection, but Please note:
1. The inferior figures for Indian practice at Sl. No. 1 & 2 are
that has to depend upon the availability mainly due to the higher ash levels in coal and coke.
of recycled scrap, and therefore, can 2. Generation of electricity for the electrical furnaces would
normally involve CO generation, which has not been
meet only 30 to 40% of the steel 2
accounted for in this Table.

requirement. For making steel through
extraction of iron from oxides, DR-EAF industries, accounting for about 60% of the
is the better option at present. total CO 2 released. Steel industry (iron
In international forums, there has been a making + steel making) comes next at about
proposal to put a “Carbon Tax” on the 10%. Out of this, sponge iron making
industries. As per this a tax should be levied constitutes less than 2%.
based on the quantity of CO2 the industries
release into atmosphere. Imposition of this 15.2 OTHER POLLUTANTS AND
tax has been deferred so far due to the ARISINGS IN SPONGE IRON
skewed nature of the details attached to the AND STEEL INDUSTRIES
proposals so far, which are weighted heavily
Table 15.3 lists the estimates of amounts of
against developing countries, but we can
safely assume that this tax in some form will solid arisings in Indian coal based rotary kiln
be levied sooner than later. sponge iron plants as per the present practice.
Table 15.4 compares these arisings with
Sponge iron industry has an advantage
similar arisings in blast furnace based
in this regard. But coal based DR stands far
behind gas based DR. This fact also indicates integrated steel plants. It can be seen that
that an opportunity exists for coal based DR sponge iron plants are already ahead in
industry. If all the energy conservation utilization of its solid arisings even though
measures are adopted as mentioned earlier the specific quantity of arisings is higher.
in this book (adoption of conical preheating Table 15.5 lists the pollutants in exhaust
kiln, air-preheater, char-recirculation etc.) the gases of rotary kiln sponge iron plants and
Figure of 1.8 would come down in the range compares them with those found in blast
of 1.1 to 1.2 tonnes of CO2 per tonne DRI. furnace stack. As far as the liquid effluents
To put matters in proper perspective, the are concerned, virtually no such effluents go
maximum amount of carbon dioxide which out of the plant, except where wet scrubber
is released into atmosphere by Indian is used. Effluent quantity in case of wet
industries, is the thermal power producing scrubbers range from 800 to 1000 litres per
Environmental Aspects of Sponge Iron Making in Rotary Kiln and Future Prospects // 139
Table 15. 2 Carbon dioxide emissions at different stages under Indian conditions
Tonnes per tonne iron Tonnes per tonne steel
Basic
Blast Smelting Scrap based electric
Coal based DR Gas based DR oxygen
furnace reduction arc furnace
furnace
1.8 to 2.0 0.80 to 1.0 3.3 to 3.5 2.3 to 3.0 0.15 to 0.30 0.03 to 0.05*
*Once again, CO2 emission during generation of electricity has not been considered
tonne sponge iron. Typical values of adoption of zero CO2 emission technology of
suspended solids in such effluents are 32 mg/ steelmaking in the next 100 years about which
litre (national norm up to 100 mg/litre); we would discuss towards the end of this
biological oxygen demand (BOD) 20 mg/litre chapter.
(national norm up to 30 mg/l) and chemical
oxygen demand (COD) 50 to 96 mg/l (national 15.3 SPONGE IRON IN INDIA
norm 250 mg/litre). Even though sponge iron making in rotary
Thus we see that, from the point of view kiln made significant starts in South Africa,
of emission of pollutants, as also CO 2 South American countries Peru and Brazil and
emission, sponge iron making in rotary kiln in New Zealand, India is now the main centre
is better placed than blast furnace iron with much more than half of world
making. As stated earlier, if all the energy production concentrated here.
conservation measures mentioned in earlier Presently, sponge iron accounts for about
chapters, are adopted (conical preheating kiln, 40% of total iron production in India and
air preheater, char recirculation, etc.), the CO2 more than half of it is from rotary kiln. World
emission is poised to go down to as as 1.1 to production figures are not so significant
1.2 tonnes per tonne sponge iron. which underscores the position India occupies
Environmental concerns may lead to in rotary kiln sponge iron making.
Table 15.3 Estimated average specific arisings from Indian coal based
sponge iron plants
Sl. No. Items Quanity Remarks
1. Iron ore fines 153 Fully sold or utilized
2. Kiln backflow 240 Partly sold or utilized; rest dumped
3. Dolo-char and ash 355 Partly utilized or sold; rest is dumped
4. Agglomerates and accretion 52 Mostly dumped
5. Fly ash 300 Average for plants having power plants.

Table 15.4 Comparison of solid arisings from blast furnaces vis-a-vis rotary kiln consumption is
sponge iron plants strongly depen-
Quantity of arisings
dent on the state
Total quantity of arisings of development
recycled/reused and sold
Percent
Million Kg/tonne Million Kg/tonne utilized
of the economy
tonnes per steel or tonnes per steel or where it is
annum sponge iron annum sponge iron consumed. In this
Blast furnaces
5.072 479 2.916 275 57.5 connection, the
(integrated plants)
Coal based rotary kiln
economies of
800 638 78.5 Germany, UK,
sponge iron plants
Does not include fly ash or sinter plant wastes USA and Japan
can be considered
Table 15.5 Comparison of pollutants in exaust gases in blast furnace iron making and
coal based rotary kiln sponge iron plants as developed
Maximum Levels :
economies where
Suspended Particulate Matter (SPM - National Norm) = 150 mg/Nm 3 the consumption
Oxides of Sulphur (SOx - International Norm) = 50 mg/Nm 3 had risen to a
Oxides of Nitrogen (NOx-International Norm) = 150 mg/Nm3 mg/Nm3
peak and then
SPM SOx NOx
declined and now
Coal based rotary kiln sponge
80 to 120 Up to 110 UP to 25 tends to fluctuate
iron plants - Exhaust Gas
around a stag-
Blast Furnace Stack Gas 6 to 26 9 to 124 3 to 48
nating curve.
If we plot the normalised consumption
Almost the entire sponge iron production of steel (i.e., ratio of consumption to peak
is consumed for steel making purposes. The
consumption value) against normalised time
following projection indicates that steel
of peak consumption (ratio of years elapsed
consumption in India is likely to peak in 2050
from a judiciously chosen starting year to the
AD at about 170 million tonnes.
years taken to reach peak consumption) we
get a plot of the type shown in Fig. 15.1.
15.4 PROJECTIONS OF STEEL
CONSUMPTION World consumption has not been
showing such a trend, or rather the peak has
Steel has so far proved to be the single key
not been reached yet. Since world
factor responsible for industrial production,
consumption would actually be
and thereby, for economic growth. And it is
superimposition of many such curves in a
growing from strength to strength with
newer developments – both within steel staggered manner, the peak is expected to
making practice as well as engineering be erratic like a mountain range. In Fig. 15.2
developments, which ask for more usage of is plotted the production figure, which should
steel. So much so, that economic development be consistent with consumption, although
has become almost synonymous with steel with correction factors for yield, stock build-
consumption for a developing country like up and releases.
India. China is presently the dominant player
Trend studies have revealed that steel in steel consumption pattern of the world.
Apart from being the largest steel consumer, 5.5%, which suggested direct dependence of
it is also the largest importer, and the latter industrial growth on steel consumption.
has contributed maximum in stimulating steel
production all over the world – as also in 15.5 PROJECTIONS FOR INDIA
India. In-house projection indicated that India’s steel
consumption would peak to 170 million
tonnes by 2050 AD (Fig. 15.3). This projection
was made in 1997 based on trend analysis
and the outcome appeared very improbable.
But since then the changes occurring in the
c
steel industry of China has raised hopes that
this type of change is very much likely.
For this to happen, the steel consumption
in India will be increasing by about four times
in the next four decades.
15.6 POSSIBLE SOURCES OF

