U.S. Environment.il Protection Agency Industrial Environnn-nl.il Rcsciirch El P A-600/7-76~0 2 2
Offi<:e of Rfjsp.m.n .UK) Devolopmcnt Laboratory •
Rf-search Trianqle Park. North Cdrolm.i 27711 October 1976
SURVEY OF
EMISSIONS CONTROL AND
COMBUSTION EQUIPMENT
DATA IN
INDUSTRIAL PROCESS HEATING
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies *
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Government, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use. i
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
EPA-600/7-76-022
October 1976
SURVEY OF EMISSIONS CONTROL
AND COMBUSTION EQUIPMENT DATA
IN INDUSTRIAL PROCESS HEATING
by
Peter A. Ketels, John D. Nesbitt, and R. Don Oberle
Institute of Gas Technology
ITT Center
Chicago, niinois 60616
Contract No. 68-02-1821
Program Element No. EHE624
EPA Project Officer: John A. Wasser
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
TABLE OF CONTENTS
Page
SUMMARY AND CONCLUSIONS 1
INTRODUCTION 3
PHASE I. INDUSTRY SELECTION 4
Restraints 4
PHASE II. DATA COLLECTION 8
Iron and Steel Industry — Processes and Equipment— Current
State-of-the-Art 8
Coke Ovens 8
Blast Furnace Plant 9
Sinter Plant 10
Blast Stoves IQ
Blast Furnace n
Pollutants From Iron-Ma king 12
Steelmaking 13
Open-Hearth Steelmaking 13
Basic Oxygen Furnace (EOF) Steelmaking 14
Electric Furnace Steelmaking 15
Direct Reduction 16
Ingot and Continuous Casting 16
Trends in Process Modifications 21
Iron Making 21
. Coke Ovens— Current and Near Term 21
Coke Ovens — Long-Term 21
Sinter Plant— Current and Near-Term 22
*f f ~*
Sinter Plant -3- Long~T^rm 23
•*"
Blast Furnace — Current and Near-Term 23
Blast Furnace — Long-Term 24
Steelmaking 24
Open Hearth— Current and Near-Term 24
Open Hearth— Long-Term 24
Basic Oxygen Furnace (BOF) — Current and Near-Term 24
Basic Oxygen Furnace — Long-Term 25
ui
-------
TABLE OF CONTENTS, Cont.
Electric Furnace — Current and Near-Term 25
Electric Furnace — Long-Term 25
Trends in Energy Utilization 26
Ironmaking — Current and Near-Term 28
Ironmaking — Long-Term 28
Steelmaking — Current and Near-Term 28
Steelmaking — Long-Term 28
Soaking Pits and Reheat Furnaces — Current and Near-Term 29
Soaking Pits and Reheat Furnaces — Long-Term 29
Heat-Treating and Finishing Operations — Current and
Near-Term 29
Trends in Emissions 29
• Data Analysis 30
Pelletizing 31
Sintering 32
Coke-Oven Underfiring 32
Blast-Stove Heating 33
Open-Hearth Furnace 33
Soaking Pits and Reheat Furnaces 34
Heat-Treat and Finishing Operations 35
Recommendations 35
Recommendations for General Emission Control and Energy
Conservation According to the Industry 35
Recommendations Based on Evaluation of Data Collected
During Program 36
Glass Industry— Current State-of-the-Art 37
Melting 37
Finishing 41
Fabrication 41
Annealing 42
Inspection 42
Energy Utilization in Glass-Manufacturing Processes 44
Air Pollutant Emissions in the Glass Industry 47
Factors Affecting Air Pollutant Emissions 47
IV
-------
TABLE OF CONTENTS, Cont.
Current Methods of Air Pollution Control 49
Trends in Process Modifications in the Glass Industry 50
Expansion of Process Monitoring and Control Capacity 50
Electric Melting (and Boosting) 51
Oxygen Enrichment 51
Raw-Batch Pretreatment 51
Submerged Combustion 52
Augmentation of Heat Transfer From Flames 52
Use of Low-Grade Thermal Energy 52
Improvements in Equipment Design 53
Trends in Energy Utilization in the Glass Industry 53
Trends in Air Pollutant Emissions From Glass-Manufacturing
Processes 55
Analysis 55
Recommendations 58
References Cited 59
Cement Industry 61
Summary 61
Portland Cement Manufacturing Processes 61
Raw Material Preparation 64
Raw Material Burning 64
Clinker Processing 66
Cement Industry Energy Requirements 66
Air Pollutant Emissions From Cement-Manufacturing
Processes 69
Trends in Industrial Process Modifications as Estimated by
Industry 72
Chain Systems 74
Enlargement of Kiln Feed End 73
Oxygen Enrichment 74
Waste-Heat Recovery 75
Preheater Installation 76
Conversion to Dry Process 73
Vertical Kilns . 73
Trends in Energy Utilization 79
Trends in Emissions v 33
-------
TABLE OF CONTENTS, Cont.
Page
Energy Cost of Emission Compliance 84
Analysis 85
Recommendations 87
Reference Cited 88
Aluminum Industry — State-of-the-Art 88
Primary Aluminum. Production 89
Mining 89
Refining 89
Smelting 9 ^
Secondary Aluminum Processes 93
Melting and Reheating 93
Melting-Furnace Design 94
Energy Requirements 95
Air Pollutant Emissions 96
Trends in Aluminum Process Modification 97
Flash Calcining - 99
Stack Charging 99
Hot-Metal Pumping 100
Recuperation f 100
Oxygen-Fuel Melting 101
Improved Air/Gas Ratio Controls 101
Infrared Heating — Crucible Melting 102
Trends in Energy Utilization 102
Energy Use Based on Availability 103
Trends in Emissions 104
Analysis of Data 104
Recommendations 105
Petroleum Refining Fired Heaters — State-of-the-Art 106
Catalytic Cracking 107
Catalytic Reforming HO
Delayed Coker HO
Hydrotreating • \\\
Energy Requirements H2
vi
-------
TABLE OF CONTENTS, Cont.
Page
Emissions 113
SOX 114
NOX 115
Trends in Process Modification 115
Trends in Energy Utilization 117
Trends in Emissions 118
Analysis 118
Recommendations HQ
APPENDIX A. Weighting System 12JL
Vll
-------
.LIST OF FIGURES
Figure No. Page
1 Flow Diagram for Soda-Lime Glass Manufacture 38
2 Plan View of Side-Port, Flat Glass Furnace 40
3 Longitudinal Section of Typical End-Port Furnace ' 40
4 Typical Annealing Cycle for a Controlled Recircu-
lation Lehr ' 43
5 Annual Cement Production With Projection to 1985 62
6 Fuel Usage of U.S. Cement Industry . 63
7 Steps in the Manufacture of Portland Cement 65
8 Thermal Efficiency For Chain Systems in Long
Wet and Dry Kilns 74
9 Air-Suspension Preheater Kiln - / 77
10 Short-Term Energy-Saving Concepts in the U. S.
Cement Industry 80
11 Schematic Diagram of Primary Aluminum Pro-
duction Process 90
12 Aluminum Melter With Sloping Hearth 100.
13 Basic Types of Pipe still Heaters 108
14 Schematic Diagram of Orthoflow Catalytic
Converter and Adjuncts 109
15 Delayed Coking Process 111
16 Types of Fuel Consumed by Petroleum Refineries 113
viii
-------
LIST OF TABLES
Table No. Page
1 Profile and Weighting Factors for Industries
Considered for Study in This Program 7
2 Coke-Plant Material Balance 9
3 Estimated NC* Emissions From Steel-Mill
Processes ana Equipment 32
4 Breakdown of Energy Consumption by the Glass
Industry 44
5 Air Pollutant Emissions From Various Production
Glass Melters 48
6 Recommended Programs for Reducing Emissions
and Energy Consumption in the Glass Industry 60
7 U. S. Cement Industry Kiln Fuel Usage (1973) 67
8 Energy Consumption in Cement Processing 67
9 Emission Factors for Cement Manufacturing
Without Controls 70
10 Estimated Total Emissions From Cement Plants
in 1972 71
11 Fuel Savings for Kilns With Chain Systems
Compared to Kilns Without Chain Systems 73
12 Energy Consumption of Various Cement-
Manufacturing Processes 76
13 Summary of Available Information on Emissions
From Aluminum-Manufacturing Processes 98
14 Summary of Available Process Modifications and
Their Potential for Reducing Fuel Consumption 99
15 Potential Sources of Specific Emissions From
Oil Refineries 114
A-l Weighting System Factors 122
ix
-------
SUMMARY AND CONCLUSIONS
The subject of the investigations in this program was the interaction be-
tween current and potential energy conservation measures and emission pro-
grams in a number of select industries. By interviewing representatives
of these industries, information in both areas was obtained. Where the goals
of energy conservation are in conflict with emission control goals, as assess-
ment of the problems involved was obtained. Subsequently, this information
•was evaluated at the process level for each industry investigated and an inde-
pendent assessment of the problem areas was made. Based on the results of
this assessment, R 4 D programs were recommended for each industry to
solve the problems involved. The scope of the program was limited to pro-
cesses whereby heat was obtained through the direct combustion of fossil fuels
(as opposed, for example, to electricity), and it was limited to emissions that
are affected by combustion itself.
The persons visited appeared to be well acquainted with both emission
problems and control technology associated with particulate emissions and
stack plume opacity, as well as stack gas acidity, carbon monoxide, and
unburned hydrocarbons. However, there did not appear to be much known
about the level of the NO emissions from industrial heating processes.
.X.
During these visits, several areas of concern were expressed by the per-
sons contacted, all of which related to the apparent conflict between energy
conservation goals and emission control goals. First, an almost universal
concern was that the ability to meet voluntary energy conservation goals is
substantially decreased by the energy consumption of emission control de-
vices required by more stringent future emission standards. A related con-
cern was that application of energy conservation measures would lead to in-
creased air pollutant emissions, particularly where NO emissions are con-
.At
cerned. No specific examples involving NO emissions were offered be-
cause there are no standards against which data might be compared, although
several areas where the goals might be in conflict were mentioned; for ex-
ample, the preheating of combustion air using heat will increase NO emis-
j£
sions. However, there is some evidence to indicate that, at least in the
case of NO emissions, emissions will not increase in situations traditionally
J*.
thought to result in increases. For example, the use of oxygen enrichment
-------
for melting has been shown not to result in increased NO emissions, al-
3C
though current information concerning NO formation would suggest that it
ji
should. Consequently, one general recommendation is that programs be
undertaken in each industry (where applicable) to demonstrate the conformity
of available energy-conserving process modifications with the goals of re-
duced air pollutant emission controls. As the data indicate, most of these
problems are concerned with NO emissions, although there are select in-
stances within each industry involving other emissions. A second recom-
mendation is that programs be undertaken to develop more efficient, lower
energy-consuming pollution control devices.
Another area of common concern among the industries investigated is the
trend away from clean-burning gas to liquid and solid fuels. This trend will
result in increased emissions from combustion, which in turn will require
additional control equipment. The additional control equipment, as already
noted, will require additional energy, and it is claimed that the use of these
fuels results in a loss of process efficiency. Although there is some skepti-
cism as to the latter point, it nevertheless deserves investigation. There
are some processes in some of the industries that are capable of using liquid
and solid fuels at increased efficiency. Programs should thus be undertaken
to clearly demonstrate the efficiency of such processes for both energy conger
vation and emission control. Where there are real problems involving effi-
ciency of liquid and solid fuels, programs should be undertaken to resolve
them, even to the point of developing hardware, the use of which would have
widespread impact — for example, the development of a low-NO , oil- or
JL
coal-fired burner that meets the requirements of both air pollutant control
and efficient fuel utilization. Beyond these general recommendations, speci-
fic recommendations for each industry are presented in the respective sec-
tions of this report. It is clear from the data gathered that there are many
processes in industry where energy consumption can be reduced without de-
trimental environmental impact, and that the demonstration of these pro-
cesses will result in an increase in their implementation by industry.
-------
INTRODUCTION
The purpose of this study was to identify the significant emission sources
within the industrial process-combustion field, to investigate the potential
•
for effective emissions control for industrial process combustion, and to
compile information on combustion equipment in use and future trends in
process and equipment design. This program investigated the interactions
between current and potential energy conservation measures and emission
control programs, assessed the potential for RAD work to advance emissions
control, and recommended R&D programs to solve the problems involved.
The program was divided into three distinct phases for the purpose of
industry selection, data collection, and data analysis.
The first phase of this program (Phase I) was to determine those industries
and industrial processes which have a high potential for conserving energy
and reducing emissions. Their selection was based on the relative amounts
of energy consumed and the relative emissions of pollutants. The selection
process included consideration of the potential for simultaneous energy con-
servation and air pollutant emission controls.
The second phase of this program (Phase II) consisted of an intensive
series of visits to industrial plants, corporate headquarters, trade associa-
tions, and builders of process heating equipment. The purposes of these
visits were to validate and fill the voids in the information collected in Phase I
and to assess a number of elements relating to existing and future trends in
process technology and equipment design affecting energy consumption and
air pollutant emissions.
The information obtained from the trade associations, equipment builders,
and industrial plants was correlated and analyzed in Phase III to provide an
assessment of process and equipment design and operating factors that are
major influences on both fuel efficiency and air pollutant emission rates.
This data analysis resulted in estimates of the potential for improvement in
fuel-use efficiency and reduction of emissions for existing equipment designs
and for new processes and equipment designs that may be instituted in the
future. Upon completion of the data analysis, recommendations for specific
programs aimed at reducing energy consumption and air pollution emissions
were made, including the identification of potential sponsors for these programs.
-------
PHASE I. INDUSTRY SELECTION
i
Hi a program of this type, it is generally not feasible to consider all
industries. Consequently, a system was developed for selecting pertinent
industries. The system related energy consumption, pollution emissions,
and the potential for reducing both for each industry, and then determined
the relative merits of studying each industry in terms of achieving the pro-
gram objectives. Toward this end, the following system of restraints and
weighting factors had been developed for selecting the industries to be studied
in this program.
Restraints '
The first step in the development of this system was to define the restraints
as determined by the problems to be solved. The following restraints were
selected:
1. Total energy use in process heating. This program is primarily concerned
with those industries that consume large amounts of energy because it is
in these industries that the most significant results of implementation of
energy conservation technology will be realized. Some industries, however,
consume large amounts of fossil fuels for feedstock and steam generation —
ostensibly, the paper industry, the chemical industry, the petroleum refining
industry, and the food-processing industry. In the areas of feedstock and
steam generation, little can be done to alter the energy utilization patterns.
Consequently, only those industries that consume large amounts of energy
for process heat are being considered in this program.
2. Combustion-related uses of energy. A further refinement of the first re-
straint is to limit the investigation to combustion-related aspects of energy
consumption. In so doing, industries -which consume large amounts of
purchased'electricity for process heating will be eliminated from consider-
ation. Electricity consumption is not within the scope of this program,
although in-plant generation may be considered if the relative load is large
enough.
3. Relative number of processes within an industry. In some industries, energy
is consumed by a large number of relatively small-scale processes, which,
when added together, show a very large total energy consumption figure for
the industry as a whole. Typical of such an industry is the industrial chemi-
cal industry. The total amount of energy consumed for process heating in
the chemical industry in 1972 was about 2800 trillion Btu. However, mb£t
of this energy was consumed in the manufacture of thousands of chemicals
which incorporate myriads of individual processes with varying energy re-
quirements. To study an industry such as this would give minimal return
toward the project objectives in that reductions in energy consumption of a
process in this industry would have a minimal effect on the national energy
picture. On the other hand, in an industry with a limited number of pro-
cesses, such as the cement industry, in which more than 90% of the
-------
581 trillion Btu of energy consumed in 1972 was consumed in kilns, even
a small (5% to 10%) increase in fuel utilization efficiency of the kiln would
have a significant, widespread impact on the entire industry as well as
the national energy picture.
4. Known and/or estimated total emissions. Two of the stated purposes of
this program are to identify the significant sources of emissions within
the industrial process combustion field and to determine the potential for
effective emissions control for industrial process combustion. There-
fore, the emissions of an industry must be a restraint in the industry-
selection process. Emissions from industrial process combustion can
be broken down into three categories:
a. Emissions directly related to combustion — typically NOX, SOX, CO,
and particulates.
b. Emissions indirectly related and therefore somewhat controllable by
combustion techniques —typically such emissions as particulates from
a glass melter or cement kiln where the combustion gases "pick up"
the dust of the raw material and carry it out the stack.
c. Emissions independent of the combustion process —typically emissions
from chemical reactors or electrolytic processes.
5. Types of emissions — NO , SO , CO, and particulates. These are the
emissions of primary concern in this program as they are the emissions
most closely related to the combustion process. Secondary emissions
of concern are fluorides, chlorides, and other halides, although these
are only indirectly related to the combustion process.
6. Potential for energy conservation. Energy conservation is one of the
primary purposes of this program. Consequently, only those industries
which have a high potential for reducing energy consumption should be
considered. The paper and paperboard industries, although consumers of
large amounts of energy (more than 1300 trillion Btu in 1972), will be
eliminated from consideration because 95% of this energy is consumed as
boiler fuel, an application which already has a direct fuel utilization ef-
ficiency greater than 80%. The potential for increasing the efficiency of
direct utilization of fuel is low.
7. Potential for reducing emissions. Reducing combustion-related air pol-
lutant emissions is the other primary objective of this program. Conse-
quently, as in the case of energy consumption, those industries with a
low potential for reducing emissions will not be considered in this program.
Upon applying the restraints and a weighting system (Appendix A)}
the following industries were selected for study in this program:
• Iron and steel (all energy-using processes)
• Cement (primarily kiln operation)
• Glass (melting, forehearth operation, annealing)
-------
• Aluminum (primary, reheating, secondary production)
• Petroleum refining.
These industries ranked highest in terms of the evaluation characteristics
used in this program. That is, all of these industries use large amounts of
process heat directly from the combustion of fossil fuels; all of these in-
dustries are relatively large emitters of combustion-related air pollutants;
and all of these industries have a high potential for reducing energy con-
sumption and air pollutant emissions. Note that these industries have been
selected based on the combination of all of the above factors. Consequently,
there are industries, such as the paper and paperboard manufacturing in-
dustry, that consume twice as much energy as the cement industry and 4
times as much as the glass industry, but, because 95% of that energy is
consumed in boilers to generate steam (a process efficiency in excess of
80%), the potential for reducing energy consumed directly by the boilers is
very low. Thus, this industry received a low priority rating and is not being
considered in this program. Information related to the industries considered
for this study is shown in Table 1.
-------
Table 1. PROFILE AND WEIGHTING FACTORS FOR INDUSTRIES
CONSIDERED FOR STUDY IN THIS PROGRAM
Cement
Food
Glass
Plastics
Aluminum
lotai energy
Consumed,
Industry 1012 Btu/yr
d Steel
Lai Chemicals
Ash
um Refining
and Paperboard
t
3 and Synthetics
im
3100
3764
161
2861
1307
581
911
250
755
352
% Consumed for
Process Heat
80
55
75
60
5
90
5
90
10
80
Base Energy
Number
22
14
20
12
14
24
11
22
11
15
Emission Potential
Index No. Energy Consumed
10
4
4
5
7
8
5
7
2
7
10
5
6
3
2
10
2
8
2
5
Potential
Emissions Recorded
10
5
6
3
5
7
3
7
2
3
B-94-1766
-------
PHASE II. DATA COLLECTION
Iron and Steel Industry— Processes and Equipment — Current State-of-the-Art
The major operations in an integrated steel plant are the coke oven, blast
furnace, basic steelmaking, rolling mill, and finishing.
Coke Ovens
Coke is the product of destructive distillation of some types of bituminous
coal, commonly referred to as metallurgical coal. It is produced in by-product
coke ovens, sometimes called slot ovens, because of their shape. Each oven
is a narrow, rectangular retort closed at the top except for charging ports,
with gas off-take ports in the sides and bottom. Each end of the oven is fitted
with a tightly sealed removable door. One end is the "pusher end" from which
a ram is inserted to push out the coke; the other end is called the "coking end"
from which the hot coke is pushed into a quenching car. Adjacent retorts are
separated from each other by heating flues and are called underfired ovens
because the air and fuel ports are located at the bottom of the heating flues.
A large number of the retorts and associated flues are stacked together, and
the entire series is called a battery.
High-, medium-, and low-volatile coals are crushed, ground, and blended
to obtain the desired coking characteristics. About 65% to 80% of the charge
is high-volatile coal containing 30% to 40% volatile matter. After charging
the retort, the coal is converted into coke in approximately 16 to 18 hours
in older ovens and in about 12 hours in modern, taller ovens equipped with
improved combustion and control systems.
By-products of the coking operation include coke-oven gas, tar, ammonium
sulphate, and light oil products (BTX). About 60% of the sulfur content of the
coal is carried over into the coke. The balance goes to the coke-oven gas as
hydrogen sulfide. A representative material balance for a coke-oven plant is
shown in Table 2.
The fuels used for heating, the coke-ovens are selected on the basis of cost,
availability, and combustion characteristics. Top to bottom temperature
gradients must be minimized. Coke-oven gas and a natural gas-air mixture
provide the desired high flame temperature. Blast-furnace gas has a much
lower heating value and a lower flame temperature, but a high concentration
of inerts that act as diluents to retard combustion and lengthen the flame.
8
-------
Table 2. COKE-PLANT MATERIAL BALANC.
Coke
Coke-Oven Gas
Tar
Ammonium Sulfate
Light Oil Derivatives
Total
Quantity
78.5%
10, 600 cu ft
(540 Btu)
1. 55 gal
21.0 Ib
2.85 gal
Pounds/Ton
of Dry Coal
1574. 0
312. 3
74.8
21.0
20.9
2000. 0
The primary fuel for underfiring coke ovens has been coke-oven gas. About
one-third of the coke-oven gas produced is used for this purpose. Blast-
furnace gas and natural gas together account for 15% to 20% of the total in-
put for underfiring coke-ovens. (Total fuel input for underfiring has been
about 3. 0 million Btu per ton of coke produced.)
The major technological trends affecting coke-oven energy requirements
are
Higher and wider ovens using high conductivity refractories. Coking rates
and productivity have increased; however, increases anticipated in efficiency
have been offset by higher coking temperatures and greater heat losses. •
Thermal efficiencies of coke ovens are reported in the range of 60% to 70%.
Predrying of Coal Charge. Predrying of coke increases coke-oven produc-
tivity and reduces heat losses resulting in a saving of about 100, 000 Btu
per ton of coke. Emissions during charging of predried coal are also
reduced.
Use of Formeoke. Formcoke is a calcined agglomerate of bituminous coals
(noncaking). Pollution in formcoke processes is also reduced.
Improved Blast Furnace Operations. The quantity of coke required per ton
of pig iron produced is reduced by using a higher top pressure, hotter blast
air temperature, oxygen enrichment of blast air, and increased use of
fuel injection.
Direct Reduction of Agglomerated Iron Ore. Direct reduction and electric
furnace melting produce steel with no pig-iron requirement and prereduction
of part of the blast furnace burden will reduce the coke requirements and
increase blast furnace output.
Blast Furnace Plant
The blast furnace and its associated stoves for preheating blast air, the
9
-------
sinter plant, and storage, sizing and handling facilities for raw materials
are the major components of the blast furnace area in an integrated steel
plant.
Sinter Plant
Some of the iron ore and flue dusts are available in particle sizes less
than 1/4 inch and cannot be charged directly to the blast furnace. These
products are mixed with flux and coke breeze and loaded onto a traveling
grate-sintering machine. An auxiliary fuel such as natural gas, coke-oven
gas, or oil is used to initiate combustion on the surface of the mixture and
is referred to as ignition fuel. Combustion is continued over the length of
travel by forcing air through the mixture on the grates. The mixture is
heated to a fusion temperature, which causes agglomeration of the iron-
bearing particles. The sinter discharged is cooled, crushed, and screened
prior to transfer to the blast furnace charging oven.
The major source of energy used in the production of sinter is the carbon
content of coke breeze and flue dust. The amount of ignition fuel required
is about 0. 12 million Btu per ton of sinter. The total fuel requirement,
including coke breeze, is about 2. 0 million Btu per ton of sinter.
Blast Stoves
Blast stoves are checker-brick-lined chambers used to supply heated air
to the blast furnace. The stoves are regenerative heat exchangers, during
part of the cycle being heated by combustion of blast-furnace top gas and
during another part of the cycle heating the blast air. Three or four stoves
are required for each blast furnace. In operation, a stove heats air until
the air preheat temperature drops to a preset value. The stove is then iso-
lated from the hot blast main and reheated by combustion of blast-furnace gas
to the temperature required to heat blast air. Regenerative air preheating
results in a cyclic variation in blast-air temperature, and the stoves may be
used as staged temperature preheaters in various combinations with the
objective of reducing the cyclic variations in blast temperature.
The thermal efficiency of the blast stove is quite high— about 72% — as
would be expected from a regenerative heat exchanger. However, efficiency
has decreased from about 87% in I960 'because of a continuous increase in
blast temperature. Average blast temperature in 1969 was in the 1500° to
10
-------
1600°F range and in I960 was below 1300°F. Some installations are now
approaching a 2000°F blast temperature.
The major reasons for increasing blast-air temperature are to reduce
coke consumption in the blast furnace and to increase the output of pig iron.
Auxiliary fuel injection in the blast furnace also requires higher blast-air
temperature to maintain temperature in the melting zone. Other factors
also affect coke rate and production. These will be discussed in the blast-
furnace section of this report.
About 25% of the blast-furnace top gas is used for stove heating. In
1969, this amounted to 165 trillion Btu. Increases in blast-air temperature
and a reduction in blast furnace gas heating value have required the use of
coke-oven gas and natural gas for higher heat release in blast stove heating.
Current technology trends in blast stove design and operation include —
• Improved designs for burners and combustion chambers
• Improved hot valve designs
• Better refractories and checker-brick design
• Automatic stove charging equipment
• Improved modes of stove operation.
