EPA-450/3 -74-044
July 1974
STUDY OF INDUSTRIAL USES
OF ENERGY RELATIVE
TO ENVIRONMENTAL EFFECTS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park North Carolina 27711
-------
EPA-450/3-74-044
STUDY OF INDUSTRIAL USES
OF ENERGY RELATIVE
TO ENVIRONMENTAL EFFECTS
by
M.E. Fejer and D.H. Larson
Institute of Gas Technology,
IIT Center,
Chicago, Illinois 60616
Contract Number: 68-02-0643
EPA Project Officer: John R. O'Connor
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1974
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - as supplies permit - from the
Air Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711, or, for a fee,
from the National Technical Information Service, 5285 Port Royal Road,
Springfield, Virginia 22151.
This report was furnished to the Environmental Protection Agency by
the Institute of Gas Technology, IIT Center, Chicago, Illinois, in ful-
fillment of Contract No. 68-02-0643. The contents of this report are re-
produced herein as received from the Institute of Gas Technology. The
opinions, findings, and conclusions expressed are those of the author
and not necessarily those of the Environmental Protection Agency. Men-
tion of company or product names is not to be considered as an endorse-
ment by the Environmental Protection Agency.
Publication No. EPA-450/3-74-044
11
-------
TABLE OF CONTENTS
I. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
II. SIC CODES 262 AND 263 - PAPER MILLS AND
PAPERBOARD MILLS II-1
Summary II-1
Paper and Paperboard-Manufacturing Processes II-2
Energy Requirements for the Manufacture of Paper
and Paperboard II-7
Energy Utilization Pattern 11-10
New Technology and Its Effects on Energy Consumption 11-11
Air Pollutant Emissions From Paper and
Paperboard, Mills 11-13
References Cited 11-20
III. SIC CODE 281 - INDUSTRIAL CHEMICALS III-l
Summary III-l
Alkalies and Chlorine III-3
Industrial Gases III-13
Industrial Inorganic Chemicals 111-23
Industrial Organic Chemicals ni-38
Conclusions III-41
References Cited 111-41
IV. SIC CODE 282 - PLASTICS MATERIALS AND SYNTHETICS IV-1
Summary IV-1
Resin-Manufacturing Processes IV-3
Energy Utilization Pattern IV-5
Air Pollutant Emissions IV-5
Trends in Plastics Manufacture IV-6
References Cited IV-6
V. SIC CODE 291 - PETROLEUM REFINING V-l
Summary . V-l
Petroleum Refinery Processes V-5
Energy Consumption in Petroleum Refining V-10
111
-------
TABLE OF CONTENTS, Cont.
Page
Air Pollutant Emissions From Petroleum Refineries V-14
Trends in Air Pollutant Emissions V-20
References Cited V-21
VI. SIC CODES 3211, 3221, AND 3299 - FLAT GLASS,
CONTAINER GLASS, AND PRESSED AND BLOWN
GLASS AND GLASSWARE VI-1
Summary VI-1
Glass-Manufacturing Processes Vl-2
Energy Utilization Pattern Vl-5
Energy Requirements of the Glass-Melting Process Vl-8
Energy Requirements of Annealing VI-15
Air Pollutant Emissions in the Glass Industry VI-16
Factors Affecting Air Pollutant Emissions Vl-16
Effect of New Technology on Air Pollutant Emissions
and Energy Usage Vl-18
References Cited Vl-22
VII. SIC CODE 324 - HYDRAULIC CEMENT VII-1
Summary VII-1
Portland-Cement-Manufacturing Processes VH-4
Raw Material Preparation VII-4
New Technologies in the Manufacture of Portlant Cement VII-6
Energy Utilization Pattern VII-12
Air Pollutant Emissions From Cement-Manufacturing
Processes VII-15
Effect of New Technologies on Air Pollutant Emissions VII-18
References Cited VII-19
VIII. SIC CODE 325 - STRUCTURAL CLAY PRODUCTS VIII-1
Summary Vm-1
Structural-Clay-Manufacturing Processes VIII-2
Energy Utilization Pattern VIH-5
New Technologies and Their Effects on Energy Consumption VIH-7
Air Pollutant Emissions VIU-8
References Cited VIH-10
iv
-------
TABLE OF CONTENTS, Cont.
IX. SIC CODE 331 - BLAST FURNACES, STEEL WORKS,
AND ROLLING AND FINISHING MILLS IX-1
Summary IX-1
Energy Utilization Background Information IX-3
Effect of New Technology on Energy Consumption IX-31
Air Pollutant Emissions Background Information IX-35
Effect of New Technology on Emissions IX-40
Energy Utilization Patterns Projections to 1985 IX-40
Environmental Impact Pattern Projections to 1985 IX-79
References Cited IX-88
X. SIC CODE 3331 - PRIMARY COPPER X-l
Primary Copper-Manufacturing Processes X-Z
Energy Utilization Pattern X-6
New Technologies in Copper Manufacturing X-6
Effect of New Technology on Energy Usage X-10
Air Pollutant Emissions in Copper1 Manufacturing X-ll
Effect of New Technologies on Air Pollutant Emissions X-12
References Cited X-l6
XI. SIC CODE 3332 - PRIMARY LEAD XI-1
Primary Lead-Manufacturing Processes XI-1
Energy Utilization Pattern XI-6
New Technologies in Lead Manufacturing XI-6
Effect of New Technologies on Energy Utilization Pattern XI-9
Air Pollutant Emissions From Lead-Manufacturing Processes XI-9
Effect of New Technology on Air Pollutant Emissions XI-12
References Cited XI-13
XII. SIC CODE 3333 - PRIMARY ZINC XII-1
Primary Zinc-Manufacturing Processes XII-2
Energy Utilization Pattern XII-9
New Technologies in Zinc Manufacturing XII-10
Effect of New Technology on Energy Consumption XII-11
-------
TABLE OF CONTENTS, Cont.
Air Pollutant Emissions From Zinc Smelters XII-12
Effect of New Technology on Air Pollutant Emissions XII-15
References Cited XII-15
XIII. SIC CODE 3334 - PRIMARY ALUMINUM XHI-1
Primary Aluminum-Manufacturing Processes XHI-3
Energy Utilization Pattern XIII-7
New Technologies in the Manufacture of Aluminum XHI-8
Effect of New Technologies on Energy Consumption XTII-IO
Air Pollutant Emissions From Aluminum-
Manufacturing Processes XIII-11
Effect of New Technologies on Air Pollutant Emissions XIII-12
References Cited XIII-15
XIV. ELECTRICITY AND STEAM GENERATION XIV-1
Industrial Boilers XIV-1
Air Pollutant Emissions From Industrial Boilers XIV-3
Electricity Generation XIV-6
Air Pollutant Emissions From Electricity Generation XIV-9
References Cited XIV-12
VI
-------
I. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
The objectives of this study were 1) to identify where and how energy
is being used by industry, 2) to determine the relationship between energy
use and its impact on the environment, and 3) to determine how new
technology or changes in existing technology will affect energy utilization
and air pollutant emissions by industry on a nationwide basis.
The 12 industries examined in this study account for more than 60%
of the primary energy consumed by industry in the U.S., or about 20%
of the total energy consumption of all energy-consuming sectors. This
study has determined the energy consumption of each industry on an overall
basis and on a process-by-process basis, indicating those processes in
which the energy utilization efficiency can be increased either by replace-
ment with an alternative process or by improvement of the existing pro-
cess. For example, the use of oxygen in a fuel-fired reverberatory
melter reduces total fuel consumption, even when the energy consumed in
making the oxygen is considered.
In addition to energy utilization, air pollutant emissions were con-
sidered on a process-by-process basis. Actual emission rates as well as
methods of reducing emissions are presented for those types of emissions
and processes for which information is available. Where applicable,
these emissions are related to the types of fuel consumed by each pro-
cess, and the effects of changes in process technology on these emissions
are considered.
By using the data collected during this study, the energy consumption
and air pollutant emissions of each industry are projected to 1985. In
general, these projections are based on historical trends and projections
published by each industry and, as such, may be subject to much debate,
The projections have been made only to 1985 because the uncertainty
increases as the time covered increases. The uncertainty is much less
for a period between now and 1985; industry is essentially locked into
its current pattern of energy consumption, in large part, because of the
long time lags in implementing new processes on a large enough scale
to affect the total industrial-energy-consumption pattern. In addition, much
of the research and development in industry has been aimed at objectives
other than energy conservation, resulting in the availability of only a very
1-1
-------
limited number of new energy-conservative technologies. Even conserva-
tion techniques that are more easily implemented, such as proper
maintenance and operating practices, can require years before any sig-
nificant decrease in energy consumption is observed. However, after
1985, the options for altering industrial energy consumption patterns
increase considerably, and the future pattern of industrial energy con-
sumption depends upon whether or not industry exercises these options.
The results of this study are valuable because they represent a
reasonably compact compilation of data on current industrial process
energy utilization and air pollutant emissions that can readily be used
as a starting point for determining the future impact of new processes,
energy conservation techniques, and air pollution control methods. The
study also presents a large amount of information on new processes that
could affect energy utilization and air pollutant emissions in the future
if they are implemented. Because this study was based solely on pub-
lished information, the acceptability of new processes to industry could
not be determined. Consequently, the projections that consider the ef-
fects of new processes on energy utilization and the environment are only
illustrative.
In 1972, the latest year for which information is readily available,
industry in the U.S. consumed slightly more than 20.6 quadrillion Btu of
energy, approximately 28. 5% of the total national energy consumption.
Table 1 summarizes the energy consumed in the U.S. by major sector.
The industries investigated in this program consume approximately 12. 7
quadrillion (1015) Btu/yr, which is about 61.5% of the total energy con-
sumed by all industry. The energy consumption of these industries in
1985, as determined by information collected during this program, is
projected to be approximately 17.2 quadrillion Btu/yr, an increase of
about 35. 6%. This agrees closely with the 35% increase in energy con-
sumption forecast by the Department of the Interior for the entire industrial
sector for 1985. Table 2 summarizes the 1972 energy consumption of
' ''\
the industries examined during this program and contains projections to
1985. For the purpose of comparison, energy consumption data for the
entire industrial sector also are presented. The projections of energy
consumption in this table do not reflect changes due either to energy
conservation programs or to replacement of current processes by new
processes. T _
JL Ci
-------
Table 1. ANNUAL CONSUMPTION OF ENERGY IN THE U.S. IN 1972*
Natural Purchased
Coal Gas (Dry) Petroleum Other Total (Gross) Electricity Total
1 rt!2 -DJ_-
4,
7,
12,
395
377
4
827
0
603
7, 642
10,591
790
4, 102
0
23, 125
6,
5,
17,
3,
32,
735
606
318
141
165
965
0
36
18
3, 513
0
3, 567
14,
20,
18,
18,
72,
772
610
130
583
165
260
3,482
2,488
18
_
0
5, 988
A-54-911
1
8,
23,
1
1
8,
8,
78,
254
098
148
583
165
248
Household and Commercial
Industrial
Transportation
Electricity Generation
Miscellaneous
Total
*
Source: Division of Fossil Fuels, Bureau of Mines, U.S. Department of the Interior,
press release of March 1973.
-------
Table 2. ENERGY CONSUMPTION BY INDUSTRY
FOR 1972 WITH PROJECTIONS FOR 1985
Total Annual Energy Consumption
1972 1985
Industry 1012 Btu
Iron and Steel 3,040 4,100
Glass 250 450
Primary Nonferrous
Copper 71.4 130
Lead 13.3 16.5
Zinc 38 51
Aluminum 352 1, 100
'Industrial Chemicals 3,764 4,900
Petroleum Refining 2, 861 3, 700
Plastics Materials and Synthetics* 134 700
Structural Clay 175 205
Paper and Paperboard 1,307 1,700
Hydraulic Cement 581 700
Total 12,685 17,209
Total Industrial Sector 20,731 28,170
#
Excluding feedstocks.
In addition to energy utilization, this study also considered the effects
of new processes and energy conservation methods on air pollutant emis-
sions. As in the case of energy utilization, this study presents a base
case of emissions by process. Unfortunately, the amount of available
published data is very limited, and there are many gaps in this informa-
tion, particularly for oxides 'of nitrogen (NO ) and carbon monoxide.
Furthermore, when the potential effects of a new process on energy util-
ization can be determined by calculation, the effects of a new process on
air pollutant emissions generally cannot be determined because emission
rates cannot be determined except under actual operating conditions.
There are some exceptions, such as new hydrometallurgical processes
from which the air pollutant emissions are extremely small because these
are basically wet chemical processes. Table 3 summarizes the total
annual air pollutant emissions from all sources in 1968, the latest year
1-4
-------
Table 3. ESTIMATED NATIONWIDE AIR POLLUTANT EMISSIONS IN 1970
Source
Transportation
Fuel Combustion in
Stationary Sources
Industrial Processes
Solid Waste Disposal
Miscellaneous
Carbon
Monoxide
SO * NO J
Particulates x Hydrocarbons x
111.0
0. 8
11.4
7.2
18.3
0.8 1.0
6.7
13. 3
1.4
4. 0
26.4
6.4
0. 1
0. 2
19.5
0. 6
5.5
2. 0
7. 3
11.7
10. 0
0.2
0.4
0. 5
Total
149. 0
26.2
34. 1
34. 9
22. 8
SO expressed as SO2.
NO expressed as NO2.
-------
for which this specific information is available. Industrial air pollutant
emissions cannot be determined from information collected on each in-
dustry because not enough information on actual operating conditions,
such as the number of plants using a specific type of pollution control
device, is available.
The data collected during this program suggest that the implementa-
tion of control devices will have the greatest impact on air pollutant
emissions from industrial processes. Implementation of new processes
is not likely to have a reducing effect on the emissions. On the other
hand, if a new process aggravates the pollution problem by creating
higher emissions, even if it reduces energy consumption, pollution control
devices would have to be used. There is a tendency, however, to develop
new processes in which air pollutant emissions, although not reduced
per se, are more readily controllable than those from an older process.
Numerous examples of this trend exist in those industries in which sulfur
emissions are a serious problem. The trend in these industries is to
develop a new process that actually causes higher sulfur emissions than
the old process, but because of the higher concentrations of sulfur, con-
trol in the form of sulfuric acid plants becomes feasible. In other in-
dustries, equipment is redesigned to allow for air pollutant emission
controls that could not be used with the older equipment. In spite of
these steps, reduction of air pollutant emissions from industrial processes
is likely to be realized only over a long period. Note that the type of
fuel consumed will affect process air pollutant emissions, but only if
proper control methods are not implemented. Certainly the type of fuel
consumed affects the method of control used, but the end result should
not be affected by fuel type.
The utilization of energy by industry can be broken down into four
basic categories: process heat, steam generation, electricity generation,
and feedstock. Most of the electricity used by industry is purchased.
Table 4 summarizes the energy consumption by each of the industries
investigated during this program according to the category in which the
energy was consumed. The amount of energy consumed by an industry
in each of these categories varies from as little as 5% to as much as
95% . Thus, the energy utilization pattern of industry is diverse, and
few solutions can be common to all industries.
1-6
-------
Table 4. ENERGY UTILIZATION BY SPECIFIC INDUSTRIES
Industry
Industrial Chemicals
Iron and Steel
Petroleum Refining
Paper and Paperboard
Hydraulic Cement
Primary Nonferrous
Glass
Structural Clay Products
Plastics and Synthetics
Annual Energy
Consumption, 1972
10t2 Btu
3764
3040
2861
1307
581
497
250
175
755
Percentage
Consumed
for Process
Heat
55
80
60
5
90
80
90
90
10
Percentage
Consumed
for Steam
Generation
&
/"
16
20
20
95
10
20
10
10
45
Percentage
Consumed
as Feedstock
29
0
20
0
0
0
0
0
45
-------
Examination of these data indicates that some industries, because of
their particular energy utilization pattern, are more likely to experience
reductions in energy consumption in the future than other industries. For
example, the paper and paperboard industries consume about 95% of their
total energy in the generation of steam. Because the fuel efficiency of
steam generation is high (generally in excess of 80% ), significant im-
provements are not likely in this area. Yet the energy consumption per
unit of product probably can be reduced by improving upon the efficiency
of steam utilization. The means for reducing steam utilization range
from insulation of steam lines to substitution of direct-fired for indirect-
heated processes. In those industries that generate large quantities of
steam, therefore, reductions in energy consumption can be achieved by
only a limited number of relatively simple means. However, data on the
current operating conditions within each industry are not available; con-
sequently, estimates of the potential reductions that can be achieved would
. be pure speculation.
Unlike steam generation production, energy consumed for process
heat is very sensitive to reduction. In the past, American business
economics dictated that productivity was of prime importance in equipment
design and that efficient utilization of cheap energy was second, but the
cost of energy is increasing. Consequently, the potential for new pro-
cesses that will reduce energy consumption is great for many industries.
In addition, the potential for improved fuel efficiency merely by improving
operating practices is great. In many cases, only minor alterations are
necessary.
Another way in which fuel efficiency can be improved is by proper
maintenance of existing equipment and replacement of obsolete, inefficient
equipment by modern, more efficient units. In many cases, the mainten-
ance requirements for achieving significant improvements in fuel efficiency
are relatively simple, such as the sealing of air leaks on a furnace.
With only minor modifications to current operations and with improved
maintenance schedules, an estimated 10% or greater reduction in energy
consumption can be achieved by industry. The effect of replacing obsolete,
inefficient equipment cannot readily be determined without an accurate
inventory of existing equipment, but such replacements are likely to re-
sult in substantial reductions in energy consumption.
1-8
-------
During this program, several new processes that were found to exist
would, if implemented, result in substantial reductions in energy consump-
tion without detrimental effects on the environment. Yet many of these
processes are being ignored by the respective industries for reasons that
cannot readily be determined without the input of industrial representatives.
A case in point is the use of oxygen enrichment to reduce fuel consump-
tion in the glass industry. Although the effectiveness of this technique
has been demonstrated on a full-size production facility, it has not gained
widespread acceptance by the industry. As a result, only the company
that successfully tested this fuel reduction method is actually using it.
Because of this type of situation, it would be improper to advocate a
changeover to this process or other new processes by U.S. manufacturers
without first determining and evaluating the reasons for their reluctance.
The information presented in this report can and should be used to
establish the basis for energy conservation programs within industry.
Although it is very difficult to evaluate these programs on a nationwide
basis, the information is complete enough that planning can be effectively
accomplished for individual industries. Given these considerations, we
recommend that the following tasks be undertaken:
1. The information contained in this report should be distributed to
industry representatives to obtain the feedback necessary for com-
plete evaluation of new processes or methods for reducing energy
consumption. We think that many of the new processes discussed
within this report offer viable alternatives to current processes with
their high energy consumption patterns and, as such, should be im-
plemented. However, industry feedback is required to determine
other factors, such as economics, not covered by this study.
2. In areas in which new technology has been developed and subsequently
ignored, programs should be implemented to demonstrate the effective-
ness of the technology. This report presents information, when
possible, that permits a decision to be made on which new process,
if implemented, offers the greatest benefits. For processes for which
specific information is lacking, programs should be implemented to
obtain this information, such as economic factors and air pollutant
emissions.
3. As indicated in this report, certain industries are more sensitive to
efforts to reduce energy consumption than others. These industries,
such as the primary metals, cement, and glass industries, are char-
acterized as large consumers of energy in direct-fired heating pro-
cesses and, consequently, are the industries upon which efforts should
1-9
-------
be concentrated. Industries characterized as large consumers of
such energy as boiler fuel should receive only secondary levels of
effort because the potential for reducing their energy consumption
is low.
4. Industries that consume large amounts of electricity or that propose
to implement new processes that consume electricity in favor of
older fuel-fired processes should be examined closely. In many
cases, the amount of energy consumed at the industry level is sig-
nificantly lower with an electric process than with a fuel-fired process,
but primary fuel consumption due to the relatively low efficiency
of electricity generation and transmission is actually higher.*
However, in a number of cases, such as induction heating and re-
sistance heating, energy consumption including fuel consumed for
electricity generation is lower than that for the corresponding fuel-
fired furnace. Air pollutant emissions also will be affected by changes
from fuel-fired processes to electric processes, although the extent
of the effect will vary according to the source of electricity. In
those cases in which the fuel used in the industrial process is cleaner
than the utility fuel, the potential for increased emissions by conver-
sion to an electric process is high. In those cases in which the
industrial process and the utility boiler are oil- or coal-fired, the
effect of such a conversion depends to a large extent on the controls
imposed on a power-gene rating plant. In general, the data suggest that
each process for which conversion to electricity is being considered
be carefully evaluated on the individual merits of that process
for its impact on national energy consumption and air pollutant
emissions.
Note that this study did not attempt to deal with the rapidly changing
fuel availability situation. We only indicated what processes could switch
fuels and what fuel is currently favored.
*
Fuel values used throughout this report are the same as those used
in the 1972 Census of Manufactures Special Report, "Fuels and Electric
Energy Consumed. Unless otherwise stated, the energy consumed
for electricity generation was not included in the figures for industrial
electricity consumption.
1-10
-------
II. SIC CODES 262 AND 263 - PAPER
MILLS AND PAPERBOARD MILLS
Summary
SIC Code 262 pertains to establishments primarily engaged in
manufacturing paper (except building paper) from wood pulp and other
fibers and also in manufacturing converted paper products. Pulp mills
combined with paper mills, and not separately reported, are included
in this industry. Establishments engaged in the manufacturing of con-
verted paper products from purchased paper stock are classified elsewhere.
Similarly, SIC Code 263 pertains to establishments primarily engaged
in manufacturing paperboard, including paperboard coated on the paper-
board machine, from wood pulp and other fibers and also in manufac-
turing converted paperboard products. Pulp mills combined with paperboard
mills, and not separately reported, are included in this industry.
Establishments primarily engaged in manufacturing converted paperboard
products from purchased paperboard are classified elsewhere.
Paper and paperboard mills consume nearly 10% of the total primary
energy consumed by industry in the United States.5 In 1971, approxi-
mately 56 million tons of paper and paperboard was produced, requiring
an estimated 1310 trillion Btu of energy. Major energy sources are
natural gas (29.3%), electricity (6.1%), coal (17.6%), oil (25.0%), and
by-product fuels (22% ), such as bark and black liquor. Approximately
93% of the energy is used in boilers for steam production, and the re-
maining 7% is used for mechanical drives.
The growth rate of the paper and paperboard industry (Figure 1) has
been greater, on the average, than that of the rest of industry; conse-
quently, its relative energy consumption also has increased. In 1947,
its share of the total primary energy consumed by industry was about
4.1%, compared with its current level of almost 10%.4 Because the fuel
utilization efficiencies of boilers are very high, usually 80% or better,
fuel consumption per ton of paper or paperboard produced is not likely
to decrease in the future as a result of improvements in boiler efficiencies.
However, certain heat requirements, in a given process for a given type
of paper, can be reduced if the plants using the process become more
efficient in their production.
n-i
-------
s
X
X
o 80
f 70
O
P 60
O
X
X
O
cr
o.
ffi
Q.
£
8
X
8
1947 '52 '57 '62 '67 '72 '77
YEAR
'82 '87
A-44-698
Figure 1. ANNUAL PAPER AND PAPERBOARD
PRODUCTION WITH PROJECTION TO 1985
These increases in efficiency usually are the result of improved
methods of heat recovery and heat reuse and conversion to direct-fired
processes when possible. These improvements reduce the amount of
direct heat lost or wasted. Following this analysis one step further
indicates that the new plants are the most efficient plants and that older
plants that have undergone revisions in equipment and processing are
less efficient but still significantly more efficient than older plants that
have undergone no improvements. In general, the paper and paperboard
industry tends to keep abreast of new developments, maintaining a very
high degree of efficiency. In addition, competition is very strong so
that inefficiently operated plants do not last long unless they produce
highly specialized products. Thus, the efficiency of energy utilization
probably will improve in the future.
Paper- and Paperboard-Manufacturing Processes1*4*7
The manufacture of paper is divided into two basic operations:
manufacture of pulp from the raw material and transformation of pulp
into the finished product.
Pulping
The pulping process is most often a chemical process, although
nearly 12% of total pulp production is by mechanical or semichemical
processes. The chemical pulping processes are classified by the type
11-2
-------
of liquor used to digest the wood chips. Of the available processes,
the sulfate process (also known as the kraft process) and the sulfite
process are the most widely used. These two processes account for
nearly 88% of the pulp produced.3 The kraft process is by far the most
popular. Figure 2 is a flow sheet of a kraft process for pulping and
papermaking.
The mechanical processes are groundwood, in which wood blocks are
held against a grindstone; steamed groundwood, which is similar to the
groundwood process except that the wood is steamed for 6-12 hours prior
to grinding; disk refining, in which the wood is cut into chips and ground
between two disks rotating in opposite directions; and defibrator, which
is similar to disk refining except that the chips are presoftened by
steaming, soaking, or a mild chemical cook. This report deals only
with the chemical processes because they consume the largest percentage
of energy in the paper and paperboard industry.
The first stage in preparing logs for pulping is removing the bark.
The discarded bark, which is saturated with water, is dried to a 50%
moisture content and used as fuel for steam-raising in bark boilers.
The calorific value of the bark depends upon the type of bark and the
time of year when it was harvested, but a typical calorific value is
3500 Btu/lb of dry bark.7
The wood chips produced in a machine that cuts the debarked logs
are cooked at high temperature and pressure in a liquor. In the sulfate
process, this liquor is made up of sodium sulfate and sulfide. In the
sulfite process, this liquor is bisulfite of lime. This process dissolves
the lignin in the wood, leaving the fibers free. The spent liquor plus
lignin (or concentrated black liquor), which is obtained from reduction
of the water content to about 50% by flash evaporation, is fired in
special boilers to raise steam. This black liquor is by far the most
important by-product fuel in the paper industry; it has a calorific value
of about 8000 Btu/lb.7 The black liquor, at about 190°F, is sprayed
onto the flat sidewalls of the boiler, where further water removal and
finally combustion take place. In the kraft process, the smelt is tapped
from the furnace, quenched, and ground to recover chemicals, which
after further preparation can be recycled into the cooking process.
II-3
-------
LOGS BY RAIL,
TRUCK, OR WATER
BARK SHREDDER
BARK BURNING
BOILER
BLOW TANK
BROWN
STOCK
SUPPLEMENTARY
FUEL
BLACK LIQUOR
RECOVERY UNIT
PRECIPITATOR
SUPPLEMENTARY
FUEL, SMALL
AMOUNT
DIRECT CONTACT
TO EVAPORATOR
SMELT DISSOLVING
TANK
SHIPMENT
AS PULP
MUD
CONCENTRATOR
PAPER
MACHINE
Figure 2. FLOW SHEET FOR SULFATE PULP AND PAPER1
11-4
-------
The pulp then is treated in a variety of processes, depending upon
the end use of the product. In some cases, the water must be removed,
although in an integrated mill, where pulping and papermaking are accom-
plished on the same site, very little water is removed. In general, the
pulp is screened, separating coarse fibers from fine fibers and removing
dirt and foreign objects. Depending upon the final product, the pulp may
also be bleached.
Papermaking
Papermaking consists of four major mechanical steps: preparing the
stock, forming the sheet, removing the water, and finishing the sheet.
(The same is true in the manufacture of paperboard. )
Preparing the Stock
The fibers produced in the pulping process are mixed with water to
form a very thin mixture containing 1 part fiber to 200 parts water.
This paper stock is further refined in a process known as beating. The
pulp is passed between two sets of bars, forming two rubbing surfaces.
In this process, the fibers are shortened, the ends shredded, and the
outer walls ruptured.
In many instances, again depending on the end product, mineral
fillers are added to the stock; these fillers increase the paper opacity,
improve printing quality, and increase brightness. The more important
fillers are clay, calcium carbonate, and titanium dioxide. Finally, to
retard the flow of ink in the paper, a "sizing" material is added to the
stock.
Forming the Sheet
Upon completion of the stock preparation step, the stock is fed into
a papermaking machine for forming the sheet. Figure 3 is a cross
section of a typical paper machine. The prepared stock is stored in the
headbox at a constant head and issued from the bottom of the headbox
through a fine horizontal opening across the entire width of the machine
onto a moving wire mesh. The wire mesh runs over rollers that remove
the water from the sheet by gravity and capillary action. Thus, a
vacuum is applied to the sheet for additional water removal. The water
content of the sheet at this point is about 80% by weight.
n-s
-------
StecK preparation r- Mechanical water removal _ Steam drying
#
in
s
Head box
i r--, 8 n°
£ 3*
3 CJ
1
Suction
Boxes
1
.B ft
i *
Press part
t>
3
'5 ?!
E Drier port
I *
1 91
Felt ,,-y-f y
driers (^J °"^ Size press
! i?
n
E After driers
-MC.
calendar
I I I i
PAPER MACHINE ELEVATION
Cylinder
Collector shoes
Figure 3. PAPER MACHINE ELEVATION WITH DETAIL OF
TYPICAL STEAM SUPPLY AND CONDENSATE REMOVAL NOZZLE7
Removing the Water
The newly formed sheet is picked off the wire screen and supported
on a continuous band of felt; it is passed through a series of mechanical
presses that squeeze the water from the sheet. Usually two or three
presses in series remove as much water from the web of paper as pos-
sible; yet at the end of the press cycle, the paper is still 62% water by
weight. Upon leaving the press section, the paper enters the drying
section, which is made up of a series of large-diameter, hollow rolls,
which are heated with steam. The paper is passed over these rotating
drums, supported on a system of felts that are designed to ensure con-
tinuous contact between the sheet and cylinder surface and to carry away
from the sheet a part of the moisture evaporated. After leaving the
dryer section, the sheet is only 6-7% water by weight and is capable of
supporting itself.
H-6
-------
Finishing the Sheet
The dried sheet enters the finishing section, where it is passed
through the calendar section. Here, it is compressed to give it a smooth
finish. In some cases, coloring may be added by the calendar rolls or
a latex or a starch surface coating may be applied as a filler and glazing
agent to enhance the printing quality. The coating usually is applied as
.an aqueous suspension; consequently, the sheet may be wetted up to 35%.
Therefore, the sheet must be redried in a section that is identical to the
main drying section, but is referred to as the after-dryer. After the
finishing operation is completed, the paper is wound onto a reel, where
it can be trimmed to the desired width.
Papermaking is a continuous process from the introduction of the
stock into the paper machine through the finishing rolls. Although there
are several variations and combinations of paper machines, there are
two general types: the "Fburdrinier" and the "cylinder. "
The Fourdrinier machine allows the pulp to flow onto a horizontal
wire screen made in the form of an endless belt. The water in the
pulp drains through the wire, assisted by suction boxes at certain points.
The matted sheet of fibers is transferred to a wool felt belt, which
carries it to the press section.
The cylinder machine consists of a cylinder covered with a wire
screen revolving in a vat. The fibers are deposited on the screen as
the water passes through the screen. The fibers are picked up on a
felt belt and carried to the presses. Usually several cylinders are
served by one "pickup felt" so that the sheet formed consists of several
plies.
Eneirgy Requirements for the Manufacture
of Paper and Paperboard
Energy in paper and paperboard mills typically is consumed in two
forms: steam and electricity. Process steam, which is required pri-
marily for drying the paper, consists mainly of steam supplied to the
drying cylinders, heat for air being supplied to the machine, heat for
conditioning the air in the machine house to maintain satisfactory ambient
air conditions, and direct steam heating for stock preparation as nec-
essary. In the pulping process, steam is used by the digesters, the
E-7
-------
washers, and the evaporators. Electricity is used primarily for mech-
anical drives in barking and chipping processes, in chemical pulping,
as well as in mechanical pulping processes. Electricity also is used to
drive the paper machines.
Table 1 is an estimated breakdown of the energy consumed by fuel
type and end use in 1971.
Table 1. ESTIMATED FUEL CONSUMPTION BY
PAPER AND PAPERBOARD MILLS IN 1971
Consumption,
Fuel Type IP12 Btu Usage
Coal 230 Steam generation*
Fuel Oil 327 Steam generation
Natural Gas 383 Steam generation, heat
Other Fuels (Bark
and Black Liquor) 288 Steam generation
Purchased Electricity* 80 Mechanical drives
#
110 trillion Btu was consumed as steam for electricity
generation.
901 trillion Btu was recovered and reused for steam
generation.
The table shows that all forms of energy are used by the industry,
mainly for steam generation. However, because steam must be gener-
ated in any case, a mill typically generates much of its own electricity.
In 1971, an estimated 53% of the electricity consumed by paper and
paperboard mills was self-generated.
Although information quantitatively relating energy type to steam
generation is available, accurate figures on steam consumption by process
are not readily obtainable because the same fuel is used several times.
For example, if a plant requires quantities of electricity, steam, direct
firing, etc., the statistics on paper production in the plant may reflect
uses of electricity, steam, and gas that all originate from the same
source. As previously stated, much of the energy consumed in a specific
process is energy that has been recovered from other processes in the
plant, which in turn can be traced to a specific fuel. If energy consumptions
II-8
-------
by process are simply added together to determine the total energy
consumed for the production of a unit weight of paper, the resulting
figures on the actual amount of energy consumed the energy input
would be inflated. Furthermore, papermaking processes vary from
plant to plant, depending upon the age of the plant and the specific pro-
cesses being used. And finally, statistics on the processes used in a
plant are well guarded because the industry is very competitive.
Pulping
Although specific energy consumption figures are not available for
each step in the pulping and papermaking processes, generalizations can
be made. In the kraft pulping process, energy is consumed 1) for bark-
ing, 2) for chipping, 3) for digesting, 4) for washing, 5) in the evaporators,
6) in the recovery plant, and 7) in the recausticizing operation. Steps 1
through 4 deal with the actual pulping process, and steps 5 through 7
deal with the recovery of the chemicals used in the pulping process. In
other pulping processes, the chemical operations are different or, in the
case of mechanical processes, are eliminated altogether, in which case
the energy requirements change. Of all the pulping processes used, the
kraft process not only is the most widely used but also consumes, in
general, less energy than the other pulping processes. Based on an
energy input to a pulp and paper mill of about 23 million Btu/ton of paper
produced, the pulping process consumes about 15 million Btu/ton of
paper produced. However, approximately 25% of paper production is
accounted for by recycled paper, which requires only 25% of the steam
and 10% of the electricity required for wood pulping processes. A more
realistic energy consumption for pulp .manufacturing is about 18 million
Btu/ton of pulp produced. On this basis, approximately 766 trillion Btu
of energy was consumed for pulping processes in 1971.
Paper and Paperboard Machines
The typical paper machine uses energy, primarily as steam, for
drying the paper as it is fed through the machine. Electricity is used
for driving the machines. Although the energy consumed by the paper
#
machine depends upon several factors, an average value is about 8. 0
Purchased electricity included.
II-9
-------
million Btu/ton of paper. 7 On this basis, approximately 450 trillion Btu
of energy was consumed by paper and paperboard machines in 1971.
In a paper machine, the efficiency of energy utilization depends upon
the ability to dry the paper as fast as possible, yet maintain the quality
of the paper. This, in turn, also depends upon the effective transfer of
heat from the steam in the cylinders to the paper web. The factors that
affect the heat-transfer rate are the condensing temperature of the steam
and the condition of the internal cylinder wall. The main requirement
in a paper machine is that the temperature be evenly distributed across
the width of the cylinder presented to the paper. How this is done often
dictates the efficiency of heat transfer. For example, early machines
utilized a steam system in which the steam entered a cylinder, was con-
densed, and was discharged through steam traps on each cylinder. On
many installations, a common steam header supplied steam to all cylinders
and a common condensate header discharged into a flash tank. More
recently, it has been found that, by allowing up to 30% of the steam to
blow through the cylinders, heat transfer and uniformity can be improved.
Another factor that affects the energy utilization of a paper machine is
the use of after-dryers. After-dryers can increase the steam conserva-
tion of a machine by as much as 30%.
Energy Utilization Pattern
As previously stated, the energy consumption pattern of the paper
industry by process cannot be determined from a fuel input basis because
of the large amount of waste heat that typically is recovered and reused
in a paper mill. Still, generalizations about energy usage can be made.
Since 1958, the unit consumption of energy in the paper industry has
been decreasing. (See Figure 4. ) Several trends and developments are
responsible for this decrease: 1) the increase in kraft paper production,
a process that generally uses less energy than the other pulping pro-
cesses, and 2) the increase in the use of more efficient continuous
digesters, rather than the batch types previously used. Other trends
that have appeared tend to increase energy consumption.