ADDITIONAL IRON
For meeting the corresponding increased
demand of iron in India, blast furnaces will
1100
1000
900
For a long time India’s steel production
and consumption figures were roughly 800
matching, but in FY 2002-03 India produced 700

about 32.5 million tonnes of finished steel,
3.6 million tonnes more than its consumption, 600
largely because of increased exports. In FY
500
2003-04 these figures increased to 36.2 and
5.8. In spite of the recession in the domestic 400
steel production thereafter, yearly con-
sumption rate increased consistently. Many 300
times, the actual consumption has been falling
200
short of projected figures. But in 1991-93
period the consumption increased by 21%, 100
averaging 10.5% per year, as against a six-
decade average growth of 5.5% per year till 0
1880 1900 1920 1940 1960 1980 2000 2020
that time. The six-decade average growth in Year
the index of Industrial Production was also Fig. 15.2 World crude steel production
180 can run efficiently – would

use less and less of hard
160
coke, may use even formed
140 coke – would inject more of
Steel demand, Million tonnes
alternative fuels like

120 beneficiated and
pulverised non-coking
100
coal, tar, etc.
80 In Table 15.1, we
attempt to predict the
60 situation in India after
about 4 decades. The
40
overall global situation
20 would be roughly com-
patible with the situation in
0 India as regards the
1940 1960 1980 2000 2020 2040 2060 2080
contribution of different
Years, AD
routes.
Fig. 15.3 Long term steel demand projection for India
continue to play significant role. But its 15.7 ROLE OF SPONGE IRON AND
dominance is likely to be eroded largely NEWER PROCESSES
because of shortage of metallurgical coal. The
While changes in rest of the world are going
availability of the same may remain restricted
to be in similar pattern, the DRI or sponge
within the country. Blast furnaces may not
iron part would be of coal based variety in
only be surviving but the numbers may grow
Indian sub-continent while in rest of the
based on imported metallurgical coal. But
world (with possible exception of African
even these imports may gradually prove cost
continent) the natural gas based sponge iron
wise prohibitive. In spite of these problems,
would predominate. Smelting reduction
mini (and micro – 23m3) blast furnaces are
processes like Corex, Romelt and/or other
now proliferating.
newer processes would supplement the hot
About 55% of iron in India is extracted in metal production from blast furnaces. While
blast furnaces, while 40% comes from the DRI blast furnaces would be using a minimum
route (Table 15.1). In India, scarcity of coking amount of hard coke, even this may be
coal has been forcing adoption of processes supplemented by formed coke or something
involving non-coking coal. The Indian non- similar. On the other hand, hard coke may
coking coals are unsuitable for many be made differently. There is a promising
applications, but appear well suited for trend of adopting heat recovery type non-
treatment in rotary kiln sponge iron making recovery and stamp charged coke ovens. Of
processes – well suited for treatment in rotary the various attempts on making good quality
kiln sponge iron making processes. The blast coke from inferior coals, the route of
furnace iron making would stay as long as it increasing the bulk density of blend (through
Table 15.6 Likely Indian steel making scenario in mid nearly 400 sponge iron rotary
twenty first century kilns will be operating in India.
(Inferred from 'Steel Demand Forecasts: India 1999-2000'–Steel Presently, sponge iron
Times International, 23 (4), pp. 14-17 (July '99)
(Million tonees)
production in India by the coal-
Present Expected after
based route is nearly 4 times the
four decades rest of the world put together
Total steel production 50 170 (Fig. 15.4).
Steel production by Hot Metal-BOF Route 30 85 However, for sponge iron
(Contribution of SR-BOF likely to increase makers in rotary kiln, the quality
from about 7% to about 20%)
of coal, especially its ash content
Steel production in Electric Furnaces (Hot 20 85
metal/Pig Iron + DRI + Scrap) (Contribution and reactivity, is a real challenge.
of DRI is likely to be maintained at about They are sometimes forced to use
70%)
‘F’ (Moisture + Ash = 40 to 47%)
Matching DRI Production 18 75
and ‘G’ grade coals with
Matching BF hot metal 27 65
consequent loss in productivity
stamp charging or otherwise) has proved and quality of sponge iron. Many of the
most economically viable. We may see producers are hoping that a suitable coal
alternatives based on this principle being gasification process would be developed,
pursued further. which would help them in eliminating ash
from the sponge iron making reactor. While
15.8 ROTARY KILN SPONGE IRON such a process has been technically proven,
MAKING: STATUS AND economic viability has eluded it so far.
PROSPECTS Gasification of coal in a rotary kiln itself may
In India the growth of coal based rotary kiln prove to be the answer to this problem. No
sponge iron making has been phenomenal. other metallurgical reactor can handle ash of
From a modest start in 1980 with coal better than a rotary kiln. While coal gas
demonstration plant of 100 tpd of Sponge Iron based sponge iron would be made in shaft
India Limited in the southern Indian state of furnaces, rotary kiln may have an important
Andhra Pradesh, the Indian sponge iron role here as well.
industry struggled to take off for nearly 10
years. Between 1990 and 1996, three large gas 15.9 USE OF SPONGE IRON IN
based sponge iron units were commissioned, BLAST FURNACES
which boosted India’s sponge iron The implications of charging sponge iron in
production. The production remained blast furnace have been covered in chapter
stagnated, although did not decline under 12. We reproduce a summary below, along
adverse circumstances, till the turn of the with projections.
century. In the 21st century, the coal-based The steel consumption in India is
units have got a boost and have catapulted increasing at such a rate that all of the produce
India to the top of sponge iron producers from both blast furnace and DR units are fully
overtaking Venezuela, Mexico and Iran. It is consumed, and there is demand for more.
expected that very soon more than 200 coal Therefore, blast furnaces do not as yet see
based rotary kiln sponge iron plants and DR units as a threat. In fact, Indian blast
India & Rest of the World India & Rest of the World India & Rest of the World
– All processes – Coal based processes – Gas based processes
50 5
40
40 4
30
ROW
India
Million tonnes
Million tonnes
Million tonnes
30 3
ROW
20
20 2
10 1 10
India ROW
India
0 0 0
1980 1990 2000 2010 1980 1990 2000 2010 1980 1990 2000 2010
Year Year Year
Fig. 15.4 Sponge iron production over the years

(Charts depict the phenomenal growth of coal based sponge iron making in India)
furnaces have started supplementing its feed sponge iron goes down to below about 40%
with sponge iron. Blast furnaces of Tata Steel of coke price. That time, the use of sponge
and one blast furnace each of the integrated iron in blast furnace may reach an average
steel plants at Bokaro, Rourkela, Bhilai and level of 10 to 15% of the charge. The demand
Durgapur have experimented with regular use projection of sponge iron, which is presented
of sponge iron. Some of them have touched in Table 15.1, does not take into account the
the usage rate of as much as 180-kg/tonne possibility of this consumption of sponge iron
hot metal. This is being done in spite of the in blast furnace. Use of sponge iron in blast
fact that, till hot metal stage, the benefits furnace at low rates is expected to boost both
obtained are always significantly below the productivity and coke saving by 1% for each
cost incurred. percent metallisation of the burden. Under
Sponge iron charging in blast furnaces is Indian context with use of coal based sponge
being resorted to when there is large demand iron, the values come to 8% of coke saving
for hot metal within the steel works and the and as much percent of productivity increase
for 10% metallization of burden,
availability of coke and coking coal is
corresponding to about 9% sponge iron of
substantially reduced. The overall plant
the total normal iron oxide in burden.
economics justifies the use of sponge iron in
the blast furnace burden. Projections indicate
that the use of sponge iron in blast furnace 15.10 SMELTING REDUCTION IN
will very soon reach a plateau and could then INDIA
start coming down – may even stop. It is Contribution by smelting reduction (SR)
expected that after a couple of years the processes to iron making is much less
charging of sponge iron in blast furnaces compared to both blast furnace and DR. But
would be started once again when cost of we would like to make a special mention of
Other smelting reduction processes have

not been seriously considered in India till now.
We, however, need to watch the develop-
ments related to CONTIFUR (acronym for
continuous induction furnace) Process being
developed by M/s Electrotherm (India)
Limited in the western Indian state of Gujrat.
15.11 MODERN TRENDS IN IRON