Blast Furnace
The blast furnace is a cylindrical, tapered, refractory-lined, vertical
furnace. It is charged with the basic iron-making raw materials, iron ore,
pellets, sinter, coke, and limestone, and discharges hot metal or pig iron
and a slag that contains the mineral gangue and other impurities. It is a
counterflow process in that the iron-making materials are charged at the top
through a double-bell seal and travel down through the furnace reacting with
an ascending stream of reducing gases, primarily carbon monoxide and hy-
drogen. The reducing gases are produced by blowing preheated air through
ports, called tuyeres, into the bottom of the furnace, where reaction with
the incandescent coke takes place. Auxiliary fuels such as fuel oil, natural
gas, tar, or powdered coal are frequently injected at the tuyeres to increase
iron output and decrease the coke rate. With auxiliary fuel injection, higher
blast-air temperature or oxygen enrichment is usually required to maintain
the iron melting temperature. Exceptionally good results have been achieved
11
-------
on a pilot-plant basis by injection of reformed natural gas at the tuyeres;
Theoretical calculations have shown that even better results could have been
obtained by injection of hot reformed gas above the tuyeres in the gaseous
reduction zone. Although these programs have been curtailed by reduced
availability of natural gas, the results indicate that blast-furnace injection
may become a potential application for use of low-Btu gas produced by gasi-
fication of coal, assuming a favorable economic justification. A suitable
reducing gas can also be produced by partial oxidation of residual oil, but
the present cost and availability of low-sulfur oil make this approach un-
economical.
Pollutants From Iron-Ma king
The coke-oven plant has been identified as one of the major sources of air
and water pollution from integrated steel plants. Some of the sources of
emissions are —
• Coal handling, processing, and storage
• Charging the coal into the ovens
• Leakage during coking, particularly from the doors
• Discharge of incandescent coke
• Water quenching of coke
• Contaminated water discharged.
A variety of methods has been incorporated in recent coke-oven plant designs
to effectively control emission of pollutants.
Sinter plants are inefficient as users of energy and are major contributors
to steel mill particulate and gaseous emissions. Emission control is tech-
nically difficult and expensive. Some steel plants have phased-out sinter lines,
and pelletized iron ore has now assumed a dominant position as a major blast"
furnace charge. Pelletizing plants are located near the iron ore sources and
the emissions from the pelletizing process do not contribute to the steel mill
emission control problems.
The major pollutants from sinter plants are particulates, sulfur oxides,
and carbon monoxide. Leakage from the seals between the exhaust ducts
and the sintering furnace structure accounts for most of the emissions.
12
-------
Disposal of in-plant fines and flue dust as process feedstocks will be
required by means other than landfill. Both pelletizing and hot-briquetting
will probably be used.
The emission of pollutants from blast-furnace stoves consists of dust
carried over from the blast furnace. With properly designed combustion
systems and air/gas ratio controls, carbon monoxide should not be considered
a significant emission. Particulate emission from blast-furnace stoves is
not treated separately in the literature, but is considered part of blast-furnace
particulate emissions.
Steelmaking
In steelmaking, iron from the blast furnace, commonly called hot metal,
and scrap are charged along with fluxes to melting process equipment. Im-
purities such as carbon, manganese, silicon, and phosphorous are oxidized
to or below specified levels during the melting process. These oxidation
reactions are exothermic and contribute part of the total process energy
requirement. The fluxes and some of the oxidized impurities form a slag
layer, and the rest of the oxidized components leave as gases. The major
types of melting process equipment are the open-hearth, the basic oxygen
furnace (EOF), and the electric furnace.
Open-Hearth Steelmaking
The open-hearth furnace is a large reverberatory vessel heated from
the top by burners located at each end of a long, narrow chamber. A long,
luminous flame is produced so that a high percentage of heat transfer to
the bath is by radiation. The heating process is regenerative in that refrac-
tory checker-brick are located downstream of each firing port so that com-
bustion air is being preheated at one end, while exhaust gases are reheating
the checker-brick at the other end. After a timed interval the flow is re-
versed. Scrap metal, fluxing agents, iron ore, and hot metal are charged
through doors located above the bath level along the length of the furnace.
Open-hearth furnace capacities range from 100 to about 600 tons per heat.
The high productivity of competitive steelmaking processes has brought
about major improvements in open-hearth furnace productivity. Some of
these improvements are oxygen-enrichment of combustion air, roof-mounted
oxygen lances, roof-mounted oxygen-fuel lances to accelerate scrap melting,
13
-------
and higher combustion-air p-reheat temperature. However, the use of oxygen
lancing has resulted in increased dust loading of the waste gases and this has
required installation of more costly air pollution control equipment.
Fuels used in open-hearth furnaces include residual oil, natural gas, coke-
oven gas, tar, and pitch. Part of the thermal requirement is provided by
exothermic oxidation of impurities contained in the hot metal portion of the
charge. Fuel required per ton of steel produced is in the 1. 5 to 4 million
Btu range, depending on the hot metal-scrap ratio and the amount of oxygen
lancing used.
Basic Oxygen Furnace (BOF) Steelmaking
In the BOF process, oxygen is blown downward through a water-cooled
lance into a bath containing scrap and hot metal. Heat produced by oxidation
of carbon, silicon, manganese}and phosphorous is sufficient to bring the metal
to pouring temperature and auxiliary fuel is not required. The furnace is an
open top, tiltable, refractory-lined vessel shaped somewhat like the old-
fashioned glass milk bottle. Furnace capacities range up to 300 tons and the
time required per cycle is very short — from 45 to 60 minutes.
The BOF has displaced the open-hearth as the major steel production
process, but is much less flexible because of the inherent limitation of 25%
to 30% scrap in the charge. The amount of BOF capacity in an integrated
steel plant is, therefore, closely associated with hot metal availability.
Additional flexibility in scrap use can be obtained by preheating the scrap
with an oxygen-fuel burner. In many steel plants, an open-hearth shop is
modernized and equipped with appropriate pollution control equipment so that
it can be used in conjunction with BOF shops to provide the required flexibility
to accommodate variations in hot metal-scrap ratio. A combination of BOF
shops and electric furnace shops provides the maximum in flexibility and may
represent the makeup of future steelmaking facilities.
Excluding fuel use for scrap preheating, other uses are for refractory
dryout and to keep the BOF vessel from cooling between heats. Their uses
amount to about 200, 000 Btu per ton of steel produced. '
Decarburization of the iron charged to the BOF produces about 400, 000
Btu of carbon monoxide per ton of steel. The off-gases also contain large
amounts of particulates, which must be removed before discharge into the
14
-------
atmosphere.. Typical American practice is to burn the combustible gases
in water-cooled hoods mounted above the BOF vessel, cool with excess air
or water sprays, and pass the cooled gases through high-energy scrubbers
or electrostatic precipitators. In most cases, the BOF vessels are equipped
with open hoods that admit air for combustion of carbon monoxide on a rela-
tively uncontrolled basis. If additional steam can be used in the plant, the
combustion hood can be used as a steam generation device, although the steam
production will only be cyclic.
In Japan, a closed-hood gas collection system known as the OG system is
used, wherein the waste gases containing CO are scrubbed, collected, and
used as fuel gas. The basic OG system is used at a few steel plants in the
United States, but the CO-rich off-gas is burned in a flame stack rather than
stored and used for fuel. A major advantage of the OG system, other than
recovering a fuel gas, is the large reduction in the volume of waste-gas re-
quired to be handled by the scrubbing system.
Electric Furnace Steelmaking
Production of steel in electric-arc furnaces has grown rapidly since
World War II and is currently estimated to be more than 15% of total steel
production. Because of the phase-out of open-hearth steelmaking, the in-
crease in BOF steel production, and the associated scrap-use limitation,
the amount of steel produced in electric-arc furnaces is expected to increase
even more.
The combustion of fossil fuels currently plays a very small role in electric
steelmaking. This may change in the future as advances in technology per-
mit the increased use of scrap preheating. Most authorities agree that scrap
preheating will be accomplished outside the electric-arc furnace in a specially
designed charging bucket, probably equipped for bottom discharge. Many of
the designs use excess-air burners to limit flame temperature and minimize
oxidation of the scrap. Associated air-pollution problems include particu-
lates from dirty scrap, iron oxide, and oil vapors. The requirement for
both incineration at or above 1400°F and particulate removal has caused
shutdown of several scrap preheating installations because of economic con-
siderations.
15
-------
Direct Reduction
A number of direct reduction processes are currently in use to provide
an alternative to the blast furnace in the production of feed material for
steelmaking furnaces. Both batch and continuous processes are in use,
involving such equipment as rotary kiln, counterflow vertical shaft, traveling
grate-kiln, and multistage fluid bed. Various types of raw materials are
processed, such as lump ore, oxide pellets, and beneficiated concentrate.
The degree of metallization ranges up to 95%, and a highly metallized pro-
duct can be charged to an electric melter without scrap addition. Prereduced
charge can be used in the blast furnace and steelmaking equipment such as
the BOF. Some advantages of direct reduction are —
• Lower capital requirements than for coke-oven blast-furnace ironmaking
• Reduced dependence on hot metal for steelmaking
• Consistent chemical analysis of the feed material to the steelmaking
process
• Reduced dependence on scrap.
In the past, direct reduction plants in the United States have been asso-
ciated with mini-steel plants which are not fully integrated. These plants
are dependent upon a reliable supply of natural gas at a low price. Currently,
both availability and price are adverse to the process and it seems doubtful
that additional plants will be constructed. Natural gas consumption in the
most efficient plant is in the range of 13 to 14 million Btu per ton of product;
less efficient plants may require as much as 20 million Btu per ton.
Hot metal from the blast furnace is transferred to the steel-making fur-
nace in large, refractory-lined vessels or ladles. These are also used
for transferring molten steel to the various casting operations. The refrac-
tory linings of the vessels must be cured after replacement and preheated
before each use. Typically, open-flame, burners are used for this at a
very low thermal efficiency. Fuel requirements for ladle drying and heating
range from 200, 000 to 400, 000 Btu per ton of metal.
Ingot and Continuous Casting
After removal from the furnace, the steel is poured into ingot molds or
transferred to the continuous caster. Forming steel into ingots is an
16
-------
intermediate step prior to further processing. The steel is cooled to a
solid, whereupon the ingot mold is removed. The finished ingot can be sent
directly onto additional processing or stored indefinitely to meet scheduling
needs.
In the continue us-casting process, the ingot stage is bypassed, and the
molten steel is placed into one or more streams or strands that are fed
from a holding chamber called a tundish. The molten steel is transformed
directly into slabs, blooms, or billets and cut to the desired length by a
traveling torch. The continuous casting process eliminates intermediate
ingot formation, along with the energy required to reheat the ingot prior to
the formation of slabs, blooms, or billets. The amount of steel output pro-
duced by continuous casting is increasing.
Soaking pits are used to reheat ingots to rolling temperature (about 2340°
to 2400°F for carbon steel). Ingots are charged at a variable temperature
because of varying time intervals between ingot pouring and charging into
the soaking pits, brought about by conflicts in rolling schedules or other
delays. The amount of energy required for reheating ingots varies a great
deal because of the temperature variation in heating a cold ingot and one at
intermediate temperature levels.
*r '.£-
Soaking pits are simple round-7 square-, or rectangular-shaped refractory-
lined chambers having a retractable cover. Several pits grouped together
are referred to as batteries, having common flue ducts and a single recu-
perator and stack. Fuels used in soaking pits include mixed blast-furnace
gas and coke-oven gas, mixed blast-furnace gas and natural gas-air, straight
coke-oven gas, straight natural gas, and residual oil. Single, nozzle, or
port-mixing burners are used* and the burners are used and designed to mini-
mize the temperature gradient between the burner wall and the back wall of
the pit. Problems in fuel utilization occur when changing from one fuel to
another, particularly from gaseous to liquid fuels, because of major varia-
tions in flame length and heat-release profiles.
Mill operating practices have a major effect on soaking pit fuel economy.
Among them are —
• Ingot charging temperature
• Percentage of cold ingots charged
17 , .
-------
• Pit loading
• Holding time at temperature.
Fuel requirements vary from 545, 000 to 2 million Btu per ton. The
national average fuel consumption in soaking pit operations is estimated at
1 million Btu per ton.
Good combustion control is essential for maintaining fuel efficiency in
soaking pits. This includes temperature control-fuel input, pit pressure,
and air/fuel ratio. Substantial fuel reductions can be obtained by installing
recuperators on soaking pits now operating on cold combustion air. Im-
proved recuperator design can deliver combustion air preheated to higher
levels than the current range of 700° to 800°F. The potential for fuel re-
duction is about 15%, taking into account the large number of soaking pits
operating with cold combustion air or inefficient recuperators.
Most of the fuel used by the steel industry in reheat furnaces is for slab,
bloom, and billet heating. The evolution of the modern, continuous, reheat
furnace has been forced by rolling-mill improvements, steel quality require-
ments, floor space limitations, and low-cost energy, toward the direction
of increased heating rates and higher mean-effective-thermal head tempera-
tures. Most of these furnaces are continuous-pusher types with the load
supported on water-cooled skids over most of the furnace length and on a
refractory hearth in a final soak zone, which is overfired. Both three-zone
and five-zone furnaces are in use. More of the fuel input is progressively
burned near the charge end to increase furnace capacity. As a result, flue-
gas temperatures have increased, and this factor, as well as the heat lost
to water-cooled surfaces, has resulted in increased fuel requirements per
ton of steel heated. In recent furnace designs, the soak zone has been elim-
inated by using a walking-beam design, wherein the load is alternately sup-
ported on stationary and moving water-cooled supports. This furnace pro-
duces a uniformly heated product, without the cooler regions associated
with the water-cooled support skids, and has the highest output per unit of
floor space, but accomplishes this at the expense of fuel economy.
Average fuel consumption for reheat furnaces with preheated combustion
air is in the range of 2. 0 to 2. 2 million Btu per ton for three-zone furnaces
and 2. 7 to 2. 8 million Btu per ton for five-zone furnaces. Fuel economy
18
-------
for steel mill reheat furnaces is effected by furnace design, firing arrange-
ment, heat losses, heat recovery, combustion and process controls, and
operating practices. Although the furnaces are continuous, frequent delays
at the rolling mill have an adverse effect on fuel efficiency and on the tem-
perature uniformity of the product. Another important factor is the number
of operating turns per week.
Estimates of total fuel requirements of steel mill reheat furnaces are
frequently based on an average consumption value of 2. 5 million Btu per
ton of steel heated. Some of the rolled products may be reheated more than
once, and, consequently, estimates of total fuel consumption for reheating
range up to 300 trillion Btu per year.
Fuel conservation measures for reheat furnaces include those previously
mentioned for soaking pits:
• Retrofitting recuperators to furnaces using cold combustion air or cur-
rently using inefficient recuperators
• Improved combustion control
• Improved operating practice
• Programmed input control.
Additional measures specifically for reheat furnaces include the following:
• Improved maintenance of skid rail insulation
• Control of air infiltration
• Increased temperature of preheated combustion air.
Emissions from soaking pits and reheat furnaces may include carbon
monoxide and unburned hydrocarbons produced by inadequate or poorly
maintained ratio controls. These can be eliminated by the installation of
better equipment or by improving maintenance practices. The most signifi-
cant emission will be NOX, produced by the high flame temperatures required
and combustion-air preheat.
Annealing, heat treating, and finishing operations follow in sequence the
reheating and rolling mill operations. The major portion of the fuel used in
the steel industry for annealing is for cold-rolled products in strip form.
Approximately 35% of total United States steel production is in the form of
19
-------
strip and other cold-rolled products. Estimates of the amount of energy-
consumed for annealing, heat treating, and finishing operations range from
50 to 60 trillion Btu per year, including electricity. About one-half of this
quantity is for annealing.
Fuels used for these processes are mainly natural gas and coke-oven gas.
Natural gas is preferred because the sulfur content of a fuel has an adverse
effect on product quality and on the maintenance of furnace alloy components.
Temperatures employed in heat-treating and finishing operations are
much lower than those required in other types of process equipment and,
consequently, NOX levels should be correspondingly lower. Other emissions,
such as carbon monoxide and hydrocarbons, are associated with the quality
and maintenance of air/fuel ratio-control equipment. Many of the heat treat-
ing and annealing operations require the use of a protective atmosphere,
primarily reformed natural gas. Disposal or leakage of protective atmospheres
can result in locally high concentrations of carbon monoxide.
The equipment used in annealing, heat treating,, and finishing operations
includes the following:
• Batch and continuous coil and strip annealers
• Car bottom, roller hearth pusher tray, and other types of heat treat
furnaces
• Tin-plating lines
• Galvanizing lines.
Some of the energy conservation measures being used on annealing equip-
ment are —
• Conversion from radiant tube to direct firing
• Recuperation
• Improved combustion controls
• Use of ceramic fiber furnace linings, replacing brick refractories
• Substitution of nitrogen from oxygen plants for natural gas-based atmospheres.
Because of the relatively small percentage of total fuel use in this area and
the low level of NOX produced, any conservation measures adopted will not
have a material effect on overall steel plant emission levels.
20
-------
Trends in Process Modifications
Iron Making
Coke Ovens — Current and Near-Term
The major sources of emissions from coke ovens are the rapid evolution
of steam when moist coal is charged, the discharge of gases and particulates
from the charging openings, and the emissions during the coke push and subse-
quent quenching. Recent coke-oven battery designs have reduced the emissions
to an acceptable level by predrying and preheating coal sized for pneumatic
/
transport. Steam is used as the conveying medium and, with pipeline charging,
the charging ports are sealed.
During the coking process, leakage from the push side and coke side door
seals account for most of the emissions. New coke ovens feature improved
door seals which reduce door leakage substantially.
New battery designs feature a hood over the coke side door to collect
the emissions occurring during the coke push. The hood is ventilated and
the gases conveyed to suitable air pollution control equipment. In another
design coke is discharged to sealed container cars.
Emissions during coke quenching can be minimized by use of a process
involving both wet and dry quenching on a continuous basis. Waste heat re-
covery can also be incorporated.
Hydrogen sulfide in the coke-oven gas is removed in some modern ovens
by molten carbonate treatment. This process requires about 500 pounds of steam
per ton of coke produced. The availability of better refractories has led to
the design of batteries having wider and higher coking chambers, thus in-
creasing productivity. Improved control over heat-release profiles is re-
quired and this is obtained by staged combustion and increased use of blast-
furnace gas that contains a high percentage of inerts. This practice makes
available to other steel mill furnaces a larger amount of coke-oven gas to
replace natural gas or oil.
Coke Ovens — Long-Term
Dry coke quenching is claimed to improve the quality and yield of the coke
produced and to achieve a major reduction in the emission of pollutants, as
well as a potential for recovery of waste heat. The process is employed at
21
-------
several installations in Russia and uses counterflow shafb-furnace technology.
Each quenching tower has a capacity of 90 tons per hour. A typical 3000-ton-
per-day coke oven battery will require five quenching towers and will cost
about $11 million. It is the general opinion of the American steel industry
that significant dry coke quenching capacity will not be installed before 1985,
although a demonstration plant might be in operation by 1980.
Some steel industry people think that the current replacement of existing
coke-oven batteries by the modern coke ovens is the last "generation" to be
installed. They expect that form coke or other types of compacted forms of
calcined coal will take the place of coke ovens in the future, but not to a
significant degree before 1990. Process off-gases yield the equivalent of
coke-oven gas in total heating value, but have a lower heating value.
Sinter Plant — Current and Near-Term
Traveling-grate sinter lines are used to produce an agglomerate suitable
for charging to a blast furnace from flue dust, coke breeze, fluxes, and other
iron and carbon bearing particulates. Sinter lines are inherently a dirty
operation and must be equipped with suitable stack-gas cleaning equipment
to comply with emission regulations. Any combustible vapors in the stack
gases present major problems because incineration requirements both in-
crease overall fuel input and the stack-gas volume. This results in higher
capital and operating costs for the emission control equipment, usually
baghouses.
The economics of bringing the smaller and older sinter lines into com-
pliance are unfavorable and, consequently, many of these are being shut down
and removed. Some steel industry people think that sinter lines will be
phased out altogether and replaced by alternative agglomeration processes,
but others are currently installing very large sinter lines designed so as
to avoid the major problems associated with the process. Some of these
improvements are better seal designs to reduce air infiltration,, improved
seal maintenance, design of gas recirculating systems to reduce stack-gas
volume and to eliminate carbon monoxide in the stack gases without incinera-
tion, and control over stack-gas temperature to prevent condensation in the
baghouse.
22
-------
Sinter Plant — Long-Term
Existing sinter plants cannot handle rolling mill scale, which is produced
in large quantities, but is contaminated with rolling oil. The oil vapor is
not removed by baghouses and passes to the exhaust stack as a pollutant.
Consequently, the rolling mill scale is used as landfill, although it has a
u-
71% iron content and could be a very good blast-furnace charge.
Both pelletizing and briquetting are under development and will, in the
future, provide a hot-compaction process to replace sinter lines. These
systems have a potential for handling rolling mill scale and for large Deduc-
tions in both energy requirements and emissions.
Blast Furnace — Current and Near-Term
Because of the decreased availability of coking quality coal in the United
States, major emphasis in recent years has been placed on increasing blast-
furnace output and reducing the coke required per ton of hot metal. Some of
the process and equipment modifications employed have been —
• Larger blast furnaces
• Higher blast temperature and oxygen enrichment
• Improved burden, pelletized and fluxed sinter
• Higher top pressure
• Improved blast-furnace stove designs and scheduling
• Auxiliary fuel injection, primarily oil and natural gas.
Increased cost and decreased availability of the auxiliary injection fuels
have led to the development of systems for injection of pulverized coal or
oil-coal slurries or emulsions.
The increased adoption of the basic oxygen process for steelmaking and
the adverse hot metal-scrap ratio required for the BOF has made the capa-
city of some steel mills limited by the availability of hot metal. Additional
mill capacity will require installation of new, modern blast furnaces and
give the beneficial effect of increased dependence on coal. Additional reduc-
tion in coke consumption will be achieved by use of higher blast temperature,
oxygen enrichment, and auxiliary coal injection.
23
-------
Blast Furnace — Long-Term
The decreased availability of high quality coking coals will lead to a
gradual adoption of processes for production of synthetic coke, such as
form-coke, which can use high-sulfur, noncoking coals. The trend toward
larger blast furnaces, higher top pressures, and additional improvements
in burden will continue. If a satisfactory process for direct-reduction with
coal-based fuels is developed, prereduced iron will probably become a major
portion of the blast-furnace charge.
The development of efficient, dry-type, high-temperature, blast-furnace
gas cleaning systems will make energy recovery by the use of off-gas tur-
bines feasible.
Steelmakinff
Open Hearth — Current and Near-Term
The use of open-hearth furnaces will continue to decline, both because of
the higher production capabilities of the basic oxygen furnace (BOF) and also
because of emission problems with existing open-hearth furnace designs.
Steel mills with capacity limited by hot metal availability will continue to use open
hearths because of greater flexibility in the hot metal-scrap ratio. These
open hearths will be modernized so as to increase production rates and com-
ply with emission control regulations. Increased use of oxygen roof-lancing
to increase furnace output will probably occur. Increased NOX emission
may result.
Open Hearth— Long-Term
The use of open-hearth furnaces is expected to continue to decline and
will probably amount to about 10% of total ste«l production by 1985.
Basic Oxygen Furnace (BOF) — Current and Near-Term
Use of BOF's for steelmaking will continue to, grow because of the high
productivity and favorable economics. The size of the BOF vessels is con-
tinuously increasing. A major advantage of BOF steelmaking is the very
large reduction in dependence on external fuels. Only about 200, 000 Btu
per ton of steel produced is required for vessel dryout and preheating.
Preheating of the scrap fed to a, BOS" will reduce dependence on hot metal
because only about a 68% hot-metal charge is required with the scrap
24
-------
preheated to 1200°F, whereas a 75% hot-metal charge is required with
cold scrap.
The BOF carbon monoxide-rich off-gases have an average heat content
of 250 Btu/CF or 400, 000 Btu/ton of steel, and in most installations, the car-
bon monoxide is burned in the vessel, which produces a large volume in-
crease in the stack gases flowing to the air pollution control devices. Some
recent installations have been equipped with an OG collection system, which
includes a sealed collection hood, two-stage venturi scrubbers, and a flare
stack for carbon monoxide combustion. No information is available con-
cerning the combustion efficiency of the flare stacks.
Submerged injection of oxygen has been successfully applied to BOF steel-
making and is expected to be used quite extensively in the near future.
Basic Oxygen Furnace — Long-Term
From both the energy conservation and pollution control points of view,
recovery of the carbon monoxide-rich BOF off-gases will have a more at-
tractive potential as the cost of purchased fuels increases and availability
decreases. Economic considerations are the major inhibiting factors at the
present time.
Development of improved scrap preheating systems with suitable emission
controls will come about. The scrap preheating systems may be fueled by
off-gas carbon monoxide.
Electric Furnace — Current and Near-Term
The use of electric furnaces for steelmaking will continue to grow, par-
ticularly in the nonintegrated steel mills. Some steel industry executives
believe that future integrated plants will use a combination of BOF and elec-
tric steelmaking, giving maximum ability to vary the hot metal-scrap ratio.
The productivity and efficiency of electric steelmaking will be improved by
development of practical scrap-preheating systems.
t\
Electric Furnace — Long-Term
Development of a coal-based process for direct reduction of iron oxide
pellets or lump ore may bring about extensive use of electric melting of a
prereduced charge.
25
-------
Trends in Energy Utilization
The decreased availability and increasing cost of natural gas and low-
sulfur oil has forced the steel industry to depend more on coal as the primary
energy supply in the production of steel. A major trend in recent years has
been to reduce the coke required per ton of hot metal because of the diminish-
ing supply of coking coal. This has increased dependence on purchased fuels
because of the reduced output of coke-oven gas. A current offsetting trend
has been the increased use of BOF steelmakingj which requires more hot
metal than the open hearth.