H-10
-------
UNIT ENERGY
CONSUMPTION,
!06Btu
ro rv> c
3° « <
^^^^
... -
-" **
^^
^*>
^
47 '49 '51 '53 '55 '57 '59 '61 '63 '65 '6'
YEAR
A- 44 -699
Figure 4. ENERGY CONSUMPTION PER TON OF
PAPER PRODUCED FROM 1947 TO 1967
The use of waste or recycled paper in the production of paper and
paperboard has been decreasing, resulting in an increase in the use of
wood pulp. Because paper made from wood pulp requires roughly 4
times as much energy as that made from waste paper, energy consumption
increases. Current indications are that this trend may be reversing
itself for ecological reasons. Another trend has been established with
the increased interest in air and water pollution control: the installation
of electrostatic precipitators, wet scrubbers, and other control devices,
which has resulted in an increase in energy consumption.
In spite of these trends, unit energy cbnsumption has decreased and
is likely to continue to decrease as more recycled waste paper is used
and more efficient means for drying the paper web in the paper machine
are developed.
New Technology and Its Effects on Energy Consumption
Because almost all the energy consumed by the paper industry is
converted to steam and then used, the technology for energy utilization
is quite restrictive. New technologies for improving energy utilization
are limited primarily to the improvement of boiler efficiencies, which
would directly influence energy consumption, or to the improvement of
steam utilization, which would indirectly influence energy consumption.
In this industry, most of the development work is being directed toward
the improvement of steam utilization. Some work is being done to
develop direct-fired processes; such processes also would result in energy
savings.
II- 11
-------
The paper machine is continually undergoing changes to improve the
rate of drying, which in turn reduces energy consumption. In recent
years, two important developments have resulted in reductions in energy
consumption per ton of paper produced. Simply by using modern mate-
rials and equipment, the rate of mechanical water removal from the sheet
of paper in the vacuum and press sections of the machine was increased.
As a result, less water must be evaporated from the sheet in the drying
cylinder section. The effect on energy consumption is an increase in
electric consumption because of the increased contact loading of the presses
and increased vacuum at the section boxes. But this increase is more
than offset by the reduction in steam consumption. A comparison of
machinery performance before and after redesign is given in Table 2.
These data indicate that a 10% reduction in energy consumption is possible.
Table 2. THE EFFECT OF IMPROVED MECHANICAL
WATER REMOVAL ON MACHINERY PERFORMANCE7
Maximum Output, tons/hr
Electricity Used, kW/ton
Total Steam Used in Cylinders,
Ib/ton of paper produced
Total Steam Used, Ib/lb of
water removed
Increase in Energy Consumption
Due to Increase in Electricity
Used, Btu/ton
Decrease in Energy Consumptipn
Due to Decrease in Steam Used,
Btu/ton (at 1350 Btu/lb of steam)
Net Decrease in Energy Consumption
Due to Redesign, Btu/ton
Before
Redesign
5.9
357
8180
2. 06
After
Redesign
7.1
462
6000
1.43
358, 000
2, 943, 000
2, 585, 000
The second development of importance is the use of forced-convection
heat transfer for drying paper in conjunction with the drying cylinders.
In this type of system, heated air is blown through a pattern of nozzles
onto the outside surface of the paper. This increases the evaporation
rate of the water in the paper. The extent of the increase depends on
the temperature of the air and its coefficient of heat transfer to the paper
surface being dried. The use of such a system can double or even triple
11-12
-------
the normal maximum evaporation rate of 3 Ib of water per sq ft of
heated cylinder surface per hour, which is obtained by using only steam-
fed cylinders. However, because of the high air temperatures (400°F+),
direct-fired heaters that burn light fuel oil or natural gas are used. There
is no information on the extent to which these processes are used in the
industry, nor is there any basis for determining the real effects of these
developments on the industry.
In addition to the developments just discussed, much work is being
done with the objective of improving product quality and, at the same
time, reducing costs. The work is still only in the experimental stages;
thus, no operational data are available for analysis. However, once
applied, these developments probably will decrease unit energy consumption.
At present, a direct heating application is being tested in a plant.
This application utilizes two gas-fired infrared burners, each about 1-5
million Btu/hr, which cover the width of the paper machine. They are
installed between the mechanical press section and the first drying cylinder
to provide sensible heat to the wet paper web. This allows the first
drying cylinder to be used for direct evaporation of moisture, which in
turn increases machine output.
Much work has been done to develop a technique for using fluidized
beds for drying paper webs. Although the results have been promising,
the consumption of mechanical energy seems to be excessive.
Finally, work is being done to develop heat recovery systems that
can simply be attached to older machines. Such a system would no doubt
have a favorable impact on unit energy consumption in the industry.
Air Pollutant Emissions From Paper and Paperboard Mills2*6
The major emissions from a kraft pulp and paper mill are particu-
lates and odor. Odor is by far the most serious problem. These odors,
which are characteristic of all kraft mills, are caused by various sulfur
compounds released from several sources. Table 3 summarizes the
average emissions from a kraft pulp mill without controls. The emissions
are presented as discharges from the major mill components per ton of
pulp produced. Note that, according to these data, the paper machine
is not a source of emissions. The following discussion addresses the
emissions from each process and the possible systems that can be used
to control them. 11-13
-------
Table 3. AVERAGE KRAFT PULP MILL
POTENTIAL EMISSIONS WITHOUT CONTROL2
Particulates Sulfur ^
Department Ib /ton-
Digester
Batch 0 2. 5
Continuous 0 1.5
Washers 0 0. 5
Bleach Plant* 0 2. 0*
Evaporators 0 3. 5
Recovery Furnace 170 10. 0
Dissolving Tank 5 0. 15
Lime Kiln 45 1. 0
Power Boiler (Hog Fuel) 35 0. 01
Paper Machine 0 0
it-
Six-stage bleach plant CEHDED.
Does not include sulfur as SOj.
Chlorine as Ib/ton.
Digesters
Digesters are a large source of odor in a pulp mill. The major ,
malodorous compounds are methyl mercaptan, dimethyl sulfide, dimethyl
disulfide, and traces of hydrogen sulfide. 6 (These compounds are included1
as sulfur in Table 3.) These emissions are released during the cooking
and discharging of the pulp to the blow tanks. A large quantity of steam
is released during the blow of a batch digester, but normally it is
condensed.
The control systems used on the digesters depend on whether the
process is batch or continuous. Part of the problem with a batch
digester is that the flow of gases to be treated is highly variable.
Figure 5 is a diagram of a system used for controlling emissions from
a batch digester. In this system, the relief gases are directed to a
relief condenser, where the turpentine is condensed and stored for dis-
posal by sale or burning. The noncondensable vapors are eventually piped
11-14
-------
o^a »ce uUULJ.TQIV
Figure 5. FLOW DIAGRAM OF AIR EMISSIONS
FROM BATCH DIGESTER2
to the lime kiln, where they are completely burned. The other sources
of odor in the digester area are the condensates from the turpentine re-
covery system and from the accumulator blow heat recovery system.
The odorous gases that are dissolved in these liquids can be stripped
with steam or air arid then discharged with the stage seal tank overflow
from the bleach plant.
In the case of a continuous digester, the control system is simplified,
as shown in Figure 6. Here, the relief gases from the low-pressure
steaming vessel go to a condenser, where turpentine is condensed. It
is then stored for sale or burning at a later time. The noncondensables
go directly to the lime kilns. The flash from the black-liquor flash
tank goes to a condenser, and noneondensable gases containing sulfur
compounds are likewise destroyed in the lime kiln.
11-15
-------
Figure 6. FLOW DIAGRAM OF AIR EMISSIONS
FROM A CONTINUOUS DIGEST ER*
Washers
The main emissions from the washers are the sulfur compounds
mixed with vapors from the pulp-washing cycle. Conventional systems
using vacuum filters expose the pulp and liquor to the air, releasing
sulfur compounds, which are discharged to the atmosphere. If pressure-
washing systems are used, emissions come only from liquor storage
tanks. In this case, the volume is considerably less.
Emission controls for this process involve enclosure of the washers
in vented hoods. The discharges then can be collected and directed
toward a thermal oxidation unit, such as a recovery boiler, where the
offensive vapors are destroyed. All black-liquor storage tanks also can
be vented to a common duct, which then directs the discharge to a
thermal oxidation unit.
n-16
-------
Evaporators
The black liquor that is fed to the evaporators normally is oxidized
to stabilize sulfur compounds, thus preventing their loss during subse-
quent evaporation and burning operations. The discharge from the oxida-
tion system contains sulfur compounds.
The objectionable emissions from the evaporators also are sulfur
compounds. These compounds are distilled off in the evaporation process
in low volumes, but they contain high concentrations of hydrogen sulfide,
mercaptans, dimethyl sulfide, and dimethyl disulfide.
The noncondensable gases evolved from the evaporators may be
scrubbed with white liquor or piped to the lime kiln for incineration. If
they are scrubbed, the hydrogen sulfide and mercaptans can be recovered
and returned to the liquor system. The remaining sulfides then are
destroyed by burning.
Recovery Boilers
Both particulates and sulfur compounds are emitted from the recovery
boilers as a result of the combustion of the black liquor. Improper fur-
nace operation can result in the emission of large quantities of sulfur
compounds. In general, the emissions are controllable as before. Par-
ticulates are controlled by electrostatic precipitators or wet scrubbers
or both. An efficiency of at least 99% can be achieved with electrostatic
precipitators.
Odor control of effluents from the recovery boilers can be achieved
in two ways. In one case, the black liquor is oxidized, thus fixing the
sulfur compounds and causing them to remain in the chemical recovery
system rather than to be discharged into the atmosphere. The alternative
is to eliminate the direct-contact vaporizer at the recovery furnace.
Normally the hot flue gases from the furnace release sulfur compounds
from the black liquor in the direct-contact evaporator. By eliminating
this system, the emissions also would be eliminated.
11-17
-------
Lime Kiln, Dissolving Tank, and Power Boilers
The major emission from lime kilns, dissolving tanks, and power
boilers, which burn hog fuel, is particulate matter. Particulates from
lime kilns result from entrainment of the dust in the flue gases. In
the dissolving tank, vapors are generated as the result of the reaction
between the smelt from the recovery boilers and water in the tank. These
vapors entrain chemicals as particulate matter and are vented from the
tank. The preferred systems for controlling particulate emissions from
both the lime kiln and the dissolving tanks are scrubbers. In the case of
the power boilers, cyclone filters or cinder collectors can be used .'
satisfactorily.
In general, air pollutant emissions from paper mills can be controlled.
Table 4 summarizes the emissions from a typical plant after suitable
control equipment has been installed.
Table 4. AVERAGE KRAFT PULP MILL
AIR EMISSIONS WITH CONTROL2
Particulates Sulfur
Department Ib/ton-
Digester
Batch
Continuous
Washers -- --
Bleach Plant -- --
Black Liquor Oxidation -- 0. 3
Evaporators -- --
Recovery Furnace 3. 5 1.0
Dissolving Tank 0.5 0.1
Lime Kiln 1 0.2
Power Boiler (Hog Fuel) 5 0. 01
Paper Machine --
11-18
-------
(O
UJ
H
£
z>
o
£L
900
800
700
600
500
400
300
200
100
t 1971 '72 '73 74 '75 '76 '77 '78 79 80 '81 '82 '83 '84 '85
Figure 7. PROJECTED ANNUAL AIR POLLUTANT EMISSIONS
FROM. KRAFT PULP MILLS, WITH CONTROLS
II -19
-------
These data indicate that most of the erhissions, particularly the most
objectionable ones, are virtually eliminated by more or less conventional
means. In spite of these controls, an estimated 560 million pounds of
participates and 90 million pounds of sulfur were emitted in 1971. These
figures, as shown in Figure 7, are expected to climb to 890 million and
143 million pounds, respectively, by 1985.
Improved fuel utilization efficiency within this industry will occur
primarily as the result of improved steam utilization. Consequently,
the less fuel burned for steam generation, the less air pollutants will
be emitted from the boilers as a result of this combustion. These
emission factors cannot be quantified because they depend on the specifics
of the fuel, such as ash and sulfur content of coal and sulfur content of
oil. Other emissions are not expected to be affected by implementation
of new technologies.
References Cited
1. American Gas Association, A Study of Process Energy Requirements
in the Paper and Pulp Industry, Catalog No. C200037 Arlington, Va.
n.d.
2. Hough, G. W. and Guoss, L. J., 1'Air Emission Control in a Modern
Pulp and Paper Mill," Am. Pap. Ind. 51, 36-44 (1969) February.
3. Miller, R. L. , "Kraft Pulpers and Pollution Problems and Prescrip-
tions," Chem. Eng. 1^, 52-56 (1972) December 11.
4. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, D. C. : U.S. Government Printing
Office, January 1972.
5. U.S. Department of Commerce, Bureau of the Census, "Fuels and Elec-
tric Energy Consumed," 1972 Census of Manufactures, Special Report
No. MC72(SR)-6. Washington, D.C.: U.S. Government Printing Office,
July 1973.
, 6. Walther, J. E. and Amberg, H. R. , "A Positive Air Quality Control
Program at a New Kraft Mill, " J. Air Pollut. Control Assoc. 20,
9-18 (1970) January.
7. Webzell, A. B., "Energy Utilization in the Paper Industry,"
Combustion 44, 36-42 (1973) February.
11-20
-------
m. SIC CODE 281 -INDUSTRIAL CHEMICALS
Summary
The industrial chemicals industry covers a very -wide range of
products; consequently, the survey of this industry can be approached in
a multiplicity of ways. For this report, the major chemical groupings
(i.e., alkalies and chlorine, gases, inorganic chemicals) are discussed,
with emphasis on the most important chemicals from the point of view of
production and energy consumption. Although SIC Code 281 applies to
industrial inorganic chemicals only, this section also covers certain or-
ganic chemicals (SIC Code 286) such as benzene and toluene. Finally,
this section covers certain miscellaneous chemicals (SIC Code 289) such
as carbon black.
The industrial chemicals industry is the largest energy-consuming
sec'f:or of industry. In 1971, the industries classified under SIC Code 28
(chemicals and allied products) consumed an estimated 5107 trillion Btu
of energy for process heat, mechanical drives, and feedstock. Of this
total, 3764 trillion Btu of energy, about 74% of the total, was consumed
by the industries producing the chemicals discussed in this report.
The industrial chemicals industry utilizes a large number of pro-
cesses for producing a wide variety of chemicals. In addition, companies
producing the same chemicals may use different processes in manufac-
turing, depending upon local conditions. It would therefore be impossible
to examine every process used for manufacturing every chemical. The
chemicals studied are those produced in the greatest quantities based on
the assumption that they consume the greatest amounts of energy. Care
was taken to note the effects of using fuel as feedstock. The minimum
amount of a chemical that must be produced before -we considered it here
was arbitrarily selected as 4 million tons/yr. However, some chemicals
are not produced in such great quantities, but do consume unusually large
amounts of energy per unit of production. These select chemicals also
have been included here. A list of the chemicals studied is presented in
Table 1, with an estimate of their annual production and total energy con-
sumption in 1971 or, in some cases, 1972. These are the latest years for
which figures on individual chemicals are available.
in-i
-------
Table 1. ANNUAL PRODUCTION AND ENERGY CONSUMPTION OF
SPECIFIC INDUSTRIAL CHEMICALS*
Chemical
Group
Alkalies and
Chlorines
Industrial
Gases
Inorganic
Chemicals
Organic
Chemicals
Chemical
Chlorine
Sodium carbonate (syn.
Sodium hydroxide
Acetylene
Oxygen
Hydrogen
Nitrogen
Ammonia
Aluminum oxide
Ammonium nitrate
Nitric acid
Sulfuric acid
Sulfur
Carbon black
Phosphoric acid
Methanol
Benzene
Toluene
31
7
Annual Production
9, 352,437 tons
4, 274, 763 tons
9, 666, 521 tons
11, 568 X 106 CF
353,000 X 106 CF
58, 890 X 106 CF
193, 540 X 106 CF
14,028, 795
6,461,978
6, 605, 296
6,742, 130
300,000
868,000
1, 508, 568
6,240,041
tons
tons
tons
tons
tons
tons
tons
tons
2,474, 952 tons
3,923,408 tons
3, 152, 485 tons
(g) (1971)
(1971)
(1971)
(1972) .
(1972)
(1972)
(1972)
(1971)
(1971)
(1971)
(1971)
(1972)
(1971)
(1971)
(1971)
(1971)
(1971)
(1971)
Annual Energy Consumption,
101Z Btu
285
161
252
35.0
15.2
28. 1
6.6
491
82. 3
51.1
9.0
72. 1
262.4
60
92.6
A-54-792
Including feedstocks.
-------
Alkalies and Chlorine
The alkalies and chlorine segment of the industrial chemicals
industry produces chemicals such as potassium and sodium carbonates,
caustic soda, soda ash, potassium hydroxide, and chlorine. The manu-
facture of chlorine, soda ash (sodium carbonate), and caustic soda (sodium
hydroxide) consumes most of the energy within this group. In 1971, energy
consumption in the manufacture of these chemicals is estimated at
446 trillion Btu. Energy consumption is expected to increase to about
595 trillion Btu by 1980.
The demand for these chemicals has been high, resulting in a
historical growth rate in production of nearly 9%.1 Currently, the
industry is producing at or very near capacity. This -will result in a re-
duction in growth rate in the future because growth will depend on building
new plants to increase capacity. Limited power and fuel supplies also
will slow this industry's growth. Finally, natural sources of soda ash are
increasing, reducing the need for synthetic soda ash and, in turn, reducing
the growth rate slightly. In spite of these conditions, the industry is ex-
pected to continue to grow at a rate of 7-8% per year. (See Figure 1. )
Chlorine and Caustic Soda
Chlorine and caustic soda are produced simultaneously by the elec-
trolysis of sodium chloride in either diaphragm cells or mercury cells.
One ton of chlorine is produced for every 1. 13 tons of sodium hydroxide.
In the U.S. , the diaphragm cell is used for more than 65% of the total
production of these chemicals.9 Figures 2 and 3 are flow diagrams of the
production of these chemicals using the two types of cells.
In the production of these chemicals, a saturated solution of sodium
chloride is steam-heated and then purified by treatment with sodium car-
bonate and caustic soda. After purification, the solution is neutralized
with hydrochloric acid, reheated, and then fed to the electrolytic cells.
An electric current is passed through the salt solution, decomposing the
salt to form a 10-12% sodium hydroxide solution. In addition, hydrogen
gas is formed at the cathode and chlorine gas at the anode.
Ill-3
-------
IO,U
Q 12,000
0
z"
l_ 10,000
o
1
g: 8,000
6,000
4,000
2.000
^/
/^
I
CAUSTIC SODA/
-~Jf
f^1
J?
^y
//
//
//
y
/
^
'CHLORINE
/
//
//
/
/ ,
/ /
/
I960 '55 '60 '65 '70 '75 '80 '85
YEAR
A-54-823
Figure 1. ANNUAL PRODUCTION OF CHLORINE
AND CAUSTIC SODA WITH PROJECTION TO 1985
III-4
-------
8941
SALT
SALT
HANDLING
TREATING
CHEMICALS
^ BRINE
PREPARATION
,_^ 4 PURIFICATION
PURE
BRINE
A.C. P
RECTI
ELECTR
\
OWER
COMPRESSION LIQUID
< LIQUEFACTION CI2 *"
FIEHS
°L:SIS C,z _ COOLING
MERCURY } « DRYING
PRECIPITATES
PARTIALLY DEPLETED BRINE i
HYDROGEN
NoOH
Figure Z. FLOW DIAGRAM OF MERCURY CELL
CHLOR-ALKALI PLANT9
TREATING
CHEMICALS
BRINE
PURIFICATION
A.C. POWER
RECTIFIERS Hz
1 t
PURE
BRINE
ELECTROLYSIS
PRECIPITATES
CELL LIQUOR
STEAM
EVAPORATION
PLANT
No OH
SOLID NoCI
Figure 3. FLOW DIAGRAM OF DIAPHRAGM CELL PROCESS9
ni-5
-------
The amount of energy consumed in. the manufacture of these chemicals
varies considerably, depending upon the type df cell used. Typically, the
diaphragm cell process consumes about 35 million Btu/ton of caustic soda
produced, and the mercury cell process consumes only about 12. 1 million
Btu/ton of caustic soda produced. These are equivalent to 39.5 million
Btu/ton of chlorine and 13.7 million Btd/ton of chlorine, respectively.
One difference between these two processes is that the mercury cell
actually consumes 15% more electricity than the diaphragm cell (9.4 million
Btu/ton of caustic soda versus 8. 1 million Btu/ton of caustic soda, re-
spectively). However, the caustic soda solution produced in mercury cells
does not need to be concentrated in evaporators, as is the case with the
caustic soda produced in diaphragm cells. As a result, the steam required
in the mercury cell process is about one-tenth the amount required in the
diaphragm process. Note that the figures presented for energy consumed
as electricity do not include the energy losses in generation and trans-
mission. At the site of generation, mercury cells consume about
3. 9 million Btu more energy than diaphragm cells, but this is still con-
siderably less than the amount of energy that the mercury cell process
saves by using so little steam.
The Kel-Chlor Process is a new process for the manufacture of
chlorine. u The process involves oxidation of hydrogen chloride -with a
catalyst to form chlorine and water. The process consumes very little
energy for power or heat; the energy requirements, however, are not
known because no commercial-scale plants exist as yet. An additional
benefit of this process, from the industry's point of view, is that hydrogen
chloride is readily available for feedstock because it is a by-product of a
very large number of chlorination reactions. As such, it usually presents
a great disposal problem.
Air pollutant emissions from the manufacture of chlorine and caustic
soda comprise chlorine gas, carbon dioxide, carbon monoxide, and hydro-
gen. Mercury vapors also are emitted when mercury cells are used in the
electrolysis. About 1. 5 Ib of mercury is lost for every 100 tons of chlorine
produced. Of these emissions, chlorine has been regarded as the most -
serious, although the mercury emissions are being scrutinized. Hydrogen,
III-6
-------
which is one of the by-products of the electrolytic process, is usually
collected and either burned elsewhere in the plant or sold. Table 2 sum-
marizes the chlorine emissions and shows that, uncontrolled, chlorine
emissions are 60-100% greater for mercury cell plants than for diaphragm
cell plants. But Table 2 also indicates, in part, that these emissions can
be effectively controlled by scrubbers. Other methods of control include
the use of the chlorine in other plant processes and recovery of the
chlorine from the effluent gas stream.
Table 2. EMISSION FACTORS FOR CHLOR-ALKALI PLANTS18
Chlorine gas,
Type of Source lb/100 tons
Liquefaction Blow Gases
Diaphragm Cell, uncontrolled 2, 000 to 10, 000
Mercury Cell, * uncontrolled 4,000 to 16,000
Water Absorber 25 to 1,000
Caustic or Lime Scrubber 1
Loading of Chlorine
Tank Car Vents 450
Storage Tank Vents 1,200
Air-Blowing of Mercury-Cell 500
Brine
»i»
Mercury cells lose about 1. 5 pounds mercury per 100 tons of
chlorine liquefied.
Trends in Chlorine-Caustic Soda Manufacturing
Figure 1 shows the production rates of both chlorine and caustic soda15
with projections to 1985. Fuel consumption per unit of production for each
type of electrolytic cell has remained constant during the last 20 years and
will probably continue unchanged in the future. However, total fuel con-
sumption is likely to decrease in the future, in spite of the expected growth
in production. One potential factor affecting future energy consumption
rates is the possibility of increasing the number of mercury cells in
operation. However, the emission of mercury and new water -quality
standards threaten to foreclose new mercury cells (as discussed below).
But if cleanup of the mercury emissions can be effected, large savings
HI-7
-------
in energy consumption would result. Figure 4 shows the effect of in-
creasing the number of mercury cells to account for 65% of production
rather than the 35% for which they are currently accountable. If this 65%
utilization were reached by 1985, the projected annual energy consumption
would be reduced by 25%. If mercury cells were used for 100% of
production, annual energy consumption in 1985 would be about 14% lower
than that in 1971.
The other major factor that could affect the energy consumption of
this industry is utilization of the Kel-Chlor Process. Apparently, 100%
utilization of this process would reduce energy consumption to almost zero.
However, this process is not likely to be in wide enough use by 1985 to
have any significant effect on energy utilization patterns.
In addition to their effects on energy consumption, the directions
that the industry takes in the future also will affect air pollutant emissions.
Growth in the use of mercury cells will significantly increase mercury
emissions. At present, mercury losses are about 49,000 Ib/yr, based on
a 35% utilization of mercury cells. Figure 5 shows the projected effects
of mercury cell utilization in the future if a) the current 35% utilization is
maintained and if b) utilization increases to 65%. If utilization were to
increase to 100%, 270,000 pounds of mercury would be emitted annually.
Switching to mercury cells would not affect chlorine emissions because
effective cleanup measures exist regardless of the type of cell that is used.
If the Kel-Chlor Process becomes widely used, it is likely to cause
a significant increase in emissions of nitrogen compounds (such as, but
not exclusively, NO ), because NO are used as the catalyst. Although
-X X.
methods for controlling these emissions are available, data on their
effectiveness relative to this process are not.
Sodium Carbonate (Synthetic Soda Ash)
The other major chemical manufactured by the alkalies and chlorine
industry is sodium carbonate. There are two sources: natural deposits,
from which nearly 2.9 million tons was obtained in 1971, and synthetic
manufacturing, which accounted for nearly 4. 3 million tons in 1971. Soda
ash production in the U.S. has been increasing gradually, and this trend
is expected to continue in the future. (See Figure 6.) However, there has
III-8
-------
00%MERCURY CELLS
UJ 300
200
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-54-826
Figure 4. EFFECT OF MERCURY CELL UTILIZATION
ON ANNUAL ENERGY CONSUMPTION OF
CHLORINE-CAUSTIC SQEA^MANUFACTURE
HI
-------
"
$
3
or
o
UJ
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-54'827
Figure 5. EFFECT OF USING MERCURY CELLS
IN CHLORINE-CAUSTIC SODA MANUFACTURE
ON MERCURY EMISSIONS
III-10
-------
o
Q
1950 '55 '60
TOTAL
^SYNTHETIC
y
NATURAL
'65 ,'70 '75
YEAR
'80 1985
A-54-824
Figure 6. ANNUAL PRODUCTION OF SODA ASH
WITH PROJECTION TO 1985
III-11
-------
been a shift away from synthetic soda ash toward natural sources (also
shown in Figure 6). As a result, the total energy consumed in the manu-
facture of this chemical is likely to drop gradually from its 1971 estimated
level of 161 trillion Btu/yr.
The primary process for manufacturing soda ash is the Solvay
Process.6 The first step is purification of saturated salt brine with
ammonia and carbon dioxide waste process gases in absorbers. During
this step, calcium, magnesium, and other heavy metals are removed. The
purified brine is ammoniated and then carbonated to form sodium bicar-
bonate. The resulting slurry is filtered and then heated to about 350° F in
a. rotary kiln. The product is cooled and stored or packaged.
Primary consumption of energy occurs in the calcination step, in
boilers, and in the lime kilns that produce the calcium carbonate used in
the carbonation step. The primary fuel used is coal. However, there has
been a. trend toward the use of fuel oil in recent years. Energy consump-
tion is estimated to be about 38 million Btu/ton of soda ash produced.
Because no new synthetic processes are being developed, the unit energy
consumed in this sector is likely to remain unchanged in the near future.
Air pollutants from the manufacture of soda ash comprise primarily
ammonia (as shown in Table 3) and dust.
Table 3. EMISSION FACTORS FOR SODA-ASH
PLANTS WITHOUT CONTROLS18
Type of source
Particulates,
Ib/ton
Ammonia,
Ib/ton
Ammonia Recovery*
Conveying, Transferring,
Loading, etc.
Represents ammonia loss following the recovery system.
The ammonia results from venting the gases from the brine purification
system. Additional losses also can be incurred during the transfer of the
ammonia from tank trucks to storage tanks. Dust emissions are primarily
-------
the result of handling operations, that is, conveying, transferring, and
loading. Some dust also is emitted from the rotary kilns.
Trends in the Manufacture of Soda Ash
The only change that can be anticipated in this industry is a gradual
shift from synthetic soda ash to natural soda ash. At present, more than
80% of the synthetic soda ash production is by the Solvay Process. How-
ever, the greater availability of natural soda ash accounts for the shift.
'!'
In addition, energy consumption is very low' in mining the soda ash
another likely cause for the shift. As a result of this shift, which will be
gradual, annual ammonia emissions also will decrease; however, particu-
late emissions probably will increase.
Industrial Gases
Nine principal gases are produced by the industrial gases segment
of the industrial chemicals industry. The gases of most concern are
acetylene, hydrogen, nitrogen, and oxygen because they are produced in
the greatest abundance of all the industrial gases and consume the major
part of the energy consumed by the entire segment. Figure 7 shows the
trends in production of these gases since I960. With the exception of
acetylene, the industrial gases presented have exhibited tremendous growth
in the last 10 years because the market for these gases has rapidly ex-
panded. At present, industrial gases are the fastest growing segment of
the inorganic chemicals industry, primarily because of new techniques in
the primary metals industries and the chemical industry. In the U.S.,
oxygen and nitrogen growth rates have been 19.9 and 26% per year,
respectively, since I960.11 This rapid growth is expected to continue for
some time to come.
Energy consumption in 1972 for the production of these gases is es-
timated at 85 trillion Btu. The sources of energy were primarily natural
gas and electricy; however, small amounts of fuel oil and coal; comprising
less than 5% of the total energy consumption, were used. The fuels are
used for raising steam, for mechanical drives, and for feedstock.
No actual figures are available.
IH-13
-------
375
350
325
300
275
*EXCLUDES REFINERY
AMOUNTS AFTER 1968
I960 '62 '64 '66 *68 '70 '72 '74
A-34-830
Figure 7. ANNUAL PRODUCTION OF SELECTED
INDUSTRIAL GASES
III-14
-------
Acetylene
There are several methods for manufacturing acetylene, but the most
commonly used method is by reaction of calcium carbide with water. This
reaction is exothermic, releasing about 165 Btu/CF of acetylene produced,
and as such does not consume energy. The major amount of energy con-
sumed by this industry is in the production of the reactant, calcium
carbide.
To form calcium carbide, lime and coke are mixed in a 6:4 ratio and
heated to 3600°-3800°F in an electric furnace. The liquid calcium carbide
thus formed is cooled and subsequently solidified. The solid mass is
crushed and screened to size, an operation that is carried out in a nitrogen
atmosphere to prevent explosion of acetylene that would be generated by
moisture in air. The energy consumed to form 1 ton of carbide is
approximately 28 million Btu. Approximately 18 million Btu of coke and
10 million Btu of electricity are consumed.6 If the energy used to generate
the electricity is considered, energy consumption becomes 48 million
Btu/ton of calcium carbide produced. Because lime is a major reactant
in the formation of calcium carbide, a lime kiln is an integral part of the
acetylene plant. Lime kilns typically consume 6-8 million Btu/ton of lime
produced. In the reaction of lime with coke, carbon monoxide as -well as
calcium carbide is formed. The carbon monoxide usually is recovered and
used as a fuel in the lime kiln or as a raw material for chemical synthesis.
Acetylene is produced from calcium carbide by one of two processes:
the wet process or the dry process. In both processes, calcium carbide
is combined with water to form acetylene and calcium hydrate. In the wet
process, a large amount of water is used, and the calcium hydrate is dis-
charged in the form of a slurry containing 90% water. In the dry process,
a limited amount of -water is used, and the heat of reaction is used to dry
the calcium hydrate, making it suitable for reuse as a lime source.6
There are other processes for manufacturing acetylene utilizing
natural gas as a feedstock, including incomplete combustion with oxygen,
such as the Sachase Process, and pyrolysis on a hot surface, such as in
the Wulff Process. Neither of these processes is economically feasible,
and the energy requirements are very high. Natural gas consumption is in
III-15
-------
excess of 200 million Btu/ton of acetylene produced. Work is being done
to make such processes attractive, but so far the calcium carbide process
is the most widely used.
Air pollutant emissions from acetylene-manufacturing plants occur
during the manufacture of calcium carbide. Table 4 summarizes these
emissions.
Table 4. AIR POLLUTANT EMISSIONS FROM THE
MANUFACTURE OF ACETYLENE18
Source Particulates Sulfur Oxides Acetylene
lb/ton*
Electric Furnace
Hoods 18
Main Stack 20 3
Coke Dryer 2 3 - -
Furnace Room Vents 26 - - 18
Lime Kiln
Vertical 7. 6f
Rotary 190t - - - -
*!'
Emissions per ton of calcium carbide produced.
Emissions can be reduced by conventional means by 99%.
In addition, carbon monoxide probably is emitted during the formation of
the calcium carbide, and acetylene is emitted from the recovery process
during the reaction of calcium carbide and water to form acetylene. How-
ever, no data are available on these emissions.
Trends in Acetylene Manufacturing
The manufacture of acetylene has been declining gradually for a
number of years, primarily because the chemical industry that previously
used large amounts of acetylene for feedstock is switching to other, less
hazardous and more available hydrocarbons. This trend is likely to con-
tinue in the future, resulting in a gradual reduction in annual energy
III-16
-------
consumption and air pollutant emissions. Although other processes exist,
they probably will not replace the calcium carbide rbute.
Oxygen and Nitrogen
The manufacture of oxygen and nitrogen is increasing faster than the
manufacture of any other chemical. In 1972, 353 billion CF of oxygen and
194 billion CF of nitrogen were produced, 15 an increase over 1971 of 11 %
and 15%, respectively. Energy consumed in the manufacture of these
gases in 1972 is estimated at 86. 8 trillion Btu.
The primary method of manufacture is the same for both of these
gases, liquefaction and subsequent fractioiiation of air. In fact, both gases
are produced simultaneously. Several cycles are used to obtain the final
product; they all involve the basic processes of air purification, com-
pression, and refrigeration, but the equipment differs. The specifics of
the production of these gases are not important, nor is differentiation of
the various processes, because the energy consumed is about the same.
The energy consumed in the manufacture of both oxygen and nitrogen
is electricity and steam, 22% and 78%, respectively. The amount of
energy consumed is 245,000 Btu/1000 CF of oxygen produced, or
60,000 Btu/lOOOCF of nitrogen produced. However, if the energy consumed
to generate the electricity is considered, the actual energy consumption is
3 times as much. The energy is used primarily for mechanical drives.
Air pollutants are not emitted from oxygen or nitrogen plants. If
electricity generation is considered, a small amount of emissions can
be attributed to the manufacture of oxygen and nitrogen. However, emissions
from electricity generation depend upon the source of energy (coal, oil,
natural gas, nuclear, etc.). (See the section on electricity and steam gen-
eration.) Because we cannot ascertain the combination of fuels that are
consumed to generate the electricity specifically used in these industrial
gas plants, we cannot attribute quantitative emission numbers to the manu-
facture of these gases.
No major advances in the processes for manufacturing these two
gases have occurred, so it is not likely that any significant changes will
alter the energy-consuming characteristics of these industries. Production
III-17
-------
of these gases should continue to grow at a great rate. By 1975, annual
oxygen production in the U.S. is expected to be 600 billion CF, a 70%
increase over 1972 production.7 Oxygen, 60% of which is used in the steel
industry, is beginning to be used in applications for which higher tempera-
tures, greater productivity, and reduced fuel consumption are desired,
such as the glass industry and some nonferrous smelting operations. As
fuels become scarcer, this trend is likely to increase, causing even
greater growth of this industry. Nitrogen is used as a feedstock in many
chemical-manufacturing processes, ammonia-manufacturing being a large
user. However, it is becoming widely used as a refrigerant as well; thus,
its future growth/would seem to be ensured.
Hydrogen
The other major industrial gas produced in large volume and con-
suming large amounts of energy in the manufacturing process is hydrogen.
Figure 7 shows that the annual production of hydrogen totaled 58 billion/CF
in 1972. This total does not include hydrogen produced and used in
petroleum refineries, which accounts for more than 75% of the total hy-
drogen produced. Prior to 1969, hydrogen produced in refineries was
included in the statistics, thus accounting for the apparent drop in produc-
tion in 1970 shown in Figure 7. Including the hydrogen produced in
refineries, production is estimated to have been about 250 billion/CF in
1972 and is expected to increase to 312 billion CF/yr by 1975.