MAKING
In-spite of problems related to availability of
hard coke, blast furnace iron making is still
showing a reasonable growth. In-spite of the
fact that coke cost may further increase, we
may still see installation of more mini blast
furnaces and thus further boost quantity of
Fig. 15.5 Future iron making blast furnace iron. This is because large blast
furnaces had high capital invested therein
these SR processes, not because they pose a and need to operate at full capacity. Coke
threat to blast furnace but because they are ovens catering to these blast furnaces have
being seen as forerunners of potential to continue to operate even at higher cost. In
processes of the future. These can consume the process, coke ovens would be producing
iron oxide feed, which cannot easily be taken a fraction of coke, which cannot be used in
up by either blast furnace or sponge iron the large blast furnaces, and these, mostly
making processes. Examples of these iron smaller size coke, will remain available for
oxide feed are iron ore slimes, other iron
mini blast furnaces to use. Further, doubts
oxide wastes from steel plants, lower grade
about the availability of coke for the mini blast
iron ore fines accumulated at mine heads, Red
furnaces and sudden spurts in coke prices can
Mud from aluminium extraction process, etc.
be mitigated by partially using sponge iron
At present two Corex units are operating
in India under the Jindal Vijayanagar Steel in their charge, thereby reducing the quantity
Limited (JVSL–now renamed JSW). The hot of coke usage.
metal production is in excess of 1.5 mtpy. A As mentioned earlier, iron making in
0.3 mtpy capacity plant based on Romelt India is still dominated by blast furnaces,
Process was being considered by National accounting for nearly 55% of the total iron
Mineral Development Corporation (NMDC). production.
Another Romelt unit of 0.243 mtpy capacity Availability of sponge iron has led to a
was being planned by National Aluminium situation in India where sponge iron making
Company (NALCO) in Orissa based on their and blast furnaces are partners in industrial
waste product “red mud” which is basically growth. As mentioned earlier in chapter 12,
iron oxide. Both the proposals have now been Fig. 12.2 schematically presents a typical
kept on hold. concept, which newer mini steel plants of
India may adopt. In this concept, blast furnace of water. Most of the electric power needs of
iron making and sponge iron making have the entire steel making chain would come
been judiciously married for economic from non-CO 2 producing environment
viability. friendly route. Hydro electricity, sun power,
Even though smelting reduction wind power and geo-thermal power would
processes are considered forerunners of all be contributors. Power from the waves of
processes of the future, still their adoption in seashore may also be contributing.
the near future remains uncertain and they Thus, even though there is major
are unlikely to occupy a dominant position. competition among the newer processes
described earlier, and many of them may
15.12 A PEEP INTO THE IRON AND succeed commercially, their life is not going
STEEL MAKING PRACTICE OF to be long after all. The story of iron and steel
TWENTY SECOND CENTURY making had started with sponge iron and at
Making iron and steel causes incremental the end it would still be the sponge iron,
damage to environment, which are which would predominate, even though it
cumulating over the years. Green house effect would be made differently.
of accumulated carbon dioxide in the
atmosphere is already affecting global climatic 15.13 WHAT MAKES ROTARY KILN
conditions. Strict restrictions are coming into SPONGE IRON MAKING
force for emission control in automobiles and CLICK IN INDIA?
in production plants. The reason for the above lies mainly in the
We expect that within this century the
type of coal available in India. Most of the
problem is likely to grow to the question of
Indian coals are such that the ash contributing
survival of human race. With this imperative
mineral matter is well dispersed in the
in mind, the next century is likely to permit
matrix. Thus most of the coals are high in ash,
only those iron and steel making processes,
which cannot be beneficiated without large
which are safe for the environment. With this
reduction in size. Grinding coal to very fine
compulsion in mind, it should be possible to
hazard a guess regarding the iron and steel size renders it unsuitable for many
making processes of the twenty-second applications. Earlier, very high ash coals were
century. all diverted for use in thermal power plants
for burning in the pulverised coal burners,
Environmental concerns would force
minimisation of CO2 emissions from iron and but sponge iron making in rotary kiln is
steel making. Possibly, only the CO2 emitted another reactor where high ash coals can be
in electric melting of scrap or equivalent may tolerated. Rotary kiln has large freeboard,
be permitted. About 40% of the steel making which allows heavily dust-laden gases to pass
would be from recycled scrap. The balance without problem.
metallic input has to come from DRI, which The sponge iron Pilot Plant of RDCIS
would be made from hydrogen reduction of SAIL could be operated for full one week on
iron oxide (Fig. 15.5). This hydrogen again a coal with average ash content of 45%, which
may be mostly generated from the electrolysis occasionally went as high as 53%. Although
productivity dropped sharply to less that 60% 15.14 ANOTHER OPPORTUNITY FOR
of the normal value, satisfactory metallisation ROTARY KILN SPONGE IRON
level could be maintained. MAKING?
Rotary kiln sponge iron making process Another opportunity is waiting in the wings
appears to be tailor-made for Indian coals – for rotary kiln sponge iron making. The blast
and is poised to occupy centre stage in iron furnaces in the integrated steel plant of India
production in India. are presently using considerable quantities of
In the coming years, technology policies imported coals taking advantage of low
of governments are going to be increasingly international prices. But attempts have been
dominated by concern for the environment. focussed in reducing this dependence on
Emphasis would be more on environment imported coal through the following
friendly technologies and here also rotary measures:
kiln sponge iron making has a definite edge 1. Reducing consumption of metall-
over other coal based iron making processes. urgical coke in blast furnaces
If the optimised concept of preheating of 2. Using coke breeze along with sinter
charge and of process air by waste gas is in blast furnace
adopted, the edge could widen to an 3. Replacing metallurgical coke through
unsurpassable gap. coal dust injection through tuyeres
(using beneficiated indigenous non- designed for using high ash coal, there is no
coking coal) metallurgical reactor which is better suited
4. Using indigenous blendable coal in for handling high ash coal than the rotary
the coke blend. kiln. The large free board in the sponge iron
The option 3 listed above is a very rotary kiln makes it amenable to handle abuse
important option, which is likely to be exer- with large quantities of flying ash. The gas
cised, in stages, by most of the blast furnaces turbulence existing inside the sponge iron
in India. rotary kiln ensures that heat transfer is not
This would necessitate setting up of large hindered by increase in the dust load of free
capacity (or large number of) coal washeries board gas.
for making available the lower ash coal
In conclusion, we would like to present
needed for coal dust injection. The middlings
pictorially a scenario of coal utilisation by
generated from such coal washing plants may
integrated steel plants, which is presented in
be of such quantity that it may not be possible
to consume all of it in the traditional Fig. 15.6. This Figure points towards the
consuming point – the pulverised coal burners availability of massive quantities of coal
of thermal power plants. Rotary kiln sponge middlings in the coming days. This is an
iron plants would have the opportunity to use opportunity, which the sponge iron industry
this high ash coal fraction. If properly can ill afford to miss.
Bibliography
♦ Abraham, M.C. and A. Ghosh, Kinetics Making Shaping & Treating of Steel, 11th
of reduction of iron oxide by carbon, Iron edition (1999).
making and Steel making, 7(1), 14-23
♦ Ang, H.M., M.O. Tade & M.W. Sze,
(1979).
Residence time distribution for a cold
♦ Agariya, Dalganjan and K.K. Prasad, Iron model rotary kiln, AUSIMM Proceedings
making technique of the Agariyas: In the 303, (1), Nov. 11-16, 1998 (Australia).
words of an Agariya (in Hindi), Ispat
♦ Balasubramaniam. D., Radio isotopic
Vikas (First issue of 1995) (Published by
RDCIS, SAIL), 4-6, Ranchi. determination of charge movement in a
Rotary Kiln, M. Tech. Thesis, IIT Kanpur,
♦ Agrawal, B.B. and K.K. Prasad, Evolution India, (Jan. 1981).
of sponge iron technology - Invention
Intelligence, 23(8), 211-216 (August 1988), ♦ Balasubramaniam. R., Story of the Delhi
New Delhi. Iron Pillar, Foundation Books, New
Delhi, 2005.
♦ Agrawal, B.B., K.K. Prasad and G.I.S.
Chauhan, Utilisation of iron ore fines to ♦ Bandopadhyay, A., A. Ganguly, K.K.
produce ore-coal composite pellets for Prasad, S.B. Sarkar and H.S. Ray,
improved sponge iron making in rotary Determination of Kinetic Parameters for
kiln, Proc. Symp. Potential of sponge iron the Re-oxidation of Direct Reduced Iron
in Indian steel industry, Ranchi, 2005. under rising temperature conditions,
Thermochimica Acta, 228, 271-281 (1993).
♦ Agrawal, B.B., K.K. Prasad, G.I.S.
Chauhan, P.K. Chaudhuri & S.M. Aeron, ♦ Bandopadhyay, A., A.K. Ray, K.K.
Evaluation of earlier iron making Prasad, P.K. Bandyopadhyay and B.R.
processes and their relevance in Indian Chowdhury, Modelling of Heat Transfer
context, presented in the National in the Preheating Zone of a Rotary Kiln
Seminar on Indian Archaeometallurgy, for Sponge Ironmaking, Indian Chemical
Oct.1991, The Banaras Hindu University, Engineer, 31(4), 8-11 (Oct-Dec. 1989).
Varanasi. ♦ Bandopadhyay, A., A.K. Ray, M.P.
♦ AISE Steel Foundation (Pub.), The Srivastava, K.K. Prasad, S.V.B. Subba
Rao, P.K. Bandyopadhyay, R. Haque and ♦ Bandopadhyay, A., M.P. Srivastava, A.K.
B.R. Choudhary, Selection of Coals for Ray and K.K. Prasad, Effects of operating
Rotary Kiln Sponge Ironmaking, Trans. parameters on movement of charge in a
Indian Inst. Metals, 40(3), 209-218 (June rotary kiln for sponge iron making, Steel
1987). India, 11 (1), 9-17 (April 1988), Ranchi.
♦ Bandopadhyay, A., Amit Ganguly, K.K. ♦ Bandopadhyay, A., M.P. Srivastava, A.K.
Prasad, S.B. Sarkar and H.S. Ray, On the Ray & K.K. Prasad, Mathematical
Determination of Ignition Temperature modelling of charge movement in rotary
of Sponge Iron, Thermochimica Acta, 149, kiln for sponge ironmaking, Trans. Indian
Inst. Metals, 39(3), 181-186 (June 1986),
27-31 (1989).
Kolkata.
♦ Bandopadhyay, A., Amit Ganguly, K.K.
♦ Bandopadhyay, Amitava, Amit Ganguly,
Prasad, S.B. Sarkar and H.S. Ray, Low K.K. Prasad, A.K. Singhi and S.K. Gupta,
and High Temperature Re-oxidation of DRI on Fire: A Case Study, Steel India,
Direct Reduced Iron: A Relative 10(1), 58-63 (April 1987).
Assessment, Reactivity of Solids, 8, 77-
♦ Bandopadhyay, Amitava, Amit Ganguly,
89 (1990).
K.K. Prasad, P.K. Bandyopadhyay, S.B.
♦ Bandopadhyay, A., Amit Ganguly, K.K. Sarkar and H.S. Ray, Determination of
Prasad, S.B. Sarkar and H.S. Ray, Low Ignition Temperature of Sponge Iron
and High Temperature Re-oxidation of Samples by Thermal Analysis, Proc. Sixth
Direct Reduced Iron: A Relative National Symp. Thermal Analysis, Delhi
Assessment, Reactivity of Solids, 8, 77- (Dec.1987), V16-V21.
89 (1990). ♦ Bandopadhyay, Amitava, Amit Ganguly,
♦ Bandopadhyay, A., Amit Ganguly, K.K. K.K. Prasad, S.B. Sarkar and H.S. Ray,
Prasad, S.B. Sarkar and H.S. Ray, Newer Approaches to the Study of Re-
Thermo-gravimetric Studies on the Re- oxidation of Sponge Iron - Trans. Indian
oxidation of Direct Reduced Iron, ISIJ Inst. Metals, 42(1), 21-26 (Feb. 1989).
International, 29(9), 753-760 (Sept.1989). ♦ Bandyopadhyay, D., N. Chakraborti and
A. Ghosh, A Study on the Kinetics of Iron
♦ Bandopadhyay, A., Amit Ganguly, K.K.
Oxide Reduction by Solid Carbon, Steel
Prasad, S.B. Sarkar and H.S. Ray, On the
Res., 64 (7), 340-345 (July 1993).
Determination of Ignition Temperature
♦ Barr, P.V., J.K. Brimacombe and A.P.
of Sponge Iron, Thermochimica Acta, 149,
27-31 (1989). Watkinson, Heat transfer model for
rotary kiln: Part I, Metall. Trans. B, 20B,
♦ Bandopadhyay, A., Amit Ganguly, K.K. 391-402 (1989).
Prasad, S.B. Sarkar and H.S. Ray,
♦ Barr, P.V., J.K. Brimacombe and A.P.
Protective techniques to combat sponge Watkinson, Heat transfer model for
iron re-oxidation, Trans. Indian Inst. rotary kiln: Part II, Metall. Trans. B, 20B,
Metals, 44(4), 299-304 (August 1991). 403-419 (1989).
Bibliography // 151
♦ Batelle Institute e.v., Frankfurt, Report ♦ Choudhary, S.K. and A. Ghosh, Study of
on Reduction processes outside the blast coal reactivity with reference to direct
furnace and their effect on future iron reduction processes, Trans. Indian Inst.
and steel production in the world, 1975. Metals, 42 (6), 541-552 (Dec. 1989).
♦ Bayard, R.A., Chemical and Metallurgical ♦ Crawford, G.P., Iron & Steel Making in
Engineering, 52, p.100 (1945). Soderberg electric arc furnace, Iron
making & Stee lmaking, 16 (5), 314-319
♦ Biswas, D.K., Some design aspects of
(October 1989), London.
pneumatic coal injection for coal based
direct reduction process in rotary kiln, ♦ Dey, Shobhana, Flotation responses of
Powder Handling and Processing, 5(2), Gondwana coal in release analysis,
145-152 (June 1993). Mineral Processing and Extractive
Metallurgy, 114(4), 248-254, Dec. 2005.
♦ Cartaxo, S.J.M. and S.C.S. Rocha, Object
oriented simulation of pneumatic ♦ Direct from Midrex, Quarterly
conveying: Application to a turbulent publication from Midrex Corporation,
flow, Braz. J. Chem. Engg. 16(4), Dec. U.S.A., various issues since 1993,
1999. Charlotte.
♦ Chatterjee, A., A.V. Sathe, M.P. ♦ Feinman, Jerome and Donald R. Mac Rae
Srivastava & P.K. Mukhopadhyay, Met. (Ed.), Direct Reduced Iron: Technology
Trans. B. 14B, 375-381 (September 1983). and economics of production and use,
♦ Chatterjee, A., Beyond the blast furnace, Iron & Steel Society, Warrandale, P.A.,
CRC Press, USA, 1994. 1999.
♦ Chatterjee, A. and B.D. Pandey, Iron ♦ Ganguly, A. and A. Bandopadhyay, Self-