Part of the reduction in coke requirement has been achieved by injection
of natural gas or oil at the blast-furnace tuyeres. The major injection fuel
in the future will probably be powdered coal or high-sulfur oil, which cannot
be used in direct combustion applications because of sulfur oxide emission
regulations. Low-sulfur oil and coke-oven tar and pitch will probably be
used for the more critical applications, such as soaking pits, reheat furnaces,
and modernized open-hearth furnaces retained to provide hot metal-scrap
flexibility.
Fuel cost, at least for the more critical heating applications, appears to
have become secondary to fuel availability. Natural gas cost is currently
in the $0. 90 to $1. 00 per million Btu range in the Chicago-Cleveland-Pittsburgh
areas, and low-sulfur oil is estimated to cost $2. 10 to $2. 25 per million Btu
at the burner. Many integrated steel plants are converting soaking pits and
reheat furnaces to oil firing because the plants cannot obtain firm contracts
for natural gas. Some of these companies will continue to use natural gas
when it is available during the summer months.
Serious consideration is being given to the in-plant production of synthetic
fuels from coal even though current estimates of cost are in the $3. 00 to $4. 00
per million Btu range.
One of the problems in assessing the effects of fuel availability, fuel cost,
and changes in process technology is that most energy conservation estimates
are based on overall energy requirements, including coal, rather than that
for critical fuels such as natural gas and low-sulfur oil.
For example, on an overall energy basis, an integrated plant with BOF
steelmaking requires more energy than one with open hearths because of the
26
-------
increased hot metal-scrap ratio. However, the requirement for purchased
fuels will decrease by more than 3 million Btu per ton, and this decrease
has the greater significance.
Future integrated steel plants can be designed to be independent of the
need for purchased critical fuels by using both BOF and electric steelmaking
for hot metal-scrap flexibility and continuous casting. The rate at which
this objective can be met will be quite slow because of the large investment
in existing facilities and enormous capital requirements for new plant con-
struction. The slow rate of change is shown by the AISI estimate of the
steel industry fuel consumption, by source, for 1972 and 1980:
Year
_ Source _ 1972 1980
Coal, % 67.3 68.5
Purchased Gas and Oil, % 28.8 26.9
Purchased Electricity, % 3. 9 4. 6
Btu X 106 per Net Ton Shipped 32. 6 30. 1
During the field interviews, various integrated steel plants reported the
following energy consumptions,, by source, for 1974:
Plant ±
Coal 75 60 70
Oil 10 22 5
Natural Gas 15 18 20
Purchased Electricity 4. 7
Other 0. 3
As indicated, general fuel-use trends in the industry will be to increase
the relative amount of coal consumed per ton of steel shipped and to decrease
dependence on purchased oil and gas. The amount of oil used compared with
natural gas will increase because of the decreased availability of natural
gas. Major shifts toward increased use of coal will require very large
amounts of capital, both for new production facilities and for the associated
emission control equipment.
The following list presents a more detailed view, by process, of the trends
in fuel utilization.
27
-------
Ironmaking — Current and Near-Term
• Increased use of coal as a primary fuel, both in increased coke-oven
output for blast-furnace injection and for boiler firing.
• Increased availability of coke-oven gas.
e Decreased use of coke-oven gas for under-firing coke ovens, partial
replacement by increased use of blast-furnace gas.
• Coal and high-sulfur oil will be used as blast-furnace injectants.
• Use of natural gas as a blast-furnace injectant, for blast stove heating
and other auxiliaries -will decline sharply.
• Direct combustion of coal in pelletizing facilities, both traveling-grate
and shaft types.
Ironmaking — Long-Term
• Coke ovens will be replaced, possible to a major extent, by form-coke or
some other pyrolyzed coal product. Heating value of the gas output may
be much less than that of coke-oven gas, possibly less than 200 Btu/CF.
• Coal gasification plants producing low-Btu gas may be installed in steel
mills or the product gas made available from a nearby source.
• Heating value will be in the 175 to 300 Btu range, depending on the process
used. This gas maybe a good blast-furnace injectant if the economics
are favorable.
• Direct reduction to a highly metalized iron product using solid reductants
may replace part of the coke oven-blast furnace ironmaking capacity. The
availability of low-sulfur fuels at a competitive price will be essential and
appears to be doubtful at the present time.
Steelmaking — Current and Near-Term
• Use of oil, natural gas, or in-plant gases for steelmaking furnaces will
continue to decline as additional open-hearth furnaces are replaced by
BOF and electric furnace steelmaking.
• Use of natural gas as an open-hearth fuel will decline sharply, replaced
by oil or coke-oven by-products.
Steelmaking — Long-Term
• Recovery of carbon monoxide waste gas from BOF steelmaking and pos-
sibly using it to preheat the scrap charge,
• Development of direct reductive processes using solid reductant.
28
-------
• Application of nuclear energy to steelmaking with availability of reducing
gases for direct reduction or blast-furnace injection.
• Development of continuous steelmaking.
Soaking Pits and Reheat Furnaces — Current and Near-Term
• Shift from natural gas and coke-oven gas to oil firing or blast-furnace
gas and other coke-oven by-products.
• Fuel conservation by use of techniques previously discussed.
• Increased use of induction heating for slabs and billets.
• Increased use of continuous casting.
Soaking Pits and Reheat Furnaces — Long-Term
• Use of coal, solvent refined coal or a slurry, or emulsion of oil and solid
fuel.
• Major reduction in fuel requirements by development of a process for
production of steel strip directly from molten steel.
Heat-Treating and Finishing Operations — Current and Near-Term
• Available natural gas and coke-oven gas conserved in other mill areas
will be used.
• Some installations will be converted to oil firing. In most cases, distillate
oil will be required because of sulfur limits.
• Gas atmospheres produced from natural gas will be replaced to a large
extent by by-product nitrogen from captive oxygen plants.
• Electric heating will replace natural gas in cases where alternatives are
not feasible.
Trends in Emissions
In most cases, those interviewed were not disposed to discuss specifics
of emission control problems except to state that every effort is being made
to comply with local standards at very heavy costs. In many cases, attention
was drawn to the large energy requirements for compliance with future stan-
dards, particularly those for fugitive emissions and building evacuation.
It was generally agreed that not much is known about NOX emissions from
steel-mill equipment. In one case, concern was expressed that flare-stack
combustion of carbon monoxide might not be effective. Another concern was
that baghouses and precipitators or scrubbers are not effective in the
29
-------
elimination of oil vapor from oily scrap. Combustion of such vapors or
carbon monoxide adds greatly to the volume of gases handled in demonstration
cleanup devices.
Major combustion-related emission problem areas were given as —
• Coke ovens
• Sinter lines
• Open hearth
• BOF and electric furnace
• Acid recovery from waste pickle liquor.
Data Analysis
The emission of pollutants from iron and steel process equipment involved
in direct combustion of fossil fuels includes carbon monoxide, nitrogen oxides,
sulfur oxides, and unburned hydrocarbons. Particulates result from the
direct combustion of coal, but are mostly present in flue products as fine
particulates picked up from process materials. Unburned hydrocarbons
usually occur as the result of inadequate or improperly adjusted combustion
controls. Sulfur oxides are formed primarily from process materials con-
taining sulfur and secondarily from fuel sulfur and cannot be controlled by .
modifications in direct combustion technology. Carbon monoxide can result
from improper adjustment of combustion controls, but in the iron and steel
industry the major sources are the blast furnace, sinter and pellet lines,
and steelmaking furnaces.
The primary sources of nitrogen oxides are chemically bound nitrogen in
the fuels and that formed in flames at high temperature by combination of
active species of nitrogen and oxygen. The latter type is strongly affected
by combustion process modifications, and, in the case of utility power boilers,
substantial reductions ranging up to 90% have been achieved. The modifica-
tions used have included flue-gas recirculation, staged combustion, excess
air control, water injection, fuel-air mixing rate control, and internal re-
circulation.
Steel companies do not, in general, appear to have much specific knowledge
concerning emissions of nitrogen oxides from process equipment, except that
from boilers. Some concern was expressed about future regulations concerning
30
-------
NO emissions and the state-of-the-art concerning emissions from specific
Jt
types of process equipment and means for control.
The most frequently expressed concern was about the high energy input
required for emission control equipment and the corresponding adverse effect
on the ability to meet energy conservation goals. This concern was reflected
in a request by several of the steel companies for R&D programs to develop
emission control systems requiring less energy input.
Several types of process equipment in use involve collection of process
off-gases with a substantial carbon monoxide content. Direct combustion or
incineration as a control technology increases the volume of gases discharged
to emission control equipment thus increasing capital cost, operating cost, and
the energy requirements for emission control. The sealed collection hood sys-
tem developed by the Japanese (OG system) for basic oxygen furnaces prevents
or minimizes the infiltration of air at the collection point. The collected gases
are cleaned and used as auxiliary plant fuel. In American installations using
this system, the cleaned off-gases are flared. The effectiveness of flare stacks
in burning carbon monoxide was questioned by some steel industry representatives.
The logical end use for BOF off-gas is for preheating the scrap charged to the
furnace, but up to the present time both technical and economic considerations
have prevented such use.
The emission of nitrogen oxides from iron and steelmaking and processing
equipment does not appear to have been extensively investigated. However,
reasonable estimates can be made by assuming a relationship between known
operating temperatures and NOX concentration in stack gases. This relation-
ship is affected by other variables, such as combustion air preheat tempera-
ture and oxygen enrichment of combustion air.
Table 3 shows the estimated NOX concentrations for the major energy-
intensive processes and the resulting total annual combustion-related NO.,
jC
production based on 1971 steel production energy consumption data.
Pelletizing
Pelletizing is carried out at or near the iron mines because low-grade ores
are used and ore beneficiation precedes the pelletizing operation. The bene-
ficiated product is extremely fine and would be very difficult to handle and
transport to the steel mills. The cost of shipping unbeneficiated ore would
31
-------
Table 3. ESTIMATED NOX EMISSIONS FROM
STEEL-MILL PROCESSES AND EQUIPMENT
Fuel
Fuel NOX
Consumption, Emission/ Tons NOX/
1012 Btu/yr* Level, ppm 1012 Btu Tons NOy/yr
Pelletizing
Sintering
Coke Oven Underfiring
Blast Stove Heating
Open-Hearth Furnace
Soaking Pit and Reheat
Furnaces
Heat Treating and
Finishing
27
93
212
200
127
510
60
300
500
200
100
600
300
200
210
350
140
70
420
210
140
5,670
32,550
29,680
14,000
53,340
107, 100
8,400
Basis: 1971 steel industry data.
be almost double that of the beneficiated and pelletized product. Previous
IGT studies have shown that pelletized ore production will be about 60 million
tons/yr by 1985. The fuel consumed by the pelletizing furnaces has remained
about constant at 600, 000 Btu/ton. This indicates that total NOX emission
from pelletizing furnaces will reach about 8500 tons/yr by 1985. The steel
industry and equipment builders are considering coal firing the pelletizing
furnace combustion chambers. If this is done, it will probably bring about
an increase of about 5Q% in NO emission.
A,
Sintering
The use of sinter machines to agglomerate ore fines, flue dust, and coke
breeze has been declining since 1966 and amounted to 43 million tons in
1971. If the present rate of decline continues, the 1985 production of sinter
would be about 27 million tons. The attitude of the steel industry is mixed
because many steel plants are phasing out sinter lines, while at least one
major producer has replaced several small sinter lines with a large machine
designed to meet pollution control regulations. In either case, the NOX out-
put -will continue to be a major pollutant, if our estimates are correct.
Coke-Oven Underfiring
Although the current practice of firing coke ovens with a mixture of
blast-furnace gas and coke-oven gas and slow mixing in the combustion
chambers should tend to minimize NOX production, the estimated total is
32
-------
quite large because of the large quantity of fuel consumed.
The reduction in the coke required per ton of hot metal achieved during
the 1960's will continue, but steel mills are currently installing new coke
ovens because of the increased need for hot metal due to the high BOF hot
metal-scrap ratio. The 1985 projection for coke-oven underfiring fuel is
458 trillion Btu. If the NOX concentration remains constant, the resulting
total emission of NOX will reach 64, 120 tons.
Although it is reasonable to assume that substitution of form coke may
result in a substantial reduction in NO production, the general opinion in
j£ ,
the steel industry is that form coke will not be a significant factor in 1985.
Blast-Stove Heating
As in the case of coke-oven underfiring, the blast stoves require very
large quantities of fuel for heating. However, since the stoves are heated
primarily with blast-furnace gas (80 to 95 Btu/CF) the NOX concentration
is lower due to the presence of diluents and a low flame temperature.
The projected need for hot metal in 1985 is 124 million tons. This amount
of hot metal will require 280 trillion Btu for blast-stove heating. Assuming
no reduction in NO., stack-gas concentration, the NO... emission in 1985 will
X, J»-
be 19, 600 tons/yr. Because of the low estimated NOX concentration and the
presence of inerts in the fuel gas, equivalent to flue-gas recirculation, the
potential for NOX reduction is probably small.
Open-Hearth Furnace
NOX emissions from open-hearth furnaces are very high because of the
high combustion air preheat temperature, high operating temperature, -and
the use of oxygen lances to increase production rates. The data available
and our estimating technique indicate that NOX concentration will be in the
609,-to 800-ppm range. Although many open1 hearths are being phased out
because of emission control difficulties and better economics of steel pro-
duction with the BOF process, several steel mills are modernizing open-
hearth shops, including pollution control equipment to provide flexibility
in the hot metal-scrap ratio, particularly those mills with a hot-metal de-
ficiency. Therefore, predictions that the open hearths will be phased out
entirely by 1985 are unrealistic, and it is anticipated that about 15 million
tons will still be made by the open-hearth process in 1985. Fuel consumption
33
-------
has been decreasing and may reach 2. 5 million Btu/ton in 1985. This will
require a fuel consumption of 37. 5 trillion Btu for open-hearth steel pro-
duction and result in an NOX emission level of 15, 750 tons.
Soaking Pits and Reheat Furnaces
These are large furnaces with fuel inputs ranging from 1. 0 to 3. 5
million Btu/ton heated. Fuel efficiency is affected by many factors such
as furnace size, design, combustion controls, combustion air temperature,
furnace scheduling, and downtime. Improved efficiency measures, which
do not increase flame temperature, will, in general, reduce NO emission
in proportion to the reduction in fuel usage.
Existing fuel-conservation measures in soaking-pit heating include im-
proved scheduling so as to charge at a higher ingot temperature, programmed
input control, improved burner designs, air/fuel ratio control responsive to
stack-gas oxygen content, addition of recuperators to existing cold combustion
air installations, and use of recuperators designed to give higher preheat tem-
perature. Of these, the use of high-mixing-rate burners and an increase in
combustion air preheat are likely to increase the NO emission level. At
Ji,
the present time, only experimental information is available concerning the
effect of these parameters on NOX levels.
Soaking-pit and reheat-furnace operating temperatures are such that the
estimated NOX levels should fall in the 250^to 350-ppm range. However,
the very large amounts of fuel used result in a total NOX output estimated
at 107,000 tons in 1971.
A major factor that will reduce consumption of purchased and in-plant
fuels and thereby decrease NOX output is the trend toward use of continuous
casting. In this process, finished shapes are produced from molten steel,
thus eliminating soaking pits and most of the reheat requirement. About
20% of total steel production, or 40 million tons, is estimated to be pro-
duced by continuous casting in 1985. In spite of this, soaking-pit and reheat-
furnace steel capacity will have to be increased during the 1975 to 1985
period to provide for the expected growth in steel production. According to
the IGT projection, conventional steel processing will account for 160 million
tons in 1985. At the present fuel consumption of 4. 7 million Btu/ton, the
total fuel consumed for soaking pits and reheat furnaces in 1985 will be 750
34
-------
trillion Btu. This fuel consumption will result in an estimated NOX emission
of 157,900 tons.
Heat-Treating and Finishing Operations
This category includes annealing, hardening, carburizing and normalizing
of most of the steel industry cold-rolled products, as well as production of
/\
coated products. Fuel consumption in 1971 was about 600 trillion Btu for
the production of cold-rolled products (about 25% of total steel production).
NO emission levels are assumed to be in the 15Q"to 250-ppm range. On
•*» / *
this basis, total NO emission in 1971 for this category will be about 8400
•'' Ji
tons. Assuming that production of cold-rolled products remains at about
25% of total steel production, the 1985 NO emission will amount to 11,200
ji
tons/yr.
Recommendations
This program is concerned with air pollutant emissions from combustion
processes and the control of these emissions by combustion modifications.
The iron and steel industry has assembled a list of recommended programs
for dealing with the problems of energy utilization efficiency and air pollutant
emissions. However, most of these recommendations involve means other
than combustion modifications and thus fall outside the scope of this program.
Recommendations for General Emission Control and Energy Conserva-
tion According to the Industry
• Development of stack-gas and air emission control devices requiring
less energy input,
• Development of means other than incineration for emission control of
carbon monoxide and oil vapors..
• A study of the interrelationship between energy use and pollution control
in new steel plants, where energy self-sufficiency is now a major objec-
tive. The capital requirements to meet these objectives are now almost
prohibitive.
• Systems for conversion of low-temperature-level stack and other heat
losses to a useful form.
• Design, construction and operation of a 3000 ton/7 day form-coke
demonstration plant^ A'
• Design, construct!on^and operation of a 3000 ton/day eotee
quenching plant* /
35
-------
• Development of an improved process for desulfurization of coke-oven gas.
• Development of an improved process for treatment of coke-oven water
discharge, particularly for cyanide removal.
• Completion of the development of a system for injection of coal at the
tuyeres of blast furnaces.
• Development of a process for hot-briquetting of in-plant fines to replace
sinter lines.
• Development of a process for agglomeration of mill scale.
• Development of an air-recirculating system for sinter lines to reduce the
volume of stack gases requiring emission control.
• Development of practical means for control of carbon monoxide in sinter
lines.
• Development of processes for practical control of NOX from steelmaking
processes.
• Development of processes for practical collection or suppression of oil
vapors from charging of oily scrap in electric furnaces and BO vessels.
• Determination of the effectiveness of flare-stack disposal of carbon
monoxide from BO vessels.
Recommendations Based on Evaluation of Data Collected During
Program
The major pollutants that can be affected or controlled by combustion
technology are carbon monoxide, unburned hydrocarbons, and nitrogen
oxides. Sulfur oxides and particulates come from materials in the process
and the fuel itself and usually cannot be reduced by combustion technology.
Many of the listed research or development recommendations by the steel
industry are undoubtedly worthy of consideration, but are not amenable to
control by combustion modifications and are, therefore, not within the
scope of this investigation.
There does not appear to be much specific information available on NOX
produced by steel mill process equipment, particularly the types operating
at high-temperature levels. The important types would be pelletizing-furnace
combustion chambers, open-hearth furnaces, soaking pits, and large reheat
furnaces.
We recommend that a field investigation be undertaken to determine the
NOX emission levels from the listed types of equipment. Following this
36
-------
investigation, or possibly concurrent with it, we also recommend a modi-
fication program to determine the effectiveness in NOX reduction of the
techniques already in use on power boiler firing such as:
• Reduced excess air
• Staged combustion
• Stack-gas recirculation
• Aerodynamic control over mixing and recirculation.
Glass Industry— Current State-of-the-Art
The glass industry can be broken down into three segments, based on the
type of glass produced. The largest segment is the container glass industry,
which produces about 64% of the total amount of glass (by weight) produced
by the entire industry. The next largest segment, pressed and blown glass
and glassware, produces nearly 21% of the total amount manufactured by
the entire industry. The third segment is the flat glass industry, which
produces about 15% of the total industry output.
While the specific processes used within each segment of the industry
vary according to the product being manufactured, glass manufacturing
involves three major energy-consuming processes: melting the raw materials,
refining the molten glass, and finishing the formed products (Figure 1).
Typically, about 80% of the energy consumed by the glass industry is for
melting and refining, 15% is for finishing, and 5% is for mechanical drives
and conveyors. The primary differences in processes used among the various
segments occur in the refining and finishing operations.
Melting
Melting practices are essentially the same throughout the glass industry,
regardless of the end product. The first step is the preparation of the raw
glass batch. In this step, the various raw materials, such as sand, lime-
stone, soda ash, and various minor ingredients (fluxes, colorizers, and
decolorizers) are weighed in a batch hopper and dumped into a mixer. Water
is usually added at this point and everything is thoroughly mixed. The ma-
terial is transferred to a storage hopper, from which it is charged into the
furnace. The energy consumed by this process is electrical and is used
to drive the mixers and conveyors.
37
-------
SILICA SAND
SODA ASH
LIMESTONE
MgO • CaO
FELDSPAR
R20-AI203-6Si02
MISC
AGENTS
REGENERATIVE FURNACE
SUBMERGED THROAT
1472°- 20I2°F
-H FABRICATION
1 FINISHING
930°F
PACKING |
GULLET
CRUSHING
A-83-1249
Figure 1. FLOW DIAGRAM FOR SODA-LIME
GLASS MANUFACTURE
38
-------
The second step is the actual melting of the glass batch. The mixed raw
materials are mechanically charged through a port, known as the "doghouse, "
into the melter. Most of the melters in use within the industry are continuous
reverberatory furnaces equipped with checker-brick regenerators for pre-
heating the combustion air. Some unit melters are also in use at a savings
in investment cost but at a substantial sacrifice in fuel-utilization efficiency.
Reverberatory furnaces in this industry break down into two classifications.
according to the firing arrangement used: end-port and side-port melters. In
the operation of a side-port-fixed furnace (Figure 2), the preheated combus-
tion air mixes with the fuel in the port, resulting in a flame that burns over
the glass surface. The products of combustion exit via the opposite port,
down through the checker-bricks, and oub through the reversing valve to the
exhaust stack. Typically, there are several ports situated along each side
of the furnace. In contrast, there are only two ports in an end-port-fired
furnace (Figure 3), located on the rear wall of the furnace. The flame is
ignited in one port, travels out over the glass toward the bridgewall, and
"horseshoes" back to the exit port — the other port in the rear of the furnace.
In both types of furnaces, the firing pattern is reversed every 20 to 30 minutes,
depending upon the specific furnace. During this reversal period, the flame
is extinguished, the furnace is purged of combustion gases by reversing the
flow of combustion air and exhaust gases passing through the reversal valve,
and combustion is then reestablished in what was previously the exhaust port.
Both types of melters are operated continuously throughout a campaign that
normally lasts 4 to 5 years, at sustained temperatures up to 2900°F.
In addition to the reverberatory-type melters, day tanks, unit melters,
and pot melters are used, mostly in the pressed and blown glass industry.
Many of these melters are batch-type, as opposed to continuous, resulting
in a substantial reduction in fuel-utilization efficiency. Much of the fuel that
is wasted is due to the antiquated methods of operation and associated equip-
ment used with these melters.
From the melter, the molten glass passes through a refractory-lined
throat into the refinery. The refinery may be an integral part of the melter,
separated by the bridgewall, but taking its heat secondarily from the melter,
or the refinery may be a separate entity. In the latter case, the refiner is
equipped with auxiliary burners. The purpose of the refining step is to allow
39
-------
REGENERATOR
_T T_
Figure 2. PLAN VIEW OF SIDE-PORT, FLAT GLASS FURNACE
REGENERATOR
PORT;
FURNACE
MELTING END
FURNACE
WDRNNGEND
BRIDGE WALL
THROMV.
Figure 3. LONGITUDINAL SECTION OF TYPICAL
END-PORT FURNACE
40
-------
the gas bubbles trapped in the molten glass to escape and to reduce the
temperature of the glass to a workable range. Upon completion of the re-
fining step, the molten glass is ready for processing.
Finishing
Within the process of finishing, there are the fabrication and annealing
steps. The specific processes used within each step depend on the product
that is being made. This, in turn, will determine the energy requirements
for the entire finishing process.
Fabrication
Product fabrication within the container-glass industry is relatively
straightforward, with minimal variation from one process to another. The
two primary processes in use for container fabrication are the Owens pro-
cess and the flow process. In the Owens process, conditioning of the glass
prior to forming takes place in pots. In the flow process, conditioning occurs
in the forehearth and feeder. In both cases, a small amount of energy,
usually natural gas, is consumed to bring the glass to the temperature re-
quired for forming.
After the glass has been properly conditioned, it is fed into a forming
machine. The machines used in the two processes vary and for purposes
herein are not significant. The energy consumed by the machines is electri-
cal and is required not only to drive the machines, but also to drive com-
pressors to obtain the required compressed air for blowing.
Product fabrication within the flat-glass industry varies according to
the type of glass (sheet, plate, or float) produced. Sheet glass is produced
by one of several continuous processes in which the glass is drawn verti-
cally out of the drawing chamber that contains the properly refined and con-
ditioned molten glass. The surface of the product made in this manner, al-
though slightly wavy in appearance due to variances in viscosity, is very
brilliant and thus requires no polishing.
Plate glass, in contrast to sheet glass, is rolled. Furthermore, the
glass is not "finished" upon exit from the rolling operation. Rather, a
substantial amount of grinding and polishing is required to produce the smooth
surface needed for quality purposes. Because of the sizable investment
that must be made to install a grinding and polishing operation, the
41
-------
float-glass process has been developed, which produces superior quality
plate glass without grinding and polishing.
In the float-glass process, the molten glass flows from the melting cham-
ber into a float chamber, which is a molten tin pool. As the glass flows
over the molten tin, it assumes the smooth and perfectly flat surface charac-
teristics of the tin surface, along with uniformity of thickness. The tin is
maintained in the molten state with electric heating elements. Oxidation of
the tin is prevented by the use of a controlled atmosphere. In terms of
energy utilization, the float-glass process reduces the amount of energy
consumed for power. However, it is not clear whether the overall process
consumes less energy than the typical plate glass production line.