There are several methods for manufacturing hydrogen, but the most
predominantly used method is the catalytic steam reforming of hydrocar-
bons, including methane, oil refinery gas, liquefied petroleum gas, natural
gasoline, naphtha, and fuel oil. 12 In catalytic steam-hydrocarbon re-
forming, desulfurized feed is mixed with steam and passed over a nickel-
based catalyst in a reforming furnace. The effluent gases containing
carbon monoxide and hydrogen are cooled with steam or condensate to the
point at which the carbon monoxide reacts with the steam to form carbon
dioxide and additional hydrogen. The process requires about 450, 000 Btu
of energy to produce 1000 CF of hydrogen. About half of the energy is
consumed as feedstock; the other half is consumed in raising steam.
Ill-18
-------
Other methods of manufacturing hydrogen are dissociation of ammonia,
the steam-iron process, and electrolysis of water. The main deterrent to
the use of these processes is the high cost of the raw materials and power.
Thus, steam-methane reforming probably will dominate in the manufacture
of hydrogen in the near future. However, steam-reforming processes that
use hydrocarbons other than methane have been developed, and they might
replace the steam-methane reforming process when methane becomes
scarce.
Specific data on the air pollutant emissions from the manufacture of
hydrogen are not available. However, small amounts of methane and hy-
drogen probably are lost during processing. In addition, emissions occur
as a result of the combustion in boilers to raise steam. The emissions,
of course, depend on the fuel used. Table 5 summarizes the average
emissions from refinery boilers for the two most commonly used fuels. '"
Some of these emissions, such as NO , probably can be reduced by using
X.
new combustion techniques that are being developed. In general, these
emissions can be controlled by proper treatment of the effluent gases.
One new process for manufacturing hydrogen recently has been
developed. This process involves the partial oxidation of hydrocarbons and
is said to require less energy for heat than the steam-hydrocarbon re-
forming process. The energy consumed for heat in this new process is
estimated at 35,000 Btu/lOOOCF of hydrogen produced, which represents
an 80% reduction in that required in the steam-hydrocarbon reforming
process. In spite of this reduction, this new process is not expected to
gain prominence until its economics become favorable. At present, the
cost of most liquid hydrocarbons is too high, compared with the cost of
present feedstocks, to justify the switch.
A second method for manufacturing hydrogen on a large scale is the
electrolysis of water, for which several schemes have been developed.
For more complete information on boiler emissions, see Section XIV,
Electricity and Steam Generation.
HI-19
-------
Table 5. AVERAGE AIR POLLUTANT EMISSIONS FROM
REFINERY BOILERS USED IN PRODUCTION OF HYDROGEN4
Unit Emission Rates
1972 Emission Rates
1 1
R
t\»
o
Emissions
NO
X
Aldehydes
Hydrocarbons, as hexane
Organic Acids, as acetic acid
Particulate Matter
S02
SO3
Fuel Gas
0.
0.
0.
0.
0.
IK / 1 nnn P TT TT
05
0007
0055
0029
0044
jf
'i-
*
Fuel Oil
0.
0.
0.
0.
0.
095
0008
0048
0154
0280
o-
jf
ri"
Fuel
12,700
175
1,375
725
1, 100
*
*
Gas
,000
,000
,000
,000
,000
23
1
3
7
Fuel
,800
200
,200
,850
,000
Oil
,000
,000
,000
,000
,000
->*
'!=
Sulfur dioxide and sulfur trioxide emissions are dependent upon sulfur in fuel.
-------
The primary objection to this method in the past has been the high cost of
electricity; consequently, the target of most of the research has been to
reduce electricity consumption to the absolute minimum. Toward this
end, a high-temperature cell that uses a zirconia electrolyte7'8 has been
developed, but as yet has not become economical. Another proposal is
the production of hydrogen by electrolysis at nuclear power plants, which
would be an economical source of electricity in the future.
Trends in Hydrogen Manufacturing
The consumption of hydrogen is expected to continue to increase as
its demand in ammonia-manufacturing, petroleum-refining, and
other chemical-manufacturing processes increases. By 1975, hydrogen
consumption is expected to increase by 20% over 1972 consumption.
Demand for hydrogen probably will increase even more as its potential use
as a fuel in industrial and commercial applications is realized. On the
current energy requirement basis; energy consumption would be expected
to grow linearly with the growth in production. However, if the partial
oxidation process becomes widely used, it would have a significant effect
on annual energy consumption (Figure 8). Figure 8 assumes a linear
growth of the use of partial oxidation to 100% by 1985. Although data on
air pollutant emissions from the partial oxidation process do not exist,
hydrocarbon emissions could increase over current levels.
A switch to an electrolytic process for manufacturing hydrogen would
increase energy consumption if the energy for electricity generation is
considered. A perfect electrolytic cell would require about 270, 000 Btu of
electric energy to produce 1000 CF of hydrogen.8 (in practice, the energy
consumed is about 540,000 Btu/1000 CF.) At a 30% generation efficiency
for electricity, the actual energy consumption is about 1. 8 million Btu/
1000 CF, 200% greater than the energy required for steam-hydrocarbon
reforming. Air pollutant emissions also would be affected, but the degree
would depend, of course, on the source of electricity and the cleanup
methods employed.
111-21
-------
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-54-8Z8
Figure 8. PROJECTED ANNUAL ENERGY CONSUMPTION
IN MANUFACTURE OF HYDROGEN
IH-22
-------
Industrial Inorganic Chemicals
The industrial inorganic chemicals industry includes establishments
engaged in the manufacture of literally hundreds of chemicals. This in-
dustry consumes about 26% of the total energy consumed by the industrial
chemicals sector of the chemical and allied products sector of industry.
In 1971, the total energy consumed is estimated to have been 985 trillion
Btu. 14»16 The majority of the fossil-fuel energy was consumed either in
boilers or as feedstock. Only a relatively small amount was consumed in
direct-fired processes. As would be expected for such a large diversifi-
cation of processes, all forms of energy are consumed, although more than
half of the energy consumed is electricity. The most important chemicals
produced within this sector are shown in Table 1. The manufacture of
these chemicals alone accounts for more than 50% of the total energy con-
sumed by the industrial inorganic chemicals sector. Growth of the
industrial chemicals industry has been steady, averaging about 10% per
year in recent years. The industry output is expected to increase by 50%
over current levels by 1980, 17 as -will the annual energy consumed by the
industry.
Air pollutant emissions from the manufacture of inorganic chemicals
also are diverse. Table 6 summarizes the air pollutant emissions from
the manufacture of the most important chemicals within this group. As
production increases, these emissions will increase, but with an increase
in the use of available control methods and hardware, the amount of
emissions per unit of production can be reduced.
Ammonia
The manufacture of ammonia consumes more energy annually than
the manufacture of any other chemical in this sector. Table 1 shows that,
in 1971, more than 14 million tons of ammonia was produced.15 Of this
amount, 80% was used in fertilizers and the rest in industrial and military
applications. Although fertilizers are classified separately (SIC Code 287),
the U. S. Department of Commerce includes ammonia produced for
fertilizer in the inorganic chemicals sector of industry.
IH-23
-------
Table 6. AIR POLLUTANT EMISSIONS FROM THE MANUFACTURE
OF SPECIFIC INDUSTRIAL INORGANIC CHEMICAL18
Chemical
Source or
Process
Air Pollutant Emissions
Particulates Hydrocarbons x
Ib/ton
NO SO
x
Carbon
Monoxide
Other
Ammonia
H
n
I
fO
Plants -with
methanator
Purge gas
Storage and load
Plants with CO
and regeneration
systems
90
Negl
3 (Ammonia)
200
Carbon
Black
Regenerator exit
Purge gas
Storage and loading
Channel
Thermal
Furnace
Gas
Oil
Oil or gas
- -
2300
Negl
220
60
10
90 - -
11,500
as methane
Negl
1, 800
400
200
Negl
33, 500
Negl
5, 300
4, 500
7
3
200
385
-
(H2S)
-------
Annomia production in the U.S. has been increasing at a rate such
that total annual production is currently 3 times as great as in I960. (See
Figure 9.) However, ammonia plants, which rely on natural gas for energy
and feedstock, are being restricted because of the shortage of this fuel;
consequently, little increase in production is forecast in the near future.
Instead, the U.S. is becoming increasingly dependent on foreign sources.17
The most widely used process; for manufacturing ammonia is the
catalytic reaction of nitrogen and hydrogen at high temperatures and
pressures.10'13 The actual manufacture is a series of six consecutive
processes incorporated into one plant: 1) reforming of methane with steam
to give hydrogen and carbon oxides, 2) secondary reforming of the product
gas from step 1 to further reduce its methane content, 3) high-temperature
conversion of carbon monoxide to carbon dioxide with shift catalyst,
4) low-temperature shift conversion, 5) methanation, and 6) ammonia
synthesis. Energy consumption in an ammonia plant is about 35 million
Btu/ton of ammonia produced, including the energy used as a raw material. 10
About 20 million Btu/ton of ammonia produced is consumed as feedstock,
8 million Btu/ton for raising steam, and 7 million Btu/ton as fuel for
process heat. For economic reasons, natural gas is the primary source
of energy, but if natural gas is unavailable, other sources can be used.
In countries -where natural gas is not available cheaply, light naphtha from
petroleum refineries is used for feedstock. In spite of this,, the energy
requirements (excluding feedstock) are about.the same as those for the
process in which natural gas is used. Approximately 3% more energy is
required to reform the naphtha than the methane.
Table 6, which summarizes air pollutant emissions from the manu-
facture of certain inorganic chemicals, shows that carbon monoxide,
hydrocarbons, and ammonia are the primary emissions from the manu-
facture of ammonia. The values given are those for plants without controls.
However, through the use of scrubbers, about 99% of the ammonia in the
gas streams can be recovered. The carbon monoxide, which makes up
about 75% of the regenerator gases, is recovered and burned for fuel when
needed, usually for driving compressors.
TH-25
-------
It
13
12
II
10
£
0
zf 8
O
S3 7
Q
0 c
I 6
5
4
3
2
1
19
y
^
/
/
/
/
/
y
1
J
1
1
50 '55 '60 '65 1970
YEAR A-54-825
Figure 9. ANNUAL PRODUCTION OF AMMONIA
III-2 6
-------
Although no major new processes have been developed for manu-
facturing ammonia, certain trends have affected the energy consumption
pattern of this industry in recent years. Among the most significant de-
velopments have been high-pressure reforming, new catalysts, new carbon-
dioxide-removal processes, and the use of new feedstock materials. As
a result, energy consumption has decreased from about 71 million Btu/ton
of ammonia produced with coke or coke-oven gas to the current level of
35 million Btu/ton of ammonia with natural gas. Pressures in the re-
former are expected to continue to increase, thus increasing the process
efficiency. The trend toward higher pressures should also make partial
oxidation of hydrocarbons competitive with reforming as a source of
hydrogen. However, the ability of partial oxidation to gain advantage over
reforming will depend on the future cost and availability of the various
feedstocks.
Ammonium Nitrate
The manufacture of ammonium nitrate consumes about 7.7 million
Btu/ton of product. (See Figure 10.) In 1971, approximately 51.1 trillion
Btu was consumed to produce about 6. 6 million tons. The primary forms
of energy used are electricity (10%) and steam (90%), 6 the fossil-fuel
sources of which cannot be determined. (See the section on electricity and
steam generation.)
Although there are several variations, depending upon the desired
end product, the basic process involves the reaction of ammonia and nitric
acid to yield ammonium nitrate. The resulting product then is processed
to form crystals or granules.
No information is available on the emissions from the manufacture of
ammonium nitrate. However, there are, no doubt, some ammonia
emissions. Indirectly, the emissions from the production of ammonia and
nitric acid (Table 6) also could be considered, but this would result in a
double-counting of emissions because they are considered elsewhere.
IH-27
-------
10,000
9,000
8,000
o 7,000
10
g
2 6,000
o
Q
5,000
4,000
3,000
2,000
1,000
1950 '55 '60 '65 '70 '75 '80 '85
YEAR
A-54-831
Figure 10. ANNUAL PRODUCTION OF AMMONIUM NITRATE
WITH PROJECTION TO 1985
IH-28
-------
The production of ammonium nitrate is expected to continue to in-
crease at about the same rate as in the past. Thus, annual energy con-
sumption also will increase in proportion to the increase in production.
Because the future of ammonium nitrate production depends upon the
production of ammonia for use as a reactant, ammonium nitrate production
may decline. If this happens, the effect on energy consumption will be
directly related to the decline in production.
Sulfur
The manufacture of sulfur consumes about 9. 1 million Btu/ton of
sulfur produced. In 1971, about 72.1 trillion Btu of energy was consumed
to produce nearly 8.0 million tons of sulfur.6*15
About 70% of the sulfur consumed in the U.S. is produced by the
Frasch Process,2 15% is imported, and the remaining 15% is produced
from sour gas and petroleum refinery streams. In the Frasch Process, a
well is drilled into a sulfur-bearing deposit, and three stringers of
concentric pipe, perforated near the bottom, are lowered to the bottom.
Superheated water is pumped into the well to melt the sulfur, which, when
molten, settles to the bottom of the well. Compressed air is forced down
the innermost pipe and forces the sulfur to the surface through one of the
annular spaces.
The majority of the energy consumed in this process is for heating
the water. By using heat exchangers, the thermal efficiency is about 80%,
leaving little room for improvement. A very small amount of energy is
consumed for driving compressors and other mechanical devices. Thus,
process energy requirements will probably not change in the near future.
Sulfur production in the U.S. has been increasing at a relatively low
rate during the past two decades (Figure 11) and is expected to continue at
about the same rate in the future. Asa result, total annual energy con-
sumption also will increase from its present level of 72 trillion Btu/yr
to about 80 trillion Btu/yr by 1985.
Ill-2 9
-------
10
2 8
0
o
cc
Q.
-J 6
1950 '55 '60 '65 '70 '75 '80 '85
YEAR
A-54-833
Figure 11. ANNUAL PRODUCTION OF SULFUR
BY FRASCH PROCESS
Data on air pollutant emissions from the Frasch Process are not
available. However, the emissions can be assumed to be primarily the
result of fossil-fuel combustion in the boilers used for superheating the
water, including some hydrocarbons, carbon monoxide, NO , and SO ,
X X
depending on the sulfur content of the fuel. Because no breakdown of fuel
usage by this industry is available, estimates of these emissions cannot
be made. (See the section on electricity and steam generation.)
Carbon Black .-
The carbon-black industry is one of the larger consumers of energy
in the industrial chemicals sector of industry. In 1971, carbon-black plants
consumed an estimated 262 trillion Btu of energy in the production of
1. 5 million tons of product.2 The types of energy consumed were primar-
ily natural gas and fuel oil, which are used as feedstock. Air pollutant
emissions, although previously a problem, have been significantly reduced;
in some plants, reductions of 99.5% have been achieved.
Ill-30
-------
Carbon black is produced by one of three processes: the furnace
process, the channel process, and the thermal process. Figure 12 shows
the relative importance of each of these processes on a production basis
from 1950 to the present.3'5 At present, the furnace process is the most
widely used. In this process, fuel is fed into a furnace, in which it is
mixed with an amount of air insufficient for complete combustion, resulting
in the formation of carbon particles, carbon monoxide, hydrogen, nitrogen,
and water vapor. The mixture is cooled, and the carbon-black particles
are separated out in an electrostatic precipitator or in cyclone collectors.
The feedstock determines the specifics of the plant design. Fuel oil is
used as feedstock for about 75% of the total carbon black produced by this
process.2
The channel process, which used to account for more than 90% of the
carbon black produced in this country, now accounts for less than 5%.
This process is similar to the furnace process in that the mechanism of
production is incomplete combustion of natural gas. However, many small
flames are used, as opposed to one big flame in the furnace process, and
the carbon black is deposited by impingement upon the flat under surf ace of
moving channels. It is removed from the channels by a series of scrapers
and collected for packaging. When the furnace burns natural gas only, the
yield of carbon black is about 2. 5 lb/1000 CF of gas, about one-third the
yield of the furnace process.5 Some producers that still use this process
enrich the gas with oil and thus raise the yield over that available with
gas alone.
In the relatively new thermal process, carbon black is produced by
thermal decomposition of natural gas rather than incomplete combustion.
At present, this process accounts for perhaps 15% of the total carbon-black
production. Unlike the other two processes, energy is consumed not only
as feedstock, but also as a source of heat for the decomposition. Opera-
ting temperatures range from 2400° to 2800°F, and the yield is about
16 lb/1000 CF of gas. Although this yield is significantly higher than that
III-31
-------
(O
O
Q
TOTAL PRODUCTION^
\CHANNEL PROCESS
7
f
7
FURNACE PROCESS
THERMAL PROCESS
7
1.6
1.5
1.4
1.3
1.2
I.I
1.0
0.9
0.8
0.6
0.5
0.4
0.3
0.2
O.I
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72
YEAR
A-54-829
Figure 12. CARBON-BLACK PRODUCTION BY PROCESS
IN THE U. S.
Ill-32
-------
from the channel process, the coarser particles that are produced have
neither the reinforcing properties nor the color of channel black.5
Figure 13 summarizes the energy utilization patterns of this industry
since 1950. The data show a large reduction in energy consumption (and
natural gas consumption) between 1950 and 1954, continuing at a lesser
rate after 1954. This reduction is a result of the conversion of the
industry from the utilization of the inefficient channel process to the much
more efficient furnace process. The data also show that liquid hydrocar-
bons are assuming the burden as primary feedstocks in the furnace process
because the carbon black yield per million Btu consumed in the furnace
process is 2 to 3 times greater when using liquid hydrocarbons than when
using natural gas.
Air pollutant emissions have been a problem for the carbon-black
industry for a number of years. In fact, one reason for the conversion
from the channel process to the furnace process -was the fewer air pollu-
tant emissions from the furnace process. (See Table 6.) The data also
indicate that emissions from an oil-fired furnace are lower than those from
a gas-fired furnace. The only exception is hydrogen sulfide, the emission
rate of -which depends upon the sulfur content of the fuel. Finally, the data
show that emissions from the thermal process are negligible. The reason
is that the exit gases, -which are primarily hydrogen (85%), methane (5%),
and nitrogen, are either recycled to process burners or used in boilers
to generate steam.
The pollution problems of a carbon-black plant can be solved in
several ways, including recycling the carbon monoxide and hydrocarbons
and burning them elsewhere in the plant, as is done in plants using the
thermal process. The use of a fabric filter system is 99. 5% efficient in
cleaning up particulates. Less efficient cleanup systems such as cyclones
and scrubbers also are in use. Efficiencies of these systems range from
90 to 97%, depending upon the specific combinations used. If the whole
industry used the fabric filter system, particulate emissions in 1971 would
have been about 7500 tons. No information is available on the actual amount
of particulate emissions in 1971.
Ill-3 3
-------
CO
C\J
g
z"
g
Q.
5
ID
CO
2
O
o
>
o
a:
475
450
425
400
375
350
325
300
275
250
225
200
175
150
125
100
75
50
25
\
/N
^TOTAL ENERGY CONSUMPTION
i NATURAL GAS
'LIQUID HYDROCARBONS
\
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72
YEAR
A-54-834
Figure 13. ANNUAL ENERGY CONSUMED IN THE
MANUFACTURE OF CARBON BLACK
in-34
-------
Trends in Carbon-Black Production
By 1980, carbon-black production is expected to increase by 60% or
more over 1971 production to an estimated 2.4 million tons. On this basis,
the total annual energy consumption will be about 417 trillion Btu, based
on an average unit fuel consumption of 173 million Btu/ton. In recent
years, the trend has been toward liquid hydrocarbon feedstocks in the
furnace process. Because the yield of carbon black per million Btu of
energy consumed is highest for this process, its growth, in view of the
shortages of other feedstocks, is virtually ensured. In addition, carbon
black from oil is superior in quality to carbon black from gas, and the
liquid hydrocarbon feedstock is cheaper than natural gas. Liquid hydro-
carbons will be used to produce an estimated 85% of the total carbon black
produced by 1985, barring any unforeseen changes. The effect of this
trend on air pollutant emissions will be minimal if adequate cleanup pro-
cedures are used. However, as more oil is used, adequate measures for
controlling hydrogen sulfide emissions (not a problem with natural gas)
will have to be taken either in the form of cleanup or in the form of re-
moving sulfur from the fuel prior to feeding it into the furnace.
Phosphoric Acid
The manufacture of phosphoric acid requires from 0. 5 million to
25 million Btu/ton of acid, depending upon which one of three processes
is used. In 1971, 6.25 million tons of phosphoric acid was manufactured
in the U.S. 15 Energy consumed by this industry is estimated at 60 trillion
Btu.
The most widely used process for manufacturing phosphoric acid is
the wet process, in which sulfuric acid is reacted with phosphate rock to
produce the acid. The solution resulting from this reaction is separated
from the gypsum, -which also is formed as a precipitate, and then concen-
trated by evaporation. The entire process consumes about 0.5 million
Btu/ton of acid produced, primarily for mechanical operations and in
boilers. Based on an estimate of 4.0 million tons of acid produced in this
manner, 2.0 trillion Btu of energy was consumed in 1971. 6
III-3 5
-------
The process that used the majority of the energy consumed by this
industry but only produced about 1. 3 million tons of acid was the electric-
arc furnace process ("two-step thermal process"). In this process,
phosphate rock is reduced to elemental phosphorus by the action of coke
and heat in the presence of sand and subsequently oxidized by air to
phosphorus pentoxide. The phosphorus pentoxide then is hydrated to form
the acid. The energy required by this process is about 25 million Btu/ton
of acid produced. 6 The energy is used as electricity to supply the
necessary heat for breakdown of the rock to phosphorus. Energy also is
used as coke. This process consumes approximately 14 million Btu of
electricity and 11 million Btu of coke per ton of acid produced.
The third process for manufacturing phosphoric acid is actually a
"one-step" version of the thermal process previously described. In this
process, molten phosphorus is sprayed into a combustion chamber along
with air and steam. The reaction produces phosphorus pentoxide, which
then is hydrated to form the acid. The process is used where phosphate
rock is not available for conversion to phosphorus. The primary use of
energy in this process is for steam generation; the amount of steam used
varies considerably. Approximately 900,000 tons of acid was produced
by this process in 1971.
Air pollutant emissions from wet-process phosphoric acid manu-
facture consist of rock dust, fluoride gases, particulate fluoride, and
phosphoric acid mist, depending upon the design and condition of the plant.
(See Table 6. ) In the thermal process (electric furnace and phosphorus
burning), the primary emission is phosphorus pentoxide acid mist. How-
ever, most plants are equipped with cleanup devices, the relative effects
of which are shown in Table 6.
Emissions from the wet process, shown as uncontrolled in Table 6,
are controllable. Vapor scrubbers are universally employed to control
fluoride emissions. Such devices include venturi scrubbers, impingement
scrubbers, and various spray towers. These devices are capable of re-
ducing emissions up to 99%. Thus, whereas total fluoride emissions
from uncontrolled wet process plants would be about 156 million Ib/yr, total
emissions from effectively controlled plants would be about 1. 6 million
Ib/yr. This number can be expected to increase as production increases.
,111-36
-------
Trends in Phosphoric Acid Manufacture
In recent years, the technology of phosphoric acid manufacture has
undergone several changes. First, there was a shift from the wet process
to the electric furnace process, which produces a purer product, followed
by shifts in plant sites due to the development of phosphate rock deposits.
Second, as plant equipment improved, there was a rebirth of the wet
process. As a result of these changes, all processes are in use today.
The wet process predominates, however, and it will probably remain
predominant because of fertilizer requirements. Because this process
consumes virtually no energy, total energy consumed per year by this
industry will probably decrease. Also, several investigations are under
way to change the reactants within the wet process from sulfuric acid to
hydrochloric acid. However, if this change actually takes place, it will
affect neither the energy requirements of the process nor the pollutant
emissions.
Other Inorganic Chemicals
Two chemicals are produced in excess of 4.0 million tons/yr, shown
in Table 1, but, as it turns out, they are not large consumers of energy:
nitric acid and sulfuric acid. In nitric-acid production, the primary re-
action is the oxidation of ammonia, the products of which are passed
through a catalyst to form NO . These, in turn, are hydrated to form the
3C
acid. The heat of reaction of ammonia, and oxygen is used as the source
of heat required in other phases of the process. The manufacture of sul-
furic acid is similar. In this case, most of the process heat is obtained
from the heat of combustion of sulfur.
The only other chemical presented in Table 1 but not previously dis-
cussed is aluminum oxide. Discussion of the manufacture of this chemical
is not presented here because this information is given in the section of
this report on aluminum manufacturing. Most of the 6. 4 million tons of
aluminum oxide produced is used in the manufacture of aluminum.
Ill-3 7
-------
Industrial Organic Chemicals
The industrial organic chemicals industry is a heavy user of fossil
fuels because all the chemicals produced are hydrocarbon-based. Most
of the fossil fuel is used for feedstock; only a small percentage is used for
process heat. Electricity supplies what little energy is required for
mechanical drives. As a result, it is difficult to believe that the energy
required per unit of production by this industry will change very much. As
production increases, so too -will total energy consumption. Although the
manufacture of all these chemicals consumes large amounts of energy,
three chemicals within this group methahol, benzene, and toluene stand
out as exceptionally large consumers because of their high annual pro-
duction relative to others within this group.
Methanol
Methanol is produced by the reaction of hydrogen and carbon
monoxide under high pressures. Although there are a number of ways to
obtain the reactants, the most common way in the U.S. is by reforming
natural gas. For every ton of methanol produced, about 28 million Btu of
natural gas is reformed to supply the reactants.6 Additional energy is con-
sumed for process heating. In 1971, nearly 93 trillion Btu of energy,
primarily natural gas, was consumed in the production of 2.47 million tons
of methanol. The chief outlet for methanol is in the manufacture of
formaldehyde, -which is expected to increase in the future. However, if
the methanol-based fuel cell were to become an economic reality, a
drastic increase in methanol production would result.
In the production of methanol, carbon monoxide and hydrogen are
mixed (1:2), compressed to pressures of 3000-5000 psi, and heated in heat
exchangers by the product gases. The heated gases pass through a
catalytic converter, which must be cooled to maintain a temperature of
about 550° F. The methanol-containing gases leaving the converter are
cooled and condensed to liquid methanol. Further purification, if desired,
is accomplished by distillation.
IH-38
-------
As previously stated, the primary source of the reactants is the
reforming of natural gas. However, there are other sources. Hydrogen is
sometimes obtained by electrolysis, and carbon monoxide is sometimes
obtained from by-product gases from calcium carbide furnaces. Other
sources, which produce both hydrogen and carbon monoxide in roughly
equal proportions, are fermentation by-product gases and the manufacture
of coke and blue-water gas from coal and water. In spite of these other
sources, natural gas is still the primary feedstock. Process heat require-
ments are generally fulfilled by utilizing the heat of reaction of carbon
monoxide and hydrogen.
There is no specific information on the air pollutant emissions from
a methanol plant other than for the reforming of natural gas. (See the
discussion of ammonia production.) However, it would be reasonable to
assume that some carbon monoxide, some hydrogen, and some hydrocar-
bons are emitted during the manufacturing processes. If methanol pro-
duction increases, then so too will the total annual emissions, but air
pollutant emissions per unit of methanol produced are not likely to decrease
because no new processes that could affect them are anticipated.
Benzene and Toluene
Benzene is one of the most abundantly produced organic chemicals in
the U.S. In 1971, approximately 3.9 million tons was produced. Benzene
can be produced from the destructive distillation of coal, from the dehy-
drogenation of selected petroleum stocks, and from the hydrodealkylation
of toluene. Whereas coal distillation was the only method of production
before World War II, hydrodealkylation is the primary manufacturing
process today.
In this process, toluene and occasionally xylene are reacted with
hydrogen to produce benzene and methane. There are several ways in
which this reaction is carried out, both catalytic and thermal. Typically,
toluene and hydrogen are mixed and preheated to 1000°-1200° F and fed to
the dealkylation reactor. The products then are fed to a flash drum, where
pressure is reduced and hydrogen removed and recycled. The remaining
products go into a stabilizer, where fuel gas is removed. The products
are cooled, and the liquid product from the stabilizer then is distilled to
III-3 9
-------
yield benzene. Most of the energy consumed by this process is for feed-
stock. Approximately 2500 pounds of toluene and about 55 pounds of
hydrogen are required to produce 1 ton of benzne. 6 Figures for other
energy requirements are not available.
Because toluene is the major feedstock for benzene production, the
energy required to produce this chemical should be added to the energy re-
quired for benzene production. However, no quantitative data on this
subject have been found. Although naphthenes are used as feedstock, the
amount of naphthene used depends upon its quality. The primary process
for manufacturing toluene is the catalytic dehydrogenation of the naphthenes
in the presence of hydrogen, which yields a mixture of aromatic hydro-
carbons, chiefly toluene.
Although benzene and toluene are classified as organic chemicals,
production occurs primarily in petroleum refineries. Thus, no specific
data on air pollutant emissions from the manufacture of these chemicals
are available. A general discussion of air pollutant emissions from
refineries is presented in the section on petroleum refineries.
Trends in Benzene and Toluene Manufacture
The annual production of benzene and toluene will continue to grow in
the future. Benzene is a primary feedstock for a number of chemicals,
including styrene, phenol, and nylon via cyclohexane. Toluene is used in
gasoline and in the production of benzene. Several processes are used
for manufacturing these chemicals, but the ones .described above are the
most widely used. Barring any great shortage of raw materials, these
processes probably will remain the most important. If a shortage does
develop, the industry could revert to coal, but the amount of coal required
to maintain reasonable production rates would be large. For benzene
along, 1 ton of coal is required to produce about 2 gallons of benzene.
Based on 1971 production figures, more than 500 million tons of coal, or
13,000 trillion Btu of energy, was consumed. The effect on air pollutant
emissions also would be great because the process is basically that of
producing coke. (See the discussion of coke ovens in the steel section. )
III-40
-------
Conclusions
There is undoubtedly some room for reducing energy consumption
per unit of production within the industrial chemicals industry, but the
amount that can be expected is not great. About 55% of the energy con-
sumed by this industry is for process heat, most of it as steam. The
remaining 45% is consumed for mechanical drives and feedstock, mainly
the latter. With these facts, large savings of energy are not conceivable.
For the air pollutant emissions of this industry, the major emphasis
should be on control. Because most of the fuel is used by this industry in
boilers, changes in the type of fuel used will alter the emission character-
istics. However, the effect of such changes cannot be determined without
an accurate assessment of emissions at present. (For a more complete
discussion, see the section on electricity and steam generation.)
Similarly, changes in electricity consumption will alter emissions, but
only at the source of electricity. Emissions directly attributable to
specific processes are numerous as are the methods of control. However,
good maintenance and usage of available control equipment should ensure
minimum emissions in the future.
References Cited
1. American Gas Association, A Study of Process Energy Requirements
in the Chemical Industry, Catalog No. CZ0006. Arlington, Va. , n. d.
2. Bureau of Mines, 1971 Minerals Yearbook. I., Metals, Minerals,
and Fuels. Washington, D. C. : U.S. Government Printing Office,
1973.
3. "Carbon Black Plant Sends Air Pollution Packing, " Chem. Week 112.
55, 56 (1972) November.
4. Division of Air Pollution, Public Health Service, U.S. Department of
Health, Education, and Welfare, "Atmospheric Emissions From
Petroleum Refineries A Guide for Measurement and Control, "
Publication No. 763. Washington, D. C.: U.S. Government Printing
Office, I960.
5. Drogin, T. , "Carbon Black, " J. Air Pollut. Control Assoc. 18,
216-28 (1968) April. :
6. Faith, W. L. , Keyes, D. B. and Clark, R. L. , Industrial Chemicals,
3rd Ed. New York: John Wiley, 1966.
7. Fischbacker, C. A. and Grant, W. J., "Industrial Gases, " Rep.
Prog. Appl. Chem. 54, 17-37 (1969).
111-41
-------
8. Gregory, D. P., Anderson, P. J. , Dufour, R. J. , Elkins, R. H. ,
Escher, W.J.D. , Foster, R. B., Long, G. M. , Warm, J. and
Yie, G. G. , A Hydrogen-Energy System. Arlington, Va. : American
Gas Association, 1973. A. G. A. Catalog No. LZ1173.
9. Drotheer, M. P. and Currey, J. E. , "Production of Solid Salt From
: Diaphragm Cell Liquor, " Symp. Salt, 3rd, 1969, 2, 34-38(1970).
10. Haldor, F. A. et al. , "Catalytic Processes and Ammonia Plants, "
Chem. Eng. Prog. 63, 67-73 (1967) October.
11. Oblad, A. G., "The Kel-Chlor Process, " Ind. Eng. Chem. M_,
23-Z6 (1969) July. ;
12. Reed, R. M., "Hydrogen," in Kirk-Othmer Encyclopedia of
Chemical Technology, Vol. llj 2nd Ed. , 338-379. New York:
John Wiley, 1966.
13. Snyder, J. L.,Jr., and Burnett, J. A., Jr., "Manufacturing
Processes for Ammonia, " Agr. Anhydrous Amm. Technol. Uses
Proc. Symp., St. Louis, Mo., 1965, 1 -20 (1966).
14. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, D. C. : U.S. Government Printing
Office, January 1972.
15. U.S. Department of Commerce, "Inorganic Chemicals 1971,"
Current Industrial Reports Series: M28A(71)-14. Washington, D. C. :
Bureau of the Census, 1972.
16. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C.: U.S. Government
Printing Office, July 1973.
17. U.S. Department of Commerce, Domestic and International Business
Administration, U.S. Industrial Outlook 1974 With Projections to 1980.
Washington, D. C.: U.S. Government Printing Office, October 1973.
18. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C., April 1973.
Ill-4 2
-------
IV. SIC CODE 282 - PLASTICS MATERIALS AND SYNTHETICS
Summary
SIC Code 282 includes chemical establishments primarily engaged in
manufacturing plastics materials and synthetic resins, synthetic rubbers,
and cellulosic and man-made organic fibers. Establishments that manu-
facture rubber products and that compound purchased resins or fabricate
plastic sheets, rods, or other plastic products are not included -within this
group.
The plastics materials and synthetics industry consumes more than 3. 5%
of the total energy consumed by industry in the U. S. , excluding fossil fuels
consumed as feedstock. 6 The most important industry segment within this
classification is plastics materials and resins, because plants within this
segment consume the greatest portion of energy and manufacture the most
abundantly produced plastics: polyethylene (3.2 million tons in 1971),
polystyrene (1. 6 million tons in 1971), and polyvinyl chloride (1. 7 million
tons in 1971). 8 The energy consumed by this industry in 1971 was 134
trillion Btu, excluding feedstock; 38. 9% of the total energy consumed was
natural gas, 27.2% was coal, 22.7% was oil, and 11.2% was electricity.6
(See Figure 1. ) . ,
Because plastics are a hydrocarbon-based product, much of the fossil
fuel consumed by this industry'is consumed as feedstock after having
been converted to the primary raw materials actually used by this industry.
These primary materials are ethylene, styrene, and vinyl chloride, all of
which require fossil fuels (natural gas, fuel oil, and coal) for manufacture.
Even though the plastics industry does not actually produce the raw materials,
the fuels consumed in their production, should be considered when analyzing
.s? -"
<; - .
this industry. Table 1 summarizes the consumption of these primary
materials in 1967, the most recent year for which this information is avail-
able.
Most of the energy consumed in the manufacture of the raw materials is
consumed as feedstock, which, in turn, also was manufactured. That is, one
chemical is produced for feedstock in the manufacture of a second chemical,
which is then used in the manufacture of a third chemical,and so on. Thus,
it is not realistic to consider the energy consumed in the manufacture of
IV-1
-------
IOD
130
125
120
115
110
105
100
95
3 90
CD
-o85
|80
t 75
5
w 70
§65
0 60
tr
£55
UJ
0 45
40
35
30
25
20
15
10
5
0
/
^
-yX
/
^
n
>
/
^»
X^
X
X
""
/
/
/
X
/
/
/
/
/
/
r
/COAL
/
/
x--'
^
f
^
i
-^
/
N/
« «.