making and Steel making, Mechanism of ignition tendency – A hazard for sponge
ring formation in rotary kilns used for iron producer and user, Tool & Alloy
sponge iron manufacture, 8(6), 250-263 Steels, 22(1), 19-25 (1988).
(1981). ♦ Ginstling, A.M, D.M. Zil’berman & N.V.
♦ Chatterjee, A., Direct reduction of iron Gvozdev, Khim. Mashinostroenic, 8, p.8
ores, Prentice Hall, New Delhi, 2008. (1939); as quoted by Friedman, S.J. &
Marshall, W.R., Jr., Chemical Engineering
♦ Chatterjee, A. and A. Ghosh, Iron making
Progress, 45 (8), pp. 482-493 (1945).
and steel making. Theory and Practice,
Prentice Hall, New Delhi, 2008. ♦ Gupta, A.K., Ed., Technology of sponge
iron, Metallurgical Research Institute,
♦ Chen, Y., H. Zhang and Y. Xie, Study on
Roorkee, India, 1979.
reduction mechanism of chromium ore
pellets mixed with carbonaceous breeze ♦ Hobhouse, Steve, Iron and Steel Making
reductants, Sel. Pap. Eng. Chem. Metall. at New Zealand Steel, JPC Bulletin,
(China), 41-49 (1993). Kolkata, 5(1), 72-75 (January 2005).
♦ Jiang, Tao, Guanzhou Qiu, Jingkang Xu, Rotary Kilns using the TDR Process,
Deqing Zhu and Ram Singh, Direct Trans. Indian Inst. Metals, 37(6), 721-729
reduction of composite binder pellets (December 1984).
and use of DRI, Electrotherm (India) Ltd., ♦ Prakash, S. and H.S. Ray, Reduction of
72-Palodia, Ahmedabad, India. iron ore under rising temperature and
♦ Kong, L., T. Martinovic and W.K. Lu, fluctuating temperature conditions,
Reducibility of DR pellets, 58th Iron Thermochimica Acta, 111, 143-166(1987).
making Conference Proceedings 1999, ♦ Prasad, K.K., Iron making by indigenous
383-390 (Mar. 1999). communities in India, Part I: The scene
♦ Kubaschewski, O. and C.B. Alcock, and the Orissa and Bishunpur Processes,
Steel Times, 227 (12), p. 461 (December
Metallurgical Thermochemistry, 5th
1999), London.
Edition, Pergamon Press, London, 1979.
♦ Prasad, K.K., Iron making practice of
♦ Kumar, S.K., B.B. Agrawal, T.K. Datta,
indigenous communities in India, Part II:
K.K. Prasad, M.S.R. Iyengar & H.S.
Embers survive, Steel Times, 228 (1), p.
Ashwath, Impact of Sponge Iron on the
35 (January 2000), London.
Growth of Steel Industry, Proc. Symp.
Use of Sponge Iron in Electric and Arc ♦ Prasad, K.K., Traditional ironmaking by
Melting Units, Hyderabad, 295-312 (July indigenous communities of India,
1984). Ironmaking & Steelmaking, 27 (6), 408-
410 (December 2000), London.
♦ Levin, Ernest M., Carl R. Robbins &
♦ Prasad, K.K., Residence time of solids in
Howard F. McMurdie, Phase Diagrams
a sponge iron rotary kiln: Modelling and
for Ceramists, 3rd edition, American
measurement techniques, Proceedings of
Ceramic Society, Columbus, Ohio, 1974.
National Seminar on Experimental
♦ Lowry, H.H. (Ed.), Chemistry of Coal Approaches in pyrometallurgical
Utilisation, Supplementary volume, 521- research, Regional Research Laboratory,
522 (1976). Bhubaneswar, India, April 20-21, 2000,
pp. 85-96, Allied Publishers, New Delhi.
♦ Mazumder, A.K., T. Sharma & G.V.
Reddy, Reduction kinetics of iron oxide ♦ Prasad, K.K., Steel Demand Forecasts:
pellet by non-coking coal, Trans. Indian India 1999-2000, Steel Times Inter-
Inst. Metals, 42 (3), 299-305 (June 1989). national, 23 (4), pp. 14-17 (July ’99),
London.
♦ Mott, R.A. and R.V. Wheeler, The quality ♦ Prasad, K.K., Alternate Iron Making -
of coke, Chapman and Hall, London 1939, India is the World Leader, Steel and
p. 464. Metallurgy, 6(3), 41-44 (January 2004).
♦ Mukhopadhyay, P.K., A.V. Sathe and ♦ Prasad, K.K., Sponge Iron in Indian Iron
Amit Chatterjee, A Mathematical Model and Steel Industry, JPC Bulletin, Kolkata,
of Direct Reduction of Iron Oxides in 4(5), 11-18 (May 2004).
Bibliography // 153
♦ Prasad, K.K., Key issues in the ♦ Prasad, K.K. (Ed.), Steel Plant Techno-
Production and Utilisation of Sponge logies: An overview, RDCIS SAIL,
Iron, Proc Symp. Sponge Iron Industry – Ranchi, 2004.
Problems and Prospects, Bhubaneshwar, ♦ Prasad, K.K., A. Bandopadhyay & P.K.
7-20 (June 25-26, 2004). Bandopadhyay, Sponge Ironmaking in
♦ Prasad, K.K., Implications of charging Rotary Kiln: Strengths and Weaknesses,
sponge iron in blast furnace, Steel and Invention Intelligence, 23(10), 313-316
Metallurgy, 6(10), 12-17 (August 2004). (October 1988).
♦ Prasad, K.K., What the future holds for ♦ Prasad, K.K., B.B. Agrawal, & H.D.
iron making – Newer processes, Steel Pandey, Some environment aspects of
Times International, 26 (8), 30-34 coal based Sponge Iron making in India,
(September 2004). Proc. Symp. Environmental Management
in Metallurgical Industries, Institute of
♦ Prasad, K.K., Auto-ignition of Sponge
Technology, Banaras Hindu University,
Iron - A Review, Steel and Metallurgy,
Varanasi, India, December 14-16, 2000,
7(3), 16-23 (January 2005).
pp. 275-281.
♦ Prasad, K.K., Meeting iron needs of
♦ Prasad, K.K. and S.R. Mediratta,
India: Next four decades, Proc. Symp. Forecasting Demand for Steel, Steel
Iron making: Technology and Raw Technical Report, November ’97, pp.16-
Materials, Kolkata, 26-36 (July 1-2, 2005). 23, Lucknow.
♦ Prasad, K.K., Sponge Iron: Key to India’s ♦ Prasad, K.K. and A.K. Ray, Sponge Iron
development in coming decades, JPC Making in Rotary Kiln, R&D Centre for
Bulletin, Kolkata, 5(9), 6-13 (September Iron & Steel, SAIL, Ranchi, April 2002.
2005). ♦ Prasad, K.K. and H.S. Ray, The making
♦ Prasad, K.K., Sponge Iron as a Raw of (and attempts at breaking) the iron
Material for Iron and Steel Making, JPC pillar of Delhi, Steel World, 51-56, June
Bulletin, Kolkata, 5(11), 19-25 (November 2001.
2005). ♦ Prasad, K.K. and H.S. Ray, A pillar of the
♦ Prasad, K.K., Mysteries of the Delhi Iron community from the first to the third
Pillar, Part 1: History and Properties, millennium, Materials World, 8(7), 11-13,
Steel Times International, 30 (5), 52 (July/ July 2000.
August 2006). ♦ Prasad, K.K., H.S. Ray & K.P. Abraham,
♦ Prasad, K.K., Mysteries of the Delhi Iron Chemical and Metallurgical Thermo-
Pillar, Part 2: Making the Pillar, Steel dynamics, New Age International (P)
Times International, 30 (6), 70 (September Ltd., Publishers, New Delhi, 2007.
2006). ♦ Prasad, K.K., S.P.S. Sabharwal, M.P.
♦ Prasad, K.K., Iron and steel making: How Srivastava, B.B. Agrawal, G.I.S. Chauhan,
we reached there, Everyman’s Science, Mahesh Sharma, P.K. Chaudhuri, S.M.
41(4), 249-254 (October-November 2006). Aeron and C.R. Srinivasan. An attempt
on revival of rural iron making, Metal ♦ Saeman, W.C., Passage time of solids
News of Indian Inst. Metals, 12(1), 1-6 through rotary kilns; Factors affecting
(Feb.1990), Kolkata. time of passage, Chemical Engineering
♦ Ramakrishnan, V. and P.S.T. Sai, Progress, 47 (10), 508 –514 (1951).
Mathematical modelling of pneumatic ♦ Santra, A.K., A. Rajasekharan, V.M.
char injection in a direct reduction rotary Daptardar & V.L.N. Murthy, Statistical
kiln, Met. Trans. B, 30B, (5), 969-977B Modelling of Rotary Kiln Process,
(Oct. 1999). Presented in the 38th ATM of IIM,
♦ Ray, A.K., K.K. Prasad and P.K. Sen, Bangalore, (November 1984).
Residence-time in rotary kiln for sponge ♦ Schnabel, Wolfram, On the heat transfer
iron making: Prediction from kinetic data, during direct reduction in a rotary kiln,
Steel India 17(1), 16-20 (April 1994), Ph.D. Thesis, Rheinisch-Westfaelischen
Ranchi. Technischen Hochschule, Aachen, 1977.
♦ Ray, Asim Kumar, Ph.D. Thesis, IIT ♦ Serbent, H. & R. Dayal, Basic principles,
Kharagpur (1988). process characteristics, and industrial
♦ Ray; A.K., K.K. Prasad, H.S. Ray and P.K. application of rotary kiln iron ore
Sen, Residence time in a rotary kiln for reduction, Trans. Indian Inst. Metals,
sponge iron making: Prediction from 30(4), 246-255 (August 1977).
kinetic data, Steel India 17 (1), 16-20 ♦ Sharma, T., Reduction of Iron Ore Pellets
(April 1994). by Char Fines, Trans. Indian Inst. Metals,
♦ Ray, H.S., Kinetics of Metallurgical 43 (6), 380-382 (Dec. 1990).
Reactions, Oxford & IBH Publishing Co., ♦ Singh, R.K., A.K. Ray & K.K. Prasad, CFD
New Delhi, 1993. Analysis of combustion air flow inside a
♦ Ray, H.S., Non-isothermal kinetics – A sponge iron rotary kiln, Steel Tech
generalised approach and simulation of Journal, Kolkata (in press).
some industrial reactors, Proc. Symp. ♦ Sinhamahapatra, A.K., K.N. Gupta, S.D.
Thermodynamics and Kinetics of Singh & M. Ram, Kinetics and Mechanism
Metallurgical Processes (ICMS-81, of Formation of Direct Reduced Iron Rod
Bangalore), Indian Institute of Metals, from Ore and Coal Fines, Proceedings of
Calcutta (1985). 3rd International Symposium on Benefi-
♦ RDCIS SAIL, Trends in beneficiation and ciation and Agglomeration, Bhubaneswar,
Agglomeration of iron ore, Ranchi, 2001. India, 506-512 (16-18 Jan. 1991).
♦ Reuter G. and H. Serbent, Raw material ♦ Srinivasan, N.S. and A.K. Lahiri, Rotary
selection criterion for the SL/RN Process, Kiln Process: A simplified analysis, Trans.
Proc. Symp. The utilisation of pre- Indian Inst. Metals, 30(5), 309-312
reduced material in iron & steel making, (October 1977).
Bucharest, May 1976.
Bibliography // 155
♦ Srinivasan, Sharada and S. Ranganathan, cylinders, Chemical Engineering Science,