Annealing
After the glass has been formed into its final shape, it must be cooled to
room temperature. In order to prevent strains from occurring, the glass
is annealed, a process whereby the temperature is controlled and gradually
reduced. Figure 4 shows a typical annealing cycle for a controlled recircu-
lation lehr. While the actual cooling cycle varies from one type of glass to
another, the annealing procedure is essentially the same for all types of
glass. The glass is placed on a continuous conveyor and passed through a
tunnel-type oven, called a lehr, which is zoned so that the cooling curve
precisely matches that required to obtain a strain-free product.
The annealing lehrs can be heated by convection, radiation, or a combi-
nation of both. The most effective method is to use zoned, convection lehrs
with internal distributors to obtain lateral temperature uniformity. These
require external fans and heater boxes, but have the advantage that both
gaseous and liquid fuels can be used to heat the recirculated air. Some
lehrs are direct-fired by atmospheric or premix burners or by excess-air
burners. Compared with the melting and firing processes, energy consump-
tion is relatively low.
Inspection
The final phase of the finishing operation is the inspection of the finished
product. Glass that does not meet the quality standards is recycled back to
the melter and remelted. The amount of glass recycled varies considerably,
but a good approximation is 10% to 12% of the total glass production. While
42
-------
(JO
1300
1100
1000
900
MO
700
MO
500
400
300
300
100
10T.Q-
30-.0"
SO'O"
Figure 4. TYPICAL ANNEALING CYCLE FOR A CONTROLLED RE CIRCULATION LEHR
-------
remelting requires additional energy and would thus seem undesirable, in
fact, remelting consumes less energy than melting of the batch material.
Thus, adding cullet, as the recycled glass is called, to the raw batch lowers
the amount of energy required for melting the batch. The energy consumed
for remelting is included with the energy used for the melting process and
is not considered separately. For purposes of this report, there are no
other major combustion-related uses of energy in the glass industry.
Energy Utilization in Glass-Manufacturing Processes
Glass melting is the major energy-consuming process in the glass industry.
Efficiencies of continuous melters vary considerably, depending upon basic
design as -well as furnace age, type of glass being melted, and the end use
of the product. In the container glass industry, most glass companies claim
to operate at a fossil-fuel consumption level of 4. 0 to 7. 0 million Btu/ton of
glass melted. In the flat glass industry, fossil-fuel consumption is 6. 0 to
12. 0 million Btu/ton of glass melted. In the pressed and blown glass industry,
fossil-fuel consumption varies from 6.0 to 40. 0 million Btu/ton of glass melted,
and many companies consume between 30. 0 and 40. 0 million Btu/ton of glass
melted.
Note that these fuel consumption figures apply to furnaces during melting
operations. However, data on energy utilization (Table 4) indicate that
Table 4. BREAKDOWN OF ENERGY CONSUMPTION
BY THE GLASS INDUSTRY IN 1971*
Pressed and
Flat Glass Containers Blown Glass
(SIC Code 3Z11) (SIC Code 3ZZ1) (SIC Code 3ZZ9] Total
Glass Produced, 106 tons Z. 56 10.90 3.50 16.96
Energy Consumption, 101Z Btxi
Melting 44.7 104.3 50.5 199.5
Annealing 8.4 19.6 9.5 37.5
Other Z. 8 6. 5 3. Z i'z. 5
Total Energy Consumption,
1012 Btu 55.9 ,130.4 63.1 249.4
Average Energy Consumption
for Melting, 106 Btu/ton of
glass 17.5 9.6 14.4 11.8
Average Energy Conrumption for
Entire Production, 106 Btu/ton Zl. 8 1Z. 0 18.0 14.7
Excludes energy consumed for electricity generation. A-84-1402
44
-------
energy consumption is actually higher because operation is suspended for
breakdowns and holidays many times during the year. During these periods,
the furnaces are idled to maintain furnace temperature while production
stops. Thus, energy is consumed but no glass is produced, thereby increasing
the overall average energy consumption per ton of glass produced. An even
greater source of discrepancy is the variation in the amount of cullet used
from one glass plant to the next. The amount of cullet charged varies from
10% to 30% of the total raw material charged to the melter. As previously
stated, lower percentages of cullet charged result in higher fuel consump-
tion. In general, economic considerations prevent higher percentages of
cullet from being charged.
These figures indicate that the container glass melters are operated most
efficiently. The glass in the flat glass industry is of higher quality than that
in the container glass industry; thus, gas bubbles (seeds) and unmelted raw
batch (stones) cannot be tolerated. Consequently, furnaces in the flat glass
•A.
industry'1* are very large compared with other glass industry furnaces to
allow for the long soak periods required to remove all impurities such as
entrapped gases and solid batch particles. In the container industry, the
furnaces are smaller, and shorter retentiontim.es are used because glass
quality is not so critical. Over the length of a campaign, fuel consumption
is independent of the type of furnace used — ij e. , end-port on side-port fired.
Electric melters require less energy to melt a ton of glass than fuel-fired
melters. However, if the energy, which is usually fossil fuel, used to gen-
erate the electricity also is considered, total energy consumption per ton
of glass is higher than for fuel-fired melters. Typically, an all-electric
melter requires about 2. 9 million Btu to melt a ton of glass; this is a melting
efficiency of about 65%. But the fuel that is consumed to generate this elec-
tricity is about 8. 7 million Btu, based on a 30% efficiency of generation.
Consequently, the "real" amount of energy consumed to melt glass in an all-
electric melter is 8. 7 million Btu/ton.
In the pressed and blown glass industry, glassware is made by machine
and by hand. The furnaces vary from the large continuous type to small
* Perhaps 8 sq ft of melting area per ton of glass melted per day in the flat
glass industry versus 4 sq ft of melting area per ton of glass melted per
day in the container industry.
45
-------
pot furnaces. As a result, energy consumption per ton of glass produced
varies widely.
The type of glass being melted also affects the energy consumption of
a furnace. For example, glass color affects furnace efficiency. Most of
the heat transferred to the glass from the flame and refractories is by
radiation. However, as the color of glass changes, its absorptivity also
changes. Consequently, certain colored glasses actually receive less heat
from a given flame than others. Glasses that contain a high percentage of
silica (90% and above) generally require very high melting temperatures
(3000 °F). Consequently, energy consumption is quite high. However, very
little of this glass is produced because of the expense involved in building
and operating a facility at these high temperatures. Soda-lime glass,
which melts at considerably lower temperatures (2500°F), is the most
commonly melted glass. Theoretical energy requirements are about 2. 0
million Btu/ton of glass melted. Other types of glasses that are manufac-
tured in the United States are borosilicate, lead silicate, and aluminosilicate.
All these glasses require more than 2. 0 million Btu/ton to melt. However,
the changes in glass composition that normally occur during mixing of the
raw materials are not likely to cause changes in energy consumption.
Approximately 15% of the total energy consumed by the glass industry is
for annealing. Annealing-lehr efficiency is difficult to determine: It varies
depending upon many factors, but is generally about 25%. Efficiency is af-
fected by such factors as incoming glass temperature, lehr design, glass
loading pattern and burner operation. The leakage of unwanted cold air
into the lehr and the loss of heated air through unwanted openings also con-
tribute to cause lower than desired operating efficiency.
Recently constructed annealing lehrs are equipped with radiant gas burners
that eliminate open burning of natural gas in the lehr. These lehrs are gen-
erally more efficient than the older ones. As previously indicated, natural
gas is the preferred fuel for annealing from the point of view of glass quality,
economics, and efficiency of utilization. Oil can only be used in indirectly
fired lehrs since direct firing would result in discoloration of the final pro-
duct. Electricity is also used, but is more costly in terms of energy cost
and equipment.
46
-------
Air Pollutant Emissions in the Glass Industry
In addition to being the primary energy consumer in the glass industry,
the glass-melting furnace also is the primary source of air pollutant emissions.
The primary emissions are particulates, sulfur oxides (SOX, sulfur dioxide,
and sulfur trioxide), nitrogen oxides (NOX, nitric oxide, and nitrogen dioxide),
and carbon monoxide. Hydrocarbons are not a problem if proper combustion
conditions are maintained. Table 5 summarizes the emissions from several
glass melters as measured by a number of investigators.
Factors Affecting Air Pollutant Emissions '
Several factors influence the emission rate of particulates from a glass-
melting furnace, including batch composition, batch preparation, and type of
fuel. The production rate of the furnace also is a factor.
Measurements of stack emissions from a glass melter have shown that
the particulates emitted are primarily sodium sulfate, which is a minor in-
gredient of most glass batch. In the furnace, it vaporizes and decomposes
to form elemental sodium and sulfate. 5 When these gases pass through the
checker-brick and are cooled, sodium sulfate is re-formed. Only about 40%
of the sodium sulfate charged into the furnace is vaporized; the remainder
goes into the glass. In addition to the sodium sulfate, a small amount of
raw batch that is carried out of the furnace by the flue gases is emitted. This
emission can be minimized by proper batch preparation, which consists
primarily of wetting the material before charging it into the furnace.
The amount of SOX emitted from a furnace depends on 1) the sulfur con-
tent of the fuel and 2) the amount of sulfur-bearing compounds in the raw
materials. Consequently, natural-gas-fired furnaces generally exhibit lower
V, •**
SO... emissions than oil-fired furnaces unless the sulfur has been removed
X. ^j
from the oil. Measurements of SOX emissions from a batch melter charged
with batches having various sulfur contents showed a direct correlation be-
tween sulfur in the batch and SOX emitted. The greater the sulfur content
of the raw batch, the higher the SOX emissions.
The amount of NOX emitted from a glass-melting furnace depends upon
several factors, some of which are not understood. One important factor
is flame temperature: NO formations in the furnace increase as flame
X*
temperature increases. For example, during a recently completed
47
-------
Table 5. AIR POLLUTANT EMISSIONS FROM VARIOUS PRODUCTION GLASS MELTERS
00
CO
NO
Arrandale1
Netzley3
8% excess- air.
b 25$ to 45$ excess air.
Q
Excess air unknown.
Variable with production rate.
Natural gas fired.
Particulates,
2 to 10
2 to 10
Halogens SO
iuvts B kj.gcLiu.1. a
IGT2
Ryder and McMackin4
Stockham5
HI
35 to 50a
0 to 5b
375°
pm ' J.D / nr
490 to 700 6 to 8
450 to 600
340
1.0
--
7. 1
ppm
28
267
-------
experimental program, NOV emissions were measured during a complete
Ji
firing cycle of a glass melter. NO emissions were highest at the beginning
j£
of the firing cycle and then, as the cycle continued, decreased by about 30%.
At the beginning of the firing cycle, the combustion air is preheated to a higher
temperature, which results in a hotter flame than at the end of the cycle when
the checker-brick and hence the air have cooled considerably. Other major
factors in NO formation in a glass melter, such as flame velocity and re-
j£
circulation patterns of flue gases, are being studied.
Current Methods of Air Pollution Control
Methods of air pollution control currently in use in the glass industry are
primarily electrostatic precipitation (ESP) and baghouses for particulates
and the use of low-sulfur oil for SOX emissions. In terms of equipment costs,
ESP and baghouses are about the same upon installation. However, less
energy is consumed by ESP. To be effective, baghouses require a substantial
pressure differential, which creates a need for a substantial amount of horse-
power to move the particulate-laden air through the baghouse. In addition,
baghouses require more maintenance than ESP to be totally effective.
One of the problem areas faced by the glass industry in its attempts to
clean up the emissions is the variances in regulations that exist from one state
to the next. More than one company interviewed indicated that because of
these variations, different solutions must be implemented to bring two plants,
located in different states, but otherwise identical, into compliance. Thus,
in one state, baghouse systems may adequately control a comapny's particulate
emissions, while in a neighboring state, a process modification, the use of
electric melters instead of fossil-fuel-fired melters, is necessary. Such
variances are not only costly to a company, but they also may dramatically
affect energy utilization, as in the above example.
Other emissions, such as carbon monoxide and hydrocarbons, can be con-
trolled easily with proper combustion conditions. If opal or green glass is
being produced, halogens such as chlorine and fluorine also are emitted in
very large quantities from a fossil-fuel melter. However, the industry has
converted completely to electric melting, and this switch has eliminated these
emissions.
49
-------
Trends in Process Modifications in the Glass Industry
According to the glass industry, several process modifications exist for
potential implementation by the industry sometime in the future. These modi-
fications -would affect energy utilization and/or air pollution emissions and
are as follows:
1. Expansion of process monitoring and control capacity
2. Electric melting
3. Electric boosting
4. Oxygen enrichment
5. Raw batch preheating
6. Raw batch agglomeration
7. Use of low temperature heat to drive compressors.
8. Augmentation of heat transfer from flames
9. Submerged combustion.
Of these modifications, only the first three are considered by the industry to
have potential for implementation in the near future. These three techniques
are economically feasible and their technical feasibility has been demonstrated
to the satisfaction of the industry, and implementation is already occurring.
According to the glass industry, the latter modifications listed are generally
considered to be economically unattractive or technically unfeasible in spite
of published data to the contrary. The following discussion presents a brief
description of the modifications involved and the industry's viewpoint con-
cerning the implementation of each.
Expansion of Process Monitoring and Control Capacity
According to persons interviewed within the industry, there are several
modifications that can be made in the area of process monitoring and control
that will favorably influence the utilization of energy. One of these modifica-
tions is the use of improved temperature-sensing devices for continuous process
monitoring. For example, infrared sensors focused on critical areas of the
melter, such as the optical block on the bridgewall, are used to gauge melter
performance. These sensors can be used not only to continuously monitor
melter temperature, but their signals can be used to control fuel input, based
50
-------
on melter temperature. Another such modification is the use of flue-gas
analyses to monitor excess air and maintain it at a minimum level. None
of these monitoring techniques is expensive, and all of them would contribute
to improving the efficiency of operation.
Electric Melting (and Boosting)
The glass industry strongly supports the implementation of electric melting
and boosting primarily for reducing air pollutant emissions. Electric melting,
as a method for producing glass, has proven to be technically and, in most
cases, economically feasible by its relatively widespread use in the industry.
(Actually, electric boosting, where a fuel-fired melter is supplemented by
electric melting, is very popular and more prevalent within the industry than
straight electric melting.)
Oxygen Enrichment
Oxygen enrichment is a technique whereby pure oxygen is added to the
combustion air of a fuel-fired melter, resulting in an increase in flame tem-
perature. This, in turn, results in a reduction in the fuel required to melt a
ton of glass, or, alternatively, allows a melter operating at design capacity
to boost its production above design capacity. Based on the results of the
interviews with glass manufacturers, oxygen enrichment is a long-term goal,
primarily because acceptance by the industry requires substantial changes
in fuel and oxygen costs to economically justify implementation.
Raw-Batch Pretreatment
The area of raw-batch pretreatment includes batch agglomeration, or
compaction, and preheating of the batch prior to charging into the furnace.
Most of the companies interviewed feel that batch agglomeration is not eco-
nomically justified within the near term for reasons of reduced energy utili-
zation or reduced particulate emissions. However, at least one major glass
manufacturer has recently put into operation several pelletizing lines to
supply pelletized batch to the melters. There is no detailed data available
on this operation at the present time.
Coupled with the compaction of glass batch is the idea of preheating the
glass batch prior to charging it into the melter. It is clear that such a pro-
cess must be coupled with a batch compaction process in order to minimize
batch losses during preheating and to minimize particulate emissions from
51
-------
the batching operation. The industry contends that preheating of the batch
will cause it to become sticky, making the charging operation next to impos-
sible.
Submerged Combustion
Submerged combustion is a melting process whereby the fossil-fuel burner
is located beneath the molten glass surface and the hot combustion products
pass through the glass resulting in a very high rate of heat transfer from the
gases to the glass. Because the product from this type of melter is foam
glass — that is, it contains millions of air bubbles — it is unacceptable to the
industry without substantial refining. The only possible use for submerged
combustion would be in a premelter, which at least one company has imple-
mented. However, because the refining step requires a substantial amount
of energy (more than usual), it is not certain that there is a reduction in the
overall amount of energy consumed to melt the glass. Consequently, substan-
tial development is still required before it will become acceptable to the indus-
try, thus making implementation long-term, at best.
Augmentation of Heat Transfer From Flames
This is a rather nebulous area for consideration. There are potentially
many things that can be done to improve heat transfer from the flame to the
molten glass. One approach is the use of devices that allow an operator to
accurately and precisely position his burners. A second approach is the
injection of water vapor into the flame, which theoretically increases the
radioactive properties of the flame, resulting in an increase in the heat-transfer
rate. The entire area of augmenting heat transfer has great potential, according
to the industry, but implementation of most of the developments is deemed to
be long-term.
Use of Low-Grade Thermal Energy
At least one company interviewed said that waste heat from the melting pro-
cess, most of which is below 1000°F and consists of 20% of the energy that
goes into melting the glass, could be used to directly drive turbines for air
compressors, which would then be used in the blowing operations. Alterna-
tively, but not as efficient, was the suggestion that this hqat be used to drive
turbines in the generation of electricity. While such practices are not cur-
rently used, with some development, usage might increase in the long-term.
52
-------
Improvements in Equipment Design
Because of the rate at which equipment is replaced within the glass industry,
implementation of improvements in equipment design are considered long-term.
Improved energy utilization is expected from the application of better insulating
techniques, improved regenerator design, and improved firing patterns by
burner placement. Annealing-lehr efficiency is affected by numerous design
considerations. Among the most prominent considerations are the use of light-
weight lehr belts, the method of belt return, proper insulation of the heating
zone, use of radiant burners in the heating coal sections, and design to prevent
forward drift. Plant layout to minimize transit time of the glass between the
forming machine and lehr is also important. As indicated, most of these de-
sign modifications are considered long-term, in terms of implementation and
impact.
It is clear from the above discussion of process modifications that the glass
industry feels that there is very little that can be done in the short-term to ease
the energy shortage and air pollution emissions problems. With the exception
of the installation of process monitoring equipment, the process modifications
discussed are generally deemed to be long-term in implementation. Even in
the area of process monitoring equipment, the opinion was expressed during
the field interviews that the available equipment is not adequate to meet the
demands of continuous operation in the climate of a typical glass plant.
Trends in Energy Utilization in the Glass Industry
In 1971, the latest year for which industry statistics covering energy utili-
zation are available, a total of approximately 250 trillion Btu of energy was
consumed by the glass industry. The breakdown according to the type of energy
consumed was as follows: coal, 1. 6%; fuel oil, 4. 5%; natural gas, 88. 7%;
and electricity, 5. 2%. Natural gas is the preferred fuel because it is clean
burning and will not affect glass characteristics, such as color. Furthermore,
the natural gas flame, when properly controlled, provides a longer furnace
life than other fuels, primarily fuel oil.
/' __i
Typically, fuel oil is a standby fuel for most major manufacturers. Con-
/ '
version to oil firing is not an extensive undertaking* requiring only that
burners be changed and certain control equipment modified. Nevertheless,
investment costs for oil-fired furnaces are substantially higher than for
53
-------
natural gas. This is due in part to the requirement for a fuel oil storage
facility and the requirement for equipment to heat the oil to reduce its vis-
cosity and thus increase the efficiency of combustion. The industry is being
forced to convert to fuel oil on the melters because of the curtailment of
natural gas, particularly during the winter months. Some companies have
experienced up to 100% curtailment of natural gas in some states. They
believe that conversion to fuel oil on a large scale will occur in the near future.
Inevitably, the deciding factor in establishing fuel^-use trends is cost. Cur-
rently, the cost of oil suitable for use in the melting operation is more than
double the cost of a comparable amount of natural gas. In order to meet the
requirements of glass quality and air pollutant emissions, the level of sulfur
present in the oil must be less than 1%, adding substantially to the fuel price.
Clearly, as the cost of natural gas continues to increase, the trend toward
the use of fuel oil as the primary fuel (rather than a standby fuel) will increase.
As indicated above, 5. 2% of the energy consumed by the glass industry is
electricity. Electricity is consumed primarily in the melting process in the
glass container industry. It is used as a production-boosting technique in
conjunction -with a fuel-fired melter or as the primary energy for melting.
The amount of glass so produced is less than 500,000 tons per year. The
majority of the electricity is used for mechanical drives, compressors, and
similar applications. In the flat glass industry, where substantial amounts
of power were required for grinding and polishing the glass, the conversion
to the float glass process has shifted the use of electricity to a heating appli-
cation — i. e. , maintaining the tin bath in a molten state. Because electricity
is such an expensive form of energy, it is not likely that it will replace the
use of fossil fuels, as in electric melting. However, the use of electricity
:. »
may increase in the flat glass industry as the use of the float process increases.
Reluctance on the part of the companies interviewed to give out any energy-
utilization figures on the float process is the cause for the uncertainty in es-
tablishing this trend.
Consideration of these factors leads to the conclusion that the trend in fuel
utilization, according to the industry, is toward increased use of fuel oil as
the primary fuel for melting. The consumption of electricity can be expected
to increase in the annealing operations, particularly in the flat glass industry.
Current projections for the availability of natural gas make its continued use
54
-------
by the industry unlikely in the long-term — if not impossible. Coal has never
been an acceptable fuel to the industry and is not expected to assume any of
the glass-manufacturing load.
Trends in Air Pollutant Emissions From Glass-Manufacturing Processes
According to representatives of the glass industry, the efforts of the industry
to reduce air pollutant emissions are severely hampered by the variations in
regulations that exist from state to state. This lack of uniformity requires
that different solutions to the problem be sought, depending on the location of
the specific plant. This, in turn, adds substantially to the cost of pollution
control. In addition, not only are the regulations variable from one location
to another, but these regulations are constantly changing. As a result, very
few air pollution control equipment installations have been made on glass
furnaces, and there is very little data available in the effectiveness and cost
of these devices. Estimates of energy cost for air pollution control range
from 2% to 15% of the energy input to the industry. Clearly then, the effort
to reduce emissions to within regulation through the installation of pollution
control equipment is significantly hampered by all of the factors above.
Similarly, the industry is moving very slowly to implement process
modifications to reduce emissions. The primary area is the modification of
the batch formulation and batch pretreatment because the batch is the major
source of pollutant emissions in the glass industry. Particulate emissions
are the primary concern. No attention is being given to NO emissions at
Ji.
the present time. There are process modifications that should be considered
for energy conservation, which will not necessarily result in an increase in
NO emissions. Currently, NO emissions are not considered to be a
•A* •"•
problem, primarily because there are no regulations relative to the glass
industry. Similarly, SO emissions, except as they contribute to the
•X
emission of particulates (as sodium sulfate), are not considered to be a,
serious problem. Since the primary source of these emissions is the fuel,
the SO emissions will be a direct function of the amount of sulfur in the
JL.
fuel. Otherwise, SO emissions will not be affected by process modifications.
j£
Analysis
The data collected during this program indicate that the glass industry feels
that only minor increases in energy-utilization efficiency can be made and that
55
-------
the only existing pollution problem is particulates, which are being controlled,
but at a substantial economic and energy-utilization cost. Clearly, economic
considerations are the primary factors in this evaluation. Some discrepancy
appears to exist due to a lack of accurate and reliable information concerning
the relevant areas, thus demonstrating the critical need for informing the in-
dustry in these areas.
In general, SOX, NO , and particulates are the primary air pollutants from
the glass manufacturing processes. Note that the concern is primarily with
the melting process because this is the largest energy consumer and the major
contributor to air pollutant emissions. However, the major pollution problem
in the combustion process and efficient utilization of energy is NO... emissions.
Jt
The SOX emissions which occur are due to the type of fuel used and they can
be reduced merely by using low-sulfur fuels. SO emissions are not affected
X,
by the process modifications to reduce fuel consumption, other than the small
reduction that would inherently occur by reducing fuel consumption. Particu-
late emissions cannot be viewed as arising in the combustion process because
the primary fuels used in the industry, fuel oil and natural gas, do not produce
particulates as effluents, as would coal, which is not used by the glass industry.
The relationship between these emissions and energy utilization by the industry,
can be seen by looking at the control method employed to determine the effect
on energy consumption. (That is, how much energy is required to operate the
control device?) It is not likely that operation of the control device will affect
the energy utilization of any particular process. Process modifications to re-
duce energy consumption will not have any significant impact on these emissions.
Care should be taken when analyzing process modifications and pollution control
methods that consume electricity because it is necessary to look at the gener-
ating power plant to determine the true process energy consumption and air
pollutant emission rates in the manufacturing plant. Thus, the only emission
of concern in the glass industry is NO . While the formation of NOX in the
combustion process is not entirely understood, it is clear that the goals of re-
ducing NO emissions and reducing energy consumption are seemingly at odds.
X.
NO formation is a temperature-related phenomenon; as temperature increases,
ji
NO... emissions increase. On the other hand, increasing available heat to a
Ji
process may result in increases in temperature, which in turn increase NOX
emissions. Analysis of the process modifications under consideration in the
56
-------
glass industry- shows that there is a possibility of increasing NO emissions,
Ji.
but if the implementation is carried out properly, this need not occur.
By way of example, oxygen enrichment as a technique for boosting produc-
tion rate and reducing fossil-fuel consumption has been proven to be very ef-
fective. The effectiveness is due to the increase in flame temperature that
occurs as the percentage of oxygen in the combustion air increases and the
percentage of nitrogen decreases. This increased temperature can then re-
sult in an increase in NO emissions. However, boosting the production rate,
assuming no increase in fuel input, requires that this additional heat be taken
from the flame and transferred to the glass. The net result is a flame at pre-
oxygen enrichment temperature levels. If there is no need for the additional
production, although uneconomical, the fuel input can be reduced to achieve
the same net result.
Modification of the batching process, specifically, raw batch preheating
and batch agglomeration, will have no impact on NOX emissions on a per unit
of energy basis. However, NO emissions are likely to be reduced due to an
jSi
overall reduction of energy consumption. If, however, these modifications
were coupled with the concept of submerged combustion, both fuel consump-
tion and NOX emissions per unit of energy consumed would decrease. Modi-
fication of the batching process would be required, particularly implementation
of batch agglomeration, in order to prevent particulate emissions in the form
of raw batch being carried out in the combustion products. The problem is
that submerged combustion is not fully developed as a viable process because
the product, foam glass, is unacceptable to the industry in that it fails to
meet the quality standards for virtually any use.