-V
^''
~~s
/-\
*TUF
.
/
X*
1
I
1
OTAI
IAL
-~~y
f
/
^
^
t
/
/
3AS
/
'
/
f
,-
/
/
/
'
/
0\L
ELECTRICITY
/
/
Figure 1.
1947 '49 '51 '53 '55 '57 '59 '61 '63 '65 '67 '69 '71 '73
YEAR
B-54-832
TOTAL ANNUAL ENERGY CONSUMPTION OF PLASTICS
MATERIALS AND RESINS INDUSTRY
IV-2
-------
Table 1. SPECIFIC MATERIALS CONSUMED IN THE MANUFACTURE
OF PLASTICS MATERIALS AND RESINS IN 1967 (Ref. 5)
Energy Consumed
A , , in Manufacture,
Amount Consumed, j
Material : 1Q6 tons 10 Btu
Ethylene 1.02 38. 7b
Styrene 0. 78 --° .
Vinyl Chloride 0.73 --d
Not including feedstock.
Feedstock is refinery gas made up of a mixture of hydrocarbons.
Most of the energy consumed is benzene (1730?b/ton) and ethylene
(6401 Ib/ton) consumed as feedstock.
Most of the energy consumed is acetylene and ethylene consumed as
feedstock.
the third chemical. But it is important to realize that this practice does go
on within this industry.
Resin-Manufacturing Processes3
The chemical conversion process used for resin manufacture is poly-
merization, in which simple molecules react to form polymers. Two
principal types of polymerization reactions exist: condensation and addition.
A condensation process reacts pairs of chemical function groups to form a
group not present in the reactants and splits out water. An addition process
opens a chemical bond in a reactant and forms similar bonds with other
reactants without the formation of side products.
Polymerization occurs in a number of ways, namely, bulk, solution,
emulsion, or suspension. Bulk polymerization is carried out in either a
gas or liquid state, but it is difficult to achieve on a large scale. Poly-
merization of gaseous monomer is likely to result in a low yield without the
aid of high operating pressures or a catalyst. Addition polymerizations are
exothermic; thus, evolved heat is removed to control the reaction. Solution
polymerization is conducted in a solvent that is suitable for both the monomer
and the reaction initiator. The polymer may remain in solution or separate.
In suspension polymerization, the monomer is suspended in water by
agitation. Talc or b.entonite is added to stabilize the suspension and prevent
polymer globules from adhering to each other.
IV-3
-------
Emulsion polymerization differs from suspension polymerization in that
soap is added to stabilize the monomer droplets and form aggregates called
micelles. These micelles take the monomer into their interior, and the
initiator dissolved in the aqueous phase diffuses into the micelle to start
polymer growth. At certain degrees of polymerization, the polymer is
ejected from the micelle, but continues to grow, and the monomer diffuses
back into the micelle to replenish it. Emulsion polymerizations are rapid
and can be carried out at relatively low .temperatures.
\
Polyethylene
The three basic processes for producing polyethylene are as follows:
1. The high-pressure or Imperial Chemicals Industries (I. C. I. ) Process
2. The medium-pressure Phillips Petroleum Process
3. The low-pressure Ziegler Process.
The I. C. I. Process is used to produce low-density polyethylene, and the
Phillips and Ziegler processes are used to produce high-density polyethylene.
In the I. C. I. Process, ethylene is passed over heated copper to remove
oxygen and then compressed to as much as 500 atm of pressure. A catalyst is
added, along with a chain modifier, before compression and polymerization
begin. The reaction is exothermic, and temperatures are maintained at
400°F. Upon completion of the reaction, the polymer is extruded as a
ribbon, cooled, and then granulated.
In the Phillips and Ziegler processes, gaseous ethylene is fed into a
hydrocarbon solvent containing a catalyst. In the Phillips Process, the
catalyst is partially reduced chromium oxide supported on activated silica
or alumina. In the Ziegler Process, the catalyst is a mixed metal alkyl/
metal halide. In the Phillips Process, polymerization occurs at about 500
psi and 350°F, whereas in the Ziegler Process polymerization occurs at
or lightly above atmospheric pressure and at temperatures of 125°-150°F.
The slurry that results from both of these processes is freed from solvent
and washed to remove catalyst residues. ;
Polystyrene4
Polystyrene is manufactured by any of several available processes,
depending upon the qualities desired in the end product. The main commercial
IV-4
-------
processes are bulk polymerization, solution polymerization, and
suspension polymerization. In all these processes, a reaction medium,
such as a solvent, is used, and in solution and suspension polymerization,
a catalyst also is used to promote the reaction. Sytrene, the feedstock,
is produced by reacting ethylene with benzene in the presence of a catalyst
to produce ethyl benzene. The ethyl benzene undergoes dehydrogenation to
yield styrene.
Polyvinyl Chloride4
Polyvinyl chloride is produced by the bulk, solution, suspension, or
emulsion polymerization of vinyl chloride. The most commonly used are
the suspension and emulsion methods. In suspension polymerization,
water-soluble suspension agents and an initiator are mixed with the mono-
mer, stirred to maintain a suspension, and heated to about 125°F to effect
polymerization. Emulsion polymerization is similar, except that an
emulsifying agent is used, the initiator system is initially contained in the
water phase, and the product obtained is much finer in particle size.
Manufacture of the monomer consumes acetylene or ethylene as feedstock,
depending upon the process used.
Energy Utilization Pattern
Several possible polymerization reactions lead to the production of the
resins discussed above, so precise data on energy utilization applicable
to the manufacture of each resin are not available. Examination of the
small amount of data that are available indicates a wide range of consump-
tion rates. For example, the energy required to produce 1 ton of poly-
ethylene varies from 1. 4 million to 7.«-0 million Btu, exclusive of feedstock. 2
In most cases, the energy is consumed as steam and electricity for process
heat and mechanical drives. Energy consumed to produce 1 ton of poly-
styrene also varies, but, according to one source, is about 3.5 million Btu.1
Energy consumption statistics concerning polyvinyl chloride manufacture
are not available.
Air Pollutant Emissions
The major sources of emissions from plastics manufacturing are the
reaction vessels. During polymerization, monomers, solvents, and other
volatile materials are emitted. However, most of these emissions are well
IV-5
-------
controlled because they are recycled into the process on a routine basis.
The controls for this purpose include vapor recovery systems, purge lines
that vent to a flare system, and vacuum exhaust line recovery systems.
Emissions from the production of plastics without controls are shown in
Table 2.
Table 2. EMISSIONS FROM THE MANUFACTURE OF
PLASTIC MATERIALS WITHOUT CONTROLS7
Type of Plastic Particulate Gases
T-> i i ^i i -j T£ Ib/ton ;
Polyvinyl Chloride 35 17
General 5-10
Particulate emissions are generally controllable with fabric filters, which
have a 98-99% collection efficiency.
Trends in Plastics Manufacture
The plastics industry is one of the fastest growing industries, increasing
production at an estimated 13% per year. This growth is likely to continue,
but at a reduced rate of about 8% per year, because raw materials are
already in short supply, as are the fuels required to produce them, and
this shortage is expected to continue for several years. Although new
processes and new plastics are being developed all the time, there is no
clear information regarding their effects on the industry.
However, based on the number of processes currently in use, new
processes requiring less energy will be developed. In addition, as fuel
costs increase, more efficient utilization of waste heat, such as the
exothermic heat of reaction, will occur. Changes that will affect the
short-term energy consumption patterns of this industry are not likely to occur.
By 1980, energy consumption will be about 400 trillion Btu, excluding
feedstock, 3 times the consumption rate in 1971.
References Cited
1. "1971 Petrochemical Handbook Issue,'" Hydrocarbon Process. 50, 202
(1971) November.
2. "1973 Petrochemical Handbook Issue, " Hydrocarbon Process. 52,
164-171 (1973) November.
3. Shreve, R. N. , Chemical Process Industries, 3rd Ed. New York:
McGraw-Hill, iWT.
IV-6 j
-------
4. Sittig,M. , Plastic Films From Petroleum Raw Materials. Park Ridge,
N. J. : Noyes Development Corp. , 1967.
5. U. S. Department of Commerce, Bureau of the Census, 1967 Census of
Manufactures. Vol. II > Industry Statistics. Part 2, Major Groups
25-33. Washington, D. C. : U. S. Government Printing Office, January
1971.
6. U. S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72 (SR)-6. Washington, D. C. : U.S. Government Print-
ing Office, July 1973.
7. U. S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-42, 2nd Ed. Research Triangle
Park, N. C. , April 1973.
8. U.S. Tariff Commission, Synthetic Organic Chemicals, U. S. Production
and Sales, 1971, TC Publication No. 614. Washington, D. C. : U. S.
Government Printing Office, 1973.
IV-7
-------
V. SIC CODE 291 - PETROLEUM REFINING
Summary
The petroleum-refining industry includes establishments primarily
engaged in the refining of crude oil to produce gasoline, kerosene, dis-
tillate fuel oils, residual fuel oils, and lubricants. Establishments that
produce natural gasoline from natural gas, establishments that produce
lubricating oils and greases by blending and compounding purchased
materials, and establishments that re-refine used lubricating products
are classified elsewhere.
The petroleum-refining industry is the third-largest energy consumer,
next to industrial chemicals and iron and steel manufacturing. An esti-
mated 2861 trillion Btu of energy was consumed for petroleum refining
in 1971. The 1972 Census of Manufactures10 indicates that energy con-
sumed for heat and power totaled 1441 trillion Btu in 1971. This figure
does not include the large amount of fuels, such as hydrogen, produced
and consumed within a refinery, nor does it include energy used for
feedstock. In 1971, an estimated 4085 million barrels of crude oil was
run to the stills.l If the industry continues to grow at its current rate,
total annual energy consumption will increase to about 3700 trillion Btu
by 1985, and the amount of crude oil run to stills will increase to about
5500 million bbl/yr. (See Figures 1-3.)
There are many sources of air pollutant emissions in a refinery,
producing large amounts and a wide variety of emissions. Because of
the complexity of a refinery, categorization of emissions by process is
very difficult. Many sources, such as gas relief valves, are found in
several processes, whereas other sources, such as waste gas flares,
cannot be associated with any single process. Thus, the level of each
emission must be determined by summation. Air pollutant emissions
include particulates, SO , carbon monoxide, hydrocarbons, NO , aldehydes,
X. X.
and ammonia. The greatest contributors to air pollution are SO , carbon
X.
monoxide, hydrocarbons, and NO .
X,
V-l
-------
2900
2800
2700
2600
2500
* 2400
00
M
~o
g
o.
2300
2200
o
o
e>
tr
LJ
2100
2000
1900
1800
1700
1600
1500
1400
1300
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72
YEAR
A-54-746
Figure 1. TOTAL ANNUAL ENERGY CONSUMPTION BY
THE PETROLEUM-REFINING INDUSTRY, 1950-1971
V-Z
-------
4100
4000
3900
3800
3700
3600
3500
3400
t 3300
3200
5 3IO°
uu
Q
DC
O
3000
2900
5 2800
2700
2600
2500
2400
2300
2200
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72
YEAR
A-54-747
Figure 2. TOTAL CRUDE OIL RUN TO STILLS, 1950-1971
V-3
-------
14
13
12
a
T3
-O
68
£7
§6
<
CJ
a 5
LJ
TOTAL CRUDE CAPACITY
-CATALYTIC CRACKING CAPACITY-
LTHERMAL OPERATIONS CAPACITY
1950 1955 I960 1965
YEAR
1970 1975
A-54-748
Figure 3. OPERATING CAPACITIES OF OIL,
REFINERY PROCESSES, 1950-1971
V-4
-------
Petroleum Refinery Processes3'4*8
As of January 1, 1973, there were 247 petroleum: refineries operating
in the U.S.,^ each one different from the other, using different processes
to produce the final products. Thus, no typical refineries can be used as
models for an energy survey. Even though distillation, catalytic crack-
ing, catalytic reforming, and thermal cracking are processes common to
all refineries, these processes differ from one refinery to the next in
terms of implementation. For example, catalytic cracking can be accom-
plished by one of several processes, including fluidized-bed cracking,
moving-bed cracking, and fixed-bed cracking. Each of these processes,
in turn, is varied from refinery to refinery. For this survey, only the
major types of processes are discussed.
Separation Processes Distillation
The refining, or manufacturing, of petroleum products is accomplished
by two means: physical means or separation operations, and chemical or
conversion processes. The first step in the refining of crude oil is dis-
tillation, in which the lighter, more volatile components are separated
from the heavier, less volatile components. Several types of equipment
can be used in the process; the basic types are shown in Figure 4.
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 process may be 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. Upon completion of distillation operations, the
products usually are prepared and fed into the conversion processes.
The energy consumption of a crude oil distillation process varies,
depending upon the particular installation, but a review of several pro-
cesses indicates that a, typical installation consumes about 100, 000 Btu
of fuel per barrel of crude oil feed for heat.7 An additional 20,000
Btu of energy is consumed as steam.
V-5
-------
TO STACK-
OUT-
TO STACK
OUT
-^BURNERS E4 BURNERS-?
(A) LARGE BOX TYPE
,**
i /< >f-KNsr
OUT-M^Ill
OUT
BURNERS
(B) SEPARATE CONVECTION
BURNERS-
TO STACK
TO STACK
.TO STACK
IN
OUT
(0)STRAIGHT-UP
TO STACK
BURNERS
(E) A-FRAME
BURNERS
OUT
(C> DOWN CONVECTION
,TO STACK
IN
-OUT
BURNERS
(H) SMALL ISOFLOW
BURNERS
(F) CIRCULAR
A TO STACK
' ' -IN
LLJTO STACK
OUT
OUT
»-
BURNERS
(G) LARGE ISOFLOW
TO STACK
BURNERS -*l Si i- F- BURNERS
l'5'l
'5'J
OUT-
BURNERS '
(DEOUIFLUX (J) DOUBLE UPFIREO (KJ RADIANT WALL
Figure 4. BASIC TYPES OF PIPESTILL HEATERS3
Conversion Processes Catalytic Cracking
Catalytic cracking is one of the most important processes used in
refineries. The estimated feed capacity of catalytic cracking units was
5.5 million bbl/day in 1972* (80% fresh feed, 20% recycle). This rep-
resents almost 52% of the total crude oil running capacity in the U.S.
(See Figure 3. ) The primary purpose of catalytic cracking is to produce
gasoline. Chemically, this is accomplished by breaking and rearranging
the chemical bonds in large hydrocarbon molecules. In addition to
catalysis, cracking also can be accomplished by heating the feed (thermal
cracking), but this method is no longer preferred (Figure 3).
As previously stated, 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.
V-6
-------
In the fluidized-bed process, the reacting vapors (feed) are forced
upward through a bed of fine granular material with finely divided solid
catalyst evenly distributed throughout the system. By so doing, a quasi-
fluid suspension is established, in which the reacting vapors come in
contact with the evenly distributed catalyst and undergo cracking. Upon
completion of the cracking, the catalyst is separated from the products
by cyclone separators. 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 5 is
a schematic diagram of one type of catalytic cracking unit in operation.
ORTHOFLOW
CONVERTER
REACTOR
AIR BLOWER
Vapor to go»
recovery unit
Liquid to gat
recovery unit
/K\ Rich oil from
^-^ absorber
Light cycle oil
product
Decanted oil
product
©Virgin gas oil
feed
Figure 5. SCHEMATIC DIAGRAM OF ORTHOFLOW CATALYTIC
CONVERTER AND ADJUNCTS8 (Stippled Areas Represent Fresh
or Regenerated Catalyst) (M.W. Kellogg Co.)
The reaction temperatures used in such a unit 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, 8-20 psi, some cracking processes
utilize pressures as high as 1000 psi. During catalyst regeneration,
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
V-7
-------
is similar, employing instead of a fine-grain catalyst, a flowing bead-type
catalyst. (See Figure 5.) 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.7* On the other
hand, 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 also is an important conversion process within
a refinery. In this process, certain ring hydrocarbons are converted
into aromatic compounds. As of January 1, 1973, the charge capacity
for catalytic reforming was almost 3. 3 million bbl/day. 2 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 prod-
ucts of the reaction are cooled in heat exchangers and then fractionated
or stabilized. The stabilized product then can be used for high-octane
gasoline, 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 consump-
tion also varies considerably. Fuel in the form of oil or gas and elec-
tricity are the primary types of energy consumed in catalytic reforming.
Fuel consumption varies from 200,000 to 450,000 Btu/bbl of feed.7
Electric consumption, primarily for driving pumps and compressors,
varies from 3000 to 20,000 Btu/bbl of feed.
M-
Estimated average based on energy consumed by various processes.
V-8
-------
Hydrotreating
Hydrotreating is used in a refinery for removing impurities, including
sulfur, nitrogen, and metallic compounds; for hydrogen saturation of
olefins and aromatic s; 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
1972, refineries had a catalytic hydrotreating charge capacity of nearly
4.4 million bbl/day.
In a typical hydrotreating process, feedstock, made up primarily of
petroleum distillates, is mixed with recycle and makeup hydrogen and
heated to temperatures of 400°-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, separated from recycle gas, and stripped of hydrogen
sulfide.
As in the other processes discussed, energy consumption varies con-
siderably, 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 approx-
imately 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.
Thermal Processes Cracking
Thermal processes are processes that decompose, rearrange, or
combine hydrocarbons by the application of heat without using catalysts.
As shown in Figure 3, the use of thermal processes in refining has been
gradually falling off in favor of catalytic processes. The charge capacity
for all thermal operations at the end of 1972 was 1.44 million bbl/day,
only about 10% of the total crude capacity of refineries.7
V-9
-------
Numerous processes for thermal cracking are in existence, but
very few are used any more. Initially, thermal cracking was used to
convert heavy charge oil into gasoline materials, but today conversion
by this method is not economical. Instead, existing units are being used
to crack heavy catalytic cycle stock that usually goes to fuel oil. In a
typical cracking process, the crude oil is separated into various fractions
directly in a combination distillation-cracking unit, and the resulting
fractions are cracked in continuous operations. Cracking temperatures
range from about 900° to 1100°F; operating pressures vary from 600 to
more than 1000 psi. Energy consumed in thermal cracking is about
260, 000 Btu/bbl of charge7; this energy is used for direct heat. Although
steam, at pressures up to 300 psig, also is required, there is an overall
net production of steam equivalent to about 100,000 Btu/bbl of feed.
Energy Consumption in Petroleum Refining
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 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 6
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 electricity, not shown in
Figure 6, 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 6, are acid sludge, petroleum coke, liquified
petroleum gas, and purchased steam. l Together these forms of energy
account for about 16% of the total energy consumed by this industry.
V-10
-------
03
w
O
-Z.
8
ir
cow
2700
2600
2500
2400
2300
2200
2100
2000
1900
1800
1700
1600
1500
1400
1300
1200
MOO
1000
900
800
700
600
500
400
300
200
100
0
/__
/
^~-
s
/
/
«-
X"
L ^
/
/
/
^
/
^
A
/ \
^f
?*
__^-
/
/
^
s
^
^
y
~-S\
NA
^*
=^
"^RE
"
|x
OTAL
/
X
ENEF
/
RGY
TURAL GAS^_
^\ S\
FINEF
UEL
-~
COAL
Y GA
OIL
.
\
sL
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70
YEAR
A-54-749
Figure 6. TYPES OF FUEL CONSUMED
BY PETROLEUM REFINERIES1
V-ll
-------
Over the years, the amount of energy consumed per barrel of crude
oil run to the stills has remained fairly constant. Since 1950, the energy
consumed per barrel of crude oil run has ranged from a low of 626,000
Btu to a high of 770, 000 Btu. (See Figure 7. ) .
ENERGY CONSUMPTION,
I03 Btu/bbI of crude run
§0» O> ~J -J -J -
ro 01 ->j o ro 01
OtnooiotnOc
\
\
/
/
/~
A
/!/
/ H
r
"^
\ /
V
\
\
n
710,000 Btu/bbI
^^
1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72
YEAR
A-54-750
Figure 7. AVERAGE ENERGY CONSUMED PER
BARREL OF CRUDE OIL RUN TO STILLS1
Although this is a wide range, most of the time this figure was within
5% of 710, 000 Btu/bbl of crude oil over the 22 years covered by these
data. During the last 3 or 4 years, this figure has hovered around
700,000 Btu/bbl of crude oil run. Prior to 1950, this figure was con-
sistently below 700,000 Btu/bbl. (See Figure 8.) These data indicate
that the energy consumed per unit of crude oil can be reduced significantly
over current values. However, information on the method of data col-
lection is not available; thus, it is conceivable that these data are the
result of changes in collection methodology from year to year and cannot
be compared. What is more likely is that the product mix has changed
over the years, which, in turn, would affect energy consumption by
forcing refineries to alter their processes accordingly.
V-12
-------
850
825
.a
_a
v.
13
s
10
O
800
775
750
CO
O
O
O
cc
LU
LL)
725
700
675
650
625
600
575
550
A-
X
\
\
\ _:_.j_
W
1925 '27 '29 '31 '33 '35 '37 '39 '41 '43 '45 '47 '49 '51
YEAR
A-54-751
Figure 8. ENERGY CONSUMPTION PER BARREL
OF CRUDE OIL RUN, 1925-19501
Short of drastic changes in the product mix, the energy consumption
of petroleum re'firieries per barrel of crude oil run is not likely to de-
.crease -very much in the future. Because fuel is a major cost item for
refineries, they do attempt to cut their needs to a minimum, .a minimum
that the data suggest is currently around 700, 000 Btu/bbl of crude oil run
to stills based on the current product mix. Fuel used by this industry
is primarily for direct heating of feedstocks (60%), process steam pro-
duction (34%), and electric generation (6%).9 Over the years, the per-
centage of energy consumed as direct heat has not changed -very much,
increasing by only 1% since i960. With this energy usage pattern, it
is difficult to foresee much of a change in the future.
V-13
-------
Of course, new processes are constantly being developed to improve
refinery operations, but they do not seem to have much effect on the
energy consumption pattern. For example, several different catalytic
cracking processes are in use, yet the energy consumed by these pro-
cesses varies over a relatively narrow range. In addition, if all the
refineries were to adopt the process that consumes the least amount of
energy, that process might be incompatible with other phases of the
operation, which might then increase energy consumption.
In short, the energy utilization of a petroleum refinery is very
closely tied to the interrelationships of the processes used. Substituting
one process for another because it consumes less energy could alter the
energy consumption characteristics of the rest of the refinery, resulting,
perhaps, in a net increase in energy consumption for that refinery.
Given these considerations and the complexity of the industry, the in-
dustry as a whole probably will not reduce its energy consumption per
barrel of crude oil run to stills without altering the product mix.
Air Pollutant Emissions From Petroleum Refineries
The major air pollutant emissions from petroleum refineries are
SO , carbon monoxide, NO , hydrocarbons, aldehydes, ammonia, and
X- -X
particulates. These emissions cannot be categorized by process because
some emission sources are found throughout a refinery, whereas others
are not associated with any one process. The refinery must be treated
as an integrated system of pumps, valves, cooling towers, process
heaters, and other equipment. Table 1 summarizes the potential sources
for each type of emission from petroleum refineries. 6
Table 1. POTENTIAL SOURCES OF SPECIFIC
EMISSIONS FROM OIL REFINERIES6
Emission
Oildea of sulfur.
Hydrocarbons.
Potsotlai taunts
Oilde* of nitrogen...
Partlculat* matter.
AMehydes
Ammonia
Odom
Carbon monoxide
Boilers, proem heaters, catalytic cracking unit regenerators, treating units,
HtS flares, deeoklng operations.
Loading ftrtlltiev turnaroundii. sampling, storage tanks, waste water leparaton.
Mow-down systems, catalyst regenerators, pumps, ralres, blind changing,
cooling towers, vacuum Jet*, barometric condensers, sir-blowing, high pressure
equipment handling volatile hydrocarbons, process heaters, boilers, compres-
sor engines.
Prooeai heaters, boilers, compressor engines, catalyst regenerators, Bares.
Catalyst regenerators, boiler.i. process heaters, deooking operations. Incinerator*.
Catalyst regenerators.
Catalyst regenerators.
Treating units (air-blowing, steam-blowing), drains, tank vents, barometric
condenser sumps, waste water separators.
Catalyst regeneration, deooking, compressor engines. Incinerators.
V-14
-------
In addition, because petroleum refineries differ so widely, emissions
can only be estimated, rather than quantified throughout the industry.
Usually these estimates can be based on the amount of fuel burned or
the amount of feed processed.
Hydrocarbons
Hydrocarbon emissions from a petroleum refinery vary according to
the throughput of crude oil. The values range from about 0. 1 to 0. 6%
by weight of the crude throughput; the lower value is representative of
refineries with extensive pollution control equipment. Smaller refineries
with distillation, treating, and blending processes only tend to be at the
lower end of this range, whereas complex refineries cover the full range.
Tables 2, 3, and 4 summarize the factors for hydrocarbon emissions
from combustion sources, equipment leakage, and miscellaneous processes
such as vacuum jets and cooling towers, respectively.
Storage tanks and loading operations are the largest sources of
hydrocarbon emissions, accounting for nearly one-half of the total
hydrocarbon emissions from refineries. Hydrocarbon emissions vary
from 100 to 1000 lb/1000 bbl of refinery capacity. The actual number
depends upon the type of equipment and controls used and, consequently,
does not lend itself to detailed analysis.
In general, combustion operations are not a major source of hydro-
carbon emissions. Table 2 shows that, of the major combustion operations,
fluid catalytic cracking units are the most offensive. Assuming that
80% of all catalytic cracking capacity is of the fluidized-bed type, about
180, 000 tons of hydrocarbons could have been emitted by these units in
1972. An additional 18,000 tons could have been emitted by moving-bed
catalytic units.
Table 2. FACTORS FOR HYDROCARBON
EMISSIONS FROM COMBUSTION SOURCES6
Source
Boilers end proean hetten.
Fluld oiUlytlc cncklng unite
Movlnf-bed eaulytle encktnf unite
Compreeaor Interact eomboftlon enf Inee.
Unite of rector
(Poundi per 1,000 oubte tot of fuel fu
{ homed.
I Pounds p«r barrel of fuel oil burned
Pounds per 1,000 berreli froth feed
Poundi per 1,000 bimti freen feed
Pound! per 1,000 cuMo teet of fuel ftt
burned.
VeJue
0.0*8
.14
BO
87
1.1
V-15
-------
Hydrocarbon emissions also occur as the result of the combustion
of fuel gas (both natural gas and refinery gas) and fuel oil in process
heaters, boilers, and internal combustion engines for driving compressors.
No accurate figures on the consumption of refinery gas and fuel oil in
1971 are available, but based on the consumption of these fuels by the
industry in the recent past, reasonable estimates of hydrocarbon emis-
sions can be made. In 1971, approximately 1.3 trillion CF of natural
gas was consumed by this industry. In recent years, refinery gas
consumption has been nearly equal to natural gas consumption. (See
Figure 6. ) Thus, a total fuel gas consumption of about 2. 6 trillion CF
in 1971 is not an unreasonable assumption. Based on this assumption
and assuming 95% of this amount was consumed in process heaters and
boilers and only 5% in internal combustion engines, total hydrocarbon
emissions as a result of the combustion of fuel gas would have been
approximately 110,000 tons in 1971. Hydrocarbon emissions as a result
of the combustion of fuel oil in boilers and process heaters are esti-
mated at about 5220 tons in 1971. This is based on an estimated fuel
oil consumption of 46 million barrels in 1971. Thus, the total amount
of hydrocarbon emissions from combustion sources in 1971 is estimated
to have been about 310, 000 tons.
As shown in Tables 3 and 4, there are sources of hydrocarbon
emissions other than combustion sources. However, detailed information
on these other sources is seldom available. Thus, in Table 3, two
factors are presented for each source: One is based on the emissions
from a specific piece of equipment and the other is an overall average
for a refinery based on feed capacity. Because these emissions result
from leaks, the degree and effectiveness of maintenance in the refinery
will have a pronounced effect on emission rates. The factors given in
Table 3 are average emission rates, but they assume a high degree of
maintenance. It also has been assumed that maintenance is the only
control activity used, although blowdown systems may be used by some
refineries, in which case emissions would be reduced. Based on the
factors presented in Table 3 and given a total U.S. refinery capacity of
13.4 million bbl of crude oil per day, approximately 150,000 tons of
hydrocarbons were emitted from these sources in 1972.
V-16
-------
Table 3. FACTORS FOR HYDROCARBON
EMISSIONS FROM EQUIPMENT LEAKAGE6
Source
Pipeline valve*
Vtawil rtllef valvw
Pipeline relief valve*
drmimmr Mate
Pipeline valves
Veawl relief valves
Unite* factor
Pound! per day per valve
Pound! per day per valve
Pounds per day per Mai
Pound* per 1,000 barreti retoery oipt
Ponndi per I.OW harrth rtfiiMry fltpi
VihM
0.1*
J.4
Na«
4.1
11
ictty. *
idty 11
Table 4 shows hydrocarbon emissions from miscellaneous process
equipment, which can be major sources of emissions. When a range
of values is given, the low numbers are indicative of a plant using
control techniques and the high numbers are estimated emissions from
a plant without extensive controls. The sources shown as "other" in
Table 4 include air-blowing, pipeline blind changing, and sampling.
Because of the high degree of uncertainty involved, attempts to summarize
these numbers should not be made.
Table 4. FACTORS FOR HYDROCARBON EMISSIONS
FROM MISCELLANEOUS PROCESS EQUIPMENT6
Sourc*
Slowdown lystenu
Proom drain* and watte water
rporaton.
Vacuum Jet«
C'oolinf towert.
Otlwr
Units of (actor
Pound! per 1,000 barrels refining capacity
Poundi per 1,000 barrrli waste water procvssed
Pounds prr 1,000 barrels vacuum dinillation
capacity.
Hounds per million gallons cooling water circu-
lated.
Pounds prr 1.000 barrels re finery capacity
Vnlne
IUnci< of
SO
SO emissions are a direct function of the sulfur content of the crude
x
oil being processed, and as such, they vary considerably and do not lend
themselves to estimation. The sulfur that goes into the refinery in crude
oil, in purchased fuel oil or gas, and as sulfuric acid leaves the refinery
in the various products, as spent sulfuric acid, as sulfides and sulfates,
and in the liquid wastes; or it is recovered in sulfur-recovery plants.
V-17
-------
The difference between what goes in and what comes out is emitted to
the atmosphere as sulfur dioxide, sulfur trioxide, and hydrogen sulfide.
Miscellaneous Emissions
Other emissions from petroleum refineries are carbon monoxide, NO ,
.X
aldehydes, organic acids, ammonia, and particulate matter. These emis-
sions are discharged from combustion sources, which include catalyst
regeneration units, boilers and process heaters, and internal combustion
engines. Tables 5 and 6 summarize these emission factors.
Table 5. FACTORS FOR PARTICULATE EMISSIONS6
Sou roe
Boilers and proem beaten.
Fluid catalytic units:
With electric precipitation
Without rtoclrlc predplUlkm
Movlni-bed catalyst unltl, hfeb efficiency cen-
trtfunl wpknton.
UnlU of factor
Poundi per 1.000 cu. ft. of fuel pu burned
Pounds per barrel of fuel oil burned
Percent of catalyst circulated.
Percent of catalyst circulated.
Percent of eauljrat circulated.
0. OX
.8
.009
.on
Table 6. FACTORS FOR EMISSIONS OF
CARBON MONOXIDES, ALDEHYDES, AND
AMMONIA FROM OIL REFINERIES6
Source
BoUen tod prooMi neaUr*
rnraprmnr Internal oomba*-
UaneacfcMi.
rwd-ted oaUlrtto orMktnc
unit*.
McrtDf^ied oMlytle enektnf
OUU.
UolU of tutor
Pound* per l OOO cu. ft. of fuel
(Mborned.
Poaode per bwrel of fuel oil
burned.
Pound* per 1 .000 eu. ft. of fuel
|M burned.
Pound* per 1.000 terreb of
fmhleed.
Poandi per UKO bvral* of
(reebfeed
Vtlue of fsctor for vtrloui emlsalom
NO.u
NOi
an
le
.M
a
5.0
CO
Net.
Net
Nef.
11.700
1800
Aldehydes
uHCHO
0.0011
0»
.11
It
12
Ammonia
u NHi
Net.
Net.
02
M
S.O
Table 7 summarizes these emissions on an annual basis using 1971
production data. The data presented in Table 7 were calculated by
assuming 365 days of operation, fuel gas consumption of 2. 6 trillion
CF/yr, and fuel oil consumption of 46 million bbl/yr. In addition, it
was assumed that 80% of the catalytic cracking capacity was of the
fluidized-bed type.
V-18
-------
Table 7. ANNUAL EMISSION RATES FOR NOX, CARBON
MONOXIDE, ALDEHYDES, AMMONIA, AND PARTICULATES
Air Pollutant
N0x (as NO*) Aldehydes (as HCHO)
350,750
41, 610
823
55, 900
449, 083
Negl
9, 026, 000
626, 000
Negl
9, 652, 000
175
12, 500
1,976
7, 150
21, 801
Ammonia (as NH3)
Negl
35, 600
823
13,000
49,423
Particulates
43, 100
--f
--f
--
43, 100
Source
Boilers and Process Heaters
<| Fluid Catalytic Cracking Units
>-i Moving-Bed Catalytic Cracking Units
vO
Compressor Internal Combustion Engines
Total
*
Rates calculated from data presented in Reference 6.
Cannot be determined. B-54-733
-------
The figures for air pollutant emissions presented here are estimates
only and should not be assumed accurate for every refinery. The only
way to determine the emissions from a given facility is to measure them.
Trends in Air Pollutant Emissions
As in the case of fuel consumption, it is difficult to assess the
effect of future production methods on air pollutant emissions. Numerous
methods can be used for controlling emissions from the various sources,
as shown in Table 8. Because air pollution controls are not 100% effi-
cient, emissions continue to occur. Consequently, total annual emissions
will increase as the feed input increases. However, this increase could
be minimized if new control equipment, with better efficiencies than those
already in use, were developed and installed.
Table 8. SUGGESTED CONTROL MEASURES FOR REDUCING
AIR CONTAMINANTS FROM PETROLEUM REFINING5
Source
Control method
Storage vessels
Catalyst regenerators
Accumulator vents
Slowdown systems
Pumps and compressor*
Vacuum jets
Equipment valves
Pressure relief valves
Effluent-waste disposal
Bulk-loading facilities
Acid treating
Acid sludge storage and
shipping
Spent-caustic handling
Doctor treating
Sour-water treating
Mercaptan disposal
Asphalt blowing
Shutdowns t turnarounds
Vapor recovery systems; floating-roof tanks; pressure tanks; vapor balance;
painting tanks white
Cyclones - precipitator - CO boiler; cyclones - water scrubber; multiple cyclones
Vapor recovery; vapor incineration
Smokeless flares - gas recovery
Mechanical seals; vapor recovery; sealing glands by oil pressure; maintenance
Vapor incineration
Inspection and maintenance
Vapor recovery; vapor incineration; rupture discs; inspection and maintenance
Enclosing separators; covering sewer boxes and using liquid seal; liquid seals
on drains
Vapor collection with recovery or incineration; submerged or bottom loading
Continuous-type agitators with mechanical mixing; replace with catalytic
hydrogenation units; incinerate all vented cases; stop sludge burning
Caustic scrubbing; incineration; vapor return system; disposal at sea
Incineration; scrubbing
Steam strip spent doctor solution to hydrocarbon recovery before air regen-
eration; replace treating unit with other, less objectionable units (Merox)
Use sour-water oxidicers and. gas incineration; conversion to ammonium
sulfate
Conversion to disulfides; adding to catalytic cracking charge stock; incin-
eration; using material in organic synthesis
Incineration; water scrubbing (nonrecirculating type) ..
Depressure and purge to vapor recovery
V-20
-------
References Cited
1. American Petroleum Institute, Petroleum Facts and Figures, 1971.
New York, 1972.
2. "Annual Refining Surveys," Oil Gas J. j>2_-7_L (1954-1973) March or
April issues.
3. Bland, W. F. and Davidson, R. L. , Ed. , Petroleum Processing
Handbook. New York: McGraw-Hill, 1967.