India’s legendary Wootz steel, IISc, 1, 253 (1952).
Bangalore, 2004. ♦ Venkateswaran, V. and J.K. Brimacombe,
♦ Stephenson, R.L. (Ed.), Direct Reduced Met. Trans., 8B, 387 (1977).
Iron: Technology and economics of ♦ Vishwakarma, Sonadhar Poyam and K.K.
production and use, Iron & Steel Society Prasad, First iron smelting: A Lohra
of AIME, Warrandale, P.A., 1980. folktale from Bastar, Invention Intelli-
♦ Sullivan, J.D., C.G. Mair and O.C. Ralston, gence, 30(10), 9-10 (January 1955), New
U.S. Bureau of Mines technical papers. 384 Delhi.
(1927).
♦ Wingfield, S.L., A. Prothero and J.B.
♦ United States Steel (Pub.), The Making Auld, Jl. Inst. Fuels, (1974), 64.
Shaping and Treating of Steel, 10th
♦ Zablotny, W.W., The movement of
edition, The Association of Iron & Steel
charge in a rotary kiln, International
Engineers (1985).
Chemical Engineering, 5(2) 360-366,
♦ Vahl, L. and W.G. Knigma, Transport of
April 1965.
solids through horizontal rotary
This page
intentionally left
blank
Index
A B
A residence time 50 Bastar 2
A.K. steel 108 Beneficiation 76, 79, 133
ABC 42, 65, 127 of coal 79
Accar 14, 122, 123 Bihar sponge iron 15
Accretions 6, 12, 63, 75, 89, 90, 92 89, 92, Bimolecular 72, 73
93 115, 123, 131 Bituminous 79
After burning and cooling 25, 126, 127, 129 coals 85
After burning chamber 126 Blast furnace 17, 18, 47, 53, 79, 96,
Agglomerated 76, 79 105, 108, 134, 111, 137, 138,
Agglomerated reduction process 122 141, 142, 143, 144
Agglomerates 6, 87, 89, 91 iron making 75, 107, 139, 142, 145
Agglomerating 8 technology 6
coal 115 Brown coal 79
Agglomeration 76, 115, 116
Air fans 24, 64 C
pre-heater 44 129, 138 Caking 87
preheating 127 Calcination 8, 41, 123, 127, 133
tubes 12, 24, 65, 123 Calcined 87, 126
Alloy steel plant 97 Calcining 8
Alumino-silicate refractories 129 Carboxy reactivity 81
Andalusite 129 Carburisation 2
Angle of repose 47, 49, 97 Carburisers 18, 106
Anthracite 79
Carburising 4
coals 79
Castables 129, 130, 131
Arc furnace 97, 105, 107, 133
Cementation 2
Ash fusion 89
Central burner 24, 63, 64, 65
Association of iron and steel engineers 19
Available heat 35 CFD analysis 25, 28, 29
Char 18, 22, 24, 35, 37, 40, 44, 46, 57, Conductive and convective heat transfer 62
79, 80, 81, 82, 83, 87, 115, 116, 124, 126, 129, Conductive heat transfer 63
137, 138 Consumption of steel 140
recirculation 40 Contifur 145
Characteristics of iron ore 91 Continuous casting 109
Charcoal 4, 7 charging 17, 18, 105, 107
Charge movement 44, 45, 46, 48 cast 107
Chemical tracers 53, 54, 55, 56 operating reactor 61
Chhattisgarh 5, 14, 15, 124 Convective heat transfer 62, 63
Chromite ore 133, 134 Corex 142, 145
Chromium 133, 134 Crossing 103
Clusters 6 Crossing point 24, 103
Coal 20, 22, 23, 25 35, 39 115, 116, 119, 123, Crossing point temperature 102, 103
142, 143, 146, 147 Crushing strength 129
caking propensity 115 Culation 40
Coal based 9, 11, 13, 14, 15, 44, 103, 105, Cylindrical shell 9
138, 142, 143, 147
DRI 95 D
sponge iron 99, 102, 103, 106, 144, Decar-burised 5
beneficiation 135 Decrepitate 75, 115, 116
char 18, 22, 43 Decrepitation tendency 115
char reactivity 85, 114, 115 Deficit 37
dust injection 147, 148 Degree of reduction 37, 41, 69, 70
gasification 16, 135, 143 Delhi pillar 5
gasifier 16 Dense phase transportation 124
requirement 127, 128 Dilute phase 124
selection 87 Direct reduced iron 1
throw 57, 59 reduction 4, 15
thrower 65 reduction kilns 129
throwing 59, 124 Dolomite 75, 133
throwing/slinging 45 Downscaling 13, 116
thrown or slung 23, 124 DR processes 6, 9, 16, 34
washeries 148 DRC 14
Coarse coal throwing or slinging 119 DRI 1, 15, 37, 40, 41, 106, 108, 138, 142, 146
Coking coal 142, 144 DRI (sponge iron). 4
Cold bonded pellets 55
Combinations 111, 115 E
Combinations of ore and coal 46 Electric furnaces 17, 18
Commissioned 123 smelting furnace 134
Commissioning 11, 119, 123 Electrically heated 83
Composite pellets 11, 55 heated tubular reactor 115
Conductive 62 Endothermic 20, 61
Index // 159
Endothermic heat 134 available 40