Only brief mention has been made concerning the analysis of fuel consump-
tion and air pollutant emissions relative to the use of electricity. Electric
melting and boosting are believed to require less energy than straight fossil-
fuel melters and are believed to virtually eliminate air pollutant emissions.
These beliefs are based on operating data that show an allfelectric melter
/i
to consume about 3 million Btu/ton of glass, compared with 6 million Btu/ton
for a comparable fuel-fired melter producing the same type of glass. Operating
data also indicate that because there are no combustion products, there are no
NO , SOX, or particulates given off. However, if the generating of that elec-
tricity is considered, the actual amount of energy consumed for melting is
57
-------
higher than for a straight fuel-fired melter. The data for air pollutant emis-
sions are not clear but electric melting is not pollution-free.
If the industry heads in the direction that it is going for the short-term,
that is, implementation of process monitoring and control devices, fuel con-
sumption rates will decrease by 5% to 10% with no effect on pollutant emis-
sions. For the long term, in spite of the great potential, it is not likely that
major process modifications will be implemented by the industry. Consequently,
the potential for reducing emissions and energy consumption is minimal.
As indicated above, the trend in fuel utilization is toward increased use of
fuel oil and away from natural gas. In addition to the effects of this trend on
particulates and SOX emissions, there is likely to be a dramatic increase in
NOV emissions. Although there is only a small amount of data on NO... emis-
X. X-
sions from glass-manufacturing processes, research on the formation of NO
in combustion processes shows two major contributing factors to the formation —
thermal factors in the furnace and the amount of nitrogen in the fuel. In the
case of natural gas, where there is nop fuel-bound nitrogen, NOX emissions are
due to thermal factors. But in the case of liquid or solid fuels, not only is
NO., produced by these same thermal factors, but additional NOV is produced
jf. t a
because of the presence of nitrogen in the fuel. Thus, an increase in the use
of fuel oil by the glass industry will increase NOX emissions.
Re c ommendations
As previously indicated, the overwhelming viewpoint of the glass industry
is that only minimal reductions in energy consumption can be achieved. With
respect to emissions, the industry feels that, considering the great variances
in regulations from state to state, it is doing an adequate job. Thus, parti -
culate emissions are seemingly well controlled with attempts to reduce SO
jfL
emissions by using low.-sulfur fuels. In the absence of an established stan-
dard, applicable to industry processes, the industry is not actively concerned
with NO emissions. As a result, if most industry pollution control recom-
Jt
mendations were implemented they would not help to reduce NO emissions.
Ji
With this in mind, the recommendations that follow are intended to inform
the industry. We are recommending, first, that programs be undertaken to
determine what the true facts are in areas where data is limited. Specifically,
there is only a minimal amount of data available on NOX emissions from glass-
manufacturing processes —most of it scattered and unreliable. Before any
58
-------
affirmative steps can be taken to deal with this problem, a base-line case
must first be established.
Having established a base-line case, the next step is to support programs
demonstrating and developing new technology that is applicable to the industry
and that provides the greatest potential for solving the problems. Table 6
lists the program areas that should be considered. The order of listing is
according to programs that afford the greatest potential for solving the problems
in the shortest period of time. The table also presents estimates of improve-
ments that may be obtained, where such estimates can reasonably be made.
The programs in these recommended areas should be devoted to developing
data, including operating and economic statistics. This data should then be
used to convince the industry of the feasibility of these concepts, providing
incentive for implementation. Clearly, in the case of submerged combustion,
the programs required are more in the way of development. At its present
state of development, submerged combustion is virtually unsuitable for use
by the industry. But in light of the potential gains to be made, programs to
further develop this concept are easily justifiable.
References Cited
1. Arrandale, R. S. , "Air Pollution Control in Glass Melting, " Proc. Symp.
Glass Melting, 619-44 (1958).
2. Nesbitt, J. D. , Larson, D. H. and Fejer, M. , "Improving Natural Gas
Utilization in a Continuous End Port Glass-Melting Furnace, " in Proceedings
of the Second Conference on Natural Gas Research and Technology, Session
IV, Paper £. Chicago: Institute of Gas Technology, 1972.
3. Netzley, A. B. and McGinnity, J. L., "Glass Manufacturing, " in Danielsen,
J. A. , Air Pollution Engineering Manual, U. S. PHS Publication No. .'999-AP-40.
Washington, D. C. : U. S. Government Printing Office, 1967. ""•
4. Ryder, R. J. and McMackin, J. J. , "Some Factors Affecting Stack Emissions
From a Glass Container Furnace, " Glass Bid. 50, 307-10, 346-50 (1969)
June and July.
5. Stockham, J. D., "The Composition of Glass Furnace Emissions. " Paper
No. 70-22 presented at the 63rd Annual Meeting of the Air Pollution Control
Association, St. Louis, June 14-19, 1970.
59
-------
Table 6. RECOMMENDED PROGRAMS FOR REDUCING EMISSIONS
AND ENERGY CONSUMPTION IN THE GLASS INDUSTRY
Program
1. Develop current emission data
2. Raw batch pretreatment — i. e.
preheating and agglomeration
3. Oxygen enrichment
4. Augmentation of heat transfer
from flames — e. g., burner
positioning
5. Use of low-temperature heat
for driving compressors
6. Development of submerged
combustion process
Expected Improvements in
Energy Consumption, %
25-50
5-15
10-20
50
Expected Improvements in
Air Pollutant Emissions
25-50% potential NO_ reduction
may reduce particulate in form
of batch carry-over
No effect on NOX, SOX, or parti-
culate s
Proportional NO reduction
Will substantially reduce NO
emissions
-------
Cement Industry
Summary
The cement industry includes all establishments engaged in the manufac-
ture of hydraulic cement (generic term: portland cement), masonry, natural,
and pozzolana cements. This study is limited to the production of portland
cement because it accounts for 95% of the total cement manufactured in the
United States, with the remaining 5% split among the other types.
Over the past 20 years, the cement industry has been growing at a steady
annual rate of about 1. 8%. According to industry estimates, this growth rate
is expected to increase, ranging from 2. 6% to 4. 1% annually. It is reasonable
to assume that actual production up to 1985 will fall within the range shown in
Figure 5. Total U.S. cement production in 1973 is estimated at 81. 8 million
tons. It is expected to reach a minimum of 97 million tons in 1980 and 109
million tons in 1985.
The amount of energy consumed by the industry in 1973 is estimated to be
573 trillion Btu, based on average unit energy consumption of 7 million Btu/
ton of cement produced. The primary sources of energy are natural gas, oil,
coal, and electricity. Figure 6 illustrates energy consumption by type up to
1972, excluding electricity, and indicates that coal and natural gas are used
more than fuel oil.
The industry is composed of 50 companies that operate 173 plants in the
United States, including Puerto Rico. The annual practical production capa-
city, based on theoretical production capacity adjusted downward about 10% to
reflect maintenance shutdowns, is in excess of 90 million tons of finished
cement per year. The number of plants operated by an individual company
may range from 1 to 14, and no single producer accounts for more than
7. 3% of the total U. S. market.
Portland Cement Manufacturing Processes
Raw materials used in the manufacture of portland cement consist of
limestone, chalk or marl, and seashells. These are combined with either
clay, shale, slate, blast furnace slag, iron ore, or silica sand. The end
product is a chemical combination of calcium, silicon, aluminum, iron, and
other trace materials. The raw materials are first ground and blended to-
gether. Depending upon which of the two processes is used, water may be
61
-------
IfU
135
130
125
120
115
no
105
100
95
90
85
80
75
70
65
60
55
50
45
40
19.
1
/
/
'
S"
I
J
/
r-
j
i
/
/
i
i
.
j
' /
/
1
1
1
t
/
/
/
50 '55 '60 '65 '70 '75 '80 196
YEAR
A76020374
Figure 5. ANNUAL CEMENT PRODUCTION
WITH PROJECTION TO 1985
62
-------
O
70
65
60
55
50
45
40
35
30
25
20
15
10
-COAL
S
GAS-
1946 '48 '50 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 1972
YEAR
A-44-669
Figure 6. FUEL USAGE OF U. S. CEMENT INDUSTRY
63
-------
added during blending (the wet process) or the ingredients can be mixed oh
a dry basis (the dry process). In general, the moisture content of the raw
materials determines the process used. If the moisture content is greater
than 18%, by weight, the wet process will be used. If the moisture content
is less than 18%, the dry process will be used. The next step is the cal-
cining or burning of the mixed raw material in a rotary kiln (pronounced
"kill"). During this step, the material is heated to approximately 2700 °F
(1500°C) and transformed into clinker, which has different chemical and
physical properties than the raw materials had initially. The clinker is dis-
charged from the kiln and cooled. The last step is the grinding of clinker to
the desired fineness, and gypsum is added to control the setting time of the
concrete.
Raw Material Preparation
The step that differentiates the wet process from the dry process is th4
initial grinding of the raw materials. Both processes are shown in Figure 7.
In the wet process, the raw materials are properly proportioned and combined
•with water to form a slurry. The slurry may be as much as 50% water. After
mixing, the slurry is dewatered to a 20% to 30% water level in a continuous
separator consisting of a large tank in -which the solid material settles to the
bottom, and water is drawn off at the top. The thickened slurry is then me-
chanically discharged from the bottom of the separator and is stored or is
transferred to the rotary kiln. In the dry process, the raw materials are
simply ground and proportioned, without the addition of water, and passed
on to the next processing step.
Raw Material Burning
Burning is the heating operation in which the raw materials are burned to
form a new product called clinker. The product is grayish-black in color,
about marble size, and has different chemical and physical properties than
the raw materials. The raw materials are charged continuously into a cylin-
drical rotary kiln, which turns at about 1 revolution per minute. The kiln
varies in diameter — generally 12 to 14 feet. It is several hundred feet in
length and lined with firebrick. Some kilns may be as large as 25 feet in
diameter and 750 feet long. They are mounted at an angle slightly off hori-
zontal, so that as they rotate, the raw material moves countercurrent to the
flow of hot gases. The raw materials are charged into the high, charging
64
-------
IAOI 1A« lUTUUt
B sraao unumr
ww (uiauu camm w
euuautiam o» UMIITOMI,
x. MAR at OTITII IMBIS.
> «*», cur, u«o. o» now oa
1 Stone Is first reduced to 5-in. size, then %-hn, and stored.
is
jv TO *a g»A**ro« •
f7 ^rAAA/ ;
3i ga._^fci a&
-------
end to the low, discharging end. A hot flame is produced at the discharge
end by the controlled burning of natural gas, oil, or coal. The raw materials
are heated in a series of three stages.
As the raw material enters the first stage, the moisture is driven off. In
the second stage the temperature gradually increases to about 1800°F (980 °C),
and the carbon dioxide is driven off. The third stage, the hottest, is where
the raw material burns to form clinker. The temperature of the third stage is
about 2700 °F (1500°C). The resulting clinker is cooled upon discharge and
stored until needed.
Clinker Processing
When needed, the clinker is ground to specification, a fine powder 90% of
which will pass through a screen with 40, 000 openings per square inch. At
this point, gypsum is also added for the purpose of controlling setting time of
the concrete -when it is mixed. The finished product is then sold to concrete
companies that mix the portland cement -with aggregate of various types to
form concrete. The cement manufacturer has no control over the composi-
tion of the aggregate with which the cement is mixed.
Cement Industry Energy Requirements
In 1972, there were a total of 454 cement kilns in operation in the United
States. Of this total, 250 (55%) were in wet-process plants, and 204 (45%)
were in dry-process plants. By year-end 1973, the number of kilns in opera-
tion decreased to 445, of which 247 were in wet-processing plants and 198
were in dry-processing plants. The average annual capacity per kiln, both
wet and dry, is about 194,000 tons/yr. The average age of all kilns is 25
years, and 110 (24%) are at least 40 years old.
In 1972, 79 million tons of cement was domestically produced, and in 1973,
domestic production rose to 81. 8 million tons. Using the estimate of 7 million
Btu/ton of cement, the industry consumed approximately 553 trillion Btu in
1972 and 573 trillion Btu in 1973. The energy consumed was in the form of
electricity, natural gas, oil, and coal. Electric power is used in all cement
plants for motors, to activate crushing and grinding equipment, for pollution
control equipment, and for general lighting. Fossil fuels, coal, oil, and
natural gas are used directly in the rotary kiln to transform the raw materials
into clinker. The information given in Table 7, as provided by the Portland
66
-------
Table 7. U.S. CEMENT INDUSTRY KILN FUEL USAGE (1973)
Type of fuel jjo. of plants* Percent of capacity
Coal 52 30
Oil 13 8
Natural gas 26 ' 10
Cos!, oil 7 5
Coal, natural gas 32 17
Oil, natural gas 38 .___ 26
Coal, oil, natural gas 5 4 .
Totals 173 100%
Source: Portland Cement Association, Market and Economic
Research Department.
'Number of plants includes all clinker-producing units, with
white and gray plants at same site counted as two plants.
Grinding-only plants are'not included.
Cement Association, shows the breakdown of fuel consumption on the basis
of plant distribution and kiln fuel use in 1973.
The energy consumption of the various stages of cement processing is
shown in Table 8.
Table 8. ENERGY CONSUMPTION IN CEMENT PROCESSING
Crushing, Finish
Process/ Milling, Milling,
Fuel Type Quarry Mixing Burning Packing Total
1000 Btu/ton of cement
Wet:
Electrical 8.2 113.5 85.9 212.8 420.4
Fuel 16.0 -- 5560.0 -- 5576.0
24.2 113.5 5645.9 212.8 5996.4
Dry:
Electrical 5.8 154.4 94.5 228.6 483.3
Fuel 16.0 600.0 4600.0 ^ 5216.0
21.8 754.4 4694.5 228.6 5699.3
Electricity is the major form of energy consumed in the initial steps of
cement manufacturing — quarrying, crushing, milling, and mixing — providing
mechanical energy to operate the crushers, grinders, and blenders. Prior
to the development of ball-and-tube mills, the dry process required more
67
-------
electrical energy than the wet process because the power requirements were
greater for crushing and grinding the harder raw materials. The fuel re-
quirements for the dry process are greater than for the wet process because
the raw materials must be dried to remove all the moisture content, which
can be as high as 18%. The amount of fuel used for drying is about 600, 000
Btu/ton of cement produced. The amount of electricity consumed for the wet
process is about 121. 7 Btu/ton of cement and about 160. 2 Btu/ton of cement
for the dry process. If electrical generating is considered (assuming gener-
ating efficiency of 33-1/3%), the Btu consumption is equivalent to 365. 1 and
480. 6, respectively, per ton of cement produced. The additional require-
ments for the dry process are attributable to increased horsepower for fans
and dust control equipment. ;
Although the average energy consumption for the burning process is about
6. 4 million Btu/ton of cement, actual consumption can range from 3. 7 to 9. 0
million Btu/ton. The information given in Table 8 is for newer, well-
maintained plants (less than 25 years old); the average energy consumed in
the wet process is 5. 6 million Btu/ton and 4. 7 million Btu/ton in the dry pro-
cess. The amount of energy consumed is determined by many factors, such
as chemical and mineral composition of the raw materials, particle-size dis-
tribution and fineness of the charge, length pitch and speed of kiln rotation,
and the efficiency of combustion and heat-recovery systems.
The most pronounced difference between the wet and dry processes is in
the burning stage. As shown in Table 8, the difference between the wet and
dry processes is 951,400 Btu/ton of cement. The wet process requires about
20% more energy for burning than the dry process because of the need to
vaporize the additional water.
As shown in Table 8, approximately 212,800 Btu of energy is consumed
in the final stage of the wet process and 228, 000 Btu in the dry process.
Although no revolutionary changes have occurred in the manufacture of
Portland cement in the last 60 years, improved efficiency and reduced fuel
consumption have been brought about on an evolutionary basis. The U. S. in-
dustry, however, had not been taking advantage of these changes until the last
year or two. The remainder of the world's cement producers have been con-
verting from the wet process to the dry process, which consumes less energy
per unit of production.
68
-------
The primary reason for this is that fuel has been cheap in the United States,
as compared with the cost of fuel in other countries. The trend toward dry
processing is beginning to occur in the United States, brought about by in-
creased fuel shortages, particularly natural gas and oil. At the present time,
approximately 45% of the fossil fuel energy consumed by the cement industry
is natural gas, 40% coal, and 15% oil.
Air Pollutant Emissions From Cement-Manufacturing Processes
The major air pollutant emission problem in the manufacture of portland
cement is particulates, which occur in all phases of cement manufacturing
from crushing and raw material storage to clinker production, clinker grind-
ing, storage, and packaging. However, emissions also include the products
of combustion of the fuel used in the rotary kilns and drying operations;
these emissions are typically NOX and small amounts of SO . Table 9 sum-
marizes the emissions from cement-manufacturing processes without con-
trols.
The largest source of emissions in cement plants is the kiln operation.
At present, about 56% of the cement kilns in operation use the wet process,
and 44% use the dry process. Based on this information, estimates of total
emissions from cement plants in 1972 are given in Table 10. These estimates,
because of a lack of data, assume the use of no controls by the industry, and
without an inventory of equipment in use, they cannot be refined.
Most efforts to control air pollutant emissions focus on particulates be-
cause they are not only the greatest problem, but also the easiest to control.
The most desirable method of control is to collect the dust and recycle it
by injecting it into the burning zone of the kiln, thus converting it to clinker.
However, because of the high alkali content of the dust and its potentially bad
effects on product quality, only a limited amount of dust can be recycled.
Because of the complications of kiln burning and the large volumes of
gases and materials that are handled, several dust collection systems have
been developed. Depending upon the temperature of the effluents and the
emission standards of the community, the industry uses mechanical collec-
tors, electrical precipitators, baghouse filters, or combinations of these
devices to control emissions. Typically, in treating the flue gases from the
kiln operation, mechanical separators are normally required ahead of either
69
-------
Table 9. EMISSION FACTORS FOR CEMENT
MANUFACTURING WITHOUT CONTROLS*' °
Pollutant
Paniculate*
Ib/ton
kg/MT
Sulfur dioxided
Mineral source*
Ib/ton
kg/MT
Gas combustion
Ib/ton
kg/MT
Oil combustion
Ib/ton
kg/MT
Coal combustion
Ib/ton
kg/MT
Nitrogen oxides
Ib/ton
kg/MT
Dry Process
Kilns
245.0
122.0
10.2
5.1
Negf
Neg
4.2S9
2.1S
6.8S
3.4S
2.6
1.3
Dryers,
grinders, etc.
96.0
48.0
-
-
-
-
-
-
-
-
-
-
Wet process
Kilns
228.0
114.0
10.2
5.1
Neg
Neg
4.2S
2.1S
6.8S
3.4S
2.6
1.3
Dryers.
grinders, etc.
32.0
16.0
-
-
-
-
-
-
—
-
-
-
aOne barrel of cement weighs 376 pounds (171 kg).
''These emission factors include emissions from fuel combustion, which should not be calculated
separately. • .. ...
cTypical collection efficiencies for kilns, dryers, grinders, etc.. are: multicyclones, 80 percent;
electrostatic precipitators, 95 percent; electrostatic precipitators with mult icy clones. 97-.S
percent; and fabric filter units. 93.8 percent.
*The sulfur dioxide factors presented take into account the reactions with the alkaline dusts
when no baghouses are used. With baghouses. approximately 50 percent more SOj u removed
because of reactions with the alkaline paniculate filter cake. Also note that the total SOy from
the kiln is determined by summing emission contributions from the mineral source and the
appropriate fuel.
'These emissions are the result of sulfur being preiont in the raw materials and are thus depend-
ent upon source of the raw materials used. The 10.2 Ib/tgn (5.1 kg/MT) lectori account for
part of the available sulfur remaining behind in the product' because of its aikzline nature and
_ affinity for SO2-
1 Negligible.
9S is the percent sulfur in fuel.
70
-------
Table 10. ESTIMATED TOTAL EMISSIONS FROM CEMENT PLANTS IN 1972
Dry Process Wet Process
Pollutant Kilns Dryers, etc. Kilns Dryers, etc. Total
Particulates, tons 4.5 X 106 1.8X106 5.3 X 106 0. 7 X 106 12.3 X 106
Sulfur Dioxide, tons* 0.19 X 106 — 0.24 X 106 -- 0.43 X 106
Nitrogen Oxides, tons 0. 047 X 106 -- 0. 060 X 106 -- 0. 107 X 106
Does not include emissions due to sulfur in fuel because these numbers depend on the sulfur content
of the fuel, which is variable.
-------
precipitators or glass bag collectors. However, wet-process kilns, which
have lower exit-gas temperatures than dry-process kilns and relatively high
moisture in the exit gases, are commonly followed directly by the precipita-
tors because the moisture contributes to more satisfactory precipitator oper-
ation.
Efficiency of particulate removal depends upon the types and combinations
of equipment used. Typically, multicyclone filters have an efficiency of 80%,
and electrostatic precipitators have an efficiency of 95%. A combination of
electrostatic precipitators and cyclones has an efficiency of 97.5%. Fabric
filters (baghouses and the like) operate at a 99. 5% removal efficiency. From
this information, adequate equipment for particulate control apparently is
readily available.
SO emissions result from burning the raw meal in the kiln. There are
Ji,
two sources of emissions in the kiln: sulfur in the raw material and sulfur
in the fuel. The amount of sulfur present varies from plant to plant and with
geographic location. Most of the sulfur dioxide emissions are inherently con-
trolled in the process of cement manufacturing because about 75% of the raw
feed is converted to calcium oxide, which reacts with sulfur dioxide. In addi-
tion, the presence of sodium and potassium compounds in the raw material
aids in the direct absorption of sulfur dioxide into the product. However, the
variable chemistry and operating conditions in U. S. cement plants affect the
amount of sulfur dioxide entrapment and, in some cases, the quality of the
product. Sulfur dioxide entrapment of this type seems to be about 75% in
those plants for which data are available. Sulfur dioxide also is removed by
this same mechanism in baghouse filters, in which the sulfur-dioxide-laden
gases contact the collected cement dust. The degree of control by sulfur
dioxide absorption depends upon the alkali and sulfur content of the raw ma-
terials and fuel.
Trends in Industrial Process Modifications as Estimated by Industry
Several developments involving the modification of the equipment used
in the manufacture of portland cement have improved energy utilization. Al-
most all of these developments are centered around the rotary kiln and the
burning process. To reduce the amount of energy consumed in this process,
the heat-transfer rate from the combustion gases to the charge must be in-
creased. The modifications can be classified into near-term, 1 to 5 years,
72
-------
and long-term, 5 to 10 years, as far as implementation is concerned. Near-
term modifications consist of:
• Chain systems
• Enlargement of kiln feed end
• Oxygen enrichment
• Waste-heat recovery
Chain Systems
Chain systems can be utilized in both wet and dry process kilns. Chains
are installed at the feed end of the kiln for the purpose of absorbing heat from
the gas stream and transferring it to the raw material mix as the kiln revolves.
As the kiln rotates, the chains emerge from the raw material mixture, thus
providing a larger surface-area contact with the hot gases. Estimates of fuel
savings derived from the implementation of chain systems are shown below
in Table 11, and in Figure 8.
Table 11. FUEL SAVINGS FOR KILNS WITH CHAIN SYSTEM
COMPARED TO KILNS WITHOUT CHAIN SYSTEM
Lb of chain per Fuel savings— Btu per ton of clinker
daily ton of clinker _ Wet Long dry
100 .................................. .................. 1,150,000 500.000
200 ........................ .;. ......................... 2.100.000 1.000.000
100 Difference 950,000 500,000
In order to achieve the maximum benefit of a chain system it must be
carefully maintained. The effectiveness of the installation is reduced in
direct proportion to the amount of reduction in chain surface area. For ex-
ample, a reduction in chain weight equivalent to 25% will reduce the heat
absorbing capability by 25%. Chain systems are designed for a specific kiln,
taking into consideration structural design along with material and tempera-
ture profiles.
Enlargement of Kiln Feed End
Seventy-five percent of the existing rotary cement kilns in the United States
are of a single diameter. Enlargement of the kiln feed end can increase ther-
mal efficiency and production capacity. Thermal efficiency is increased by
73
-------
DRY WET
250
I 200
I
i- 150
2. 100
u
50
300
250
200
150
100
60
DATA SOURCE
BPG SERVICES. INC.
WESTPORT. CONN.
WET 500 1000 1SOO 2000
DRY 2SO 500 7SO 1000
Fuel Savingi Per Ton of Clinker - 1000 Btu'i
2SCO
12SO
Figure 8. THERMAL EFFICIENCY FOR CHAIN SYSTEMS
IN LONG WET AND DRY KILNS
distributing the radiation heat loss over a greater quantity of material pro-
duced, thus resulting in reduced unit radiation losses.
The enlargement of the kiln feed end allows a larger quantity of chain to
be installed, which can further increase kiln efficiency. This is especially
applicable to dry-processing plants since most of the wet-processing plants
have a sufficient amount of chain to reduce exit gas temperatures to 400 °F,
the practical, lower limit. In one instance, a cement manufacturer was able
to increase production rates by almost 30% by increasing the kiln feed end
diameter from 12 to 14. 5 ft. Fuel consumption was reduced by about 20%.
The quantity of fuel consumed in the kiln declined from 4. 8 million Btu/ton
to 3. 9 million Btu/ton. The modification was accomplished during the course
of a 1-month shutdown for each kiln involved. The primary factors to be con-
sidered prior to such a modification include capacity, available horsepower,
and existing foundations.