4. Codd, M. A. et al., Ed. , Chemical .Technology; An Encyclopedic
Treatment. IV. Petroleum and Organic Chemicals. New York:
Barnes and Noble, 1972.
5. Danielson, J. A. , Ed. , "Air Pollution Engineering Manual, "
Publication AP-40, 2nd Ed. Research Triangle Park, N. C. :
Environmental Protection Agency, May 1973.
6. Division of Air Pollution, Public Health Service, U. S. Department
of Health, Education, and Welfare, "Atmospheric Emissions From
Petroleum Refineries A Guide for Measurement and Control, "
Publication No. 763. : Washington, D. C. : U.S. Government Printing
Office, I960.
7. "1972 Refining Processes Handbook, " Hydrocarbon Process. 51,
111-221 (1972) September.
8. Shreve, R. N. , Chemical Process Industries, 3rd Ed. , New York:
McGraw-Hill, ~
9. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, D. C. : U.S. Government Printing
Office, January 1972.
10. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C. : U.S. Government
Printing Office, July 1973.
V-21
-------
VI. SIC CODES 3211, 3221, AND 3229 - FLAT GLASS, CONTAINER
GLASS, AND PRESSED AND BLOWN GLASS AND GLASSWARE
Summary
The glass industry in the United States consumes about 1. 5% of the
total energy consumed by industry annually. The energy consumed by the
glass industry in 1971 was estimated at about 250 trillion Btu. The
primary forms of energy consumed are natural gas, fuel oil, and elec-
tricity; the percentage of coal consumed is very small. During the 1960's,
the glass industry grew at the rate of 4.5% per year, and this trend is
expected to continue through 1980. At this rate, energy consumption will
increase to about 375 trillion Btu/yr, a growth of 50%.
The potential for reducing the projected amount of energy required
to produce a ton of glass is not good. The use of oxygen in the melting
furnace can decrease energy consumption by about 10% in a fuel-fired
melter. The use of agglomerated batch can increase the melting rate of
the raw materials, thus reducing the energy required to melt a ton of
glass. Precise control of combustion in the melter through proper bur-
ner positioning and automatic furnace controls also can improve fuel
efficiency.
However, these new technologies will require considerable develop-
ment before they become commercially acceptable. Each of these pro-
cesses presents technical problems that must be overcome. Our best
estimate is that these processes for reducing fuel consumption will not
I
have a significant impact on the glass industry for at least 8-10 years.
Historically, the glass industry has been slow to both develop and imple-
ment new technology because of the high starting cost compared to the
marginal profit of producing glass.
Moreover, as the availability of fossil fuels decreases and air pollution
regulations become tighter, the industry is expected to turn toward more
electric melting. Nevertheless, electric melting ultimately consumes more
energy per ton of glass produced than the direct use of fossil fuels if the
energy consumed for generating the electricity is included. New technology
utilizing a fuel-fired submerged combustion melter also is being developed and
has the potential for reducing fuel consumption per ton of glass melted by
VI-1
-------
as much as 40% while also reducing pollutants. At the same time, the
use of agglomerated batch, which can decrease energy consumption, also
reduces air pollutant emissions. However, again, submerged combustion
and agglomerated batch are expected to have a small impact in the next
10 years compared to the growth of electric melting for pollution control.
Glass-Manufacturing Processes
The glass industry is broken down by four-digit SIC codes into three
categories: glass containers (SIC Code 3221), flat glass (SIC Code 3211),
and pressed and blown glass and glassware (SIC Code 3229).
The glass container industry is engaged primarily in the production
of glass containers for commercial packing and bottling for home con-
sumption. In 1971, an estimated 10.9 million tons of glass were melted
in the container industry. This estimate is based on the number of con-
tainers produced, as reported by the Bureau of the Census, assuming an
9
average weight of 8 oz/container. In addition, an estimated 15% of the
glass produced is recycled back to the melting process as cullet and
subsequently remelted; this also is included in the estimate of total pro-
duction for 1971. Total energy consumption by the container industry in
1971 was 130.4 trillion Btu.
The flat glass industry is engaged primarily in the production of
flat glass products, such as plate glass, sheet (window) glass, float glass,
rolled glass, and figured and wired glass. Also included within this
classification are companies producing laminated glass from glass made
by other companies. In 1972, an estimated 2.56 million tons of flat
glass was melted; 45% was classified as sheet glass, and the remaining
Q
55% was primarily plate glass, float glass, and rolled glass. These
figures include a 25% recycling of cullet back to the melting process.
This estimate is based on the production statistics reported by the Bureau
of the Census. In addition, the average glass density is estimated at
0.82 tons/1000 square feet, and in the case of sheet glass there are 50
square feet of glass to a box. The total amount of energy consumed by
the flat glass industry in 1971 was 55.9 trillion Btu.
VI-2
-------
The pressed and blown glass and glassware industry primarily pro-
duces glass and glassware (not elsewhere classified) pressed, blown, or
shaped from glass produced from the same establishment. This classifi-
cation includes companies that deal in textile-type glass fiber products
but not glass wool insulation products. Also included are companies en-
gaged in the manufacturing of pressed lenses for headlights, beacons, and
lanterns but not optical lenses. Estimated production during 1971 is 3. 5
i
million tons of glass. This estimate is based on a Z0% recycle of glass
for cullet. Energy consumption by this industry in 1971 is estimated at
63.1 trillion Btu.
The manufacturing of glass' 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 first step in the manufacturing of glass is the preparation of the
raw glass batch. In this process, the various raw materials, such as
sand, limestone, soda ash, and various minor ingredients (fluxes and
colorizers or decolorizers), are weighed in a batch hopper and dumped
into a mixer. Water usually is added at this point and thoroughly mixed.
The material is transferred to a storage hopper from which it is charged
into the furnace. Energy consumed by this process is electrical and is
used to drive the mixers and conveyors.
The second step in the manufacturing of glass is the melting of the
raw materials and refining of the molten glass. This process consumes
approximately 80% of the total energy consumed by the industry. In 1971,
energy consumed for melting was about 200 trillion Btu. Natural gas,
fuel oil, or electricity is used in this process, but natural gas is pre-
ferred at present. Approximately 75% of the energy consumed by melting
furnaces is natural gas.
Melting practices are essentially the same throughout the glass in-
dustry, regardless of the end product. Most of the raw materials are
converted to glass in continuous reverberatory furnaces equipped with
checkerbrick regenerators for preheating the combustion air. These fur-
naces generally are classified by the firing arrangement used: end-port
VI-3
-------
SILICA SAND
SiOg
SODA ASH
Na2C03
LIMESTONE
MgO- CaO
R20-AL03-6Si02
BATCH MIXING
MELTING
2700°F
REGENERATIVE FURNACE
SUBMERGED THROAT
REFINING
2300°F
1472 - 20I2T
FABRICATION
ANNEALING
INSPECTION
WAREHOUSING
A-83-1249
Figure 1. FLOW DIAGRAM FOR SODA-LIME
GLASS MANUFACTURE
VI-4
-------
or side-port melters. These melters are operated at temperature con-
tinuously throughout a campaign that normally lasts 4-5 years. The
refiner may or may not be separate from the melter. If it is a part
of the melter, the refiner is physically separated from the melting area
of the furnace by a shadow wall, but it gets its heat from the melter.
Refiners that are separate units are heated independently of the melter.
Continuous melters operate by using a reversal system. In this
system, fuel and combustion air enter oh one side of the furnace, and
the combustion products exit on the other side. After about ZO minutes,
the melter is reversed; the exit ports become the entrance ports, and the
entrance ports become the exit ports. These continuous melters usually
are used by the container and flat glass industries. Most of the melters
in the pressed and blown glass industry are day tanks, unit melters, and
pot melters, many of which are batch-type melters.
The third step in the manufacturing of glass is the processing and
finishing of the glass. After the glass has been melted and refined, it
is pulled from the furnace. In the container industry, the glass normally
is fed into forming machines that blow it into the desired container. In
the flat glass industry, the glass is drawn, rolled, or floated to form
sheets. In the pressed and blown glass industry, much of the glassware
is made by hand.
After the glassware has been formed, it is annealed in lehrs. This
process prevents strains from developing in the glass during cooling.
In the annealing process, the pressed and blown glassware and the con-
tainers are placed on a continuous, mesh-belt conveyor and passed through
a tunnel-type oven zoned such that the cooling curve of the ware precisely
matches that required to produce a strain-free product. Sheet and plate
glass are annealed in a similar manner, except for the conveying mechanism.
Energy Utilization Pattern
Figures 2 and 3 show the annual production and energy consumption
of the entire glass industry from I960 to 1972. In addition, these figures
show a 50% projected increase in production and a 50% projected increase
in energy consumption through 1980. However, this growth depends upon
several factors. The recent boom in the container industry has been
yi-5
-------
28.0
26.0
| 24.0
"Q 22.0
0 20.0
§18.0
Q
2 ,60
14.0
12.0
10.0
^
X
^x
X
/
/
'
'
/
/
/
/
/
/
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-83-1243
Figure 2. GROWTH RATE OF GLASS INDUSTRY IN TERMS
OF ANNUAL PRODUCTION RATE (Projected After 1972)
due to the popularity of the nonreturnable container, but environmental
concerns are causing the industry to revert back to returnable containers.
The growth of the flat glass industry depends upon the growth of the
construction industry, because much of the glass produced is used in
houses and buildings. In recent years, construction declined for a var-
iety of reasons, and this has led to declines in flat glass production.
This trend appears to be reversing itself. In addition, the flat glass
industry produces the glass used in automobiles. Automobiles use about
70% of the plate glass manufactured in this country. Thus, any large
reductions in automobile production will adversely affect the glass industry.
VI-6
-------
425
3 400
£
~o 375
|350
t 325
3 300
1 275
z 225
UJ
200
175
X
x
s^
^~
/
/
/
/
f
/
s
/
/
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-64-925
Figure 3. TOTAL ANNUAL ENERGY CONSUMPTION
BY GLASS INDUSTRY (Projected After 1972)
In the past, the glass industry has met its energy needs with natural
gas. Natural gas is preferred because it is a clean-burning fuel that
does not affect such glass characteristics as color. In addition, the nat-
ural gas flame, when properly controlled, provides for a longer furnace
life. Oil flames are very hard on refractories.
At present, electric melting is gaining in usage, but still represents
an extremely small amount of the total energy used to melt glass.
Figure 4 shows that, from 1961 to 1968, annual glass production by
electric melters increased from 21, 000 to 350, 000 tons. Electric melters
are used primarily in the container industry, but other segments of the
glass industry are expected to turn to electric melting. Proponents of
electric melting equipment claim lower construction costs, better melter
efficiencies, better working environments, and less air pollution. These
VI-7
-------
TVW
in
O
O
z" ^
O
3 100
Q ivw
O
a:
Q- n
01
cl
43
^*^
79
134
^
*
215,
/
35a
/
1957 '58 '59 '60 ^1 '62 '63 '64 65 66 '67 '68
YEAR
A-83-1246
Figure 4. ELECTRIC GLASS MELTERS IN USE
OR UNDER CONSTRUCTION IN U. S.
advantages are offset by higher energy costs, less turndown flexibility,
and greater refractory wear. As air pollution regulations become stricter
and as natural gas supplies diminish, the use of electric melting will
grow rapidly.
The use of coal is restricted to a few small hand shops in the pressed
and blown glass industry. In general, coal is not an acceptable fuel be-
cause of its burning characteristics, its sulfur content, and its deleterious
effect on glass quality and because, as a fuel, it causes air pollution.
Shops that use coal are expected to switch to a cleaner fuel in the near
future.
Energy Requirements of the Glass-Melting Process
Glass melting is the major energy-consuming process in the glass
industry. Efficiencies of continuous melters vary considerably, depend-
ing 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. 0-7. 0 million Btu/ton of glass melted. In the flat glass industry,
fossil-fuel consumption is 6. 0-12. 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.
VI-8
-------
Note that these fuel consumption figures apply to furnaces during
melting operations. However, data on energy utilization (Table 1) indi-
cate that energy consumption is actually higher because many times dur-
ing the year operation is suspended for breakdowns and holidays.
Table 1. BREAKDOWN OF ENERGY CONSUMPTION
BY THE GLASS INDUSTRY IN 1971*
Pressed and
Flat Glass Containers Blown Glass
(SIC Code 3211) (SIC Code 3221) (SIC Code 3229) Total
Glass Produced, 106 tons 2.56 10.90 3.50 16.96
Energy Consumption, 1012 Btu
Melting 44.7 104.3 50.5 199.5
Annealing 8.4 19.6 9.5 37.5
Other 2.8 6. 5 3.2 12. 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 Consumption for
Entire Production, 106 Btu/ton 21.8 12.0 18.0 14.7
Excludes energy consumed for electricity generation. A-84-1402
During these periods, the furnaces are idled to maintain furnace temper-
ature 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 cnllet charged varies from 10 to 30% of the total raw
material charged to the melter. Lower percentages of cullet charged
result in higher fuel consumption. 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 industry* 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 retention times
are used because glass quality is not so critical.
*
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.
VI-9
-------
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 generate the electricity also is considered, total energy consump-
tion 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 electricity is about 8. 7 million Btu, based on a 30% ef-
ficiency of generation. Consequently, the "real" amount of energy con-
sumed to melt glass in an all-electric melter is 8. 7 million Btu/ton.
In the pressed and glass blown industry, glassware is made by mach-
ine and by hand. The furnaces vary from the large continuous type to
small pot furnaces. As a result, energy consumption per ton of glass
produced varies widely.
As previously stated, the type of glass being melted affects the energy
consumption of a furnace. For example, glass color affects furnace ef-
ficiency. 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 (70% and above) generally require
very high melting temperatures (3000°F). Consequently, energy consump-
tion 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 temp-
eratures (Z500°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 manufactured in the U.S. are borosilicate, lead sili-
cate, and aluminosilicate. All these glasses require more than 2. 0
rnillion 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.
The most significant factors that affect energy consumption of a glass
melter are the combustion characteristics and equipment. Thus, mixing
characteristics of the air and fuel, the air-to-fuel ratio, the type of
burners used, and burner positioning are all important. Historically,
U.S. glass furnaces use 20% and higher excess air for combustion,, thus
VI-10
-------
ensuring complete combustion of the fuel. However, the use of excess
air is very costly in terms of thermal efficiency and energy consumption.
Figure 5 shows the effect of excess air on available heat. The available
heat is the gross quantity of heat released minus losses in the flue gases,
that is, the quantity of heat that is actually usable in the heating process.
As available heat decreases, more fuel must be consumed to supply the
necessary heat for melting. Figure 5 also shows that preheating the
combustion air increases the available heat and, consequently, results in
a decrease in the amount of fuel required for melting.
AVAILABLE HEAT AS PERCENT
OF GROSS HEATING VALUE
_ N> CM -& CJl
oo oooooc
- "
_-
I"'
'
-^
, -^
-- 1
*-^-^<
71
_^^~~
^^--*
-^z.
0% EXCESS AIR
0% EXCESS AIR
£>% EXCESS AIR
^
2000°F FLUE GAS TEMPERATURE
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
COMBUSTION AIR TEMPERATURE,°F
A-83-1245
Figure 5. AVAILABLE HEAT AS A FUNCTION OF PERCENT
EXCESS AIR AND COMBUSTION AIR TEMPERATURE
At present, the trend in glass melting is toward reduced excess air
in an effort to reduce energy consumption. Companies currently operate
at approximately 6-8% excess air. In order to achieve this low. level
without increasing air pollutant emissions (carbon monoxide and hydro-
carbons), operating procedures must be refined. One such refinement
involves the use of proper burner positioning, which requires a device
that allows for movement of the burners in very small increments. Work
done on this refinement resulted in a reduction in excess air of 10%,
from 18 to 8%, just by properly positioning the burners for optimum
air-fuel mixing. At the same time, energy consumption decreased by
8%, and carbon monoxide and hydrocarbon emissions were maintained at
acceptably low levels. Burner movements of 1 degree (the burners being
rotated on an arc) were found to be significant in minimizing fuel
VI-11
-------
consumption, maintaining proper melting conditions, and controlling re-
fractory temperatures. Burner positioning is one of the more important
ways in which energy consumption can be controlled.
Furthermore, the glass industry is seeking to increase production
rates without building new facilities. The most widely used technique is
electric boosting. In this process, electrodes are pushed through the
furnace walls below the glass surface. These electrodes supply heat
directly to the already molten glass, causing an increase in stirring, which
increases the heat transfer rate from the electrode to the glass. This
system acts as an auxiliary source of energy on fuel-fired furnaces, which
are restricted in the amount of fuel consumable because of regenerator
and stack capacities. Thus, more heat is made available to melt the
glass, and, consequently, more glass can be produced.
The net effect of electric boosting on energy consumption, if the
energy consumed in electric generation also is considered, is an increase
in the amount of energy used to melt "a ton of glass. If, on the other
hand, only the energy used directly for melting is considered, energy
consumption per ton of glass melted usually is lower than that without
electric boosting, although this will depend upon the input efficiency of
the electric energy. Although there is no specific published information
on electric consumption for boosting, 1. 6-1. 8 million Btu/ton of glass
is the range most often quoted by companies that sell the equipment.
Figures as low as 1.3 million Btu/ton are known. Because the theoretical
energy requirement for melting glass is about 2. 0 million Btu/ton and
because only 1. 6-1. 8 million Btu of electric energy is consumed per ton
of boost, additional energy must be used to make up the difference.
Furnaces that utilize electric boosting do use more fuel, but the total
energy consumption per ton of glass melted with electric boosting is
still reduced.
For clarification of these points, the following example is presented.
Consider a furnace capacity of 100 tons/day and a fuel efficiency of 6.25
million Btu/ton with no electric boost. The efficiency of this furnace is
about 32%. Assume that the electric boosting system is designed to
produce 5% of the total energy load of the furnace without electric boost-
ing, or approximately 1. 3 million Btu/hr. This assumption is reason-
able according to one glass manufacturer. Assume a boost in production
VI-12
-------
of 24 tons/day. On this basis, the electric energy consumed is 1.3
million Btu/ton of boost. This is only 65% of the theoretical energy
required to melt a ton of glass, and consequently, additional heat must
be supplied in the form of additional fossil-fuel combustion, approximately
17 million Btu/day. However, fossil-fuel efficiency is only 30%; thus,
the total additional fossil-fuel requirements are 57 million Btu/day. On
this basis, the total energy consumed is 5. 75 million Btu/ton of glass, a
net reduction in energy consumption of 8°^. This analysis assumes a
100% input efficiency of electrical energy, which is somewhat unrealistic.
Table 2 shows the additional fuel requirements for the lower input effi-
ciencies. This table also shows that net energy consumption per ton of
glass increases if the input efficiency drops below 70%.
Table 2. ENERGY NEEDED PER ADDITIONAL TON
OF GLASS MELTED IN AN ELECTRICALLY
BOOSTED CONTAINER GLASS MELTER
Total
3.66
4.37
5.30
6.50
8.07
Energy from fossil fuel available at 30% efficiency to make up
deficiency between 1. 3 and 2. 0 million Btu theoretical.
At this point, as previously described for electric melting, electricity
generation also must be considered. Thus, the amount of energy con-
sumed in generating the 1. 3 million Btu of electric energy is 4. 3 million
Btu/ton of glass, and the total amount of energy consumed by this process
increases to 6. 33 million Btu/ton of glass melted, an increase of 1. 3%
over the case in which no electric boosting is used.
Input Efficiency of
Electrical Energy, %
100
90
80
70
60
Theoretical
Energy
2. 00
2.22
2. 50
2. 86
3. 33
Electrical
Energy
Available
1 n6 T
1.3
1.3
1.3
1. 3
1.3
Supplemental
Energy From
Fossil Fuels*
J t\l'
2.33
3.07
4.00
5.20
6.77
VI-13
-------
A second method for boosting production on a glass melter is oxygen
enrichment, that is, the use of oxygen rather than air to burn the fuel.
In this process, raw oxygen is mixed into the combustion air, thus in-
creasing the percentage of oxygen in the air being fed to the flame. The
hotter flame that is produced results in an increase in available heat for
melting. As before, more available heat results in increased production.
The energy consumed in producing the oxygen is less than 2. 5% of the
total energy required to melt a ton of glass and as such is not a factor.
One advantage of this system over electric boosting is that the total energy
consumption per ton of glass melted is reduced, whereas this is not nec-
essarily true with electric boosting. The glass industry has not yet
accepted this process as a means for boosting production; consequently,
it is in very limited use at present.
The third method for boosting production is agglomerated batch. This
method is gaining a great deal of attention in the U. S. not only as a
means for boosting production, but also as a means for reducing air
pollutant emissions from glass melters. However, batch agglomeration
is not widely used in the U.S. because it is expensive to produce.
Theoretical studies have determined that agglomerated batch should melt
2-3 times faster than normal granular batch. In Japan, undocumented
increases of as much as 50% in production rates have been reported.
In the U.S., several melting trials with agglomerated batch have been
run, and in all cases, fuel consumption per ton of glass melted was re-
ported to have decreased. In one instance, on a small 40 (ton/day furnace,
fuel consumption decreased by 1.0 million Btu/ton, from 7. 5 to 6. 5
million Btu/ton, when charged with agglomerated batch. Because furnace
characteristics and operating practices vary so much from one glass
plant to another and batch agglomeration is not an established practice,
no generally applicable quantitative conclusions can be reached. How-
ever, agglomerated batch will increase production rates and
fuel utilization efficiencies, even though it does require a considerable
amount of energy to manufacture.
VI-14
-------
One very novel approach to melting glass is submerged combustion.
In a submerged combustion melter, the burners are located in the bottom
of the melter. The combustion gases pass through the glass bath, violently
stirring the bath. At the same time, heat transfer to the batch from the
gases is extremely efficient. Experimental work on a nonregenerative
submerged combustion melter showed a reduction in fuel consumption of
50%. Although data are not available for regenerative submerged combus-
tion melters, similar results are expected. The expected efficiency of
such a furnace is high, perhaps ,75%.
In spite of its advantages, submerged combustion melting is not used
in the glass industry. One such furnace, however, is used as a premelter
and reportedly still results in a saving of energy. One major problem
of submerged combustion is that the product is molten glass filled With
tiny air bubbles and therefore must be refined before it can be used.
The time required for refining ;is quite long with present methods because
the molten glass is so viscous. Solutions such as vacuum degassing have
been suggested but have not been satisfactory. The problems of sub-
merged combustion are of such severity that a, commercially viable pro-
cess is not expected to be available for another 10-15 years.
Energy Requirements of Annealing
Approximately 15% of the total energy consumed by the glass industry
is for finishing operations, primarily annealing. Glass annealing lehrs
are heated by convection, radiation, br a combination of the two. The
most effective heating means is by zoned convection lehrs that have in-
eternal distributors to obtain lateral temperature uniformity. Unlike other
types of lehrs, this design allows for inter change ability between natural
gas and oil. Some lehrs are direct-fired by atmosphere or premix bur-
ners or by excess air burners. These lehrs generally are restricted to
gaseous fuels.
Annealing lehrs in the glass industry generally operate at a thermal
efficiency of only 20%. Much of the inefficiency is due to poor mainten-
ance and operating practices. Most lehrs leak a considerable amount of
unwanted cold air into their chambers or lose heated air through unwanted
openings. However, little is done to improve annealing lehrs because
their energy consumption is low and the possible results would not be
worth the effort by the glass industry.
VI-15
-------
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 (SO ,
.X
sulfur dioxide and sulfur trioxide), nitrogen oxides (NO , nitric oxide and
X.
nitrogen dioxide), and carbon monoxide. Hydrocarbons are not a problem
if proper combustion conditions are maintained. Table 3 summarizes
emissions from various glass tanks across the country.
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 ingredient of most glass batch. In the furnace, it vaporizes and
decomposes to form elemental sodium and sulfate. When these gases
pass through the checkerbrick and are cooled, sodium sulfate is re-formed.
Only about 40% of the sodium sulfate charged into the furnace is vapor-
ized; the remainder goes into the glass. In addition to the sodium sul-
fate, 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 SO emitted from a furnace depends on 1) the sulfur
X.
content of the fuel and 2) the amount of sulfur-bearing compounds in the
raw materials. Consequently, natural-gas-fired furnaces generally ex-
hibit lower SO emissions than oil-fired furnaces unless the sulfur has
x
been removed from the oil. Measurements of SO emissions from a
x
batch melter charged with batches with various sulfur contents showed a
direct correlation between sulfur in the batch and SO emitted. The
x
greater the sulfur content of the raw batch, the higher the SO emissions.
X.
VI-16
-------
Table 3. AIR POLLUTANT EMISSIONS FROM
VARIOUS PRODUCTION GLASS MELTERS
Investigators
fG'f2
Hyler and McMackin4
Slockham5
A rrnndale1
Nrl/.ley3
CO
35-50
0-5b
375°
_
_
NOX
ppm
a 490-700
450-600
340
_ _
Parti culates,
Ib/hr
. . 6-8
--
_ _
2-10d
2-10d
Halogens SOX
ppm
1.0 28e
7.1 267
_ . .
--
H excess air.
.15-45% excess air.
Excess air unknown.
Variable with production rate.
{*
Natural gas fired.
-------
The amount of NO emitted from a glass-melting furnace depends
X.
upon several factors, some of which are not understood. One important
factor is flame temperature: NO formations in the furnace increase as
flame temperature increases. For example, during a recently completed
experimental program, NO emissions were measured during a complete
X.
firing cycle of a glass melter. NO emissions were highest at the be-
.X
ginning 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 checkerbrick and hence the air
have cooled considerably. Other major factors in NO formation in a
3C
glass melter, such as flame velocity and re circulation patterns of flue
gases, are being studied.
Other emissions, such as carbon monoxide and hydrocarbons, can
be controlled 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.
Effect of New Technology on Air Pollutant
Emissions and Energy Usage
The major change in the glass industry during the next 8-10 years
is projected to be a substantial growth in electric melting. Based on an
extrapolation of the current growth rate (69% per year) of electric melt-
ing, by 1985 all melting would have to be done electrically. However,
this is not a reasonable projection because electric melting growth was
very rapid when just introduced but now is slowing down. Consequently,
for the purpose of showing the effect of electric melting on energy con-
sumption, a more reasonable projection has been assumed based on
statements from the glass industry that electric melting growth will follow
production growth. This more reasonable growth is shown in Figure 6.
This projection indicates that, by 1984, 10 million tons of glass, or
approximately one-third of total glass production, will be produced an-
nually by electric melters. Based on this projection, energy consump-
tion also was projected (Figure 7). These data show that by switching
VI-18
-------
ou
28
26
24
22
22°
o
18
Z"I6
O
§'4
I'*
o:
°-io
8
6
4
2
0
^
^
/
X
«
/
PROJECTED
TOTAL ANNUAL
PRODUCTION
X
X"
X
x
PROJECTED
ELECTRIC MELTING X
^
^
^
x
X
x
r
/
X
^
j
/
/
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84
YEAR
A-83-1247
Figure 6. PROJECTED GROWTH OF TOTAL, ANNUAL
PRODUCTION COMPARED WITH AMOUNT PRODUCED ELECTRICALLY
VI-19
-------
600
500
CM
Q
z"400
O
^300
O
O
O
CC 200
100
PROO
CONE
LEVE
^
ECTE
JUMP'
:LSO
^
DTO
noN
FUSA
^
TALF
BASE!
GE
^-*
T
A
M
(1
G
'URCH
DON
|_ "
^_ -^"*
OTAL ENER
SSUMING 1C
ELTED ELE
NCLUDING t
ENERATION
ONSUMPTK
ASED ENEI
CURRENT
--^ ^
-
~s
<
^
GY CONSUMED
) MILLION TONS
CTRICALLY
ELEC-
IFUE
)N)
TOY/
| ,X
.^
FRICH
L
X
X
^
FY
>
.X
s'
^.
L^
x'
/
/
--- PURCHASED ENERGY CON-
SUMED ASSUMING 10 MILLION
T(
L>
El
DNS r
r(Ex
-ECT
l^ELTl
CLUDf
RICIT^
ID EL
ES FU
CGEN
ECTR
ELFC
ERAT
ICAL-
)R
ION)
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84
YEAR
A-64-923
Figure 7. EFFECT OF CONVERSION TO ELECTRIC MELTING
ON ENERGY CONSUMPTION BY GLASS INDUSTRY
to electric melting, the total purchased energy consumed by the glass
industry itself would decrease by about 18% over energy consumption based
on current levels. However, if the amount of energy consumed to gen-
erate the electricity is included, based on a 30% efficiency of generation,
total energy consumption would increase by 18%. Thus, in fact, the
switch to electric melting will cause an increase in energy consumption
nationally.
Electric melting has been gaining in usage because of its ability to
reduce air pollutant emissions. Data on total emissions from glass plants
are not available on a national basis; consequently, the effect of the pro-
jected increase in electric melting on air pollutant emissions cannot be
determined quantitatively. Certainly, overall emissions would be reduced.
VI-20
-------
However, the emissions from the electricity-generating stations also should
be considered. And on this basis, the net effect of electric melting quite
possibly could be an increase in air pollutant emissions nationally.
By using new technology, significant reductions in air pollutant emis-
sions can be obtained. Recently completed tests on agglomerated glass
batch in a submerged combustion melter showed virtually no particulate
emissions and very low NO emissions (Table 4).
Table 4. AIR POLLUTANT EMISSIONS FROM
SUBMERGED COMBUSTION MELTER
NOx> ppm 34.5-48.7
SO , ppm 71-1210
X.
Particulates, gr/SCF
Loose batch 0.461-0.527
Agglomerated batch 0.0068-0.0150
The use of agglomerated batch prevents batch carry-over by the combus-
tion gases out of the stack, so it should be beneficial in conventional
melters as well. Submerged combustion reduces NO emissions because
3C
of its high efficiency in transferring heat away from the flame and to the
glass itself. Consequently, a low-temperature flame is maintained, and
NO emissions are substantially reduced. However, the product from a
X.
submerged combustion melter is foam glass, that is, glass with air bubbles,
and is not acceptable for use in any product.
The use of oxygen enrichment of the combustion air as a means for
boosting production and reducing energy consumption does not increase
air pollutant emissions. Of particular concern in using oxygen enrich-
ment is the possibility of an increase in NO formation due to a. hotter
flame. In fact, test results on a production facility showed no increase
in NO emissions when the combustion air is enriched with 21-24% oxygen.
ji
Because the flame is theoretically hotter, heat transfer from the flame
to the glass is much higher. Because more heat is transferred from the
flame, the net effect is in reality a flame temperature comparable to
that without oxygen enrichment. As a result, NO emissions do not
ji
increase.
VI-21
-------
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 _9. 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 Ind. 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.
6. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed," 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D.C.: U.S. Government
Printing Office, July 1973.
7. U.S. Department of Commerce, "Consumer, Scientific, Technical, and
Industrial Glassware, 1971," Current Industrial Reports Series;
MA-32E(71)-1. Washington, D.C.: Bureau of the Census, July 1972.
8. U.S. Department of Commerce, "Flat Glass, Fourth Quarter 1972,"
Current Industrial Reports Series; MQ-32A(72)-4. Washington, D.C.:
Bureau of the Census, February 1973.
9. U.S. Department of Commerce, "Glass Containers, February 1973,"
Current Industrial Reports Series; M32G(73)-2. Washington, D.C.:
Bureau of the Census, April 1973.
VI-22
-------
VII. SIC CODE 324 - HYDRAULIC CEMENT
Summary
SIC Code 324 includes all establishments engaged in the manufacture
of hydraulic cement, including portland, natural, masonry, and pozzolana
cements. This report is limited to portland cement production because
95% of the cement manufactured in the U. S. is portland cement and the
remaining 5% is split between the other types.
The cement industry has been growing at a steady rate of about
1. 8% annually for more than 20 years, and this growth rate, which is
shown in Figure 1, is expected to continue in the future.
vt
O
O
O
O
1950 55 60 65 '70
YEAR
'75 '80 1985
A-44-667
Figure 1. ANNUAL CEMENT PRODUCTION
WITH PROJECTION TO 1985
VII-1
-------
Total cement production in the U.S. is estimated at about 83 million
tons and is expected to reach 93 million tons by 1980.n The total
amount of energy consumed by this industry in 1972 was about 581 trillion
Btu, based on an average unit energy consumption of 7. 0 million Btu/
ton of cement produced. The primary sources of energy are natural
gas, coal, and fuel oil. Figure 2, which breaks down energy consump-
tion by type, indicates that coal and natural gas are used much more
than fuel oil.
70
65
45
LL)
O
o:
25
20
7
s
COAL
GAS-
w
1946 '48 '50 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 1972
YEAR
A-44-669
Figure 2. FUEL USAGE OF U.S. CEMENT INDUSTRY7
However, a trend away from coal toward fuel oil has developed within
the last 3 or 4 years, primarily because of the restrictions imposed by
air pollution regulations.
VH-2
-------
In the short term (until 1985), air pollution regulations apparently
will force this trend to continue. However, if fuel availability becomes
a major problem, this trend is likely to reverse itself. In the long run,
the implementation of new processes, particularly the vertical kiln, will
tend to increase coal utilization. Although total annual energy consump-
tion will increase as production increases, the unit energy assumption
(i. e. , Btu/ton of cement) will decrease as new process techniques are
developed and implemented. In a modern, we 11-maintained cement plant
that uses the latest available technology, fuel consumption can be reduced
to less than 3. 5 million Btu/ton of cement, about one-half of the current
average consumption by the entire industry. On this basis, the industry
average could reach 4. 0 million Btu/ton of cement produced. 7
The primary air pollutant emissions from the production of cement
are particulates, SO , and NO . In general, particulate emissions can
" A.
be controlled with collection devices capable of removing 99. 8% of the
particulates from an effluent gas stream. On the other hand, SO emis-
X.
sions are only partially controllable, and NO emissions are totally un-
controllable at present. Much of the air pollutant emissions from this
industry are directly related to the type of fuel used; consequently,
changes in the industry's fuel utilization pattern will directly affect air
pollutant emissions. In particular, if the current trend away from coal
as a fuel persists, particulate emissions will decrease, but this situation
is readily reversible if a shortage of other, less polluting fuels develops
and the industry is forced to revert to coal. SO emissions also depend
on the type of fuels consumed and as such would be adversely affected
in an energy supply shortage. In this case, however, although the industry
might not be forced to burn coal, it might be forced to burn high-sulfur
oil, in which case particulate emissions would be minimized, but SO
emissions would increase. Under these circumstances, in the industry's
attempts to deal with energy shortages and to reduce energy consumption
by implementing new processes, both SO emissions and particulate
emissions will increase.
VII-3
-------
Portland-Cement-Manufacturing Processes1"4*12
Portland cement is made from a mixture of limestone and clay or
similar materials. The raw materials first are ground and mixed
together. Depending upon which of the two available processes is used,
water is added during mixing (the wet process) or the ingredients are
mixed dry (the dry process). In general, if the raw materials are moist
to begin with, the wet process is used; likewise, the dry process is used
if the raw materials are initially dry. The next step is the burning, or
calcining, of the mixed material in a rotary kiln. This burning is done
by heating the charge to approximately 2700°F to form the clinker, which
is discharged from the kiln and subsequently cooled. The final step is
the grinding of the clinker to the desired fineness. During this step, a
small amount of gypsum (2-3%) is added to the clinker to control the
setting time of the cement when wetted.
Raw Material Preparation
In preparation, the raw materials must be ground finely and mixed
in the proper proportions. This step differentiates the wet from the dry
process. In the wet process, the crushed raw materials are mixed with
water and then ground up; the product is a slurry containing up to 50%
water. The slurry is dewatered to about 20-30% water in a continuous
separator. Basically, the separator is a large tank in which the solid
material settles to the bottom. Water is drawn off the top, leaving a
thickened slurry that is mechanically discharged at the bottom. The
slurry then is transferred to the burning apparatus. In the dry process,
the raw materials are simply ground and mixed together in the proper
proportion; the material then can be charged to the kiln.
There was a time when one process might have been preferred over1
the other. Hard raw materials were used in the dry process, and soft
materials were used in the wet process. Hard materials required that
crushers and grinding mills be used to achieve the desired fineness, but
the blending was difficult. The wet process was found to be the easiest
and most economical way of producing the desired fineness of soft mate-
rials. Moreover, blending in a slurry is much easier. However, the
equipment has improved such that either process can be used as desired.