Energy balance 32, 42, 61, 117 conservation 130
Entropy 19 deficit 35, 41, 38, 128
Essar steel limited 14 demands 6, 20, 41
Estimation of heat coal available 39 engine 19
Explosives 93 flow 62
flux 63
F input 107
Feasibility 19 insulator 101, 104
report 114, 119, 120 liberated 111
Feed coal 24, 43, 57, 87, 124 losses 105
Filling degree 25, 28, 45, 46, 126 radiation 53, 91, 92
Final stages 96 recovery 142
of reduction 56, 124 required 40, 81
Fine iron ore 125 supply 23
Finmet 97 transfer 8, 18, 25, 28 105, 126, 127, 148
Fior 97 transfer calculation 127
First order reaction 68, 73 transfer co-efficients 62
Fluidising velocity 25 transfer coefficients 63
Flux 35, 41, 109, 90, 123, 126, 127, 135 transfer model 63, 125
Free energy 19, 67, 79 transfer terms 62
Friable iron ores 116 transferred 61
Furnace 5, 144 treatment 2
Fusion 14, 45, 75 Heated 5, 62, 35, 87, 115, 126
Heating 37, 85, 102
G of iron ore 125
Gas evolution 18 Heats 5, 16
Gas evolved 24 Hematite 133
Gas sealing 122 Hematite 70, 75, 76, 77, 78
sealing system, preheating system 119 Hojalata ya Lamina 6
Gasification 6, 21, 22, 82, 84, 126, 135 Hot testing, commissioning, guarantee
of coal 45, 135, 143 tests, post 119
rate 82 Hydroxy-reactivity 81
reactions 18, 20, 81 HyL 96, 97, 98, 100, 102
Geothermal heat 79 HyL process 6
Granular iron ore 123, 125
Grinding medium 126
I
GSAL 15 Implementation 32, 114
Imported coals 147
H Imported metallurgical coal 142
Haematite 76 India iron and steel 109
Heat 7, 19, 21, 35, 37, 39, 41, 53, 61, 125, 126 Initial deformation temperature 87, 89
Injection coal 23 Lost heat 61