Oxygen Enrichment
Much work has been done to determine the feasibility of enriching the
combustion air of a rotary kiln with oxygen to increase production and
74
-------
reduce fuel consumption. One such program used oxygen lancing, a tech-
nique in which oxygen is introduced directly into the combustion chamber,
rather than being introduced into the combustion air fan and mixed with
the air. The production rate of the kiln increased 40%, and the fuel con-
sumption decreased by 15% per ton of product. Another program used oxy-
fuel burners and achieved comparable results. In spite of its technical suc-
cess, oxygen enrichment is not commonly used in U. S. cement plants. The
primary deterrent is that the high cost of oxygen cannot presently be justified
by the improvements in efficiency. However, if similar results could be
achieved at lower oxygen consumption rates, or if fuel availability to this in-
dustry becomes severely restricted, the process may be given more consi-
deration.
Waste-Heat Recovery
Two methods are employed in the recovery of waste heat from cement
kilns. These are waste-heat boilers and raw material drying. The use of
waste-heat boilers is generally declining because of the low efficiency of
making steam at the far end of a kiln. To a certain extent, chain systems
have been installed in place of steam generation from waste heat and are
more efficient.
Waste heat can be used to dry the raw materials prior to entry into the
kiln. The moisture content can be reduced from 7% to 8% levels to approxi-
mately 0. 5%. The amount of fuel saved is about 500, 000 Btu/ton of cement
produced. The drying of raw materials with waste heat can be accomplished
in rotary dryers, roller mills, or air suspension preheater units. This is
discussed as part of the long-term equipment modification section.
One major problem area exists in the utilization of waste heat to dry raw
materials. The temperature of the exit gas from the dryer is in the range of
220° to 250°F. The temperature is below the dew point of sulfuric acid and
is relatively close to the water dew point. Therefore, careful design and
stringent maintenance practices are required to preclude corrosion or con-
densation in the duct work and pollution-abatement devices.
Long-term equipment modifications consist of the following:
• Installation of prcheaters
75
-------
• Conversion to dry process.
• Use of vertical kiln.
Preheater Installation
A major advance in reducing energy consumption in the burning process of
cement is the separation of the first stage from the kiln and preheating the
raw material up to near required temperatures before it enters the kiln. The
methods in which this can be accomplished are summarized below in Table 12,
•which shows the effect of each method on energy consumption by using current
practice.
Table 12. ENERGY CONSUMPTION OF VARIOUS
CEMENT-MANUFACTURING PROCESSES
Process
Wet
Long Kiln
Calcinator and Short Kiln
Semiwet
Preheater and Short Kiln
Dry
Long Kiln
Suspension Preheater and Short Kiln
Semidry
Grate Preheater and Short Kiln
Vertical Kiln
Energy
Consumption,
106 Btu/ton
5. 94
4.68
3.60
4.68
3. 15
3.42*
4. 14*
Reduction Over
Average Current
Practice, %
26.9
43.8
50.8
46.6
35.3
Includes 0. 54 million Btu/ton for drying.
The operation of the preheater shown in Figure 9 involves four parallel
flow cyclone heat exchangers. Each cyclone includes a long riser where raw
feed is introduced and conveyed upward into the cyclone in parallel flow with
the hot gas from the preceding cyclone. This system provides intimate con-
tact between raw material feed particles and the hot gases, resulting in a
rapid rate of heat exchange.
Eacy cyclone spins out the feed where the next lower cyclone receives it
by gravity discharge and the gases get successively hotter. The process
76
-------
through the four cyclones increases the temperature of raw material feed
from about 160° at the point of entry to 1450 °F just before entering the kiln.
The system achieves apparent calcination of about 45% prior to entry into
the kiln.
Gas duct to ID fen with
expansion joints end
venturi
Fuller-Kinyon
raw feed I in*
STAGE 2
Cyclone (refractory lined)
Cos duct (refractory lined)
Feed pipe (refractory fined)
Bellows type expansion joint.
Double tipping valve
STAGE 4
Cyclone (refractory lined)
Cos duet (refractory lined)
Expansion joint
Feed pipe (refractory lined)*
Single tipping valve —~<-^
Gas duct to dust collector
Preheater induced draft fan
STAGE I
Cyclone (potl. refractory lined)
Gas duet (refractory lined)
Feed pipe
.^—Bellows type expansion joint
— Double tipping volve
Splosh ploU
STAGE 3
Cyclone (refractory lined)
Gas duct (refractory lined)
Bellow* type expansion }Oint
_X\X^ee^ P'P* ('tffactory lined)
type expansion joint
Double tipping valve
Splash plate
Rotary kiln
Feed end bousing with feed
sKelMrefroctory lined)
OATASOMCC: PULLU C»V*MY
Figure 9. AIR-SUSPENSION PREHEATER KILN
Acceptance of the preheater by the U. S. cement industry has not been
favorable. About one-half of the 13 systems installed since the unit has
been developed are currently shut down. This has been attributed to inadequate
knowledge regarding the process and the effects of alkalies, chlorides, and
combustibles found in the raw materials on the operation of the system.
Another system has been developed that is a modification of the suspension
preheater and can overcome the effects of the alkalies and combustible pro-
blems. The unit is referred to as the secondary furnace (SF) system and is
basically the same as the suspension preheater with the addition of a separate
combustion chamber at the base of the unit. The system increases the cal-
cination rate from the aforementioned 45% for the suspension preheater to
77
-------
approximately 90% prior to entry into the rotary kiln. The SF system was
developed in Japan and Germany, and the anticipated benefits, based on re-
sults of units operating in Japan, include —
• Increased kiln capacity
• Reduced firebrick consumption per ton of clinker produced
• Reduced fuel consumption
• Lower NO emissions
Ji.
• Improved process control
• Improved alkali control
• Lower installed cost than the suspension preheater. .
Conversion to Dry Process
As previously discussed, the dry process of cement manufacture requires
less energy than the wet process. As the cost of fuel continues to increase,
it becomes a larger part of the overall manufacturing cost, and greater atten-
tion will be directed toward reducing it. At the end of 1973, wet-process
plants accounted for 58% of the cement-producing capacity in the United States.
The cost of converting existing wet-process kilns to the dry process is
estimated to be about $41/ton of capacity. This is regarded by the industry
as a nearly impossible task because of its recent period of depressed earnings,
making it difficult to attract new capital.
Vertical Kilns
The vertical kiln is not a new concept in cement technology, but when it
was first introduced about 75 years ago, its performance was unsatisfactory.
It was unable to produce cement of consistently good quality; fluctuations in
strength, setting, and soundness properties of the cement were very high.
However, the vertical kiln has seemingly been developed so that these pro-
blems have been solved. Satisfactory performance of the vertical kiln re-
quires that the raw materials be dampened and nodulized prior to being fed
into the kiln. It also requires the use of low-volatile fuels, such as coal.
In the vertical kiln, the nodules and fuel are fed continuously into the top of
the kiln, and the clinker is extracted, cold, from the bottom by a rotating
grate. Fuel consumption in a vertical kiln is about 3. 6 million Btu/ton of
78
-------
clinker — more than 40% less .than the average fuel consumption of rotary
kilns (6.4 million Btu/ton). Further development is expected to reduce this
figure to 2. 6 million Btu/ton.
One disadvantage of the vertical kiln is that its capacity is limited to about
300 tons/day. A number of vertical kilns in a row, however, require less
space and cost much less than a rotary kiln of corresponding capacity, so
output is not really a problem.
Trends in Energy Utilization
The energy consumed per ton of cement produced can be expected to de-
crease in the future, and most of the methods used to reduce consumption will
center around the kiln as the major area for significant energy savings. In
the short term (prior to 1985), energy consumption by the kiln process can
reasonably be expected to decrease from 7. 0 to 4. 7 million Btu/ton in all
kilns. Figure 10 summarizes the concepts for reducing energy consumption in
the short term on the typical rotary kiln. If all these concepts were imple-
mented to the maximum extent possible and if the industry were to switch over
to the dry process, the reductions in energy consumption stated above would
be realized. The figures given after each concept in Figure 10 are the re-
ductions in energy consumption that have actually been achieved. However,
these reductions are not cumulative; that is, implementation of all these con-
cepts would not reduce energy consumption by 5. 4 million Btu/ton — the sum-
mation of savings realized by implementation of each individual concept.
In the long run, after 1985, reductions in energy consumption to below
3. 5 million Btu/ton should be expected. Most of the reductions will be a re-
sult of the implementation of new kiln processes, primarily the vertical kiln
and suspension preheaters in combination with short rotary kilns. The rate
of implementation will depend on the economic climate, but the industry is
not expected to be able to amass the necessary capital for converting to these
new processes until 1985.
Until recently, the cement industry had been undergoing a trend away from
the use of coal and oil as fuels for kiln operation in the production of clinker.
The reason for this is twofold — the price of natural gas had been inexpensive
compared with other fuels, and the use of cleaner-burning natural gas enabled
the industry to meet emission regulations at lower cost. This trend, however,
79
-------
Ftcd
14 Siliten Oioiidt
4 Aluminum Ondi HJ°
1 Ftil.cO.id. JO-4S*»ET«T C°2
4 Olhrrt - MjO. N». » JiSORY «f
K. S03, ci. tu. r
CC
O
•••
jj
Jl
1
i
1
100^ °"""c*' DI« \
WATER SLURRY fgU - - wt, [
/
°<"KI PitCitcinmi
1 *•"• 7... tA '
H tit * 1 7n
1 '"« P»cr>ii '•* — ' — _
. 10«' .m
;;"« !»„,„,„, /•Co.hnrZow
A Zonc i, A
BiOO'f 'OOft. 1 II-
L I >0°-F IIWF ^ J00a >40ft ,,0»
|.| BI|--4:===in=ii=Xi!l^.--- '" ^ ^Vi m
^ . ° rrOO
I k- / i _^^. j i o^T" w
^ /n;i v
IS_' r*K L^ T '
K tt-fTE ,
/ <.„,. K,
' f3l ^i l-^1^ 1
\ ^l i <;
i — i ;oj 100
ioo>r
1
. High- Sulfur Fuels, variable 6. Air Infiltration, 0.2 X 106 Btu/ton x T,P,C*I p.oduc. •
i. Chain Systems, 1.6 X 106 Btu/ton 7. Oxygen Enrichmen
J. Feed End Enlargement, 1.0 X 106 Btu/ton 8. Process Control, 0
i. Trefoils, nominal 9. Slurry Dewatering,
>. Kiln Ledges, nominal 10. Waste Heat Utiliza
. JO 0>C«lcmm SiliCilt
t, 0.5 X 106 Btu/ton ss t,.«.i«..™ J.I..M.
CLINKER
1.3 X 106 Btu/ton
S GYPSUM
cion, u.b X 106 bcu/con ,00* PORTLAND CEMENT
Figure 10. SHORT-TERM ENERGY-SAVING CONCEPTS IN THE U. S. CEMENT INDUSTRY
-------
is reversing itself because of the scarcity and rising costs o" energy, par-
ticularly natural gas.
Results of field interviews indicate the industry is well aware of the
changes taking place in terms of energy availability and price. In light of
current energy shortfalls and price increases, the industry can pursue one
or a combination of three courses of action. These are the conversion from
natural gas and oil to coal, which is more readily available; increased re-
search and development work directed toward improving the technology of
manufacturing cement; and implementation of the more fuel-efficient tech-
nologies that are available.
The conversion of cement plants from natural gas and oil to coal is the
most practicable measure the industry can undertake in the short-term in
order to alleviate energy-supply problems. This appears to offer only a
partial solution to the problem because many plants are not located close
enough to coal fields to make conversion economically feasible. Transpor-
tation costs of the coal could override any benefit gained from such a con-
version. In addition, some plants are located in regions where coal could
become available if new mines were opened. This places dependence upon
the coal industry to make the new supplies available through additional capi-
tal investment. Until the investment can be made and the necessary lead
times met, a short-supply situation could occur in the near term. In addition,
some of the plants currently using natural gas do not have adequate space
available for stockpiling the quantity of coal necessary to maintain production
on an uninterrupted basis between coal shipments.
In some areas of the country, the use of coal has been precluded by en-
vironmental restrictions that limit the sulfur content of fuels used. This
is an additional factor that can restrict the conversion of equipment to coal.
The second course of action is to accelerate research in the area of im-
proving cement-manufacturing technology, Such activity would include basic
research into new methods of cement manufacture, with emphasis placed on
improving the efficiency of energy utilization. Research funds required
would amount to approximately $18. 5 million — about 1% of industry sales.
This is several times current industry funding levels in research programs.
81
-------
The implementation of more energy-efficient technologies is the third
option available in the near term. If energy efficiency is the only criteria
selected for process evaluation and selection, that is, wet or dry, the dry
process is more efficient. Three important factors deter widespread con-
version to the dry manufacturing process within the near-term time frame.
Two of the factors are equipment availability and the long lead time for con-
version of about 2 years. The third factor, and probably most important,
is the availability of capital for such a conversion. According to industry
estimates, the cost of such a conversion would be $41/ton of capacity, or a
total of $2. 03 billion to convert all existing wet plants.
Information obtained during the course of interviews with cement manu-
facturers was related to current energy availability and cost. Natural gas
is available on an interrupted basis only and, if available, is still subject
to curtailments that have been in the range of 10% of a base year. The base
year is 1972. The availability of natural gas is subject to many regional
variations and is dependent upon whether or not the gas is purchased from
intrastate or interstate suppliers. Generally, availability has been good
from' intrastate sources.
The cost of intrastate natural gas is much higher than interstate natural
gas because of the absence of Federal regulation. In one instance, the cost
of interstate natural gas had increased about 43% from 35^/1000 CF in 1972
to 50^/1000 CF at year-end 1974. In some regions of the country, the
intrastate cost of natural gas was reported to be as high as 90^/1000 CF.
As a result, the cost of natural gas fuel has become a larger portion of the
cost of cement manufacture and interest in alternative sources of energy
has been created.
Coal supplies have not been adversely affected as has natural gas. How-
ever, the price of coal has increased sharply and is dependent upon the point
of origin and negotiated contract terms. In one instance the contract price
of coal had increased from $10. 50/ton in 1972 to $19. 25/ton in 1974, with
purchases made on the basis of a 1-year contract. A second firm purchases
one-half of its anticipated coal requirements on the basis of a 5-year con-
tract. In 1974, the contract price was $17. 50/ton and had increased to
$27. 50/ton in early 1975. The remaining coal requirements are purchased
on the open market and, in 1974, the spot price was about $40/ton and had
-------
declined to $31/ton in early 1975. Purchases are made on u.e open market
on the premise that, historically, spot prices had been lower than contract
prices. Within the past year or two, this relationship no longer existed,
and the opinion was expressed that it may not occur again. Although the
price of coal has escalated rapidly, availability problems are not anticipated
in the immediate future.
Wide-scale conversion from natural gas and petroleum-based fuels could
crea,te shortages of coal supplies. Time would be required for the coal in-
dustry to obtain necessary capital and additional mine openings.
Trends in Emissions
To date, the cement industry has been able to comply with local, state,
and Federal emission requirements. The types of emissions have been pre-
viously discussed, as well as methods of control. The method for controlling
particulates is primarily the use of baghouses. These systems trap particu-
late matter that evolves primarily from the kiln and the clinker cooler. Elec-
trostatic precipitators have been installed on a limited basis, but were found
to be less satisfactory than baghouses. Corrosion of the unit and water dis-
posal were two of the main problems encountered.
Disposal of the particulate matter, after collection, is a major problem
in the industry. One of the companies contacted during the course of the
field survey indicated that it must dispose of 650 tons of particulate waste
per day and this is expected to double over the next 10 years. The material
must be hauled by truck to suitable landfill sites. In some states, the land-
fill site must be specially prepared to prevent seepage of salts into substrata
water. In many instances the expense of equipping an older kiln could not be
economically justified and these have been phased out of service.
Not all of the particulate matter collected can be recycled into the raw
material stream; this is especially true if low alkali cement is being made.
The alkali content of low-alkali cement cannot exceed 0. 6%/ASTM standard.
It was suggested that, if this standard were revised upward to 1. 5%, all of
the collected particulate matter could be recycled and the problem of land-
fill disposal would be solved. The alkali salts react with the aggregate used
in the final concrete mixture, and the cement manufacturer has no control
over the type of aggregate used. The opinion was given that the increased
alkalinity would not affect the quality of the cement.
83
-------
Currently, emissions of SO and NO are not a major problem for the
Ji. 2i
industry. Emissions of SO are controlled primarily by the use of low-
Jt
sulfur fuels and controlled burning practices. The sulfur contained in the
waste-heat stream reacts with the particulates collected and the raw materials
as they are fed into the kiln. As a result the suflur emissions are collected
along with the particulates in baghouses. The efficiency of this method of
collection would be reduced noticeably if the industry would be forced to uti-
lize fuels of higher sulfur content than those presently being used. About
50% of the SO2 formed is collected in the baghouse because of the reactions
with the alkaline particulates. NO emissions are not considered to be a
j£
major problem to the industry in the absence of legislation establishing mini-
mum standards.
It is anticipated that the industry, as a whole, will have to spend $280
million to be able to comply with new 1973-1975 EPA standards.
Energy Cost of Emission Compliance
The cost of emission compliance, in terms of energy expended, is not
readily quantifiable and is subject to many variables that should be evaluated
on an individual plant basis. Included in such an assessment should be the
energy requirements of the control equipment and the secondary units re-
quired to service the primary control device.
The energy required to operate the primary equipment is generally in the
form of electrical power. This power is used to drive air-circulating equip-
ment such as fans for the purpose of moving air through baghouses and other
collection devices. It should be noted that, over the past 25 years, energy
consumption in the form of fossil fuel, primarily in the operation of the ro-
tary kiln, has shown a decline. However, over the same period, the con-
sumption of electricity has remained steady at about 130 kWhr/ton of cement
produced. This has been partially attributed to air pollution control equip-
ment requirements. Estimates received during the course of field inter-
views regarding additional power requirements for primary pollution con-
trol devices were in the range of 5% to 15%. By applying the estimated
range to the 130 kWhr mentioned above, the amount of energy required is
from 6 to 17 kWhr/ton of cement produced.
An example of the secondary equipment operation required to service
primary equipment is the trucks needed for hauling collected particulate
84
-------
matter from the baghouse or other collection device to landau sites. As
time passes, the distance traveled increases as nearby sites are filled.
Fuels required for this are diesel fuel and/or gasoline, depending on the type
of trucks used. The amount of energy consumed in this application is not
available for this report.
The legislation pending, such as the establishment of an NO standard,
ji,
can increase the amount of energy required to power pollution abatement
devices. On the other hand, greater recycling of collected particulate mat-
ter into the raw material stream can reduce the energy consumed in this
area.
Analysis
The potential for reducing energy consumption and air pollutant emissions
in the cement industry is very great. In most instances, the industry is well
aware of this potential. However, because of economic restraints, it is
expected that this potential will be realized only after an extended period of
time. It is significant to note that cement manufacturing in the rest of the
world is far more efficient than in the United States, a fact that can, in part,
be attributed to the premium placed on fuels outside the United States. Be-
cause most of the process modifications discussed above for reducing energy
consumption are already in use in other countries, further technological
development does not appear to be required.
With respect to air pollutant emissions, the industry's primary concern
is with particulate emissions. Control methods are costly in terms of both
economics and energy consumption. Because of the expense involved in
purchasing and fitting pollution abatement equipment to existing facilities,
many of the older facilities are being shut down rather than being converted.
The collection of particulates in control devices creates an additional pro-
blem — that of disposal. Currently, this material is used for landfill, a
very costly solution. More than one company suggested that the standards
for product quality could be altered without affecting the characteristics of
the final product. Specifically, by increasing the maximum allowable alkali
content in the final product, a greater amount of particulate matter could be
recycled, reducing cost and reducing energy by the amount required to trans-
port this material to the landfill areas.
85
-------
SOX emissions do not present as severe a control problem as participates,
primarily because of the chemical reaction inherent in the burning process,
which act to remove SO from the effluent stream. In general, SO control
X. X
does not require any substantial add-on equipment. Consequently, control
of SOX should not result in increases in energy consumption. In fact, the use
of preheaters, as discussed above, not only increases the efficiency of ce-
ment manufacturing with the potential of reducing energy consumption by up
to 50%, but also acts to precipitate SOV out of the exhaust gases, preventing
X.
expulsion into the atmosphere. The fact that preheaters are not widely used
in this country is again a matter of economics and raw material composition.
However, in spite of the economics, a few companies are currently using
preheaters, suggesting that their application is greatly dependent upon local
conditions.
Not too much is known about NO emissions from cement kilns. But con-
Ji
sidering the process temperatures involved— greater than 1500°C in some
cases — the potential for large quantities of NO emissions is great. As dis-
J±
cussed in other sections of this report, NOX formation is a temperature -
dependent phenomenon. Process modifications that increase flame tempera-
ture without improving heat transfer to the process load will almost cer-
tainly result in increased NO emissions. Conversely, adequate removal
Ji
of the additional heat resulting from the applicable process modifications
should maintain NO,, emissions at their current level.
Ji
Of the modifications deemed to be near term, only the use of oxygen en-
richment has any great potential of increasing air pollutant emissions, pri-
marily NO... While in some applications in other industries, for example,
Ji
glass melting, oxygen enrichment can be used without increasing NO emis-
sions. However, due to the different type of load in the cement industry
and the different patterns of heat transfer, it is not clear that NO would not
increase with the implementation of oxygen enrichment. None of the other
modifications classified as near term are expected to affect air pollutant
emissions.
The potential effect of implementation of long-term modifications for re-
ducing energy consumption on air pollutant emissions is substantial. Con-
version to the dry process will increase the potential for increased particu-
late emissions. Based on current available data, if uncontrolled, particulates
86
-------
would increase by more than 10%. The technology is available for controlling
participates, but application of these control devices will increase energy
consumption. Whether the increase will totally cancel the reduction in energy
consumption achieved by converting to the dry process is not determinable in
this program.
Implementation of the vertical kiln can be expected to substantially reduce
NO emissions. The mechanism of operation is such that heat transfer to
*rL
the load is very high, and peak temperatures are lower than required to ob-
tain the formation of NOX in large amounts. The vertical kiln may also affect
SOX emissions. As discussed above, the process requires the use of low-
volatile fuels, such as coal, which may contain substantial amounts of sulfur.
While it is true that SOX emissions are eliminated, to a large extent, from
the flue gases by absorption, there is no data for how much is removed in
this manner. By increasing the amount of SO that must be removed, the
•X
potential for exceeding the absorption capacity of cement reactants is high,
resulting in an increase in SO emissions. Control devices added on to re-
move the excess SO., emissions from the effluent will require additional
Jt
energy, reducing the attractiveness of the vertical kiln as a method of re-
ducing energy consumption.
The net result of this analysis is that implementation of energy-conserving
process modifications can be accomplished without increasing air pollutant
emissions. It is not clear to what extent energy utilization will be reduced.
Control devices required to maintain the emissions at the present level or t
-------
As in the other industries studied, the first step in reducing emissions is
to determine what the emission rates are at the present time. Specifically,
a program should be undertaken to measure NO emissions from the various
•rx
heating processes in the industry; this will primarily involve sampling from
dryers, preheaters, and kilns. This information is essential and vital to
the success of other programs in this area.
The study of this industry has revealed that, underneath the economic
objections raised by the industry, there are some questions with respect to
technical feasibility of the suggested modifications. But, as already pointed
out, most of these modifications are being satisfactorily used elsewhere in
the world. Rather than set up demonstration facilities in this country, we
recommend that programs be undertaken to collect the necessary operating
data from plants already in operation in other countries. These plants af-
ford the opportunity for collecting not only fuel data but pollution data as
well without the great cost of a demonstration facility. Such information
would be most beneficiel in determining the direction to take in dealing with
energy and emission problems, when the available technology becomes more
economically feasible.
Reference Cited
1. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-42, 2nd Ed. Research Triangle
Park, N. C. , April 1973.
Aluminum Industry— State-of-the-Art
The aluminum industry can be broken down into two major segments —
primary and secondary manufacturing. The primary aluminum industry
is concerned with the manufacture of aluminum using the basic raw material
bauxite, converting it to alumina, and in turn processing the alumina into
aluminum by means of the Hall-Herault electrolytic process. The secondary
aluminum industry is concerned primarily with the recovery of aluminum from
scrap by melting down the scrap and removing impurities to obtain the re-
fined product.
The aluminum industry is growing at a greater rate than any of the other
nonferrous metal industries, doubling its production output every 10 years.
In the primary aluminum industry alone, production is expected to increase
from 4 million tons in 1970 to more than 12 million tons in 1985.
88
-------
Primary Aluminum Production
A simplified flow sheet for the production of primary aluminum by the
Bayer-Hall process is shown in Figure 11, which includes the manufacturing
unit operations for carbon anodes and cathodes. In general, primary alu-
minum production can be subdivided into three operations: mining the bauxite,
refining the bauxite into alumina, and smelting the alumina into aluminum.
Each of these operations is described briefly below. However, it should be
noted that the major operation in primary aluminum production is not the
focal point of this study as it does not involve the direct combustion of fossil
fuels. As will be seen, most of the direct fossil fuel consumption is in the
refining, melting, and reheating operations.
Mining
Bauxite is the ore from which aluminum is produced. Because most bauxite
lies at or near the surface, it usually is mined by open-pit methods. After
the ore is dug up, it is transferred by truck and rail to a point of processing.
There the bauxite is ground down to uniformly sized particles and passed
into a rotary kiln in which excess moisture is removed.
The drying process is a relatively low-temperature process; it requires
temperatures of only 1100°F. Consequently, heat losses are minimized and
energy utilization efficiency is relatively high— about 40%. The energy con-
sumed in drying is 1. 0 million Btu/ton of ore dried. Drying accounts for
only a small percentage of the energy consumed domestically by the aluminum
industry because only 15% of the bauxite consumed in the U.S. is obtained
from domestic sources. Thus, in 1971, only 2. 3 trillion Btu, less than 1%
of the industry's total annual energy consumption, was used for drying.
Refining
The refining of bauxite to obtain alumina is achieved by the Bayer Process.
The objective of this process is to separate out the impurities, which include
iron oxide, silica, and titanium dioxide.