VII-4
-------
The only energy consumed by this process is mechanically consumed
in the operation of the crushers, grinders, and blenders. Previously,
the dry process required more energy than the wet process because the
power required for crushing and grinding the hard material was greater
than the power required for washing the soft material and because the
slurry was easier to blend than the dry raw meal. However, with the
development of ball and tube mills, it became possible to grind hard
materials with water at a 10-20% lower power consumption rate than
without water. 2 Thus, the difference in power consumption between the
two. processes was minimized. Typically, preparation of the raw mate-
rials consumes about 600, 000 Btu/ton of cement (2. 0 million if elec-
tric generation is included). This energy is consumed as electrical
energy, driving the crushers, grinders, blenders, and assorted conveyors.
On this basis, approximately 8% of the total annual energy consumption,
or 50 trillion Btu, is consumed in the preparation of the raw materials.
Burning the Raw Material3
Burning is the heating operation in which the raw materials combine
chemically to develop a product with the desired hardening characteristics.
Typically, the prepared raw materials are charged continuously into a
cylindrical steel rotary kiln, turning at about 1 revolution per minute.
The kiln, which is lined with firebrick, varies in diameter from 12 to
25 feet and in length up to several hundred feet. The kiln is mounted
at an angle slightly off horizontal so that, as it rotates, the charge
moves by gravity from the high, charging end to the low, discharging
end. The discharge end of the kiln also is the firing end. The fuel
used for firing can be natural gas, powdered coal, or, in some cases,
fuel oil. After burning at 2700°F, the product, or clinker, is discharged
and cooled so that it can be handled and stored. On the average, the
burning process requires about 6.4 million Btu/ton of cement produced.
Thus, in 1972, an estimated 530 trillion Btu of energy was consumed in
the burning process.
VII-5
-------
In the burning process, the charge passes through three stages
before being discharged. In the first stage, which begins as the load
is charged into the kiln, the moisture is driven off. As the charge
moves through the kiln, temperatures increase until, at about 1800°F,
carbon dioxide is driven off. Finally, the charge moves through the
hottest temperature zone, 2700°F, and burns. The resulting clinker is
discharged and cooled.
Although the average energy consumption for burning is 6.4 million
Btu/ton of cement, energy consumption varies from 3. 7 to 9. 0 million
Btu/ton, depending upon many factors, such as chemical and mineral
composition of the charge; particle-size distribution and fineness of the
charge; diameter, length, pitch, and speed of rotation of the kiln; and
efficiency of combustion and heat-recovery systems. 5 Also, the wet
process requires about 30% more energy for burning than the dry process,
most of which is needed to vaporize the extra water.
Clinker Processing
After burning, the clinker is cooled and stored until needed. Basically,
the only processing necessary is grinding the clinker to specification.
This usually requires about 34, 000 Btu of electric energy per ton of
clinker ground, or 110,000 Btu if energy consumed for electric genera-
tion is included.
New Technologies in the Manufacture of Portland Cement
Several recent developments in the manufacture of portland cement
have improved energy utilization. All the developments are centered
around the burning process. To reduce the energy consumed in this
process, the heat-transfer rate from the combustion gases to the charge
must be increased.
Chain Systems3
In the first stage of the burning process, in which the raw materials
are dried, several types of systems, including lifters to cascade the
material through the combustion gases and plates and diaphragms to baffle
the gases, have been developed to improve the heat-transfer rate over
that which can be obtained in the plain cylindrical kiln. The method
most used now is a system of chains hanging in the kiln.
VII-6
-------
In the wet process, the chains hang so that they dip into the slurry,
thus becoming coated. As the kiln rotates, the chains emerge from the
slurry, offering large-surface-area contact with the hot gases. Further
down the kiln, in the second stage' of the burning process, a system of
festoon chains helps push the slurry and pug (nearly dried material) through
the kiln. A well-balanced system of chains can reduce the temperature
of the flue gases by about 350°F from the normal operation. Such a
reduction in temperature means that more heat has been transferred to
the load; i. e. , less heat is escaping out of the stack.
In the dry process, temperatures are much hotter than in the wet
process, resulting in the burning of the chains. To circumvent this
problem, lifting devices are used to cascade the raw meal through the
gases.
Preheaters3
A major advance in reducing energy consumption in the burning
process was separating the first stage from the kiln and preheating the
raw material up to near burning temperatures before it entered the
kiln. The several ways in which this can be done are summarized in
Table 1, which shows the effect of each method on energy consumption,
assuming an average consumption for burning of 6.4 million Btu/ton by
using current practice.
Table J. ENERGY CONSUMPTION OF VARIOUS
CEMENT-MANUFACTURING PROCESSES3
Process
Wet
Long Kiln
Calcinator and Short Kiln
Semi wet
Preheater and Short Kiln
Dry
Long Kiln ;
Suspension Preheater and Short Kiln
Semi dry
Grate Preheater and Short Kiln
Energy
Consumption,
106 Btu/ton
5. 94
4. 68
3. 60
4. 68
3.15
Vertical Kiln
Includes 0.54 X 106 Btu/ton for drying.
; VH-7
3.42*
4.14*
Reduction Over
Average Current
Practice, %
26. 9
43. 8
50. 8
46.6
35.3
-------
One method for preheating the raw material is to feed it dry, as a
powder, into a series of cyclones in which it is heated by the hot com-
bustion gases from the kiln. The hot raw material then is fed into a
kiln that is shorter than would otherwise be used. This process is still
considered to be a dry process, and the preheater is known as a suspen-
sion preheater. By using this system, energy consumption can be re-
duced by nearly 51% from the current industry average of 6.4 million
Btu/ton.
The second method, referred to as the semidry process, involves
wetting the raw materials to a moisture content of 12-14% and then
nodulizing or pelletizing the material. The nodules or pellets then are
fed onto a traveling grate, as a deep bed, through which hot gases are
passed before the solids enter the kiln. Variations of both these methods
have been developed and are widely used.
The energy savings in this process occur as the result of preheating
the raw material and as a result of the improved rate of heat transfer
that ensues when agglomerated raw materials are used. Agglomerating
the raw materials enhances heat transfer by offering a uniformity in the
materials through which the combustion gases pass, thus creating a more
uniform flow distribution. In addition, agglomerated materials offer a
greater amount of surface area for heat transfer, thus increasing its rate.
The energy saved by using this process is about 3. 0 million Btu/ton
of cement produced, compared with the industry average. In fact, the
savings for this process and the process using the suspension preheater
are slightly lower than stated because the energy consumed by a typical
dry process is lower than the industry average.
In the wet process, developments have followed along the same lines,
A preheater, known as a calcinator, that has been developed is separate
from the kiln and consists of a rotating drum filled with heat-exchange
bodies. Fuel consumption is reduced by 27% over the average fuel con-
sumption of a normal kiln. The calcinator, in addition to reducing fuel
consumption, increases the output when applied to existing kilns. Work
is continuing to develop a way of reducing the water content of the slurry
by filtration and, subsequently, producing nodules that then would be fed
to a preheater. If this process were fully developed, further reductions
VII-8
-------
in energy consumption would be possible. At present, through the use of
a complete series of batch-operated pressure filters, the moisture con-
tent can be reduced to 18-20%, but the product is difficult to handle and
nodulize. However, the material is usable in deep-bed counter-flow
preheaters and in traveling grate preheaters. This type of process is
referred to as semiwet. Potential energy savings of 44% exist.
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. 9 It was unable to produce cement of consistently good
quality; fluctuations in strengths, setting, and soundness properties of
the cement were very high. However, the vertical kiln has seemingly
been developed such that these problems have been solved. Satisfactory
performance of the vertical kiln requires 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 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. 9
One disadvantage of the vertical kiln is that its capacity is limited
to about 300 tons/day. 3 A number of vertical kilns in a row, however,
require less space and cost much less than a rotary kiln of corresponding
capacity, 9 so output capacity is not really a problem.
Oxygen Enrichment
Much work has been done in recent years to determine the feasibility
of enriching the combustion air of a rotary kiln with oxygen to increase
production and reduce fuel consumption. One such program used oxygen
lancing, a technique in which the oxygen is introduced directly into the
combustion chamber, rather than being introduced into the combustion air
fan and mixed with the air. The use of oxygen resulted in a maximum
VII-9
-------
production rate 40% higher than the rate with no oxygen; in addition,
fuel consumption per ton of product decreased by 15%. (See Figure 3. 8)
O
ID
O
O
QC
QL
1500
1400
1300
1200
1100
1000
o>
0)
O
.0
.a
m
o
o
o
o
0.
2
O
O
_l
UJ
ID
O.I 0.2 0.3 0.4 0.5 0.6
ENRICHMENT LEVEL, 02/fuel by volume
0.7
A-44-668
Figure 3. PRODUCTION AND FUEL CONSUMPTION IN
ROTARY KILN VERSUS OXYGEN ENRICHMENT LEVEL8
Another program with oxy-fuel burners achieved good results. 6 In spite
of its technical success, oxygen enrichment is not used in U.S. cement
plants. The major deterrent is that the cost of oxygen is too high and
cannot be justified, at present, by the improvements in efficiency. How-
ever, if the same results can be produced at lower oxygen consumption
rates, or if fuel availability to this industry becomes severely restricted,
this process might be more realistic. Otherwise, only a drastic reduc-
tion in oxygen costs would bring this process into use by the industry.
Cement Production Without Heat10
Aside from all the techniques for reducing fuel consumption in rotary
kilns or other heating devices, one process has been developed for pro-
ducing cement without heat treatment; Certain clays can be strengthened
by treatment with small amounts of lime to produce cementitious products
similar to those formed by hydration of portland cement. This occurs
VII-10
-------
through the slow reaction of lime with the alumina silicates of clays.
The only energy consumed is power for a specially designed grinder and
mixer that can accelerate the rate of strength development of the mix-
tures. The process uses a wide variety of available materials and can
produce a wide range of products. Although this process was designed
for use in underdeveloped areas, the fact that waste products such as
flyash, rice hulls, and siliceous mining wastes can be converted to
cementitious materials through a relatively simple process should make
it attractive to industrialized countries as well. At present, a pilot
plant is being planned to determine, among other things, the economic
factors associated with this process.
Other Concepts
One concept that has been developed recently is the use of a submerged
combustion burner that operates while submerged in the cement mix.
Although the burner was originally used in a rotary kiln, the technology
has been developed for use in a trough kiln. The trough kiln is divided
into a preheating zone and a calcining zone. In the first zone, waste-
heat gases from the calcining and clinker-cooling processes are introduced
under pressure through a high-temperature porous media, thereby fluid-
izing the mix for conveying through the sloped stationary kiln into the
calcining zone, where the mix is fluidized and calcined by the submerged
combustion burners. The fully calcined mix then is discharged into a
short rotary kiln, in which clinkering occurs. Based on experimental
laboratory data, the heat requirements for this system are about 2. 7
million Btu/ton of cement. At this fuel consumption rate, a 58% savings
in fuel over the current average consumption rate would be realized.
Other stationary systems have been developed, but, as yet, they are
unproved. Consequently, there are no definitive operating data. The ad-
vantage of a stationary system is the capacity for fully automatic process
control instrumentation; such a capability significantly enhances the oper-
ating efficiency. Automated process control has limited applicability in
a rotary kiln.
VII-11
-------
Energy Utilization Pattern
Although no revolutionary changes have occurred in portland cement
manufacture during the last 60 years, the evolutionary changes that have
occurred have improved efficiency and reduced fuel consumption. How-
ever, the industry in the U.S. apparently is not taking advantage of these
changes. While the rest of the world's cement producers are switching
from the wet process to the dry process, which consumes less energy
per unit production, the U.S. producers have been staying with the wet
process.
There are two reasons for this trend. First, the dry process re-
quires slightly more manpower than the wet process. Second, fuel has
been cheap in the U.S. Consequently, more attention has been focused
on saving labor costs than on saving fuel costs. This trend is beginning
to reverse itself, however, because possibilities for increasing production
capacity with little or no extra investment in equipment are better in
modern dry-process kilns and fuel prices are beginning to increase.
At present, approximately 45% of the energy consumed by the cement
industry is natural gas, 15% is fuel oil, and 40% is coal. Wide accept-
ance of the vertical kiln by the cement industry would cause a marked
shift away from natural gas and toward coal. Figure 4 shows that if the
cement industry -were to switch 25% of the projected production load from
rotary kilns to vertical kilns by 1985, total energy consumption would be
about 12% less than if current energy consumption levels were maintained.
Furthermore, natural gas consumption would decrease by nearly 10% from
its current level and would be 25% lower than if the current ratios of
gas-to-coal-to-oil consumption were maintained through 1985. At the
same time, under current fuel ratios, total annual coal consumption would
be 22% higher in 1985 than in 1972, and if the projected shift to vertical
kilns were to take place, total coal consumption would only be 6% higher
in 1985 than if no shift had occurred, or a total of 30% higher than in
1972. With a switch to vertical kilns, total oil consumption would be
26% lower than current projected levels.
VII-12
-------
o
CL
8
500
400
PROJECTION BASED ON CURRENT
FUEL CONSUMPTION RATES
PROJECTION BASED ON LINEAR
EXPANSION TO 25% OF CEMENT
PRODUCTION BY VERTICAL
KILNS BY 1985
O
UJ
O
cr
o.
200
100.
O TOTAL ENERGY CONSUMPTION
A NATURAL GAS
D COAL
V FUEL OIL
1972
'74 '76
'78
YEAR
'80
'82 '84 1985
A-44-670
Figure 4. EFFECT OF CONVERSION OF
ROTARY TO VERTICAL KILN BY CEMENT
INDUSTRY ON ENERGY UTILIZATION PATTERNS
The energy consumed per ton of cement produced can be expected to
decrease in the future, and most of the methods used to reduce consump-
tion 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 6.4 to 4.7
million Btu/ton. Figure 5 summarizes the concepts for reducing energy
consumption in the short term on the typical rotary kiln. If all these
concepts were implemented 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 5 are the reductions in energy consumption that
VII-13
-------
Typical Ra* Feed
Calcium Carbonate
Silicon Dioxide
Ferric O.ide
Others - MfiO. Na.
K. S03. Cl. etc.
H20
JO--1S% *ET *T
C02
JiJlDRY *f
Coolinj; Zont
LNJ
1. High-Sulfur Fuels, variable 6. Air Infiltration, 0.2 X 106 Btu/ton
Z. Chain Systems, 1.6 X 106 Btu/ton 7. Oxygen Enrichment, 0.5 X 106 Btu/ton
3. Feed End Enlargement, 1.0 X 106 Btu/ton 8. Process Control, 0.3 X 106 Btu/ton
4. Trefoils, nominal 9. Slurry Dewatering, 1.3 X 106 Btu/ton
5. Kiln Ledges, nominal 10. Waste Heat Utilization, 0.5 X 106 Btu/ton
10
Typical Product
Dicalcium Silicate
Tncalcium Silicate
Tncalcium Aluminate
CLINKER
5 GYPSUM
10051 PORTLAND CEMENT
Figure 5. SHORT-TERM ENERGY-SAVING CONCEPTS
IN THE U.S. CEMENT INDUSTRY7
-------
have actually been achieved. However, these reductions are not cumula-
tive; that is, implementation of all these concepts would not reduce energy
consumption by 5. 4 million Btu/ton, the summation 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 result 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 nec-
essary capital for converting to these new processes until 1985.
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 pro-
" duction, to clinker grinding, storage, and packaging. However, emis-
sions also include the products of combustion of the fuel used in the
rotary kilns and drying operations; these are typically NO and small
X.
amounts of SO . Table 2 summarizes the emissions from cement-
X.
manufacturing processes without controls.
The largest source of emissions in cement plants is the kiln opera-
tion. 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 3. 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
because 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 con-
verting 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.
VII-15
-------
Table 2.
EMISSION FACTORS FOR CEMENT MANUFACTURING
WITHOUT CONTROLS (Ref. I3)a»b
Pollutant
Participate6
Ib/ton
kg/MT
Sulfur dioxided
Mineral source6
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
Neg'
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. IS
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 multicyclones, 97.5
.percent; and fabric filter units, 99.8 percent.
dThe sulfur dioxide factors presented take into account the reactions with the alkaline dusts
when no baghouses are used. With baghouses, approximately 50 percent more SC>2 is removed
because of reactions with the alkaline paniculate filter cake. Also note that the total SC>2 from
the kiln is determined by summing emission contributions from the mineral source and the
appropriate fuel.
8 These emissions are the result of sulfur being present in the raw materials and are thus depend-
ent upon source of the raw materials used. The 10.2 Ib/ton (5.1 kg/MT) factors account for
part of the available sulfur remaining behind in the product because of its alkaline nature and
-affinity for SO2.
Negligible.
9S is the percent sulfur in fuel.
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 mech-
anical collectors,15 electrical precipitators,16 baghouse filters,17 or
combinations of these devices to control emissions. Typically, in treating
!the flue gases from the kiln operation, mechanical separators are normally
VII-1
-------
Table 3. 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. 8 X 106 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.
-------
required ahead of either 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 precipitators because the moisture contributes to
more satisfactory precipitator operation. 4
Efficiency of particulate removal depends upon the types and combi-
nations 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
-X
are 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 controlled in the process of cement manufacturing because
about 75% of the raw feed is converted to calcium oxide, which reacts
with sulfur dioxide. In addition, 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 en-
trapment 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.7 Sulfur dioxide also is removed by this same mech-
anism 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 mate-
rials and fuel.
Effect of New Technologies on Air Pollutant Emissions
The implementation of new processes in the cement industry could
result in an increase in air pollutant emissions, even if current control
devices continue to be used. The most serious problems are likely to
occur with implementation of the vertical kiln because of the potential
for increases in both SOx and particulate emissions. These emissions
VII-18
-------
would result from the increase in coal utilization due to implementation
of the vertical kilns. However, the sulfur dioxide absorption rate might
be higher because of the intimate contact between reactants, in which
case SO emissions could conceivably be lower than current emission
X
rates. Whether or not particulate emissions increase will depend on
the control methods used.
The use of oxygen enrichment, because it increases flame temper-
ature, may tend to increase NO . However, oxygen enrichment has been
used in other industrial processes in a similar manner and NO emissions
JC
did not increase.* NO emissions probably will be reduced if the vertical
kiln becomes widely used. Although there are no data on NOX emissions
from vertical kilns, the absence of a high-temperature flame, which
favors NO formation, tends to support this claim.
References Cited
1. American Gas Association, A Study of Process Energy Requirements
in U. S. Industries Cement^ Catalog No. C'z'OO'O^.Arlington, Va. :
n. d.
2. Barrell, K. C. , "The Manufacture of Portland Cement. 1. Introduction,"
Cem. Lime Gravel 46, 3-10 (1971) January.
3. Barrell, K. C., "The Manufacture of Portland Cement. 4. Variants
of the Burning Process," Cem. Lime Gravel 46, 137-42 (1971) June.
4. Barrell, K. C. , "The Manufacture of Portland Cement. 5. Clinker
Grinding and Cement Handling," Cem. Lime Gravel 46, 187-93
(1971) August.
5. Blanks, R. F. and Kennedy, H. L. , The Technology of Cement and
Concrete, 142. New York: John Wiley"1955.
6. Fredericks, S. , Stirling, R. and Blessing, J. , "Oxy-fuel Burners:
Minimum Heat Loss for Maximum Profits, " Rock Prod. 74, 56-58
(1971) August.
7. Garrett, H. M. and Murray, J. A. , "Energy Conservation in the
Cement Industry. " Supplement to paper presented at 9th International
Cement Industry Seminar, Chicago, 1973.
M-
See Section VI pn the glass industry (SIC Codes 3211, 3221, and
3229).
VH-19
-------
8. Gaydos, R. A., "Oxygen Enrichment of Combustion Air in Rotary
Kilns," J. Portland Gem. Assoc. Res. Dev. Lab. T_, 49-56 (1965)
September.
9. Gottlieb, S., "The Modern Vertical Kiln, " Rock Prod. 69, 82-86,
(1966) June.
10. Mehta, P. K., "Cement Production Without Heat," Rock Prod. 74,
84-87, 120, 122 (1971) May.
11. Predicasts, ' Issue No. 44. Cleveland: Predicasts, Inc., July 30,
1971.
12. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, D. C. : U.S. Government Printing
Office, January 1972.
13. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors," Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C. , April 1973.
14. van der Lyn, A. , "Prescription for Cement Plant Dust Control,
Part 2," Rock Prod. 73, 118-20, 136-38 (1970) September.
15. Walling, J. C., "Cement Plant Dust Collectors, Operation and
Maintenance. Part 1. Mechanical Collectors, " Pit Quarry 63,
107-10 (1971) June.
16. Walling, J. C. , "Cement Plant Dust Collectors, Operation and
Maintenance. Part II. Electrostatic Precipitators, " Pit Quarry 63,
143-45, 148 (1971) July.
17. Walling, J. C. , "Cement Plant Dust Collectors, Operation and
Maintenance. Part III. Bag Filters," Pit Quarry _63, 72-76 (1971)
August.
VII-20:
-------
VIII. SIC CODE 325 - STRUCTURAL CLAY PRODUCTS
Summary
The structural clay products industry comprises establishments
engaged in the manufacture of brick and structural clay tile, ceramic
wall and floor tile, clay refractories, and miscellaneous items such as
adobe brick and clay sewer pipe. Establishments engaged in the manu-
facture of china, earthenware fittings, bathroom accessories, kitchen
articles, and porcelain electrical supplies are classified in SIC Code 326.
By comparison, the latter establishments consume about one-sixth as
much energy as establishments covered by SIC Code 325.
This report section covers several types of ceramic products with
widely differing manufacturing processes and end uses. In addition, the
units of production depend on the product being manufactured; production
is reported in single units, thousands of units, thousands of dozens of
units, short tons, and so on. (The dollar value of shipments is not used
in this report because it is constantly fluctuating, making comparisons
from one year to the next difficult, if not impossible, without a complex
economic evaluation.) Consequently, general statements about the industry,
as well as comparisons between sectors of the industry, cannot readily
be made.
During the past 15 years, this industry has been very stable, in-
creasing in annual production at only about 1% per year. Consequently,
total annual energy consumption has tended to increase only slightly.
(See Figure 1.) In 1971, energy consumption did decrease, but this was
due ostensibly to a temporary decrease in construction, where many of
this industry's products are used.
The primary sources of energy consumed by this industry are nat-
ural gas, manufactured gas, still gas, coke-oven gas, blast-furnace gas,
fuel oil, coal, and electricity. About 97% of the energy consumed is
for direct process heat in the firing of kilns and dryers. The remaining
3%, practically all of which is electricity, is consumed in the drives
for the grinders, conveyors, and blenders. This report covers only the
direct heating processes because most of the energy is consumed here
and the greatest possibilities for significant reductions in energy consump-
tion exist here.
VIII-1
-------
e. \ \J
-,200
&
>a 190
oo
£izf '80
zo
^H 170
ii160
0150
140
130
/
X
^
x^
x%
\
k
\
V y
\'
/
jt
s
/
/
/
/
s
/
/
/
1958 '60 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-54-822
Figure 1. TOTAL ENERGY CONSUMED BY THE
STRUCTURAL CLAY PRODUCTS INDUSTRIES8
The industry as a whole is sensitive to the economy and tends to
grow at nearly the same rate as the economy. Thus, total energy con-
sumption is likely to increase only very slowly. No dramatic changes
that could affect energy consumption by the industry are expected in the
future. However, a slight trend toward automatic processing and shorter
firing cycles that is developing will reduce the energy consumed per unit
of production. Because of the diversity of this industry, the size of this
reduction is difficult to assess.
Structural-Clay-Manufacturing Processes
Although the large number of products manufactured by this industry
require different raw materials, different firing temperatures and cycles,
and different types of machinery for each operation, the basic sequence
of operations is very similar for all products. First, the raw materials,
which are usually various clays and other minerals, including soda ash,
kaolins, magnesite, and dolomite, are crushed and ground to the required
mesh size. Next, the material is mixed with water in a pug mill or
blunger. When the material is of the proper consistency, it is formed
into various shapes by any number of methods, such as casting, pressing,
or extrusion, and subsequently dried. The next step is firing, which is
usually done in kilns at temperatures from 1500°to 3000°F, depending on
VHI-2
-------
the product. For products requiring glazing, an additional firing cycle
is required, but for those requiring no glazing, the manufacturing is
complete after one firing cycle.
Drying1'5
The first major energy-consuming process in the manufacture of
structural clay products is drying the green product after it has been
formed. Four methods are used for drying: rack drying in the open or
in open sheds, hot floor, drying rooms, and tunnel dryers. Of these
methods, drying rooms and tunnel kilns are most commonly used because
they provide the shortest drying times. Air-drying, which is certainly
the cheapest method, also is the slowest method, generally too slow for
modern production operations. In addition, air-drying requires more
handling than other methods, subjecting the ware to dirt, possible damage,
and moisture pickup from the atmosphere. Consequently, its use is limited.
Hot floor dryers, drying rooms, and tunnel dryers are typically
heated by direct-fired air heaters, exhaust heat from tunnel kilns and
gas turbines, steam coils, and infrared heaters. Of these methods,
direct firing with auxiliary fuel is the most desirable because it allows
for the most complete control over the drying process. The waste gases
from the kilns are also used with supplementary firing, but only if doing
so does not disrupt the draft in the kiln. Waste warm air from the
cooling section of a kiln is especially desirable because it is not diluted
with combustion gases from the kiln, thus reducing the chance of con-
tamination by carbon or sulfur. Compared with steam coil drying, direct
firing can usually produce hotter drying air with a smaller sized dryer,
accomplishing faster drying with less floor space. 5
Infrared heaters, although used, are ineffective for drying and re-
moving moisture unless blowers are used for circulating the air through
the ware. An alternative to this system is a recirculating air heater,
which is cheaper to operate and requires a lower capital investment than
the infrared heating system.
Steam coils are used to heat air for the dryer, but because many
boilers and steam coil systems are limited to low-pressure operation,
the coils do not get hot enough to do a very fast job of drying. Boilers
used for this purpose are either fired with auxiliary fuels, or they utilize
waste heat from the firing kilns.
VIII-3
-------
Because of the diversity of this industry, no figures on the energy
consumed by the drying process are published, nor are there any statistics
on the usage of the different drying methods.
Firing2
Three basic types of kilns are used to fire the dried ware: field
kilns, beehive kilns, and tunnel kilns. Of these types, tunnel kilns are
preferred for plant efficiency, manpower requirements, and product
quality.
Tunnel kilns are large car-type kilns that fire the product continu-
ously. Ware is loaded up on refractory-topped cars or, in the case of
low-temperature applications, on conveyor belts and continuously pushed
through a continuously heated tunnel. Ware entering the tunnel is gradually
heated by waste heat circulated from the cooling section at the exit of
the kiln. The preheated ware is pushed into the firing zone, where it
is heated to the desired maturing temperature. The fired material then
is pushed into the cooling zone by incoming cars loaded with ware, where
heated air is removed and channeled back to the kiln entrance for pre-
heating. Natural gas is the most common fuel used for firing large
tunnel kilns, although coal, fuel oil, and on occasion electricity may be
used.
Beehive kilns are large, round downdraft kilns that are operated on
a discontinuous basis; thus, they also are termed periodic kilns. With
this type of kiln, ware is placed inside, fired to the desired temperature,
cooled, and then removed. During firing, the combustion gases circu-
late to the roof of the kiln and then down through the ware to exhaust
ports in the floor. One advantage of this type of kiln is that the temper-
ature gradient is nearly eliminated, compared with straight updraft
kilns. Another advantage is that this type of kiln can be fired when re-
quired and can be used for a wide range of heavy clay products. But
therein also lies its major drawback. In a periodic kiln, the ware is
set and the kiln fired. At the end of the firing cycle, the kiln is allowed
to cool for ware removal. This results in a tremendous waste of heat
and consequently in high fuel consumption rates.
VIII-4
-------
Historically, these kilns have been fired with coal, but in recent
years these kilns have been switched to oil and natural gas firing. 2 One
trend has been the use of instrumentation and automatic controls to im-
prove fuel economies and reduce labor costs. Fuel economies also have
been achieved with insulating firebrick linings, steel encasements, and
use of insulating refractory concretes for crowns, flues, arches, and
firebox sidewalls.
Another type of periodic kiln is known as a chamber kiln; it is used
primarily in Europe. In these kilns, chambers are located in a circular
or oval shape and alternately loaded, fired, cooled, and unloaded as the
firing zone is moved around the kiln. Waste heat from the cooling sec-
tions is used to preheat the ware ahead of the firing zone. Although
these kilns are not as thermally efficient as tunnel kilns, they are a
large improvement over the beehive kilns.
Energy Utilization Pattern
The pattern of energy utilization within this industry, other than the
total fuel consumption by type, is very difficult to assess because no
definitive figures exist on the fuel consumption rates of specific processes.
Different firing times and temperatures are required for different types
of products. Figure 2 shows the effect of temperature on energy con-
sumed by both tunnel kilns and periodic kilns for various types of ware.
For bricks fired in chamber or tunnel kilns, energy consumption varies
from 400 to about 3Z50 Btu/lb. But without specific information relating
the product mix to firing temperatures, process energy consumption
cannot be quantified for the industry as a whole.
Nevertheless, certain trends that do exist within the industry have
implications for energy consumption patterns in the future. The first
trend is a noticeable switch away from periodic kilns to continuous kilns.
Figure 2 shows that periodic kilns typically consume about twice the
amount of energy that tunnel and chamber kilns consume at comparable
temperatures. Such a shift means that the average energy consumption
per unit of production should decrease in the future. The second trend
is toward increased automation of the entire production from grinding of
the raw material to firing of the ware. The expected reduction of re-
jected ware also will reduce energy consumption. A third trend that is
developing is a shift away from steam-drying to direct-firing.
VIII-5
-------
100.000
1000
400
900
1000
1100
1200 1300 1400
MATURING TEMPERATURE °C
1500
1600
1700
Figure 2. HEAT UNITS REQUIRED FOR FIRING
IN THE CERAMIC INDUSTRY7
In general, the industry is striving to reduce energy consumption
by reducing the length of firing cycles and lowering firing temperatures
when possible. The new firing techniques that have been developed are
discussed later in this section.
It is difficult to say, with any accuracy, what changes will occur in
the types of fuels that are used. At present, between 60 and 70% of the
energy consumed in this industry in the U.S. is gas (a combination of
natural, manufactured, still, blast-furnace, and coke-oven gas) because
of the color requirements of the ware and the clean burning character-
istics of the gas. However, a large part of this load could be transferred
to oil or coal. Glaze-firing or certain special cases in which controlled
atmospheres are required would seem to be the only major exceptions.
VIII-6
-------
New Technologies and Their Effects on Energy Consumption
Although this industry is slow to change, some new firing techniques
that have been developed reduce the energy consumed per unit of product
'by kilns. One such technique, applicable to both tunnel kilns and periodic
kilns, is the use of high-velocity jet-firing.4'6 The use of jet burners
improves several aspects of the firing operation by supplying the necessary
movement of the gases in the-kiln to eliminate problems of heat stratifica-
tion, improve temperature uniformity, and provide for better control of
the kiln atmosphere. The jet burners enable the use of smaller kilns,
resulting in lower heat losses through the kiln sidewalls and improving
fuel efficiency. Information on the extent of the improvement in fuel
efficiency is not available, but a fuel consumption rate of 500 Btu/lb of
standard 8-inch three-hole brick at 2550°F (1400°C) has been reported
for tunnel kilns. 6 This fuel consumption rate appears at the low end of
the curve in Figure 2 and thus would appear to represent a substantial
improvement over current practices, perhaps as high as a 50% reduction
in fuel consumption.
A second technique that has been developed is forced-draft firing in
periodic kilns.3 A major portion of the fuel in a natural-draft kiln is
spent for heating the kiln so that an adequate draft is established. In
fact, the proper kiln conditions may require several days to achieve.
However, forced-draft firing systems drive hot, dry air through the kiln
long before a draft would be available, resulting in a significant decrease
in kiln time. Forced-draft firing also eliminates problems normally
associated with windy days. With a forced-draft firing system, the kiln
can be made to operate under slightly pressurized conditions, virtually
eliminating any effects of the wind on the kiln operation. Experiments on
various beehive kilns indicate that production can be increased by 25%,
firing time reduced by 30%, and fuel consumption per unit of production
reduced by 20% by switching to a forced-draft firing system.
VIH-7
-------
Air Pollutant Emissions
Emissions from structural-clay-manufacturing processes are primarily
particulate matter, although other pollutants, including SO , carbon
X.
monoxide, and fluorides, also are emitted. Table 1 summarizes the
emissions from the various processes. The data show that particulate
emissions are most severe during the handling of the raw materials.
All the other emissions occur during the firing of the ware.
Several factors are involved in the emission rates of the various
pollutants. Particulate emissions are affected by the type of grinding
(wet or dry), the kiln temperatures, the gas velocities, and the flow
direction of the gases in the kiln. Gaseous emissions are affected by
the composition of the clay being processed and by the fuel being used
to fire the kilns. Of primary concern are the SO and the fluorides.
X.
Both of these gases are emitted as a result of the presence of sulfur
and fluorides in the clay. In addition, sulfur results from the type of
fuel being consumed; the amount that is emitted (as SO ) depends upon
X.
the sulfur content of the fuel. Other gaseous emissions that pose prob-
lems are NO and carbon monoxide, both of which are formed during
X
combustion. NO formation is not entirely understood, but higher flame
X.
temperatures are known to increase NO formation. Carbon monoxide
usually is formed as a result of incomplete combustion of the fuel due
to insufficient combustion air or incomplete mixing of the fuel and air.
Coal, because it is a solid fuel, is more susceptible to incomplete com-
bustion than the other fossil fuels.
Methods of controlling emissions are similar to those used in other
industries. Cyclone filters used in conjunction with scrubbers can reduce
particulate emissions by 90%. Bag filters and electrostatic precipitators
also can be used effectively. Both SO and fluorides can be removed
from the effluent gases by a variety of scrubbing techniques, but the cost
is nearly prohibitive.
vni-s
-------
Table 1. EMISSION FACTORS FOR STRUCTURAL-CLAY
MANUFACTURING WITHOUT CONTROLS (Ref. 9)a
Type of process
Raw material handling
Dryers, grinders, etc.
Storage
Curing and firing
Tunnel kilns
Gas-fired
Oil-fired
Coal-fired
Periodic kilns
Gas-fired
Oil-fired
Coal-fired
Particulates
Ib/ton
96
34
0.04
0.6
1.0A
0.11
0.9
1.6A
kg/MT
48
17
0.02
0.3
0.5Ad
0.05
0.45
0.8A
Sulfur oxides
(SOX)
Ib/ton
-
kg/MT
Carbon monoxide
(CO)
Ib/ton
kgAMT
i ~ I ~
- .
Negb
4.0SC
7.2S
Neg
5.9S
12.0S
Neg
2. OS
3.6S
Neg
2.95S
6.0S
! r
0.04
Neg
1.9
0.11
Neg
3.2
0.02
Neg
0.95
0.05
Neg
1.6
Hydrocarbons
(HC)
Ib/ton
-
-
0.02
0.1
0.6
0.04
0.1
0.9
kg/MT
-
-
0.01
0.05
0.3
0.02
0.05
0.45
Nitrogen oxides
(NOX)
.. A .
Ib/ton
-
-
0.15
1.1
0.9
0.42
1.7
1.4
kg/MT
-
-
0.08
0.55
0.45
0.21
0.85
0.70
.Fluorides
(HF)
Ib/ton
-
-
1.0
1.0
1.0
1.0
1.0
1.0
kg/MT
-
-
0.5
0.5
0.5
0.5
0.5
0.5
aOne brick weighs about 6.5 pounds (2.95 kg). Emission factors expressed as units per unit weight of brick produced.
Negligible.
c S is the percent sulfur in the fuel.
A is the percent ash in the coal.
-------
Examination of techniques being applied by the structural clay industry
to reduce fuel consumption leads to the conclusion that emission rates
affected by combustion (as opposed to clay composition) will tend to de-
crease in the future. Possible exceptions are NO emissions, which
.X
may increase because of the utilization of higher operating temperatures,
and particulates, which may increase because of an increase in kiln gas
velocities, causing an increase in dust pickup from the load.