Installation 15, 107, 129, 133, 145 Luppen 6, 96
cold testing 119 Lurgi 14, 85, 98, 100, 122, 123, 135
commissioning and project cost 119
Insulating 62, 126, 129, 130, 131 M
castable 131 Maghemite 76
Iron 5, 34 Magnetising roasting 133, 135
making blast furnace 16, 75 Magnetite 70, 76, 78, 133
ore 119 Manganese as tracer 55
ore fines 54, 75, 145 Manganese ore 55, 134
ore fines as tracer 56 Martite 76
ore lumps 54, 75, 76, 133 Material 42
ore pellets 6, 33, 54, 75 and energy balance 61
ore reducibility 43, 77 balance 12, 32, 61, 117
ore reduction 6 flow 51
oxide 17, 24, 33, 108, 111, 133, 144, 145 flow diagram 117
pillar 4, 5 Mathematical 31, 34
pillar of Delhi 5 model 31, 43, 44, 116
shaft furnaces 75 Mathematical modelling 44
Ispat Industries Limited 14 Mechanisms 45, 71
of reaction 70
J of reduction 7, 73
Jharkhand 5, 15, 97 Metallurgical coal 142
Jindals 14, 124, 135, 145 reactors 53, 75, 87, 143, 148
JSPL 15 Midland Ross company 6
Midrex 15, 96, 97, 98, 100
K process 6
Kinetics of reduction 43 Molecularity of reaction 69
Krupp Renn process 6, 14 Movement of charge 47, 51, 89
Krupp-Codir 14, 96 Muffle furnace 7, 102
L N
Laboratory test 77 Natural beneficiation 76
Lamellar 25 Nickeli-ferrous overburden 134
Last stages of reduction 57 Non-coking coals 16, 113, 142, 148
Layout 117 Nova iron 15
Lean iron ores 123, 133
Lignite 79, 85 O
Lime 8 Of the raw materials 114
Liquidus 92 Ore and coal combination 43
Lloyds metals and engineers 15 reducibility 115
Index // 161
coal 11 Properties 114

coal composite 11 Proximate analysis 115
Orissa 15, 134, 135, 145 Pulverised coal 124
sponge iron 123 burners 146, 148
sponge iron, Tata sponge iron 15 Pyro-metallurgical beneficiation 8
OSIL 14, 60, 123, 124
Overbed air 122 Q
injection 122, 123 Quick response thermocouple 63, 92
Oxidation 96
Oxides of chromium 134 R
Oxy-reactivity 81 Radiation 28, 41, 51, 62, 97
Radiative 63
P heat transfer 62
Pelletisation 76, 116 transfer 62
Pelletisation of iron ore and induration 125 Radioactive tracers 53
Physical properties 54 Rate of gasification 22
tracer 54 reactions 68, 71
Pilot plant 10 reductions 7, 53, 57, 73, 77
testing 116 Reaction kinetics 18, 67, 103
Pore formation 7 rates 19, 43, 67, 69, 71
Ported kiln 12, 14, 122, 123, 133, 135 chars 22, 57, 85, 87, 124
Post commissioning 119 Reactivity of coal char 81
stabilisation 119 Reactor size 117
Pre-feasibility 119 volume 117
Pre-heat air 44 Reducibility 77, 78, 114, 119
Preheated 17, 41, 102 of iron ore 18, 78
air 126 toward carbon monoxide 77
Pre-heating 11, 25, 40, 46, 125, 126, 127, 13 Reduction 96, 126
charge 127 front 45
kiln 43, 44, 126, 127, 128, 129, 138 kiln 12, 44, 54, 126, 128
char recirculation 139 Reduction of FEO 103
of charge 147 Reduction of Fe2O3 6, 20, 73, 77, 127
of iron ore 127 of granular 45
zone 127 of iron ore 127
Pre-reduced 134 of iron oxide 5, 35, 70, 102, 146
Pre-reduction 134 process 6, 20, 53, 57, 61, 67, 122, 135
Process 145 reactions 16, 21, 37, 81
design 113, 116, 117, 119 zone 21, 23, 24, 25, 28, 57, 63, 55, 122
engineering 113, 114, 118 Refractory 123, 126, 127, 129
Production of steel 9 properties 63
Project engineering 114, 117 Re-oxidation 18, 24, 95, 96, 104
Requirement 61 a preheating kiln 127