The first step in the Bayer Process is digestion of the bauxite into a solu-
tion of hot caustic soda. The dried bauxite is mixed in slurry tanks with a
strong caustic soda solution. The mixture is pumped into large digesters in
which it is heated with steam under pressure. The alumina in the bauxite is
dissolved in the caustic soda, forming a sodium aluminate solution. The
89
-------
TO CONTROL
DEVICE
HYDROXIDE
TO CONTROL DEVICE
UXITE
DRYING
OVEN
m - Vrr»
1
BALL MILL
Y ».i
SETTLING
CHAMBER
DILUTION
WATER
RED MUD
(IMPURITIES)
DILUTE
SODIUM
HYDHOXIDE
AQUEOUS SODIUM
ALUMINATE
ALUMINA fgOE
TO CONTROL DEVICE
I
BAKING
FURNACE
ELECTROLYTE
BAKED I
ANODESj |
1 TC CONTROL Dt'VlCE
PREBAKE
REDUCTION
CELL
TO CONTROL DEVICE
ANODE PASTE
•**•
HORIZONTAL
OR VERTICAL
SODF.RBERC
REDUCTION tCLL
MOLTEN
ALUMINUM
Figure 11. SCHEMATIC DIAGRAM OF PRIMARY
ALUMINUM PRODUCTION PROCESS
90
-------
impurities, which are insoluble, remain as solids. The mr '•'.ire is passed
through a series of pressure-reducing tanks and filter presses. Cloth fil-
ters hold back the solids — known as "red mud" — but allow the liquid, which
contains the dissolved alumina, to pass.
At many plants, the red mud is discarded, but at some — where lower
grade ores are refined — this mud is sintered with limestone and soda ash to
recover a sodium aluminate solution. This solution then is returned to the
digesters for processing through the remaining steps of the refining process.
After the solution containing the alumina passes through the filters, it goes
into a cooling tower and then into precipitators, which are steel structures
resembling farm silos. As the solution cools, it is seeded with sizable
amounts of crystalline alumina hydrate to hasten crystal separation. Even-
tually, a precipitate of hydrated alumina settles out of solution. This pre-
cipitate is filtered from the solution, washed, and then heated in kilns at
1800°F. The crystals are dried, and the water is driven off, leaving com-
mercially pure alumina. The dried alumina is shipped to an aluminum
smelter for reduction.
The Bayer Process consumes from 10 to 15 million Btu of thermal energy
per ton of alumina produced. Natural gas is the preferred fuel; consequently,
most Bayer Process plants are located in areas of lower cost natural gas. .
Further discussions of this operation will be minimized since the technology
of the rotary kiln, the primary energy consuming equipment in the Bayer
Process, is similar to the technology of the cement industry, discussed else-
where in this report.
Smelting
Smelting is the process that breaks alumina down into its two components,
aluminum and oxygen. The basic smelting process is the Hall Process, which
has been used by the industry since 1886. In this process, alumina is dis-
solved in a bath of molten cryolite — sodium aluminum fluoride — in large elec-
tric furnaces. These pots, as the furnaces are called, are deep rectangular
steel shells lined with carbon and connected electrically in series to form a
"potline. "
High-amperage, low-voltage direct current is passed through the cryolite
bath — by means of carbon anodes suspended in each pot — to the bottom of the
91
-------
pot, which serves as the cathode. The molten aluminum formed as a result
of this electrolysis goes to the bottom of the pot, and the oxygen combines
with the carbon anode to release carbon dioxide. The layer of molten alum-
inum that covers the carbon lining at the bottom of the pot becomes the cathode.
Additional alumina is added to the bath to replace that consumed in the reduc-
tion process. Heat generated by the electrolysis maintains the cryolite bath
in a molten state, so that the additional alumina charges are dissolved. Peri-
odically, molten aluminum is siphoned and cast into ingots or alloyed. The
resulting aluminum is at least 99. 5% pure.
The primary source of energy for this process is electricity, which is
consumed at the rate of 55 to 60 million Btu/ton of aluminum produced. This
does not include the fuel consumed for generating the electricity.
The three types of pots used in this process are distinguished by the type
of anode configuration: prebaked, horizontal-stud Soderberg, and vertical-
stud Soderberg. The major portion of aluminum produced in the U. S. (61. 9%
of 1970 production) is processed in prebaked cells. In this type of pot, the
anode consists of blocks that are formed from a carbon paste and baked in
an oven prior to use in the cell. These blocks are attached to metal rods and
serve as replaceable anodes. As the reduction proceeds, the carbon is grad-
ually consumed. The rate of consumption is 0. 45 to 0. 55 pound of electrode
per pound of aluminum produced, which is typical of all three types of pots.
The second most commonly used pot (25. 5% of 1970 production) is the
horizontal-stud Soderberg. This type of cell uses a continuous carbon anode
in which a mixture of pitch and carbon aggregate is periodically added at the
top of the superstructure, and the entire assembly is moved downward as
the carbon burns away. The cell anode is contained by alumunum sheeting
and perforated steel channels, through which electrode connections, called
studs, are inserted in the anode paste (the pitch and carbon aggregate mix-
ture). As the baking anode is lowered, the lower row of studs and the bot-
tom channel are removed, and the flexible electrical connectors are moved
to a higher row.
The vertical-stud Soderberg is similar to the horizontal-stud pot, except
that the studs are mounted vertically in the cell. The studs must be raised
and replaced periodically, which is a relatively simple operation.
-------
Anode preparation constitutes a major energy-consuming -rocess in the
primary aluminum industry, which is often overlooked as a secondary opera-
tion. However, depending on the type of anode used, energy consumed for
anode baking varies from 1300 Btu to 5700 Btu/pound of aluminum metal pro-
duced. Each anode system currently accounts for about 50% of the aluminum
produced, resulting in an estimated 28 billion Btu of energy consumed annually
for anode baking.
Secondary Aluminum Processes
Secondary aluminum processing is essentially the process of remelting
aluminum. The equipment used is essentially the same as that used for
melting in the primary aluminum industry. Thus, the following discussion
of the melting processes is applicable to both segments of the industry.
Melting and Reheating
The types of furnaces generally used for the melting of aluminum are —
• Large and small stationary and tilting reverberatories
• Holding furnaces (reverberatory)
• Tilting, barrel-type furnaces
• Stationary and tilting crucible furnaces
• Dry hearth melters.
Primary aluminum melting furnaces range from small crucible furnaces
for foundry use, called heading pots, to very large reverberatory melters
having a capacity of 25 to 80 tons per charge. Small furnaces of the ladle-
crucible or tilting-crucible types and small reverberatories of up to 10,000-
Ib capacity are usually based on a standardized-design from equipment
builders. Some of the largest producers of aluminum and aluminum-alloy
products design and build their own melters, ranging up to 80-ton capacity.
Others may purchase even their largest melters from equipment builders,
but generally these must be designed to meet the requirements of a parti-
cular casting facility.
Although some continuous melters are in use, normal practice is to melt
on a cyclic basis because of the need to arrive at a design compromise in-
volving caster layout, need to adjust chemical analysis, arrangement of other
93
-------
plant equipment, building layout, type of charge, operating efficiency, and
production requirements. Firing rates on a melting furnace are usually
quite high so as to obtain the fastest melting-rate possible without excessive
metal loss or refractory damage. On the other hand, after melting is com-
plete, the metal is cleaned by passing mixtures of fluxing gases through the
molten metal and is brought carefully to the exact pouring and casting tem-
perature in as quiescent an atmosphere as possible to avoid oxide formation
and metal loss. This requires a very close control over heat input. Because
of the major differences between relative inputs for melting and holding and
the respective combustion requirements, two different furnaces are usually
used for the large-scale melting installations: one for melting, called a
breakdown melter, and a second for refining and holding, called a holding
furnace.
Melting-Furnace Design
A typical large-scale, reverberatory aluminum melter is quite similar
in general design to the steel mill open-hearth furnace. It is usually a
rectangular, refractory-lined box with burners at one endwall and charging
doors above the metal line along one side. Furnace size is dictated by two
key factors: melting rate and capacity. The melting rate is determined by
the thermal head between a) the metal refractory surfaces and the flame, and
b) the heat-transfer area or area of the surface of the metal bath. Metal
capacity is determined by the depth of metal below the metal line. The melting
rate can be increased by using a sloping dry hearth from which molten metal
drains to a holding bath, but this design tends to increase metal loss.
Design practice for reverberatory melters varies considerably from com-
pany to company. One of the major aluminum companies prefers the rec-
tangular design, and regards a length-to-width ratio of 2 as optimum geometry
for the 25 to 50 ton capacity melters and a square or round melter for smaller
reverbs of up to 5-ton capacity. Another major producer uses a tangentially
or radially fired round melter with sidewall charging, on the basis that the
best design geometry has the largest volume-to-wall ratio.
Aluminum melters are designed for melting rates as high as 62 Ib/sq ft-hr,
taking only melting into account. Allowing for charging, holding, and dis-
charging, a good general-design number is 35 Ib/sq ft-hr. Dry-hearth melters
94
-------
can produce at a. rate of 100 Ib/sq ft-hr, but are not used fo.. .iigh-quality
aluminum-alloy production because of high oxidation loss.
Secondary melters (metal reclaiming) are generally designed for lower
melting rates, about 15 to 20 Ib/sq ft-hr. In these furnaces, fluxing is usually
done in a charging well, a compartment separated from the main melting and
holding chamber by a partition wall and submerged throat. The melting period
in these furnaces is only about 40% of the overall cycle.
Energy Requirements
Approximately 75 million Btu of energy is consumed in an aluminum plant
for power and heat in the production of 1 ton of aluminum. However, fossil
fuels also are consumed in the manufacture of the carbon electrodes used in
the electrolytic process. And finally, because electricity is the primary
form of energy consumed, fossil-fuel consumption for electricity generation
also should be considered. Specific process energy requirements have al-
ready been presented for primary aluminum manufacturing —that is, the
Bayer-Hall process — and will not be discussed further. Attention can thus
be focused on the combustion-oriented processes, specifically, melting and
reheating.
Overall melting efficiency for a large (25 to 50 ton) aluminum melter is
normally 15% to 20%, including time required for charging, holding, and dis-
charging. Efficiency might be as low as 10% for some special alloys and, at
the other extreme, as high as 50%. Current melting practice is to melt down
as rapidly as possible with high input rates to achieve maximum productivity,
which has an adverse effect on fuel efficiency. Peak gas temperature entering
the flue is now about 2800°F.or higher; a few years ago, it was about 2200°F.
Some of the factors affecting fuel efficiency are type of alloy, amount of metal
loss due to oxidation, type of charge, charging practice, furnace geometry,
burner and flue arrangement, and ratio of melting time to total time between
charge and pour.
The relation between fuel consumption per pound of metal heated and fuel
efficiency can be shown as follows:
95
-------
Efficiency, % Natural Gas, CF/lb
15 3.37
20 2.53
25 2.02
35 1.44
50 1.01
A typical fuel consumption for large melters is 3 cubic feet of natural
gas per pound of metal, giving an efficiency of 15%.
Efficiency for the smaller crucible melters is about 5% to 10%. The
higher efficiency obtained in reverberatory melting is attributable to direct
heat transfer from the flame and surrounding refractory surfaces to the metal
surface, whereas in crucible furnaces heat must be transferred through a
partition wall.
Crucible melters heated by electric induction are reported to reach an
efficiency of 67% to 73%. On the basis of an average efficiency for power
generation of 32%, the overall fuel efficiency for electric-induction melting
will be only about 22. 4%.
Fuels used in the melting and reheating processes include natural gas,
LP gases, distillate and residual fuel oils, and electricity. However, natural
gas is the preferred fuel. According to industry representatives, the use of
oil requires modifications in operations that decrease the efficiency of opera-
tion. This decrease in efficiency is attributed to the relationship between the
size of the combustion space in the furnace and the flame. Because the oil
requires a larger combustion zone than gas, the relationship is changed to
the detriment of fuel-utilization efficiency.
Air Pollutant Emissions
Air pollutant emissions originate from several processes in an aluminum
manufacturing plant. The most objectionable'and most difficult to control
are those from the potlines. Aside from carbon monoxide and carbon dioxide,
which are released as a result of the reaction of oxygen from the aluminum
with the carbon anode, emissions from the potlines include sulfur dioxide,
derived from the sulfur in the petroleum coke and materials; hydrogen fluoride
gas, the result of hydrolysis of some fluoride salt vaporization; alumina,
cryolite, and aluminum fluoride dusts; and minor quantities of a number of
96
-------
other pollutants. However, this program is concerned with -ombustion-
related emissions, thus excluding the potlines from further consideration.
The pertinent sources of emissions are the calcination operation by way of
the rotary kiln, and the melting and reheating processes, and the anode
baking operation. Table 13 summarizes the available information on the
emissions from aluminum manufacturing processes. It should be noted that
many of the specific pollutants listed are peculiar to the particular process
and unrelated to combustion. For example, most of the air pollutants in the
effluent stream from an open-flame, reverberatory-type furnace used in re-
melting aluminum scrap are the result of residual material, such as grease
and oil, on the surface of the scrap. Given proper control of the combustion
process itself, combustion-related emissions are virtually nonexistent. The
reason for this is clear. The primary fuel used in secondary aluminum pro-
cessing is clean-burning natural gas, which precludes the emission of parti-
culates and SOX. NO emissions are not a problem because most of the pro-
cess temperatures are lower than the minimum required for formation of large
amounts of NO . However, given the trend toward higher melting rates re-
-X
quiring higher temperatures, NO emissions could become a problem.
Ji
The industry does not operate pollution-free. Stack emissions are con-
trolled in several ways. Many secondary aluminum producers have resorted
to natural gas-fired afterburners followed by venturi scrubbers for control
of chloride and fluoride emissions or baghouses for control of particulates.
Electrostatic precipitators are not generally used where the emissions are
high in carbon content because the carbonaceous materials cause the ESP
unit to short out. It should be noted that the use of this equipment for pol-
lution control adds substantially to the energy used by a company. For
example, the use of baghouses consumes about 1000 Btu/lb of product and
the afterburners consume about 2000 Btu/lb of product.
Trends in Aluminum Process Modifications
Because the Hall-Heroult electrolytic process is the major energy-
consuming process in the manufacture of aluminum, great effort is made to
improve its efficiency. The most significant improvement in this area has
been the development of a new smelting process that consumes 30% less
energy than the current technology. However, there are modifications that
can be applied to other processes in the industry, which, while not quite so
97 ,•'•'
-------
Table 13. AIR POLLUTANT EMISSIONS FROM ALUMINUM MANUFACTURING PROCESSES
Emissions
Process Particulates x_ x Hydrocarbons
Bauxite Grinding 6 lb/ton bauxite NA NA NA
Calcination of Hydroxide 200 lb/ton alumina NA NA NA
Soderberg Cells
Horizontal Stud 144 Ib/ton aluminum NA NA NA
Vertical Stud 84 Ib/ton aluminum NA NA NA
^ Prebake Cells, 63 Ib/ton aluminum NA NA NA
00 Anode Preparation NA NA NA NA
Sweating Furnaces 32 Ib/ton charged NA NA NA
Refining Furnaces (Secondary
Operations)
Reverberatory 4 Ib/ton charged NA NA NA
Pot 2 Ib/ton charged NA NA NA
Induction
Note: NA — Emissions known to occur but quantities not available.
-------
spectacular, can result in substantial reductions in energy.
Flash Calcining
Table 14 summarizes the available process modifications and the potential
for reducing fuel consumption.
Table 14. SUMMARY OF AVAILABLE PROCESS MODIFICATIONS
AND THEIR POTENTIAL FOR REDUCING FUEL CONSUMPTION
% Reduction in
Process Modification Fuel Consumption
Flash Calcination 20 to 30
Stack Charging 5 to 10
Hot-Metal Recirculation 5 to 6
Recuperation 20 to 25
Oxy-Fuel Melting 10 to 15
Excess-Air Reduction (20% to 10%) 5
Infrared Heating (Crucible Melting) 20 to 25
Using the fundamental principles of fluid-bed and dispersed-phase techno-
logy, a flash calciner improves heat exchange and reduces heat losses. This
process, to be used in place of the rotary kiln calciner in the conversion of
bauxite to alumina, consumes up to 30% less energy than the rotary kiln pro-
cess. The process has been successfully installed in several bauxite refineries
around the world, with the expectation that more will follow. It should be noted
that currently most of the alumina used in the U. S. is imported. There is a
trend toward refining the bauxite in this country. If this trend continues, it
is likely that this process will gain widespread usage in a relatively short
period of time because this phase of the industry would be developed from the
ground up — that is, independent of the phasing-out of any existing equipment.
Stack Charging
In some of the large aluminum melters, the charge is placed on a sloping
hearth at the base of the stack, as shown in Figure 12. The high-temperature
flue gas transfers heat to the charge on the "dry hearth" by convection and
raises the ingot temperature to the 600° to 750°F range before the ingot is pushed
into the molten bath. Advantages of this charging system are an increased
melting rate and a higher melting efficiency. It is estimated that fuel effi-
ciency can be increased by about 5% to 10%. With this system, scrap is
generally charged in a separate well.
99
-------
Figure 12. ALUMINUM MELTER WITH SLOPING HEARTH
Hot-Metal Pumping
Development of hot-metal pumps with refractory internals has made pos-
sible several improvements in reverberatory melting, particularly in better
utilization of charging wells for high-surface-area scrap such as chips and
foil. The increased bath velocity due to the pumping action increases the
rate of heat transfer to the scrap, thereby melting it faster and reducing
the time that the surface area is exposed to oxidation. Metal loss due to
oxidation is reduced by 50%. In addition, either the overall melting rate
can be increased by 15% to 25% or the fuel efficiency increased by 5% to
6% at the same melting rate.
Recuperation
Combustion-air preheat by recuperation of waste heat in the flue gases
is seldom used in the U. S. because of the low cost of natural gas. However,
it is used extensively in Europe, where fuel prices are considerably higher.
Because only 20% to 40% of the total melting cycle is required for the actual
melting operation, recuperation has been difficult to justify economically.
Also, for holding chambers where fluxing is done in the melting chamber,
the stack gases will contain hydrochloric acid and other corrosive components
that reduce the service life of normal recuperator internal parts. However,
100
-------
many of the largest melters are installed with separate holding furnaces,
and, at the higher prices anticipated for natural gas and fuel oil, it will be
possible to justify the installation of recuperators. Preheating combustion
air to the 600° to 800°F range will give a fuel savings of 20% to 25%, as the
following data demonstrate:
Air Preheat
Temperature, °F % Fuel Saving
400 13.5
600 19.8
800 25.5
1000 30.6
1200 35. 0
1400 39. 0
Oxygen-Fuel Melting
In recent years, tests on primary and secondary aluminum melters have
proved the effectiveness of using auxiliary oxy-fuel burners to reduce melt-
down time, metal loss, and fuel consumption. In remelting, where 97% of
the charge is scrap, a reduction in meltdown time from 2 hours to 1 hour
was achieved, Charge weights averaged 46, 000 Ib, and oxygen consumption
was 712 SCF/ton melted. The melting rate was increased by an average
of 55% and the melt loss reduced by 26%. Relative combustion efficiency
was 48% during the period of oxygen burning and 29% during air burning.
The oxy-fuel roof-mounted burners in this plant were used as auxiliary burners
to back up four air burners.
Improved Air/Gas Ratio Controls
Close control over excess air is a critical element in maintaining com-
bustion efficiency, particularly for furnaces in which combustion products
enter the flue ports at a temperature above 2000°F. In aluminum melters,
flue-gas temperatures typically range from 2200° to 2800 °F. With burner
designs that give good control over air-gas mixing, only 5% to 10% excess
air is required to obtain maximum heat release and high flame temperature,
but many melters run with 20% or more excess air. The problem of con-
trolling excess air is quite difficult because of the large reduction in input
required for firing at maximum rate during melting and for holding at mini-
mum input after melting is completed. A burner turndown of more than
10:1 is required, and a typical balanced pressure-mete ring orifice system
101
-------
cannot maintain this ratio over such a large turndown range. Also, a suffi-
cient flame velocity must be maintained so that combustion products are
not short-circuited to the flue when it is located directly below the burners.
One way to overcome this problem is to use several smaller burners rather
than one or two large burners and to turn off some of the burners when a
turndown greater than about 8:1 is required. Another method, used on steel
mill soaking pits, is to use a 2-in-l or 3-in-l burner in which the air housing
is compartmented so that a very large turndown range can be obtained by
reducing the burner area.
Fuel inputs on the largest melters range up to 40 million Btu/hr, and for
many of the large melters, installation of a control system that maintains a
precise air/fuel ratio by controlling stack-gas oxygen concentration should
be justifiable. This type of control system is also effective when combustion-
air preheat is used.
Infrared Heating — Crucible Melting
Extensive tests run by the Research Department of the Consolidated
Natural Gas Service Co. have shown that the efficiency of crucible melting
of lead, zinc, and other metals of low melting points can be materially in-
creased by using infrared radiant-heating elements around the crucible.
Efficiency with infrared heating has been in the 30% to 40% range, compared
with the 15% to 20% normal range obtained with conventional firing methods.
The melting temperature of aluminum is somewhat higher (1220°F), but a
substantial improvement should be obtained for aluminum melting — say a
20% to 25% reduction in'fuel requirement.
Trends in Energy Utilization
Because the potlines constitute the major energy consuming process in
the manufacture of aluminum, most of the efforts for reducing energy con-
sumption are in this direction. With respect to this process, implementation
of the new Alcoa process for smelting can be expected to account for the most
dramatic reductions in energy consumption in the long term.
On the other hand, the potential for reduction in energy consumption by
fossil-fuel fired processes is also large, but is often downrated in importance
when compared with the potlines. Of thecprocess modifications discussed
above, only flash calcination, oxy-fuel melting, and improved air/fuel ratio
102
-------
control are considered by the industry to be significant in terms of reducing
energy and even then not without some trepidation. For example, while
there are companies using oxygen in their melting operations, it is not deemed
to be particularly economical and is thus used only when there is a demand for
aluminum in excess of the normal production capacity. Flash calcination, al-
though fully developed, is considered long term in that its implementation
will occur only with the phasing-out of existing rotary kilns. Furthermore,
even if it were implemented immediately, it would only have minimum im-
pact on the total energy consumption pattern because most of the bauxite con-
sumed by the industry is actually processed outside the U. S.
The use of air/fuel ratio control equipment is one area of modification
in which implementation can be achieved in a short period of time with mini-
mal expenditures. In several companies, evaluation and installation of this
equipment is a continuing process. As such, on an industry-wide basis, the
impact of such implementation will be relatively small.
Recuperation, as attractive as it appears to be for reducing energy
consumption, is not viewed favorably by the industry at the present time.
Two reasons given for this attitude are that the economics do not as yet jus-
tify its use and the effluent is corrosive, thus shortening the lifetime consi-
derably. Only as fuel prices increase will there be a change in attitute
toward recuperation.
Energy Use Based on Availability
As indicated above, the use of natural gas is predominant and preferred in
'the manufacture of aluminum. However, shortages of natural gas are causing
the industry to switch to alternate fuels, primarily fuel oil. According to the
companies interviewed, the use of oil significantly reduces the efficiency of
fuel utilization, although none of the companies were able to actually quantify
the reduction.
In addition, although considerably in the future, there is a trend developing
to use coal as a fuel in aluminum melters. From the standpoint of product
quality, there does not appear to be any reason not to use coal. In fact,
during the 1930's, most reverberatory aluminum melters were coal-fired.
103
-------
Trends in Emissions
According to industry representatives, melters fired with natural gas,
propane, or oil with an acceptable sulfur level do not require stack-gas
cleaning or scrubbing to meet existing pollution codes. However, a shift
toward the use of coal requires the development of a compact, efficient, and
cost-effective process to remove SO2 from stack gases in order that existing
codes on SO2 emissions are met. In addition, shifting to coal firing will re-
quire installation of some kind of stack-gas cleaning system for removal of
fly ash from the effluents. In the alternative, treatment of the coal prior
to combustion to remove sulfur and reduce fly ash, such as in solvent re-
fined coal, may be a more attractive solution as far as this industry is con-
cerned.
According to industry representatives, process modifications for purposes
of reducing energy consumption should have no effect on the ability of the
industry to meet current existing standards. Most prominent on the list of
modifications is the flash calcination process. According to uses of the pro-
cess, there has been no difficulty in meeting local control regulations. How-
ever, no information concerning the pollution control equipment used was
offered.
As indicated above, the industry is not concerned with NO_ emissions
X.
primarily because of the absence of standards. There is no available data,
but considering that most processes operate below the temperature required
for generating large amounts of NO , NO., should not be a problem at the
jt Ji-
present time. With the implementation of greater oxygen-enrichment faci-
lities and with the installation of recuperators, NOX emissions will tend
to increase as process temperatures increase. This is, however, deemed
to be a long-term effect.
Analysis of Data
In general, analysis of the available data indicates that implementation of
most of the energy-conserving process modifications should have only mini-
mal impact on air pollutant emisssions. This is dependent upon the use of
adequate control measures as in the case of the flash calcination process.
Clearly, the use of such devices will reduce the overall impact of the energy-
efficient process on energy consumption; however, there is no data available
to quantify the impact.
104
-------
Conversion to coal, as discussed above, would have a significant impact
on air pollutant emissions in the absence of adequate control devices. Even
with suitable control devices, and assuming the pollution-free nature of the
combustion process in this industry, emissions will probably increase.
Whether or not the processes are actually free of pollutants in the effluent
streams can only be surmised; but with adequate controls on the combustion
itself and the use of clean-burning fuels at relatively low process tempera-
tures, it is, in fact, likely that the processes comply with current regulations.