References Cited
1. American Gas Association, A Sturdy of Process Energy Requirements
in the Ceramics Industry, Catalog No. C20004. Arlington, Va., n. d.
2. "Firing, Melting, and Annealing," Ceram. Age 86, 106, 107, 110,
112 (1970) July.
3. Marshall, R. W., "Forced Draft Firing for Beehive Periodic Kilns,"
Am. Ceram. Soc. Bull. 49, 518-21 (1970) May.
4. McFadden, C. A. , "High Velocity Jet Firing, " Ceram. Age 84,
28-31 (1968) September.
5. Reed, R. J. , "Drying Direct Firing of Air Heaters for Ceramic
Drying," Ceram. Age 86, 100,- 101, 104, 105 (1970) July.
6. Remmey, G. B., "Jet Firing Creates New Tunnel Kiln Design,"
Brick Clay Rec. 152, 32-34 (1968) March.
7. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, .Executive
Office of the President. Washington, D. C. : U.S. Government
Printing Office, January 1972.
8. U. S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C. : U. S. Government
Printing Office, July 1973.
9. U. S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C., April 1973.
VIII-10
-------
IX. SIC CODE 331 - BLAST FURNACES, STEEL WORKS
AND ROLLING AND FINISHING MILLS
Summary
The energy used in blast furnaces, basic steelmaking processes, and
various reheating processes, which include rolling and finishing (all of
which are encompassed by SIC Code 331Z), amounts to more than 18% of
the total energy consumed by industry in the U.S. In 1971, this amounted
to more than 3000 trillion Btu of primary energy, excluding the energy
consumed for electricity generation.2 Tables 1 and 2 show the breakdown
of primary fuel utilization in the industry.
Table 1. CONSUMPTION OF ELECTRICITY
BY THE IRON AND STEEL INDUSTRY*
Consumption,
Year 101Z Btu
1971
1970
1969
1968
1967
166.03
169.22
165.16
157.53
145.09
*
Excludes fuel consumed for generation.
However, the industry also consumes large amounts of secondary fuels,
that is, tar and pitch, coke-oven gas, and blast-furnace gas, as shown in
Table 3. These fuels are not included in the total energy consumption
figure because they are by-product fuels from processes consuming the
primary fuels. (See Figure 1.) By 1985, steel production is expected
to approach 200 million tons/yr, and energy consumption is expected to
reach 4100 trillion Btu/yr. This is a 65% increase in production and a
28% increase in energy consumption between 1971 and 1985.
The most significant changes in energy consumption by this industrial
sector will occur in the ironmaking and steelmaking processes. In iron-
making, the most significant factor will be the increase in coke rates to
the blast furnace if the supply of hydrocarbon fuels used for injection does
not diminish. By 1985, coke rates are projected to be approximately
0.5 ton/ton of pig iron produced, a reduction of about 20% from current
practice. However, such a reduction in coke rates will produce a
IX-1
-------
Table 2. CONSUMPTION OF .PRIMARY FUELS BY THE
IRON AND STEEL, INDUSTRY, 1967-1971
Purpose
1967
1968
1969
1970
1971
12
: ir 1 nil i r\ r>4...
Blast Furnace*
Steelrnaking
Heating /Annealing
Other
Total
Blast Furnace*
Steelrnaking
Heating /Annealing
Other
Total
Blast Furnace*
Steelrnaking
Heating /Annealing
Other
Total
Blast Furnace*
Steelmaking
Heating /Annealing
Other
Total
*
Includes coke-o-t
12.16 14.44 20.03
84.97 82.26
60.39 55.12
30.19 40.33
187.71 192.15
51.78 54.35
88.95 105.75
287.07 311.30
125.45 139.09
553.25 607.49
/*** i i /\^2
2166.82 2128.76
180.23 177.52
2347.05 2306.28
2230.76 2197.55
173.92 188.01
347.46 366.42
335.87 353.94
3088.01 3105.92
/en underfiring.
75.54
47.01
42.72
185.30
12
1 0 Rtn
53.42
95.70
349.17
158.55
656.84
Pfrll
2202.63
154.41
2357.04
tels, 1012 Btu
2276.08
171.24
396.18
355.68
3199.18
24.82
62.39
56.52
41.19
184.92
56.66
59.83
316.86
181.04
614.39
2285.01
132.83
2417.84
2366.49
122.22
373.38
355.06
3217.15
30.99
44.69
51.51
39.26
166.45
52.01
48.15
340.75
183.81
624.72
1960. 39f
121. 24*
2081.63
2043.39
92.84
392.26
344.31
2872.80
Includes steam generation.
IX-2
-------
BASIC STEEL PLANT
INPUTS:
COAL
NATURAL
GAS
ELECTRIC
ARC
FUEL OIL
NATURAL
GAS
ELECTRICITY
SOAKING
PITS AND
REHEATING
FURNACES
COKE OVENS
BLAST FURNACE
BLAST FURNACE
STOVES
COG = COKE-OVEN GAS
BFG=BLAST-FURNACE GAS
Figure 1. TYPICAL, ENERGY UTILIZATION PATTERN
FOR BASIC STEEL PLANT
A-64-924
deficit of in-plant-generated coke-oven gases currently used elsewhere in
the plant as a heating source. As a result of this deficit alone, the use
of purchased fuels will increase. In steelmaking, the most significant
factor in the reduction of total energy consumption will be the phasing
out of the open-hearth process and its replacement with the basic oxygen
process. Based on 1971 energy consumption data, this will result in a
4% reduction in energy consumption (approximately 124 trillion Btu annually).
However, this does not take into account the energy used to manufacture
the oxygen.
JSnergy Utilization .Background Information
Ironmaking Processes
Iron Ore Preparation
The first step in the manufacturing of steel is the refinement of
iron ore to pig iron, a process that is carried out primarily in a blast
furnace. (Direct reduction, in which iron ore is directly converted
IX-3
-------
Table 3. SECONDARY FUELS CONSUMED BY
THE IRON AND STEEL INDUSTRY
Purpose 1967 1968 1969 1970 1971
Tar and Pitch, 10n
Blast Furnaces* 3.35 3.66 6.94 6.80 2.75
Steelmaking 39-72 34.89 32.89 28.30 17.76
Heating/Annealing
Other 5.40 3.18 2.45 6.64 6.93
1 2
Blast Furnaces*
Steelmaking
Heating /Annealing
Other
'v_/na ri* f~ia o
J. \J J_> L U.'
153> 340
14,973
189, 014
94, 190
164,402
11, 028
181, 949
104,479
153,431
6, 546
161, 970
111, 079
Blast Furnaces* 255,679 270,610 290,086 269,698 244,773
Steelmaking
Heating/Annealing 15,475 15,717. 17,596 15,645 15,439
Other 121,290 136,774 144,236 144,328 126,956
«
Includes coke-oven underfiring.
to steel, is discussed later.) The iron ore (burden) is charged into the
blast furnace in one of three conventional forms: crushed and screened,
sintered, or pelletized. In terms of energy consumption by a blast fur-
nace, a pelletized charge is preferable because it requires less energy
than raw ore that has only been crushed and screened. The desirable
characteristics of a pelletized or sintered burden over a raw ore burden
are as follows:
The burden's porosity is increased for easier penetration off. the
reducing gases.
The more uniform particle size distribution of the burden results in
better gas flow patterns in the blast furnace.
Undesirable materials in the burden, such as iron ore dust and
gangue, are minimized.
The use of agglomerated materials also increases the production
rate of pig iron from the blast furnace while reducing the energy con-
sumption per ton of pig iron produced. 9 With agglomerated material,
the flow patterns of the blast through the burden are more uniform, and
the slag rate is reduced. Figure 2 shows that a 30-40% increase in
IX-4
-------
production rate occurs when the amount of agglomerates charged increases
from 1000 to 2900 Ib/net ton of pig iron.
2000
1900
1800
o 5 1700
I C-
o3 O 1600
PIG IRON
PRODUCTION
<
or
g
O
1500
1400
^ 1300
O.
1200
1100
1000
1000 1500 2000 2500 3000
AGGLOMERATES USED, Ib/net ton of pig iron
A-II3-I677
Figure 2. EFFECT OF BURDEN AGGLOMERATION ON
BLAST FURNACE PRODUCTION RATE AND COKE RATE
Over the same range, the coke rate decreases approximately 25-30%.
However, such a decrease in the coke rate will produce a similar de-
crease in the production of coke-oven gas. For example, from 1959 to 1971,
coke production decreased at the rate of approximately 27 Ib/ton of pig
iron/year. The manufacturing of 1 ton of coke produces approximately
14, 000 CF of coke-oven gas, or roughly 7 million Btu. Thus, because
of the reduced coke production, approximately 100, 000 Btu of coke-oven
gas/ton of pig iron is not produced. Furthermore, blast-furnace-gas
production is reduced at the rate of 70 CF, 66. 50 Btu/lb of coke,
and the calorific value of the gas that is produced is lower. For every
ton of coke produced, 20 million Btu of by-product gas is no longer
available. In 1971, 21.9 trillion Btu less by-product gas was produced
on this basis than in 1970. Consequently, this loss of in-plant energy
IX-5
-------
must be replaced by other purchased fuels, such as natural gas and fuel
oil, and this must be carefully considered in an energy projection.
Sintering
Sintering is a. process in which iron ore fines, mill scale, flue dust,
coke breeze, and water are mixed and processed through a traveling-
grate sintering furnace.16 The coke breeze supplies, most of the energy
needed, but 160,000 Btu of other energies (coke-oven gas, etc.) per ton
of sinter is required for ignition of the bed.1 In 1971, about 39.5 million
tons of material was sintered, and: the amount of energy consumed for
ignition alone was 6. 32 trillion Btu.
Sintering processes have been in use for nearly 30 years. However,
the production of sinter has been decreasing steadily during the last 8
years (Figure 3).
O PELLETIZING
A SINTERING
1946 '48 '50 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 1972
YEAR
A-II3-I680
Figure 3. ANNUAL CONSUMPTION OF AGGLOMERATED
MATERIALS IN U.S. BLAST FURNACES.
IX-6
-------
Sintering plants usually consume energy in the form of blast-furnace,
coke-oven, and natural gas for ignition and coke breeze for primary heat.
Because of their ready availability within the blast-furnace plant, blast-
furnace gas and coke-oven gas traditionally have supplied most of the
energy consumed for ignition. But within the last 3 or 4 years, natural
gas usage has increased such that it is being used for ignition in the
production of more than 60% of the sinter, partly because of the decreas-
ing availability of in-plant gases caused by the reduced coke production
in the blast furnaces.
Figure 3 shows that the use of sintering as a means of burden ag-
glomeration is decreasing in favor of pelletizing, a trend that is expected
to continue. The use of natural gas for ignition will continue to increase,
if available, although fuel oil can be used.
Pelletizing
Pelletizing is similar to sintering in that beneficiated fine iron ore
is formed into wet "green" balls, which are fired in a furnace at 2000°-
2400°F. Normal fuel requirements are approximately 600, 000 Btu/ton of
pellets produced.1 In 1971, 45 million tons of pellets were produced,
consuming a total of 27 trillion Btu. 2
The production of pellets for use in blast furnaces started in 1954.
By 1958, pellet production was about 10 million tons. Since then, the
use of pellets has increased rapidly, quadrupling in 10 years.2 Energy
consumption also has increased fourfold.
Approximately 75% of the fuel consumed in the production of pellets
is natural gas, and the remaining 25% is fuel oil.1 Coke-oven gas and
blast-furnace gas, although technically suitable for this process, are not
used because pelletizing usually is carried out at the iron ore mine or
on route to the mill. It is economically more attractive to ship pellets
to the mill rather than raw ore.
IX-7
-------
.The use of pellets is expected to continue to increase, primarily
because air pollution from pelletizing plants is less than that from a
sinte.r plant. As air pollution standards become higher, pelletizing might
grow at an even faster rate than in the past. A second reason for the
trend to more pelletizing is that the process overall consumes less energy
than the sintering process because it is significantly more efficient on
a thermal basis (75% , compared with 10% for sintering). The shift
from sintering to pelletizing also depends upon changes in coke-oven
operation, which have reduced the availability of by-product coke-oven
breeze as a fuel.
As the use of pellets increases, so too will total energy consumption
of purchased fuels (natural gas and fuel oil). The type of fuel used will
depend upon cost and availability and not on interchangeability because
both gas and oil can be used for firing the furnaces.
Blast-Furnace Operation :
After the iron ore has been suitably prepared (crushed or agglomer-
ated), it is charged into a blast furnace for chemical reduction to pig iron.
The blast furnace is the single largest consumer of energy in steelmaking,
consuming approximately 42% of the total energy used by the steel industry.
In the blast-furnace process, iron-bearing materials (including sinter,
pellets, raw iron ore, mill scale, slag, and iron or steel scrap) are
charged into the top of the furnace together with fuel (coke) and fluxes.
Heated air (blast) is blown into the bottom of the furnace. In some cases,
auxiliary fuels, such as gas, oil, or powdered coal, are also blown in.
The blast air burns with part of the coke and also supplies sensible heat
to produce the total heat required for the chemical reactions and for
melting the iron ore. The balance of the fuel and part of the off-gas from
the combustion remove the oxygen combined with the metal through chem-
ical reaction. Typically, an input to the furnace of 1.6 tons of iron-
bearing materials, 0. 65 ton of coke, 0. 25 ton of limestone or dolomite,
and 1. 8-2. 0 tons of air produces an output of 1. 0 ton of iron, 0. 2 to 0.4
ton of slag, 0.05 ton of flue dust, and 2.5-3.5 tons of blast-furnace gas.16
These quantities change slightly, depending upon the type of burden used.
The molten iron is removed at temperatures around 2700°F and trans-
ported directly to the steelmaking process or cast into "pigs. "
IX-8
-------
The production of pig iron has been increasing in proportion to the
demand for steel, as expected. Figure 4 shows the annual production
of pig iron in the U.S. from 1947 to 1971.2
CO
LJ
o
oc
ID
Li.
I-
CO
QD
fe
CC
LU
DQ
1946 '47 '49 '51 '53 '55 57 '59 '61 '63 '65 '67 '69 1971
YEAR
A-II3-I668
Figure 4. ANNUAL BLAST-FURNACE PRODUCTION
IN TERMS OF BLAST FURNACES IN OPERATION
Only since 1969 has pig iron production decreased, but this is due to the
decrease in the demand for steel. Figure 4 also shows the number of
blast furnaces in operation in January of each year. These data indicate
that individual blast-furnace capacities have increased by about 65% since
1947.
IX-9
-------
-'The primary source of energy in blast-furnace operation is coke.
(The coking process is discussed later.) Coke has two functions in the
production of pig iron: First, it produces the heat required for smelting,
and second, it supplies the chemical reactants (primarily carbon monoxide)
required for reducing the iron ore. The carbon monoxide is formed by
reaction of the oxygen in the hot blast with the carbon in the coke. How-
ever, the coke also supplies the carbon that dissolves into the hot metal
(about 80 Ib/ton of pig iron). Approximately 80% of the reduction in the
furnace is due to the carbon monoxide; the remaining 20% is accounted
for by the strong reducing capabilities of the incandescent carbon in the
coke.16
During the last 10 years, the amount of coke required to produce 1
ton of iron has been steadily decreasing (Figure 5). In 1961, 1440 pounds
(0. 72 ton) of coke were required to produce 1 ton of pig iron. By 1971,
coke consumption had decreased to 1Z50 pounds (0. 625 ton) per ton of
pig iron, approximately 14%. 2 Among the reasons for this decrease in
consumption are increased blast temperatures, hydrocarbon fuel injection,
and improvements in burdens. The single most important factor is prob-
ably the rapid conversion to pelletizing throughout the industry. Figure 5
shows that coke consumption decreased rapidly from 1959 to 1966 and
then stabilized after 1966. Figure 3 shows that the use of pellets in-
creased significantly from i960 to 1966, correlating well with the reduc-
tion in coke consumption.
During the past few years, the use of auxiliary fuels to produce
hydrogen, which reduces iron oxides faster than carbon monoxide, also
has become widespread in the industry. Between 1967 and 1971, the
amount of fuel oil used in blast furnaces increased threefold from 67 to
194 million gallons.2 During this same period, the amount of natural
gas injected remained constant, even though it is by far the easiest fuel
to inject.
The primary reason for injecting auxiliary fuels (fuel oil, natural
gas, coke-oven gas, and tar) into the blast furnace is to reduce the coke
rate. On an economic basis, coke consumed in the blast furnace is the
largest single item of conversion cost in the steel industry. In general,
the deciding factor on which fuel is used for injection is economics.15
IX-10
-------
1950
1946 48 '50 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 1972
A-113-1647
Figure 5. AVERAGE BLAST FURNACE COKE RATE
IX-11
-------
The fuel that is least expensive in the region in which the steel plant is
located is the fuel used. In terms of operating results with fuel injectants,
there is little difference between fuels other than in the blast tempera-
ture required to maintain operation.
The energy equivalent of hydrocarbons typically injected ranges from
0. 5 to 2. 5 million Btu/ton of iron; the calculated average is about 1. 5
million Btu/ton of iron.1 Based on the 1971 statistics of hydrocarbon
fuels injected into blast furnaces, approximately 58% of the iron produced
in blast furnaces was accomplished by using hydrocarbon injection, using
approximately 70 trillion Btu. The potential for consumption of energy
by hydrocarbon injection was 120 trillion Btu in 1971.
Production output from a blast furnace also is directly related to the
wind rate, that is, the rate (CF/min) at which the blast is blown into the
furnace. Higher wind rates produce higher production rates. Injection
of oxygen into the blast serves the same purpose as increasing the wind
rate and consequently is referred to as "equivalent wind. " Oxygen injec-
tion has no significant effect on coke rate, yet 6% oxygen added to the
blast can increase productivity by 30% and more (Figure 6).14'25 However,
oxygen injection is not widely used because of problems with temperature
control and generally questionable economics.
By 1985, iron production for use in steelmaking is projected to be
about 130 million tons, based on a 200 million ton steel production, of
which 35% is scrap. This production level is approximately a 65% in-
crease over that in 1971. Assuming that energy consumption per ton of
iron produced remains constant, total energy consumption by this process
will increase by 65%.
Blast-Furnace Stoves
Another energy-consuming process in the production of iron is the
heating of the blast air in blast-furnace stoves prior to blowing it into
the blast furnace. Prior to 1950, blast-air temperatures were usually
below 1200°F, but with the realization that higher blast-air temperatures
increase production rates, blast-air temperatures of 1400°-2000°F now are
more common. In general, the fuels used to heat the blast air are
blast-furnace gas and coke-oven gas, primarily because of their ready
IX-12
-------
40
20
ro
_Q
0
2 34 5
OXYGEN ADDED TO BLAST, %
Figure 6. EFFECT OF OXYGEN INJECTION TO
BLAST ON BLAST FURNACE PRODUCTIVITY
availability. The amount of energy consumed by a blast stove depends
upon its efficiency and the temperature to which the blast is being heated.
Typically, the efficiency of a blast-furnace stove is about 75%. However,
as the desired blast temperature increases, the efficiency decreases
because of increased heat losses in the form of higher flue gas temper-
atures. If the efficiency is assumed to be 75% and the desired blast
temperature 1300°F, the amount of energy required to heat the blast is
> 1
2.25 million Btu/ton of iron produced. In 1971, blast furnace stoves
consumed 180 trillion Btu of energy. Based on average operating data
from all U.S. furnaces using higher blast temperatures and other oper-
ating improvements, for an increase in blast temperatures from 1250° to
1550°F, coke rates decreased from 1600 Ib/ton of pig iron to 1250 lb/
ton of pig iron, while pig iron production increased from 1050 to 1600
tons/day. The coke replacement ratio has been determined to be 20
pounds of carbon for every 100°F increase in blast temperature.
IX-13
-------
Coke Ovens
The manufacturing of metallurgical coke is the primary consumer of
coal in the steel industry. Coke is produced by the destructive distilla-
tion of coal. Two processes for coking are known: the beehive process
and the by-product process. However, the beehive process has been
phased out of production in the U. S. , mainly because of the stringent
air pollution legislation in effect and because of a lack of coals suitable
for use in the process.16
The by-product process uses the combustion of gases external to the
coke ovens to supply the heat required for distillation. The primary
gases used for this process are coke-oven gas and blast-furnace gas.
When higher production rates are desired, small amounts of high-Btu
gas (such as natural gas) also may be burned. Typically, energy con-
sumption for this process is about 2.0 million Btu/ton of coke.1 This
does^ not include the coal itself. As shown in Table 4, during the years
prior to 1971, coal consumption remained fairly constant, but in 1971
there was a sharp drop.
Table 4. HISTORICAL COAL CONSUMPTION FOR
THE IRON AND STEEL INDUSTRY2
Year
196Z
1963
1964
1965
1966
1967
1968
1969
1970
1971
This drop presumably is due to the reduction in steel production during
that year. Coke consumption and hence coal usage are expected to de-
crease on a unit-of-production basis, but total coke consumption is ex-
pected to increase with increases in steel production. Based on a 1985
projection for annual pig iron production of 130 million tons, coke
IX-14
Coke
1729.2
1779.8
2062.0
2199.1
2240.8
2166.8
2128.8
2202.6
2285.0
1960.4
Steam
1012 Etu
169.2
162.1
165.8
175.2
173.8
159.8
160.0
140.5
124.5
114.2
Other
14.0
18.3
18.7
17.9
19.9
20.5
17.5
13.9
8.3
7.1
-------
consumption will be about 81 million tons based on a current coke rate
of 0. 625 ton of coke/ton of iron produced. Energy consumed for external
heat in the production of this amount of coke (neglecting the coal) will
be 160 trillion Btu.
Steelmaking Processes
The three basic steelmaking processes in use in the U.S. today are
1) the open-hearth process, 2) the basic oxygen process, and 3) the elec-
tric-arc process. In all these processes, molten pig iron, steel scrap,
and fluxes are charged into the steelmaking furnace, where they are heated
to about 2950°F. Controlled amounts of oxygen are supplied to the molten
metal undergoing refining to form, oxides with the impurities, which, in
turn, either leave the bath as gaseous metallic oxides or form a slag.
Composition of the slag is important in steelmaking; consequently, the
energy involved in producing the slag also is considered to be a require-
ment of the steelmaking process. In addition, sufficient heat must be
supplied to melt the scrap and superheat the molten bath to the final de-
sired temperature. After completion of the steelmaking process, the
molten metal is tapped from the furnace and either transferred directly
to continuous casting machines or formed into ingots for use later in
rolling mills, bloom mills, slabbing mills, and billet mills.
The Open-Hearth Steelmaking Process
The open-hearth furnace is essentially a regenerative, reverberatory
furnace in which a long, relatively shallow bath is heated by radiation
from a luminous flame. Burners and air ports are located on both sides
of the furnace. Fuel is fed into the furnace through burners on one side
of the furnace. Air for combustion is fed into the furnace through regen-
erators, where it is preheated. After combustion is completed, the flue
products leave the furnace through the air port temporarily not in use
for entrance of fuel and air and pass through the checkers in the regen-
erators, heating them as they leave. After 15 or 20 minutes, the fuel
is shut off and the air flow reversed; the port previously used for an
exit becomes an entrance, and the entrance becomes the exit port. The
fuel then is turned on again, and this time it is introduced on the opposite
side of the furnace from which it started. These reversals, as they are
called, occur every 15 or 20 minutes until the desired product is reached
and tapped from the furnace.
IX-15
-------
For the last 50 years, the open-hearth furnace has been the workhorse
of the steel industry, accounting for 90% of the total steel produced.
However, after i960, the amount of steel produced by the open hearths
declined in favor of the basic oxygen furnace (BOF) and the electric fur-
nace (Figure 7).
140
130
120
110
1959 60 61 '62 '63 '64 '65 '66 '67 '68 '69 '70 1971
O OPEN HEARTH
A BASIC OXYGEN
D ELECTRIC
O TOTAL STEEL PRODUCED
A-M3-I68I
Figure 7. PRODUCTION OF STEEL BY THE OPEN-HEARTH, BOF,
AND ELECTRIC-ARC PROCESSES BETWEEN 1959 AND 1971
IX-16
-------
Today, the open-hearth process produces less than 30% of the annual
total steel production. As would be expected, energy consumption by
open hearths has decreased proportionally. Energy consumption per ton
of steel by open hearths also has decreased because of improvements in
furnaces. In i960, open hearths consumed approximately 4. 0 million
Btu of energy per ton of steel produced. By 1971, energy consumption
had decreased to approximately 3. 5 million Btu/ton of steel produced.
In i960, open-hearth furnaces consumed a total of 344 trillion Btu
of energy. By 1971, the total energy consumption by open hearths de-
clined to 122 trillion Btu. The fuels used in open-hearth steelmaking
furnaces are of two types, usually called internal and external. The
internal fuels are the elements in the melt that are oxidizable. These
elements are introduced primarily by the hot metal portion of the charge.
The external fuels are the fuels, gaseous and liquid, that are burned in
the space over the charge and bath.
External fuels used in the open-hearth furnace process depend on
required flame characteristics for the production rate desired and econ-
omics. Open hearths can be fired with 100% liquid fuels or with liquid-
gas fuel combinations because gaseous fuel flames do not have the nec-
essary luminosity when used alone. Thus, they normally are burned with
the addition of liquid fuels to develop luminosity and increase heat transfer
to the bath. The following discussion of the fuels presents other factors
(that relate to their selection in this process.
The primary factors' affecting the use of natural gas in open hearths
'are its availability and cost. Although natural gas is not used by itself,
;.i it is used to atomize liquid fuels, thus resulting in a shorter, more
luminous, hotter flame that increases heat transfer and productivity.
1 Natural gas is an ideal fuel from metallurgical and air pollution stand-
"j points because it contains virtually no sulfur. When natural gas is avail-
able and economically justified, it is usually fired with about 20% liquid
fuel to increase its luminosity. Finally, natural gas can be used in gas-
oxygen burners mounted in the furnace roof to assist in the heating of
scrap.
IX-17
-------
Some furnaces burn only fuel oil. Fuel oil combustion produces a
highly luminous flame that has the required high heat transfer rate to
the charge. Heavy fuel oils must be heated to lower the viscosity and
then are atomized with natural gas, steam, or compressed air. Fuel
oils with 1% or less sulfur content must be used in this application. Of
all the possible liquid and gaseous fuel combinations, fuel oil plus natural
gas is the most popular.
Tar and pitch flames offer the same general characteristics as fuel
oil. One difference is that tar and pitch flames are more luminous, but
they are more difficult to burn in a controlled manner. However, more
coke tar and pitch are being used in this application because of the
decrease in sales demand for this coke-oven by-product. In some cases,
coke-oven pitch is blended with about 25% petroleum tar to lower the
viscosity of the pitch.
Coke-oven gas is used most often with liquid fuels or natural gas
and liquid fuels. However, the maintenance problems with coke-oven
gas are significantly greater than those with natural gas because of the
presence of trace amounts of corrosive materials in the gas.
Factors Affecting Energy Consumption
in Open-Hearth Furnaces
Three primary factors affect the energy consumption of open-hearth
furnaces: 1) the percentage of scrap steel charged into the furnace,
2) the combustion air preheat temperature, and 3) the oxygen used to
enrich the combustion air, used.to increase the oxidation reactions by
injection through the slag into the hot metal, and used in gas-oxygen
burners.
Scrap. The use of scrap in the open-hearth furnace process is
primarily one of economics. The economics vary, depending on location
of the plant, proximity to adequate raw materials, the amount of molten
pig iron available, and the scrap available from integrated units. Energy
is required to melt down the scrap; thus, increases in the amount of
scrap charged increase the energy consumption. For example, in 1968
the average percentage of scrap charged to open-hearth furnaces increased
3%, from 41% to 44% of the total charge. During this same period, the
average energy consumption increased by approximately 8%, from 3. 25
to 3. 55 million Btu/ton of steel produced.
IX-18
-------
Preheated Air. The use of preheated air for combustion reduces the
energy consumption in an open-hearth furnace by decreasing the amount
of heat required to raise the air temperature up to furnace temperature.
With increases in preheat air temperatures, the heat content of the com-
bustion products increases and likewise the available heat increases
(Figures 8 and 9).13
HEAT CONTENT
' FLUE GASES
^BLE ABOVE 2900 °F
= M W * CJ>
-> o o o o
l -
^^
-^-"
^- -
--
- '
_-
,
-
u- «*
$S°d
AVAI
100 300 500 700 900 1100 1300 1500 1700 1900 2100 2300 2500
PREHEATED AIR TEMPERATURE, °F
A-II3-I687
Figure 8. EFFECT OF PREHEATING COMBUSTION
AIR ON HEAT AVAILABILITY
1.8
£8 I"
? £ 00
°
1.4
uj Q
3-ai.Z
CD
1.0
A-II3-I688
Q 100 300 500 700 900 1100 1300 1500 1700 1900 2100 23002500
PREHEATED AIR TEMPERATURE, °F
Figure 9- EFFECT OF PREHEATING COMBUSTION
AIR ON HEAT CONTENT OF FLUE PRODUCTS
Oxygen. The most significant effect of the use of oxygen in open-
hearth furnaces is that from the injection of the oxygen through the slag
layer into the molten metal. Prior to oxygen injection, oxygen from the
air would diffuse through the slag layer covering the melt, causing de-
carburization by oxidation, an exothermic reaction. The speed of de-
carburization (about 0. 005% carbon per minute) was limited by the dif-
fusion rate. By injecting the oxygen through the slag, this limitation
IX-19
-------
was lifted and refining times were reduced drastically. With oxygen,
productivity increases, steel quality improves, and energy consumption
decreases.7
The amount of the reduction of energy consumption by using oxygen
injection depends, to a large extent, on the percentage of hot metal
charged into the furnace because of the "internal" fuel and sensible heat
contained in the hot metal. Typically, fuel consumption per ton of steel
can be reduced by 25-50% over a range of 30-60% hot metal in the charge.
The use of oxygen-enriched combustion air and gas-oxygen burners
generally increases flame temperature and available heat, thus improving
the melting rate. Of these two methods, gas-oxygen burners are more
popular. However, both methods increase productivity and decrease
energy consumption per ton of steel produced. Both methods also are
used when a reduction in flue products volume is required because of
regenerator capacities.
Other Factors. Several other factors in open-hearth furnaces can
affect the economy of the operation, especially fuel consumption. Most
of these factors relate to furnace design and maintenance, proper flame
control, charge composition, type and grade of steel produced, and types
of fuel and air controls and instrumentation. . The importance of these
factors is not dealt with here, as reason alone dictates the overall re-
sultant effects if these factors are ignored.
There are other uses of energy in open-hearth furnace processes,
primarily electricity for mechanical drives. A significant amount of
electricity also is used in the production of oxygen. These energy uses
are discussed in a later section.
The Basic Oxygen Furnace (BOF) Process
In the BOF steelmaking process, pure oxygen is injected through
water-cooled lances at high velocities into a bath of molten metal from
the blast furnace contained in an upright open-top vessel. A large amount
of heat is generated by the oxidation of carbon, silicon, manganese, phos-
phorus, and iron in the hot metal, and violent agitation of the bath occurs.
The amount of heat generated by these reactions is generally more than
is required, so scrap usually is added to the charge to absorb the excess
IX-20
-------
heat and lower the operating temperature. Virtually no auxiliary fuel
is required in the BOF process; all the necessary heat is supplied by
the fuel value of certain elements in the hot metal.16
During the past 15 years, the BOF process has experienced a sig-
nificant growth in usage. Figure 7 shows that from 1959 to 1971, BOF
production increased from less than 5% to more than 50% of the total
annual steel production in the U. S. 2 This growth is expected to continue
as more open-hearth furnaces are phased out of operation and as the
demand for steel increases.
Several reasons for the popularity of the BOF process exist. First
is the high speed of operation. Typically, the time required to complete
a cycle, or heat, is 30-40 minutes, and as a result, production capabilities
are high. Second, no auxiliary fuel is required for operation. However,
small amounts of fuel (about 200, 000 Btu/ton of steel) are used to main-
tain the furnace temperature between heats and to dry and "burn in" new
refractory linings.
Oxygen Utilization in the BOF and Energy
Requirements for Production
The amount of oxygen used per ton of steel produced in a BOF
depends on several factors:
1. The composition of the cold charge if cold pig iron and cast iron
scrap are charged together with steel scrap
2. The percentage of hot metal in the charge
3. The composition of the hot metal
4. The required carbon content in the steel at the end of the heat
5. The degree to which the process is under control in terms of iron
oxidation and other factors.
The average historical consumption of oxygen per ton of BOF steel
produced annually is presented in Figure 10. These data show that oxygen
consumption is slowly decreasing; it should continue to decrease as more
experience is obtained and better process control achieved.
IX-21
-------
o
I
\
o
I
2000
1900
1800
LU CJ
CO CO
>-
X
o
1700
1962 '63
'70 1971
A-113-1685
Figure 10. AVERAGE OXYGEN CONSUMPTION BY BOF's
At present, the most economical way to produce large volumes of
high-purity (99. 5% Oz) oxygen is by liquefaction of air in a compression
cooling-expansion cycle, followed by fractional distillation of the liquid
air to obtain gaseous oxygen. The production of oxygen in a large plant
requires approximately 280 kWhr of electricity per ton of oxygen pro-
duced at atmospheric pressure. To this figure must be added the energy
required (60 kWhr/ton) to compress the oxygen to steel plant requirements,
about 250 psig. If the final product is to be liquid oxygen, the total
energy required will be about 750 kWhr/ton of oxygen. Assuming the use
of gaseous oxygen at an average rate of 1900 SCF of oxygen per ton of
BOF steel, approximately 26.0 kWhr, or 90,000 Btu, of electricity are
consumed to produce the required oxygen. For those situations in which
storage of the oxygen is desired, the energy consumption increases by
nearly 100%.
Note that most steel companies purchase their oxygen from vendors
who pipe the oxygen to the steel plant from the oxygen plant. Thus,
technically the energy consumed at the oxygen plant to produce the oxygen
is not a direct factor in the energy consumption pattern of the steel in-
dustry. It is included here because of the importance of oxygen in the
steel industry and because it is a substitute for fuels that would otherwise
be used.
IX-,22
-------
BOF Efficiency
Based on the theoretical heat available in the BOF process and as-
suming no waste-heat boilers or carbon monoxide gas collection, the
thermal efficiency of a BOF is roughly 65%. Meaningful comparisons
with open-hearth processes are not possible because of the different
compositions of the charges. Typically, a BOF operates over the range
of 70-85% hot metal in the charge, whereas open hearths operate in the
30-65% range. Improvement of the efficiency of the BOF process can
be achieved by proper heat recovery treatment of the off-gases and sub-
sequent use of this heat in boilers. The overall effect, while possibly
increasing the efficiency to perhaps 75%, would be relatively small, on
the order of 150, 000-250, 000 Btu/ton of steel produced.22
A significant amount of work is being done to increase the scrap
utilization percentage of BOF's. This includes the preheating of the
scrap external to the BOF. Serious technical and engineering problems
still exist, but should such a process bedome feasible, energy consump-
tion would decrease because of the increased productivity that would
result from a shortened oxygen-refining cycle.
Electric-Arc Furnaces
The third process for producing steel is by means of an electric
steelmaking furnace. In this process, scrap iron and steel are melted
and superheated to the refining temperature by using three-phase electric-
arc power and graphite electrodes. Note that the electric steelmaking
process is not competitive with either the open-hearth or the BOF process
in integrated steel mills where molten pig iron is available. But electric-
arc furnaces are used in "cold-charge" nonintegrated steel plants when
additional melting capacity is required.
During the meltdown of the scrap charge, oxidation reactions begin
when a pool of molten metal is formed by using oxygen from rust and
scale in the charge and from the air. Iron ore also is used as a source
of oxygen if it is charged with the scrap. The refining stage is an
intensified oxidation period during which carbon, manganese, phosphorous,
iron, and aluminum are oxidized.4'5
IX-23
-------
Electric-arc production of steel has been in use in the U.S. since
1906. Since then, steel production in electric-arc furnaces has grown at
a slow but continuous pace, with rapid expansion beginning after World
War II. The annual production of more than 20 million tons in 1971 is
more than 3 times the annual production in 1948. 2 Still it represents only
a small portion of the total steel production in the U.S.