Requisite heat 79 crossing point 102
Residence 45 Stainless steel anchors 129
time 18, 42, 44, 113, 114 Steel 15, 107, 108, 109, 137, 140
Reynold’s number 25 consumption 140, 141, 143
Ring formation 6, 89, 115, 123 extraction 137
Roasting 123 industries 138
Role of coal 34, 35, 79, 142, 145 industry 138, 141
Romelt 142, 145 makers 2
Rotary process 145 making 6, 106, 107, 110, 111
furnace 7 plants 108, 110, 111, 147, 148
reactor 5, 11 produced 2
steel 9 production 141
tube 43, 114, 115 requirement 138
Rotary tube (Salvis) furnace 114, 116 scrap 18, 105, 107
Rotating reactor 17 making 109
Stelco (Steel Co. of Canada),
S Lurgi, Republic Steel 122
Salvis test 114 Steps process design and engineering 118
Sankey diagram 12, 61 Sulphur as chemical tracer 54
Scrap 107, 109 Sulphur tracer 56
Selection of coal 85 Sunflag iron 15
Selection of iron ore 18 Sunflag iron and steel 15
Sensible heat 41, 124, 134 Superheated steam 133
Shaft Swelling coals 87
furnace 9, 17, 18, 125, 135, 143
reactors 16 T
Shell mounted air fans 63, 123 Tata steel 108, 125, 134, 144
SIIL 14 TDR 14
Sintering 8, 90, 91, 103 Technical specification 119
Size reduction 133 Temperature 102
Sizing of key equipment 117 of reaction 17, 67
Slinging coal 9 of reduction 17, 123
Slung 23 Testing of the mathematical expression 50
Smelting reduction 144 Thermal conductivity 129
Smelting reduction processes 142, 145, 146 Thermochemistry 21
Solvent extraction of coal 135 Thermodynamics 18, 19
Specific heat 41, 128 Throw of coal 59
Sponge iron 1, 2, 4, 5, 6, 7, 9, 127 Throwing 9
in steel making 105 Throwing coal 56, 57, 124
kiln 46 Throwing or slinging 57
making 96 Thrown 23, 79
Index // 163
Thrown coal 23, 43, 85, 87 Unimolecular reaction 68, 70

Tisco direct reduction 125 Unique properties 134
Titaniferrous iron ore 134 United States steel 19
Titaniferrous ore 134 Upscaling 116
Titanium 134
Tracer 56 V
Tracer impulse 56 Valve-striker arm 123
Transfer of heat 61, 127 Valve-striker system 122
Trial runs 119, 125 Viability 57, 119, 125, 143, 146
Tubular reactor 115 Vikram Ispat limited 14
Turbulent 25 Volume reduction 73, 77, 126
U W
U.S. Steel Corporation, steel company of Waste
Canada 108 energy 125
Ultimate analysis 35 gas 42, 44, 61, 123, 124, 126, 147
Underbed hydrocarbon injection system— gases 9, 17, 35, 124, 125, 127, 128, 129
the ported kiln concept 122 Waste heat 137
Underbed oil injection system 122, 123 Waste heat recovery system 124
Underbed steam injection 123 Waste oxides 8

Prasad, K. K. - Ray, Hem Shankar - Advances in Rotary Kiln Sponge Iron Plant-New Age International (P) LTD (2009)

Uploaded by

Copyright:

Available Formats

Prasad, K. K. - Ray, Hem Shankar - Advances in Rotary Kiln Sponge Iron Plant-New Age International (P) LTD (2009)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Prasad, K. K. - Ray, Hem Shankar - Advances in Rotary Kiln Sponge Iron Plant-New Age International (P) LTD (2009)

Uploaded by

Copyright:

Available Formats

Advances in

All rights reserved.

ISBN (13) : 978-81-224-2892-6

PUBLISHING FOR ONE WORLD

Fe(M), Fe(m) = Metallic Iron

NOx = Oxides of nitrogen

Fig. 2.1 Key steps in sponge iron making in rotary kiln 10

Fig. 3.1 Ellingham diagram for a few key oxides 20

Fig. 5.1 Schematic representation of the process of mathematical modelling 43

Fig. 6.1 Material movement in a rotary kiln 46

Fig. 7.1 Modes of heat transfer in a rotary kiln 62

Fig. 9.1 A view of the Gorumahisani iron ore mines 77

Fig. 9.4 Determination of Damköhler number 83

Fig. 10.1 Mechanism of initiation and growth of accretion 90

Fig. 15.1 Steel consumption in present day economies 141

Table 3.1 Thermodynamic data of important reactions 21

Table 6.1 Formulations for residence time in a rotary vessel 50

Table 13.1 Steps in process design & engineering 118

Table 15.1 Carbon dioxide emissions in steel making 138

Preface to the Second Edition v

6.3 Charge Movement: Cascading Flow of Granular Material 46

11.3 First Reported Incidence 96

13.4.4 Waste Heat Recovery System— Generation of Electrical Power 124

Fig. 1.2 Accidental smelting of iron in an anthill

Fig. 1.3 Some primitive iron making furnaces of India

Fig. 1.4 Primitive iron making practice of Orissa

hunting weapons than anything they had

Fig. 1.8 A possible mechanism of reduction and pore

2.1 HISTORICAL BACKGROUND process concepts have been emerging in this

A typical process scheme for making

2.3 SPONGE IRON PILOT PLANT OF

Air Water spray

Fine coal tubes Coal

Fig. 2.5 Material balance in a rotary kiln sponge iron plant

Unit GCal = 109 calories 0.6, Radiative &

Share of India's production

20.00 – 15.00 10.70

Air (4N2+O2) introduced

Fig. 2.9 Delicate balance of oxidising and reducing conditions in a

unit volume, because of this capability 2.6.3 Weaknesses of the Process

4/3 A1 + O2 = 2/3 A12O2

Fig. 3.1 Ellingham diagram for a few key oxides

SL. Class of Reaction ∆ GoT ∆ G o298 ∆ G o 1273 ∆Ho298 ∆ H o 1273 Remarks

2FeO + CO2 ccomplished at the start of

3. C+½ O2 = CO – 26,700–20.95 T – 32,943 – 53,369 – 26,420 – 27,183

Thermodynamic Considerations: Feasibility of Reaction // 21

5. C+½ O2 = CO2 – 67,500 +20.75 T – 61,316 – 41,085 – 67,630 – 67,378

Mathematical models are being increasingly approach it with a sense of scepticism,

Sl. No. Element/Item Iron ore Kg of element/ Kg. of element /item % of

Total without sulphur

Table 5.2 Filled up chart of Table 5.1

Advances in Rotary Kiln Sponge Iron Plant

Mathematical Modelling in Rotary Kiln Sponge Iron Making // 37

Advances in Rotary Kiln Sponge Iron Plant

(1) Fe2O3+3CO = 2Fe+3CO2 ∆ Ho298 = –11,020 kCal/kg-mole

(Basis: 100 kg coal)

Mathematical Modelling in Rotary Kiln Sponge Iron Making // 39

Reactions considered: CO + 1/2 O2 = CO2 ∆ fPWV = –67,630 kCal/kg-mole

Advances in Rotary Kiln Sponge Iron Plant