In the attempt to improve energy-utilization efficiency in the secondary
aluminum industry, it is clear that NO emissions will be affected. As was
J&
indicated, when the process temperatures increase, NO... emissions will also
Jt
increase. One area that has had little attention is the use of afterburners
as pollution control devices. As they are currently used, effluent gases as
high as 1800° to 1850°F in temperature are being emitted into the atmosphere
without any attempts at waste heat recovery. Clearly, the waste heat from
these gases can be used to preheat scrap or preheat combustion air. An
additional value is found in the heating value of the waste organic material
which, if recovered, would allow some reduction in primary fuel utilization
and eliminate a solid waste-disposal problem in those cases where scrubbers
are used downstream of the afterburner to recover dry salts. However, it
is unlikely that the afterburners themselves are contributing to the emission
problem in the form of NOX emissions. The operating temperatures are
not high enough for large amounts of NO emissions to form. There are no
data to substantiate this conclusion, and thus it must remain speculation un-
til field investigations are made.
R e c ommen dation s
Under current conditions and relative to the specific area of concern —
that is, the relationship between energy conservation and combustion-related
air pollutant emissions — there are few areas of great conflict if it is assumed
that NOX emissions are low. It is possible, though, that this is illusory, based
on insufficient data, primarily in the area of actual emissions. Consequently,
the first recommendation must be to undertake a program for collecting data
from operating facilities. This is particularly true in the case of NO emis-
Jv
sions. Having obtained this data, a more precise evaluation of the effect of
process modifications can be made.
105
-------
The area that requires the greatest amount of development is waste-heat
recovery. Programs should be initiated to solve the existing problems of
recuperation, primarily the corrosion of the recuperators by the effluent
gases, and the economics of operation. In developing systems, it is impor-
tant that low-grade waste heat also be considered because much of the waste
heat of the industry is in this category.
Oxygen is already used in melting, but only on a limited scale. Programs
should be developed, as in other industries, to develop the true potential of
this method for increasing fuel-utilization efficiency. As is other industries,
a corresponding program would have to be run to determine the effects of
oxygen on NO emissions.
Jt
These recommendations are generally in agreement with those of the in-
dustry and thus should not run into opposition. The industry is eager to
cooperate, as evidenced by their participation in the FEA and Department of
Commerce program, and make long-term commitments for voluntarily re-
ducing energy consumption.
Petroleum Refining Fired Heaters —State-of-the-art
Petroleum refining is largely accomplished by distillation separation into
intermediate feedstocks and products. The intermediate feedstocks require
heating and thermal cracking or catalytic treating before separation of the
desired products. This separation usually requires additional heating after
the catalytic treatment or thermal cracking.
Total refinery energy consumption amounts to around 10% of the crude
throughput. Most of this energy is consumed as fuel in fired heaters. The
amount of energy consumption depends largely on the complexity of processing
that takes place after the initial crude distillation separation. Generally, a
gasoline-oriented refinery will have fuel consumption considerably greater
than a distillate-oriented refinery. In the long run, the refinery output is
controlled to meet consumers demand. This demand is for high gasoline
production in the summer and fall seasons, with high distillate fuel produc-
tion during the winter season.
Because of the diversity that exists among the various refineries currently
in operation, it is not practical to do a detailed analysis of the industry on a
process basis. Thus, the scope of this program was limited to an investigation
of only fired heaters.
106
-------
Fired heaters are used in many operations throughout the --rocess of re-
fining petroleum, beginning with the initial crude distillation unit, in which
the lighter, more volatile components are separated from the heavier, less
volatile components. The resulting fractions become the charging stock for
the other operating units in the refineries. Several types of equipment can
be used in the process; the basic types are shown in Figure 13. Typically,
distillation occurs when heated crude oil is admitted into a fractionating
column, which is tapped at several points, thus allowing continuous removal
of the various boiling fractions, or products. The residue from this pro-
cess maybe submitted to vacuum distillation. When the fractions within
the crude have small volatility differences or when it is desired to separate
a higher boiling constituent from other components of a system of vapors
and gases, additional distillations employing solvents or absorbers are used.
The energy consumption of a crude oil distillation process varies, depend-
ing upon the particular installation, but a review of several processes indi-
cates that a typical installation consumes about 100, 000 Btu of fuel per bar-
rel of crude oil feed for heat. An additional 20, 000 Btu of energy is consumed
as steam.
Upon completion of the distillation process, the products are fed into any
number of other processes. The processes considered in this program are
briefly described below.
Catalytic Cracking
Catalytic cracking is one of the most important processes used in re-
fineries. Its primary purpose is the production of gasoline. The three basic
types of catalytic cracking processes are fluidized bed, moving bed, and
fixed bed. At present, fluidized-bed cracking accounts for about 80% of the
refinery catalytic cracking capacity, and moving-bed cracking accounts for
the remaining 20%. The fixed-bed catalytic cracking process is virtually
extinct.
In the fluidized-bed process, the reacting vapors (feed) are forced up-
ward through a bed of fine granular material with finely divided solid catalyst
evenly distributed throughout the system. This establishes a quasi-fluid
suspension, in which the reacting vapors come in contact with the evenly dis-
tributed catalyst and undergo cracking. Upon completion of the cracking,
107
-------
TO STACK
-------
the catalyst is separated from the products by cyclone separ'tors. The '
catalyst, fouled by coke and tar during cracking, must be regenerated. This
is done in a separate space by passing air through the bed and burning off
these products. Figure 14 is a schematic diagram of one type of catalytic
cracking unit in operation. The reaction temperatures used in such a unit
Vopor to got
recMtry unit
Liquid to tot
r»cov»f7 ur«»
@2S±1*1'
product
product
AIR BLOWER
©Virgin go* oil
t»Ml
tetd
Figure 14. SCHEMATIC DIAGRAM OF ORTHOFLOW CATALYTIC
CONVERTER AND ADJUNCTS (Stippled Areas Represent Fresh
or Regenerated Catalyst) (M. W. Kellogg Co.)
range from 885° to 950 °F, depending upon the yield of products desired.
Regenerator temperatures are moderate, ranging from about 1050° to 1200°F.
Although operating pressures in this unit are low, 8 to ZO psi, some cracking
processes utilize pressures as high as 1000 psi. During catalyst regenera-
tion, when coke and tar are burned off from the catalyst, the combustion
gases (primarily carbon monoxide) are sent to carbon monoxide boilers and
burned to produce high-pressure steam. The moving-bed cracking process
is similar, employing a flowing bead-type catalyst instead of a fine-grain
catalyst. Both processes are continuous.
Energy consumption in a typical catalytic cracking process is about
150, 000 Btu of fuel and electricity per barrel of feed. * On the other hand,
* Estimated average based on energy consumed by various processes.
109
-------
there is a net steam production of about 75 pounds, which converts to a fuel
value of 100, 000 Btu/bbl of fresh feed.
Catalytic Reforming
Catalytic reforming is also an important conversion process within a
refinery. In this process, certain ring hydrocarbons are converted into
aromatic compounds. As in the case of catalytic cracking, several types
of processes are used. In a typical operation, the feed, usually naphtha, is
prepared in a prefractionator, mixed with hydrogen, and fed into a preheater,
where the temperature of the mixture is increased. The hot naphtha vapors
are fed into a reactor containing any of a number of catalysts, depending on
the process used. In the reactor, several chemical reactions take place.
The products of the reaction are cooled in heat exchangers and then fractionated
or stabilized. The stabilized product then can be used for high-octane gaso-
line, or it can be further fractionated into components (principally benzene,
toluene, and xylenes).
Operating conditions of a catalytic reformer vary considerably. Reactor
temperatures range from about 800° to 1100°F, and pressures range from
as low as 100 psi to more than 1000 psi. Energy consumption also varies
considerably. Fuel in the form of oil or gas and electricity are the primary
types of energy consumed in catalytic reforming. Fuel consumption varies
from 200, 000 to 450, 000 Btu/bbl of feed. Electric consumption, primarily
for driving pumps and compressors, varies from 3000 to 20, 000 Btu/bbl of
feed.
Delayed Coker
Coking processes are relatively severe cracking operations designed
to completely convert residual products such as pitch and tar into gas,
naphtha, heating oil, gas oil and coke. Delayed coking is a high energy-
consuming process employing fired heaters to heat the charge oil. It is a
semicontinuous process in which the heated, charge is transferred to large
coking drums that provide the long residence time needed to allow the crack-
ing reactions to proceed to completion. A flow plan for a delayed coking
process is shown in Figure 15. Feed is introduced into the product fraction-
ator, where it is heated. The fractionator bottoms are routed through a
furnace whose outlet temperature varies from 480° to 515°C. The heated
110
-------
Hydraulic
Cleaning
a
Soakers
Furnace
Coke
r-©-i
-Gas
Naphtha
-Gas Oil
Fractionator
Residue
Figure 15. DELAYED COKING PROCESS
oil enters one of a pair of coking drums, where the cracking reactions con-
tinue. The products leave at the top, and coke is formed on the inner sur-
face of the drum. The temperature in the coke drums ranges from 415° to
450 °C. Overhead products go to the fractionator, where naphtha and heating
oil fractions are taken off as products.
Hydrotreating
Hydrotreating is used in a refinery for removing impurities, including
sulfur, nitrogen, and metallic compounds; for hydrogen saturation of olefins
and aromatics; and for mild hydrocracking. Hydrotreating is used to prepare
catalytic cracker and reformer feedstocks and to upgrade middle distillates,
cracked fractions, lubricant oils, gasolines, and waxes.
In a typical hydrotreating process, feedstock, made up primarily of petro-
leum distillates, is mixed with recycle and makeup hydrogen and heated to
temperatures of 400° to 850°F. The heated charge then is fed into a fixed-
bed reactor containing a catalyst, usually cobalt molybdate on an alumina
carrier, and pressurized from 50 to 1500 psig. The effluent is cooled, sep-
arated from recycle gas, and stripped of hydrogen sulfide.
Ill
-------
As in the other processes discussed, energy consumption varies consid-
erably, depending upon the products desired and the specific process used.
Fuel consumption, primarily for heating, varies from 25, 000 to about
120, 000 Btu/bbl of feed; the majority of processes consume approximately
50, 000 Btu/bbl. Electricity consumption is minimal, usually about 7000 Btu/
bbl of feed. In addition, about 200 SCF of hydrogen is consumed per barrel
of feed. This hydrogen, which should be considered as feedstock, has a
heating value of about 65, 000 Btu. All the hydrogen used in this process is
obtained as off-gas from catalytic reforming processes. The fuels used for
heating are primarily natural gas and fuel oil.
Energy Requirements
Approximately 250 refineries are in operation in the U. S. at present, no
two of which are alike. Each refinery produces hundreds of products by
using the processes that are most suited to the needs of that refinery. The
list of factors that affect the choice of the process used by a refinery is long,
but the primary ones are crude oil source, power source, water availability,
and potential markets. As a result, the energy consumed by a refinery to
process a barrel of crude oil varies considerably. A reasonable average,
determined by taking the total energy consumption of the industry and dividing
by the number of barrels of crude oil run to stills, is about 700, 000 Btu/bbl
of crude oil processed. Figure 16 compares the various forms of energy
consumed by refineries and shows that about two-thirds of the total energy
consumption is in the form of natural gas and refinery gas. Purchased elec-
tricity, not shown in Figure 16, accounts for less than 3% of the total energy
consumption. Other forms of energy, which make up the balance of energy
consumed but are not shown in Figure 16, are acid sludge, petroleum coke,
liquified petroleum gas, and purchased steam. Together, these forms of
energy account for about 16% of the total energy consumed by this industry.
Emissions
Data on the emission of air pollutants from fired heaters (or for that mat-
ter any process in the petroleum refining industry) are not readily available.
None of the companies interviewed were eager to discuss this subject and
the small amount of data available is-extremely superficial in nature. What
follows is, therefore, merely a general discussion of the emissions from
112
-------
C\J
O
O
t-
Cfl
O
O
o:
LJ
2800
2700
2600
2500
2400
2300
2200
2100
2000
1900
1800
O 1700
1600
1500
1400
1300
1200
MOO
1000
UJ 900
800
700
600
500
400
300
200
100
0
z
7
7
TOTAL ENERGY
NATURAL GAS
I960 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70
YEAR
A-54-749
Figure 16. TYPES OF FUEL CONSUMED
BY PETROLEUM REFINERIES
113'
-------
combustion sources in the refining process. Any specific data presented
should be treated as estimates rather than hard fact.
The major air pollutant emissions from petroleum refineries are SO ,
carbon monoxide, NO , aldehydes, ammonia, and particulates. The poten-
tial sources of these emissions are summarized in Table 15.
Table 15. POTENTIAL SOURCES OF SPECIFIC
EMISSIONS FROM OIL REFINERIES
Emission
Oxides of Sulfur
Hydrocarbons
Oxides of Nitrogen
Particulate Matter
Aldehydes
Ammonia
Odors
Carbon Monoxide
Potential Sources
Boilers, process heaters, catalytic cracking unit
regenerators, treating units, H2S flares, decoking
operations.
Loading facilities, turnarounds, sampling, storage
tanks, waste water separators, blow-down systems,
catalyst regenerators, pumps, valves, blind changing,
cooling towers, vacuum jets, barometric condensers,
air-blowing, high pressure equipment handling vola-
tile hydrocarbons, process heaters, boilers, com-
pressor engines.
Process heaters, boilers, compressor engines, cata-
lyst regenerators, flares.
Catalyst regenerators, boilers, process heaters,
decoking operations, incinerators.
Catalyst regenerators.
Catalyst regenerators.
Treating units (air-blowing, steam-blowing), drains,
tank vents, barometric condenser sumps, waste water
separators.
Catalyst regeneration, decoking, compressor engines,
incinerators.
In this report, we are concerned only with emissions from combustion sources,
thus eliminating aldehydes, ammonia, and carbon monoxide from considera-
tion. Hydrocarbons will not be considered because they do not present a
serious problem.
SO emissions from combustion sources are a problem because of the
presence of H2S in the major fuel-gas streams used for firing the heaters.
In most refineries, the H2S in these fuel-gas streams is removed by means
114
-------
of a Glaus unit, which converts the H2S to free sulfur. Under normal cir-
cumstances, 95% of the potential sulfur emissions are eliminated. The re-
maining 5%, rather than being emitted from numerous fired-heater stacks,
is consolidated and emitted in the Glaus unit stack gases. Cleaning up the
remaining 5% requires an expenditure equal to or exceeding the Glaus unit
expenditure, adding increased investment and labor costs with practically
no return.
NO
x
NO emissions are estimated to be the most serious problem of all the
Jt
combustion-related emissions from fired heaters. These emissions from
fired heaters are estimated to be about 0. 34 lb/106 Btu of fuel burned. At
the present time, there are no NO standards to govern these emissions.
X.
However, there is concern that future standards, which might be imposed,
would effectively defeat efforts to improve heater efficiency.
Trends in Process Modifications
In general, the petroleum-refining industry is unable to effect substantial
process modifications because of the nature of the industry; that is, any
modification of processes in one part of the refinery will usually require
process modifications in other parts of the refinery to accomodate
the first modification. For example, in most existing refineries, the dif-
ferent process units are operated as separate, but nevertheless dependent,
entities having intermediate tankage for charge and products, thus providing
for flexibility in both operating and maintenance. To improve fuel efficiency,
the trend has been to eliminate the intermediate tankage and run hot feed
directly from one unit to another. However, such a modification substantially
increases the risk of intermediate bottlenecks or reduced stream factors,
limiting a completely integrated refinery to an on-stream factor no greater
than the lowest of the stream factors of the individual units.
As a result of this interrelationship between processes, most modifica-
tions are restricted to improved operating procedures and maintenance.
Most of the refineries efforts are concentrated in this area. Thus, refineries
are upgrading their process evaluation techniques — for example, by installing
more accurate and sensitive flue analysis equipment. Such equipment, it is
felt, will allow the process heaters to be fired more efficiently by reducing
US'
-------
the amount of excess air used in combustion. The stated goal is 10% excess
air for gas-fired heaters and 20% excess air for oil-fired heaters. These
goals have not been reached, according to industry representatives, because
of the inadequate design of existing burners and inadequate combustion volume
on existing heaters.
Two other techniques that have been given consideration for implementa-
tion in existing facilities are the use of preheated feed to the heaters and the
use of preheated combustion air. Both of these modifications can be readily-
implemented on existing facilities without requiring substantial modifications
to the overall refinery.
Long-range projects for continued improvements in fuel conservation in-
volving the replacement of obsolete equipment with equipment of improved
design are being examined by the industry. A comparison of past practice
with future design possibilities shows the potential for improved fuel effi-
ciency:
Past Future
Practice Designs
Lowest temperature for heat
removal by exchange 400°F 250°F
Furnace stack temperature 800°F 400°F
They recognize that furnace tube outside skin temperature will be a
consideration for the future designs because surfaces will have to be kept
above the dewpoint of the furnace gases.
Conservation coordinators feel that retrofit installations of air heaters
and steam coils can be justified at current high fuel prices. In completely
new installations they will expect to have a greater ratio of convection sur-
face to radiant service. Increased use of extended surface convection fur-
nace tubes is expected. One group favored a furnace design that will elimi-
nate the damper in the breeching to the stack, using instead an air supply
box servicing multiple burners. Such installations would provide dampers
on the forced air supply to the windboxes. It is clear that, while-short-
range process modifications for reducing fuel consumption are extremely
limited, long-range projections show substantial potential. However, the
actual implementation of long-range modifications is likely to be very slow
116
-------
because it depends on the required replacement of existing equipment which
have a relatively long life.
Trends in Energy Utilization
In order to combat the effects of rising fuel costs, the petroleum refining
industry, under the auspices of the American Petroleum Institute (API), has
instituted a fuel conservation program involving more than 80% of the U. S.
refinery capacity. The results of this program were reported as follows:
Energy Consumption,
weighted average for
37 companies,
thousand Btu/bbl input
1972 Base Period Total Energy Consumption 667
1972 Base Period Adjustment to 1974 +9
Adjusted 1972 Base Period Total Energy Consumption 676
Less 1974 Last Half Year Total Measured Energy
Consumption 624
Energy Conservation Improvement for last half of 1974 52
% Reduction from 1972 Base 7. 8%
The position of the industry, as alluded to above, is that only small reduc-
tions can be effected in the short term and that only over the long term can
large reductions be expected.
According to the industry representatives, fuel availability will have a
substantial impact on energy utilization, having a depressing effect on im-
proved efficiency. The use of natural gas, currently amounting to about
31% of the total refinery fuel consumption, is decreasing. It is contended
that switching to liquid fuels will require more furnace volume for combus-
tion than is presently available in order to maintain the fuel-utilization effi-
ciency at its current level. Thus, switching to liquid fuels is likely to in-
crease fuel consumption until such a time as new equipment, designed to
burn these fuels, is installed. This is obviously a long-term effect. It has
been suggested that the impact on fuel-utilization efficiency of switching fuels
could be mitigated by the development of better burners. While several com-
panies indicated that they could undertake such a development program, none
indicated that they were involved in such a program at the present time.
117
-------
Trends in Emissions
As the industry sees it, the primary emission problems from combustion
sources are SOX and NOX. The contention is that neither of these emissions
can be controlled to meet standards without severely affecting energy-
utilization efficiency. For example, the installation and operation of Glaus
tail-gas cleanup units for SO_ unquestionably results in high energy consump-
Jt
tion, but-with no compensating improvement in overall efficiency. Thus,
with respect to near-term process modifications, SO., and NO emissions,
•*t .JC
according to the industry, would increase.
On the other hand, given long-range process modifications involving the
redesign and replacement of existing equipment, the potential for reducing
energy consumption without increasing emissions is high. This is particularly
true in light of the anticipated change in fuel utilization pattern, that is, from
gaseous to liquid and solid fuels where under present conditions emissions
would be expected to increase. As in other industries, the addition of con-
trol equipment to prevent the emissions from increasing would result in addi-
tional energy expenditures.
Analysis
Although the typical petroleum refinery is a highly complex series of
processes, analysis of the relationship between energy consumption and air
pollutant emissions from fired heaters is relatively simple, because the po-
tential factors for affecting this relationship are relatively few in number.
Process modifications in the short term that affect energy-utilization efficiency
are limited to two areas: improved control of the combustion process; and
preheating the inputs, the charge material, and the combustion air, into the
relevant processes. Long-term modifications primarily involve the replace-
ment of existing equipment with new equipment designed to accommodate a
greater variety of fuels and improve heat transfer to the process loads.
Sufficient information on current emission levels is not available to per-
mit a definitive statement about the effect of the various process modifica-
tions on air pollutant emissions, but reason would dictate that improved
energy utilization need not result in increased emissions. And it should be
noted that, as in the other industries discussed, the addition of emission
control equipment will increase the energy consumption of a plant by the
118
-------
amount of energy required to. operate the equipment.
Clearly, the control of SO is a case in point. None of the process modi-
fications discussed will decrease SO emissions, thus requiring that control
A.
devices be used at an additional expenditure of energy. On the other hand,
switching to fuels other than gas, while not necessarily decreasing the effi-
ciency of energy utilization, could result in increased SO emissions, re-
Ji,
quiring additional control devices and increased energy consumption for the
overall process.
On the other hand, NO emissions have the potential of being significantly
j£
affected by the contemplated process modifications. For example, increases
in combustion air preheat temperatures typically result in increased NO
Ji
emissions. But there is enough data available from other industries to indi-
cate that this need not be the case. NO emissions will not increase if heat
Ji.
transfer within the particular process is increased, precluding an increase
in peak flame temperature, the primary cause of increased NO emissions
JL
under these circumstances.
Recommendations
The primary problem areas within the petroleum refining industry center
on two topics: NO emissions and improving dual-fuel capabilities to mini-
•JC
mize loss of energy utilization efficiency. With respect to NO emissions,
Ji
it is recommended that a program be undertaken to establish the current
status of the industry through actual measurement. This will aid in the de-
velopment of standards for the industry and in determining whether or not
the standards are reasonable in light of future process modifications.
Having established this information, it is recommended that demonstration
programs be conducted to establish operating procedures for process modi-
fications, where the amount of energy consumed is reduced without increasing
NO emissions.
x
Finally, it is recommended that a program for development of improved
dual-fuel burners be undertaken, where it can be shown that the current
technology is deficient in allowing a company to switch fuels without affecting
efficiency.
119
-------
APPENDIX A. Weighting System
Weighting Factors
The next step in the selection process is the assigning of weighting factors
to each of the restraints in terms of their relative importance. In developing
the weighting factors, the following assumptions have been made:
1. All air pollutants are equal in importance.
2. Conservation of energy and reduction of emissions are equal in importance.
3. Industries with no potential for energy conservation or reduction of emis-
sions will be excluded from consideration.
Given these assumptions, the weighting factors are assigned based on a scale
of 1 to 10, -where 1 is a low-priority and 10 is a high-priority rating. To
determine the suitability of a particular industry, the following set of numeri-
cal operations would be performed:
1. Base energy consumption x emission index number = emission weighted
base number.
2. Estimated potential for energy conservation x emission weighted base
number.
3. Estimated potential for reducing emissions x emission weighted base
number.
4. Items 2 + 3 = conservation and emission weighted number.
Based on the value obtained in 4 above, the industries for study have been
selected with the highest value given the highest priority.
Table A-l defines the weighting system we used to arrive at the numbers
used in the above set of equations.
121
-------
Table A-l. WEIGHTING SYSTEM FACTORS
1. Base energy number
The base energy number is arrived at by summing the weighting factors
for restraints Nos. 1, 2, and 3, where the following values are assigned
(scale 1 -> 10):
Restraint No. 1. Energy use in process heat
1 = Low usage; 10 = High usage
Restraint No. 2. Combustion-related uses
1 = No combustion-related uses; 10 = All combustion-related uses
Restraint No. 3. Number of processes
1 B Numerous processes; 10 = One process only
2. Emission index number
The emission index number is arrived at by summing the weighting factors
for restraints Nos. 4 and 5. However, since the primary emissions of con-
cern, as indicated in restraint No. 5, are generally directly combustion-
related, the emission index number can be based on restraint No. 4 alone
with values assigned as follows:
1 = Mostly independent emissions; 10 = mostly combustion-related emissions
3. Potential for energy conservation
1 = Low potential; 10 = High potential
4. Potential for reducing emissions
1 = Low potential; 10 = High potential
122
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-76-022
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
SURVEY OF EMISSIONS CONTROL AND COMBUSTION
EQUIPMENT DATA IN INDUSTRIAL PROCESS
HEATING
5. REPORT DATE
October 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Peter A. Ketels, John D. Nesbitt, and
R. Don Oberle
8. PERFORMING ORGANIZATION REPORT NO.
IGT Project 8949
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Institute of Gas Technology
IIT Center
Chicago, Illinois 60616
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
68-02-1821
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 6/74-11/75
14. SPONSORING AGENCY CODE
EPA-ORD
is. SUPPLEMENTARY NOTES TERL-RTP project officer for this report is J.H. Wasser, Mail
Drop 65, 919/549-8411 Ext 2476.
16. ABSTRACT
The report gives results of investigations of the interaction between present
and potential energy conservation measures and emission programs in a number of
select industries. Where energy conservation goals conflicted with emission control
goals, the problems were assessed. Based on these assessments, research and
development programs were recommended to solve each industry's problems. The
study was limited both to processes in which heat was obtained through the direct
combustion of fossil fuels, and to emissions that are affected by combustion itself.
It was concluded that information is available on emission problems and control tech-
nology associated with particulate emissions and stack plume opacity, as well as
stack gas acidity, carbon monoxide, and unburned hydrocarbons. However, not
much is apparently known about the level of NOx emissions from industrial heating
processes. There are several areas of concern, all related to the apparent conflict
between energy conservation and emission control goals: (1) the ability to meet volun-
tary energy conservation goals is substantially decreased by the energy consumption
of emission control devices required by more stringent future emission standards; (2)
application of energy conservation measures will lead to increased air pollutant emis-
sions, particularly for NOx; and (3) the trend away from clean-burning gas.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution
Combustion
Industrial Processes
Industrial Heating
Fossil Fuels
Energy
Conservation
Air Pollution Control
Stationary Sources
Energy Conservation
13B
21B
13H
13A
21D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
132
2O. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
123
------- |