The major advantage of the electric-arc furnace over the other steel-
making processes is its universal applicability as a melting device in
which the entire spectrum of modern steelmaking can be controlled inde-
pendently of the charge materials. The major drawback to this process
is centered around economics, which are , not truly competitive as yet
with other steelmaking processes.
Energy used in this process is primarily electrical; electric discharges
or arcs between graphite electrodes produce extremely high local temper-
atures for melting of the metal. Other heat inputs include heat from slow
oxidation of the electrodes, radiant heat from the resistance-heated elec-
trodes, and exothermic reaction if oxygen is injected.
Typically an electric-arc furnace consumes about 500 kWhr of elec-
tricity per ton of steel produced* Thus, in 1971, 10 billion kWhr, or
35 trillion Btu, of energy was consumed by electric-arc furnaces.*
Effect of Product Type
The electric steelmaking process produces three types of steel:
carbon, alloy, and stainless steels. Figures on total electric consump-
tion by electric-arc furnaces and by type of steel are not available.
Estimates of energy consumption by type of steel produced are that car-
bon steels require an average of 490 kWhr/ton of steel produced and
alloy and stainless steels require about 540 kWhr/ton of steel. Thus, the
production of alloy and stainless steels requires about 10% more elec-
tricity than that of carbon steels.
Power Input
In electric steelmaking, two measures of efficiency exist: 1) the
amount of steel produced per hour from a furnace as compared with
*
Excludes energy consumed for electricity generation.
IX-24
-------
that from other furnaces of the same size, and 2) the electricity con-
sumed per ton of steel produced. Energy input rate and heat time affect
both of these measurements. However, power inputs do not increase
proportionally to the increases in furnace volume. Consequently, ultra-
high power, 300-450 kW/ton of capacity, furnaces are becoming more widely
used. 23 The advantages of ultrahigh power furnaces are
Decrease in melting time, compared with that in conventional
furnaces (120-190 kW/ton of capacity)
Increased productivity
Decrease in energy requirements (kWhr/ton of steel)
Decrease in electrode costs.
No decreases in refining time can be achieved by using ultrahigh power
in furnaces. In addition, the unit cost of electric energy in terms of
demand charges increases, unless steps are taken to prevent such an
occurrence by minimizing the low-power and off-power periods. As long
as productivity is increased with increased power, this process is
economical.
Scrap Preheating
The reasons for preheating the scrap charged to an electric furnace
are obvious. Temperature is directly related to the rate at which heat
can be transferred to a load. Considering that electricity is more ex-
pensive than fossil fuels, economics dictate that the scrap should be
preheated by using cheaper fossil fuels prior to charging and that the
electricity then be used for melting and superheating the steel to the re-
quired temperature.
Work on scrap preheating indicates that significant decreases in
electric energy and subsequently in electric consumption can be achieved.
Preheating the scrap to 930°F with a base energy usage of 555 kWhr/ton
resulted in an 18% decrease in electric energy consumption, a 19.8%
increase in productivity, and a 23. 9% decrease in electrode consumption.
Indications are that increasing the preheat temperature to 1800°F would
reduce the electric energy consumption by 45%. Even so, energy then
will be required to preheat the scrap, and the total net decrease in energy
consumption will be somewhat less.20
1X^25
-------
Although scrap preheating has numerous advantages, several technical
problems exist; consequently, the practice is in very limited use. The
problems involve control of air pollution, control of oxidation of the scrap,
and handling of the material. Furthermore, the economics of the process
depend, to a large extent, on the cost of the scrap, which generally is
quite high. Under these circumstances, the use of prereduced iron ore
becomes attractive.
Prereduced Iron Ore Charges
The use of prereduced iron ore in the electric steelmaking process
is increasing because of decreases in scrap availability and increases in
scrap prices. Furthermore, in some cases, the use of prereduced ore
in an electric furnace is slightly more economical than the conventional
blast furnace BOF or open-hearth ~» process.
In considering energy consumption by an electric furnace using pre-
reduced ore, the energy requirements for the direct reduction process
also must be considered. Comparisons between direct reduction and blast-
furnace processes are difficult because the forms of the products are dif-
ferent, the prereduced ore being solid state and at room temperature,
whereas the blast-furnace product is molten and at Z900°F. In addition,
energy consumption by direct reduction processes depends on which process
is used, with a range of 13,-ZO million Btu/ton. of metal produced. A
good average consumption figure is 15 million Btu/ton. 6 In terms of
energy consumed by direct reduction, the important fact is that the process
does not require the use of coking coals. When carbon is the main re-
i
ducing agent, it usually is supplied as coal or a low-temperature char
made from coal. Some processes use carbon monoxide or hydrogen as
reducing agents, but the more common practice is to use both. In this
case, natural gas is reformed to produce both carbon monoxide and hy-
drogen. Apparently, the demand for high-quality metallurgical coking
coal will be lower, and the demand for lower grade coals, natural gas,
and petroleum will increase.
IX-2 6
-------
The use of prereduced ore in an electric-arc furnace results in a
slight increase in energy consumption and in melt-in time when charged
in a batch operation. However, if a continuous feed system is used for
charging the scrap and ore, energy consumption is still higher than if
only scrap were charged. Nevertheless, the total time required for the
process is decreased and the process becomes more economical. Because
of the relative newness of this process, reliable information on energy
consumption is not yet available. Experimental work with this process
showed that the ideal percentage of prereduced ore charged was 25%
(percentages higher than this resulting in a decrease in productivity) and
that energy consumption was about 560 kWhr/ton, or about 10% higher
than without prereduced ore.
Steel-Finishing Processes
At the end of the steelmaking processes, the molten steel is tapped
from the furnace and poured into ingots. The ingots are either stored
or sent directly to the soaking pits, where they are heated to sufficient
temperature to be formed into slabs, billets, or blooms, which are then
transferred to the reheat furnaces. Upon completion of reheating, the
steel goes directly to the mills, where it is worked and shaped into the
final physical forms. Some of these products than are heat-treated and
finished.
Soaking Pit Furnaces
Soaking pits are usually deep, square, or rectangular furnaces that
have a retractable cover to permit charging and discharging from the top.
A few circular pits are bottom or tangentially fired. The pits are usually
built in batteries (side by side with a common wall between them) and
thus are capable of sharing common air blowers, air supply ducts, waste-
gas ducts, and stacks. Soaking pits are generally characterized by the
type of firing system, such as one-way-fired, recuperative pit, top two-
way-fired regenerative, bottom two-way-fired, and circular tangentially
fired. Fuels used include coke-oven gas, blast-furnace gas, natural gas,
and gas mixtures. Residual oil primarily is used as a standby fuel.
The heat requirements of soaking pits vary, depending upon the initial
temperature of the ingots being heated. For cold ingots, heating require-
ments are typically 800, 000-1, 000, 000 Btu/ton of raw steel; for ingots at
1700°F, the heat requirement is about 400, 000-500, 000 Btu/ton of raw steel.
IXr-27
-------
Three main factors influence the heat requirements in soaking pits:
1) the type of steel being produced, Z) the size of the ingots, and 3) the
track time, that is, time from finish of pouring of ingot to charging into
soaking pit. The operator in the steel mill has no control over the first
two factors. The type of steel being processed affects track time. Some
steel ingots must be completely solidified before they can be charged,
whereas others can be charged with molten cores. The size of the ingots
depends upon the product mix of the particular plant. Track time affects
the heating requirements simply because the more heat lost before charg-
ing, the more heat will be required to bring the ingot back to temperature.
Figure 11 shows the effect of ingot-charging temperature and percentage of
cold steel charged on net Btu/ton heating to 2400°F.
600,000
500,000
400,000
CD
I-
UJ
300,000
200,000
PERCENT COLD STEEL
COLD STEEL TO 2400 "F 765,000 Btu PER TON
800
900
1000
1100
1200
1300
1400
1500
TEMPERATURE OF CHARGED
INGOTS TO SOAKING PIT
Figure 11. SOAKING PIT. EFFECT OF INGOT-
CHARGING TEMPERATURE AND PERCENTAGE OF COLD
STEEL CHARGED ON NET Btu/TON HEATING TO 2400°F
IX-28
-------
Soaking pit efficiencies are generally low, around 50%, primarily
because of the sensible heat lost in the flue gases. Preheating of the
combustion air by recuperators or regenerators is mandatory if fuel
requirements are to be reduced, but these systems are generally not
economical.
Continuous Casting and Pressure Pouring
Continuous casting and pressure pouring are two alternatives to ingot
pouring that are currently gaining widespread use. In these processes,
the molten steel from the steelmaking process is cast into slabs, billets,
or blooms, thus eliminating the pouring of ingots altogether. These
processes result in significant savings because the need for soaking pits
and rolling mills to roll the ingots to slabs is eliminated. Continuous
casting consumes a very small amount of energy (200, 000-300, 000 Btu/ton)
and, at the same time, results in a saving of close to 2 million Btu/ton
of slab. In addition, the yield of semifinished product is significantly
higher with continuous casting than via the ingot route.
Continuous casting was developed in the 1940's. By 1968, about 4.5
million tons of steel was continuously cast, approximately 3% of the total
annual steel production. By 1985, it is forecast that 40 million tons of
steel will be continuously cast, amounting to 20% of the expected total
annual production.
In continuous casting, the molten steel flows from a holding ladle
into a tundish, which serves as an intermediate reservoir. From there,
it passes into the molds, where it is partially solidified. The strand,
as it is called, is cooled to complete solidification; then it is cut into
sections by gas torches or hydraulic shears. The energy used is pri-
marily electric energy and natural gas for preheating the tundishes.
Pressure pouring is achieved by exerting air pressure on the surface
of the molten steel in an enclosed tank, forcing the molten steel up through
a tube into a graphite mold. Unlike continuous casting, which produces
a continuous strand that is then cut, pressure pouring produces individual
slabs. Energy requirements are about 60, 000 Btu/net ton of pressure-
poured steel. l8 Electric energy, used primarily for handling of the molds,
is estimated to be 5 kWhr/ton of steel cast. The remaining energy
IX-29
-------
requirement is 42, 000 Btu/net ton cast of natural gas for preheating the
pressure-pouring tubes. The energy saving by using pressure pouring is
nearly 2 million Btu/ton of steel, compared with ingot pouring and later
reheating.
In addition to the energy saving due to elimination of the soaking pits,
energy savings occur in the melting operation because continuous casting
produces a higher yield (95%) than the ingot route (85%). Consequently,
less steel is required to meet production demands via the continuous cast-
ing route. The difference is about 0. 128 net ton of steel. By assuming
an electric furnace for steel production rated at 525 kWhr/net ton,, the
energy saving would be 64.6 kWhr (0.123 X 525), or 220,415 Btu/net ton
of semifinished product.
Reheat Furnaces
Reheat furnaces, as discussed here, are those furnaces used for
heating the semifinished steel products in preparation for the final hot-
rolling operations. Reheat furnaces can be either continuous or batch
type, but most steel plants use continuous furnaces. Continuous furnaces
fall into two categories: pusher and conveyor furnaces. Walking beam
and roller hearth furnaces are two types of conveyor furnaces typically
used. Fuels used in reheat furnaces are fuel oil, coke-oven gas, and
natural gas; the last has become the predominant fuel in recent years.
Fuel consumption by reheat furnaces ranges from about 1.6 to 3. 0 million
Btu/net ton of steel. In 1971, reheating the billets, blooms, and slabs
consumed approximately 191 trillion Btu of energy.
The primary factors that affect the heat requirements for reheating
furnaces are 1) the type of furnace (batch or continuous), 2) the size of
the furnace, 3) the furnace design, 4) the temperature of the process,
and 5) the physical nature of the steel being reheated. For example, the
average fuel consumption for slab-reheating furnaces that burn oil or gas
fuels with preheated air is about 2. 2 million Btu/ton heated for three-
zone furnaces and 2. 8 million for higher capacity five-zone furnaces.
Thicker slabs require longer heating times and more energy. Fuel
economy for steel mill reheat furnaces also is affected by heat losses,
heat recovery, combustion and process controls, and operating practices.
Frequent rolling mill delays adversely affect fuel efficiency and temperature
IX-30
-------
uniformity of the product. Because of the large number of factors that
can affect fuel economy in reheat furnaces, for which specific information
is not available, a detailed fuel utilization analysis is not possible.
Effect of New Technology on Energy Consumption
Currently, four developments in technology are expected to have a
significant effect on the energy consumption of the ironmaking process;
1) the use of prereduced iron as part of the blast burden, 2) use of
formcoke, 3) injection of reformed gas, and 4) the use of nuclear reactors.
The effects of each of these are examined later in this section in several
hypothetical cases.
The single most important change in energy consumption in steelmaking
in the future will be the phasing'out of open-hearth furnaces in favor of
BOF's and electric-arc furnaces. On the basis of production statistics,2
124 trillion Btu of energy was consumed by this process in 1971. Energy
consumed by BOF's amounted to about 95 trillion Btu; 90% of this energy
is accounted for by the exothermic reactions of the process and the sensi-
ble heat of the final product, and the remaining 10% is consumed in the
production of oxygen. Electric-arc furnaces consumed about 35. 6
trillion Btu of energy. Phasing out of the open-hearth furnaces would
result in a 4% reduction in energy consumption by the industry, assuming
the BOF's produce the bulk of the steel.
Most of the "new" technology available for use on reheating furnaces
for both semifinished or final hot-rolled products includes such items as
improved burners for more uniform heating, improved combustion controls
such as programmed burner operation, oxygen enrichment of combustion,
improved skid rails, and improved monitoring devices. In such cases,
the effect of these technologies on the fuel utilization of reheating furnaces
cannot be determined. The only new developments that have shown an
ability to reduce fuel consumption are continuous casting and pressure
pouring, which have already been discussed.
The continued use of soaking pits by the steel industry is directly
tied to the continued use of slabbing and blooming mills. Thus, the degree
to which continuous casting is used (making absolute the slabbing and
blooming mills) is important. Several factors are involved. Continuous
casting is most suited for steel mills in which the product range is narrow.
IX-31
-------
Flexibility, product size, and quality limitations will slow down broad
application of this process. During the past few years, new soaking pits
and rolling mills have been built, indicating a reluctance on the part of
the steel industry to accept continuous casting. The large-tonnage continuous-
casting installations that have been built are for new capacity, not replace-
ment for existing capacity, so that the traditional type of equipment can
act as a backup and can supply a wider range of products than continuous
casting. Therefore, under present conditions the use of soaking pits is
not expected to decline and may even increase if the increase in steel
demand is greater than the increase in continuous casting. By 1985,
steel production is expected to be 200 million tons, 20% of it processed
through continuous casting.
Fuel-fired furnaces for reheating slabs, blooms, and billets are not
likely to change in the foreseeable future, partly because of the high
bapital investment in existing equipment. Furthermore, with the ability
to change fuels, a plant can take advantage of changes in cost and avail-
ability of different forms of energy. There is no new technology nor
are there other factors that are expected to affect fuel usage in reheating
furnaces. The long-range goal of the industry is to combine into one
step the continuous casting, the temperature equalization in a. continuous
soaking furnace, and the rolling operations. However, such a process
is still a long way off.
rJirect Reduction (Prereduced Iron Ore)
Direct reduction is a process in which iron ore is converted directly
to steel, thus bypassing the blast-furnace process altogether.16 Develop-
ment of a viable direct reduction process has been under investigation
for many years, but to date, no one has achieved total success. The
result has been the development of a myriad of proposals. Although
processes that are referred to as direct reduction do not produce steel
from iron ore, they do produce a material of which 85% of the contained
iron is in the metallic state. These materials are known variously as
reduced pellets, sponge iron, and metallized ore, or, for that matter,
prereduced ore. Among the several end uses of this material in the
IX-32
-------
steel plant are blast-furnace burden and electric-steelmaking-furnace
charge. For the present, only usage as a blast-furnace burden is dis-
cussed. (A complete discussion of direct reduction is presented later.)
Several trials with prereduced iron ore have been conducted on
experimental and industrial blast furnaces within recent years.3'17'19'21'30
The results have indicated that substantial increases in production rates
can be achieved and, at the same time, coke rates can be reduced.
Figure 12 shows that, as the percentage of prereduced ore in the burden
increases to 80%, production increases by 70%, or about 9.0% per 10%
metallic iron in the burden. This same figure also shows that for the
same percentages of prereduced ore, coke rates decrease by about 40%,
or 8.2% per 10% metallic iron in the burden. Apparently, as in other
methods of reducing coke rates, the amount of available energy in a steel
plant produced as off-gas is reduced and consequently must be replaced
with purchased fuels. However, the use of prereduced ore in the blast
furnace does increase the caloric value of the off-gas, in some cases as
much as 30% with an 80% metallized charge.
IJUU
1200
c 1100
o
w
a. 1000
o
£ 900
K 800
<
ac
% 700
O
0
600
5OO
OU
70
I
-$5 60
1
"t 50
O
1 40
z
-Uj 30
<
UJ
g 20
Z
10
o
\
V
/
\
^\
/
\
/
/
L>
/
>
PR
/
/
\
X
CO
DDUCTIVITY>
/
/
N
-------
Formcoke
Formcoke is an agglomerate, or briquette, prepared from bituminous
coals that do not meet the restrictive specifications of metallurgical cok-
ing coals. 3l The primary advantage of formcoke is that it would replace
the coke currently used, which is manufactured from metallurgical coking
coals that are in short supply. Although several trials have been con-
ducted, no successful process using formcoke has been developed. Interest
is high, however, because successful use of formcoke would result in a
lower cost for producing coke and, consequently, a saving in fuel costs.
Furthermore, use of formcoke would not upset the energy values recover-
able from the formcoke plant because the Btu value and gas volume pro-
duced by some of the formcoking operations are equivalent to those
obtainable from conventional coke ovens. However, if formcoke becomes
a successful substitute for coke in the blast furnace, the possibility exists
that additional energy would be required to meet the overall plant require-
ments, depending on the volume and heating value of the top gas. These
questions remain to be answered.
Reformed-Gas Injection
Reformed gas is a low-thermal-value gas that is made by the pyrolysis
and steam decomposition of a high-thermal-value fuel. Theoretical studies
have indicated that injection of 260 Ib (per net ton pig iron) of reformed
gas produced from fuel oil into the blast furnace, above the melting zone
but below the preparation zone, should result in a 38% increase in pro-
duction and a 28% decrease in coke rate. Investigations on an experi-
mental facility by the U.S. Bureau of Mines showed that production could
be increased by 40% and coke rates decreased by 29% by using 325 Ib
(per ton of pig iron) of reformed gas injected into the furnace.18 At pres-
ent, no commercial process for reformed-gas injection exists, but work
is continuing.
Nuclear Energy
Nuclear energy will be used in the steel industry to supply electric
energy when nuclear energy costs become competitive with costs of other
forms of electricity generation. Methods that have been proposed would
utilize nuclear reactors directly in combination with blast furnaces. One
IX-34
-------
proposed process would utilize gas-cooled nuclear reactors to preheat the
reducing gases to 2700°F prior to injection into the furnace. However,
such processes are still far in the future.
Air Pollution Emissions Background Information11'19'24'29
A significant factor in the future of the U. S. steel industry will be
its ability to solve the air pollution problems created by the various iron-
and steelmaking processes. Table 5 summarizes the air pollution emissions
from the various ironmaking processes. Emissions from the coke ovens
are considered to be the most serious problem facing the steel industry.
Of the two processes available for coking, the beehive process produces
excessive emissions, which are virtually uncontrollable because of the
equipment used in this process. The only means of control, therefore,
is a restriction on the location of such ovens near built-up communities.
However, because beehive coke ovens are virtually extinct in the U.S.,
they are of no real concern.
Air Pollution Emissions in Ironmaking
Coke-Oven Emissions
Emissions around a by-product coke-oven plant originate primarily
from four operations: 1) charging the coal into the ovens, 2) leakage dur-
ing carbonization, 3) pushing the coke out of the ovens, and 4) quenching
the hot coke with water.
Charging
Charging of the pulverized coal into the coke oven occurs shortly
after the oven has been emptied at the end of a coking cycle. Conse-
quently, the oven interior is still at red-hot temperatures. As the
charging begins, volatilization of the gases starts immediately, resulting
in the release to the atmosphere of visible clouds of yellow-brown smoke.
Several improved design features have been developed to reduce the amount
of emissions. These modifications include facilities for aspirating the
gases from the oven interior during charging, improved charging hole
spacing, facilities for enclosing the space between the lorry case (from
which the coal is charged) and the coke oven, and various mechanical
feed systems that increase the charging rate, thereby decreasing the total
amount of escaping smoke. However, these design features usually cannot
IX,35
-------
Table 5. AIR POLLUTION FROM METALLURGICAL COKING OPERATION
(Unit/Unit Weight of Coal Charged) AND OTHER IRONMAKING PROCESSES
X
1
O"
Type of Operation
1. By-product coking
Unloading
Charging
Coking Cycle
Discharging
Quenching
Under fir ing
2. Beehive Ovens
3. Iron Ore Sintering
Windbox
Discharge
4. Blast Furnace
Ore Charge
Agglomerated Charg<
Par*j-culates' S0 Ib/ton _ 1K ,. C H , Ib/ton NO , Ib/ton
Ib/ton x CO, Ib/ton x x x'
0.4 ....._.
1.5 0.02 0.6 2.5 0.03
0.1 -- 0.6 1.5 0.01
0.6 -- 0.07 0.2
0.9
10
200 -- 1 8 --
20 -- ---.-'
22 . -- 44 ._ .-
110
e 40 ... 1400-2100 -- . --
NH3, Ib/ton
--
0. 02
0. 06
0. 1
--
2
--
--
--
--
B-113-1707
-------
be incorporated into a coke oven, except between rebuilds, and the period
between rebuilds usually amounts to ZO-30 years.
Carbonization
The primary source of emissions during carbonization is leakage
around the oven doors. Control of emissions during carbonization takes
the form of optimum maintenance during operation.
Coking Discharge
At the end of the coking cycle, the coke is pushed from the oven
into the coke-quenching cars. The amount of emissions (smoke) depends
on the degree of completion of carbonization. Green coke produces large
quantities of smoke, whereas 'very little smoke is produced if the coke
is thoroughly carbonized.
Coke Quenching
In the quenching process, the hot coke is transported to the quench-
ing station, where it is quenched with large quantities of water in a short
time. The vast clouds of steam that result entrain small particles of
coke. However, the quantity of particulates is small and tends to be
deposited in the ground within a short distance of the quenching tower.
Consequently, the quenching operation is not considered to be a significant
source of emissions. For those particulates that are emitted, lower
impingement baffles installed in the coke quench tower can reduce the
quantity by 85-90%.
Other Sources
Other factors that affect the quantity of emissions in the coking oper-
ation include the amount of fines in the coal used and the sulfur content
of the coal used. In general, low-sulfur coals are used, thus maintain-
ing sulfur emissions at a minimum. Yet variations in the sulfur content of
the coal will vary the amount of sulfur-bearing compounds emitted.
Sintering and Pelletizing Emissions
Air pollutant emissions from sintering processes consist of dust in
the waste gases from the sinter strand and from the discharge end of the cool-
ing chamber. Additional dust is emitted by the grinding and blending
operations. In addition to the dust (particulates) emissions, sulfur dioxide
IX-37
-------
and carbon monoxide gases are emitted. The amount of sulfur dioxide
emitted depends upon the sulfur concentration of sulfur-bear ing native ores.
Air pollutant emissions from pelletizing processes are similar to
those from sintering processes. However, particulate emissions are
usually lower (numerical documentation not available) because the ore
fed into the heating plant is in the form of "green" pellets. Thus, the
amount of fines present is significantly reduced.
Blast-Furnace Emissions
Air pollutant emissions from blast furnaces comprise off-gases from
the process. As the blast-furnace gas leaves the top of the furnace, it
contains large amounts of particulate matter made up of about 30% iron,
15% carbon, 10% ^silicon dipxide, and small amounts of aluminum oxide,
manganese oxide, calcium oxide, and other materials. The gas also
contains large amounts of carbon monoxide, which is used for fuel in
other areas of the steel plant. During normal operation, air pollutants
are emitted during the charging of the burden to the furnace. However,
improvements in handling equipment and charging system design have
eased the problem.
Summary of Emission Controls in Ironmaking
Typically, emissions from the ironmaking processes are controlled
by conventional means, including scrubbers, electrostatic precipitators,
cyclones, baghouses, and settling chambers.. Table 3 summarizes the
controls used and gives the effect of these controls on emissions where
data are available.
Air Pollutant Emissions in Steelmaking
Table 6 summarizes the air pollutant emissions from steel-manufacturing
processes. In both the open-hearth process and the electric-arc process,
the use of oxygen lances to increase production rate increases particu-
late emissions by a factor of nearly 2. Particulate emissions from the
BOF are even higher. These emissions are primarily due to the violent
mixing of the bath that occurs as the oxygen is bubbled through. In open-
hearth and electric-arc furnaces where no oxygen lancing is used, par-
ticulate emissions (dust and fumes) result from the mechanical and chemical
reactions in the molten bath itself, and from the combustion of the fuels.
IX-3 8
-------
Table 6, AIR POLLUTANT EMISSIONS
FROM STEELMAKING PROCESSES
Particulates Carbon Monoxide
Type of Operation Ib/ton
Open Hearth
Oxygen Lance 22
No Lance 12
EOF 46 120-150*
Electric Arc
Oxygen Lance 11 18
No Lance 7 18
#
Represents generated emission. After ignition of gas
above furnace, the carbon monoxide drops to 0-3.0
Ib/ton of steam produced.
The highest rate of particulate emissions occurs during the first half
hour after hot metal addition and during the latter 2 hours of ore boil.
During the remaining portions of the cycle (roughly 80%), particulate
emissions are relatively low.
As is evident from the previous discussion, even with controls,
particulate emissions from steelmaking processes will increase as the
open-hearth process is phased out and the BOF assumes the major bur-
den of steel production.
Typical methods for controlling these emissions are electrostatic
precipitators (98% efficient), venturi scrubbers (85-98% efficient), and
baghouses (99% efficient). Hoods that have been developed for BOF's
and electric-arc furnaces catch the gases and channel them through the
cleanup systems. At the same time, the waste heat from these gases
is recovered for use elsewhere. Costs for gas cleanup systems are
high, running from $1.3 million for electric-arc furnaces to $5.0 million
for BOF's.27'28 However, as steel plants become more modern, thus re-
ducing emissions through greater efficiency and economy in the use of
resource inputs, these costs should decrease.
IX-39
-------
Air Pollutant Emissions From Reheating Operations
Emissions from reheating furnaces are primarily due to the combus-
tion of the fuel being used. Because the industry has turned to using
automatic combustion controls and instrumentation, smoke is ho longer
the problem that it once was. The main pollution problem at present is
caused by the presence of sulfur in the fuels used in these processes.
In order to lower these emissions, fuels with lower percentages of sulfur
are being used. However, this is only a temporary solution because the
availability of these fuels is decreasing. In addition to sulfur emissions,
carbon monoxide and nitrogen oxides are of concern to the industry. The
preferred fuel for controlling these emissions is natural gas. As in the
case of operating data, no reliable average emission levels are - available
for the reheating processes because of the large number of processes in
existence.
Effect of New Technology on Emissions
With respect to total emissions from the steelmaking process, as the
energy consumed is reduced, specifically, as coke rates decrease, total
emissions will decrease. As coke rates decrease, less coke must be
manufactured; consequently, fewer particulates will be emitted. Similar
conclusions can be reached concerning the blast-furnace operation.
Conceivably, if coke production is decreased by 50%, a total pollutant
emissions will decrease by a similar amount. In addition to particulate
femissions, carbon monoxide levels would be reduced. Note that most
of the pollution occurs in the off-gases from the coking process. Because
these gases are burned elsewhere in the plant, they usually are cleaned
before burning. Thus, as better handling equipment becomes available,
emissions, which usually occur because of improper handling of the gases,
will decrease. Again, these effects are examined in more detail later in
this section.
Energy Utilization Patterns Projections to 1985
The future energy needs of the iron and steel industry will be based
on a complex interaction of steel demand, availability and price of fuels, the
amount of reusable energy produced by the iron and steel processes, and the
implementation of new technology. The following illustrative analysis of
IX-40
-------
future energy needs is based first on projections of the further imple-
mentation of existing but still rather new technology, the steel industry's
own projections of the demand for iron and steel, and changes in pro-
cesses caused by changes in the availability of raw materials and of fuels
and in air pollution regulations. The analysis then considers the possi-
ble effects of implementing new technology still in the development state.
Later in this section, this projection of future energy need is related to
pollution emissions and changes in control technology.
Ironmaking Operations
The annual production of iron ore has been rather constant over the
5-year period between 1966 and 1971, that is, about 90 million tons/yr.
However, the steel industry predicts that, by 1985, pig iron production
will increase to 150 million tons/yr in order to meet the increasing
demand observed between 1971 and 1973. In order to predict the annual
production of pig iron for the years between 1971 and 1985, the simple
linear relationship shown in Figure 13 was assumed. The linear fit was
carried out between a production rate of 150 million tons/yr in 1985 and
90 million tons/yr in 1971, which is the average production between 1966
and 1971. This method tends to deemphasize the large drop in production
between 1970 and 1971 caused by a poor economic climate, forcing a
general decrease in the demand for steel products.
The projection for ore was based on the historical production of ore,
compared with pig iron. All iron ore, regardless of pretreatment method,
contains approximately the same amount of iron prior to charging into
the blast furnace, and the amount of iron necessary to produce 1 ton of
pig iron is constant. Hence, the amount of ore needed to produce 1 ton
of pig iron can be assumed constant. We have simply assumed (as does
the iron and steel industry) that any ore not containing a minimum standard
amount of iron at the time of mining will be treated until that standard
iron content is achieved.
Based on this assumption and published data, we determined that
about 1.47 tons of ore is required per ton of pig iron produced. Conse-
quently, 221 million tons/yr of ore will be needed in 1985 to produce 150
million tons of pig iron. Again, the production of ore between 1971 and
1985 was assumed linear with the 1971 production, which was assumed to
be 152 million tons, or the average from 1966 to 1971.
IX-41
-------
220
1966 '67 '68 '69 '70 '71 '72 '73 '74 "75 '76 '77 '78 '79 '80 '81 "82 "83 "84 1985
YEAR A-"3-'641
Figure 13. PROJECTED INCREASE OF ORE
DEMAND AND PIG IRON PRODUCTION
Iron Ore Preparation
As indicated earlier, iron ore can be mined and charged "as is" into
the blast furnace, or it can be pretreated to increase its iron content
per ton. The methods of pretreatment are pelletizing or sintering. Pre-
treated ore has the added advantage of increasing the production of a
blast furnace because of the physical structure of the charge or "burden. "
This has led to a constant rise in the use of pretreatment. However,
there is a substantial supply of high-grade ore that will continue to be
mined and charged as is. Pretreatment of high-grade ore is economically
unsound in spite of some production increases at the blast furnace. His-
torically, the annual production of crushed ore (untreated) has been main-
tained at about 40% of total requirements and pretreated ore at about 60%
of requirements. What has changed is the ratio of sintered to pelletized
IX-42
-------
ore. In Figure 14 are plotted the amounts of crushed, sintered, and
pelletized ore as percentages of total production.
to
40,
0
335
3 l
Q
O
tr
°-30
UJ
g
_i *
£25
2
u.
°20
*S
15
10
7
^
^
(
r^
r
^
^>
i
)Ji.
\
*^
^1
r
T
i^^
G
$&
^^
RUS
-------
100
90
p 80
O
o
o
UJ
CC
O
70
60
< 50
o
UJ
o
UJ
O
£E
O.
40
30
*
1971 '72 "73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I643
Figure 15. PROJECTED ANNUAL IRON ORE PRODUCTION
BY VARIOUS TREATMENT METHODS
89.3 million tons will be used as mined, and 105.2 million tons will be
pelletized.
The relative change in the amounts of pelletizing and sintering is
important because both the amounts and types of energy required by these
processes are significantly different. Pelletizing requires about 600, 000
Btu/ton of pellets and uses natural gas or fuel oil exclusively. Coke-oven
gas or blast-furnace gas could be used; however, pelletizing is always
carried at the mine where these gases are not available. (Pelletizing is
carried out at the mine because it is considerably more economical to
ship high-grade ore than low-grade untreated ore. ) Based on current
IX-44
-------
energy requirements for pelletizing and the fact that little new technology
appears likely to change this energy consumption, total energy consump-
tion was projected to 1985 (Figure 16).
so
90
85
80
75
3 70
m
M 65
0
»- 60
o
| 55
ui 50
45
40
35
30
OR
.
-
=«
*
~ »
,>
*
p=;
-^
^>
=--_
>
^J
~
^
N^
TOTAL FUEL
-"""ENERGY-SINTER
^
X
X
TOTAL
ENERGY-PELLETS
^
MATURAL
FUEL OIL
SINTER
^~-
^
***^
GAS,/
, LPG-
^
s
^
^
IN
"GA
t=r
s
^
N
^
-PL
iSES
|^^
^
\
^
ANT
-S^
^:
X
^
S
^
JTER
^
- ^
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-M3-I644
Figure 16. PROJECTED CONSUMPTION OF FUELS FOR
IRON ORE SINTERING AND PELLETIZING
Sintering of ore usually is carried out at the steel plant and hence
was developed to use either purchased fuels or in-plant gases. (Purchased
fuels as used in this report refer to natural gas, fuel oil of all grades,
and LPG, but exclude coal. ) Sintering requires about 2. 2 million Btu/
ton of sinter. Historically (1966 to 1971), the amount of purchased fuels
about equaled the amount of in-plant gases consumed by this process.
More specifically, 51% of the energy used for sintering was in-plant gas
and 49% was purchased fuels. Examining the historical split between
fuels more closely, we found that it occurred primarily from a need to
balance overall plant fuel supply and demand. For this projection, we
assumed that the historical split between purchased and in-plant fuels
IX-45
-------
would continue to 1985. Based on this assumption, the projection of
total energy requirements for sintering is shown in Figure 16. Based
on a projected decrease in sintering, energy requirements in this area
of steelmaking are expected to decrease from 93 trillion Btu/yr in 1971
to 57.5 trillion Btu/yr in 1985.
Coke and Coke-Oven Gas Production
Coke is a major component needed for the production of pig iron in
a blast furnace. The amount of coke required by the ironmaking process
has decreased steadily in the last 5-10 years because of such things as
pretreatment of the ore, higher blast-air temperatures, and better blast-
furnace design. The steel industry predicts that coke requirements will
continue to decrease from the present level of about 0.63 ton/ton of pig
iron to 0.5 ton of coke/ton of pig iron produced in 1985, if hydrocarbon
fuels remain available or the research on formcoke use is successful.
For this analysis, the decrease was assumed to occur linearly, as shown
in Figure 17.
0.64
0.62
O 0.60
0.58
(3
I058
UJ
*
8 0.54
0.52
050
kcURI
COKI
\
1
RENT AV
i-TO-IRO
\
\
ERA(
V R/
\
3E
MIO
y
\
\
S
P
i
\
S
TEE
RODl
\
. INI
JCTIC
\
\
)UST
)N P
[
\
RY
ROJE
I
\
:CTIC
\
)N^
^
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 *84 1985
YEAR
A-II3-I646
Figure 17, PROJECTION OF DECREASE IN COKE
REQUIREMENT PER TON OF PIG IRON PRODUCED
IX-46
-------
Based on the decreasing demand for coke per ton of pig iron produced
and on the predicted increased production of pig iron, the need for coke
should increase from the 56 million tons produced in 1971 to only 65
million tons in 1985, as shown in Figure 18.
oo
65
I64
& ^
"62
o
o 6I
§60
£59
UJ
s58
57
56
19
/
/
/
/
,
/
/
s
/^
^
^^
*^~
71 '73 '75 '77 '79 '81 '83 196
YEAR A-IB-KWS
Figure 18. PROJECTION OF REQUIRED COKE
PRODUCTION FROM 1970 TO 1985
In the coking operation, about 7 million Btu of coke-oven gas is
generated per ton of coke produced. Major changes in coking process
hardware are expected to reduce pollutant emissions. However, the heat
content per ton of coke of coke-oven gas generated is expected to change
very little. Therefore, the fuel available from the coking process in the
form of coke-oven gas will increase, along with expected increases in
the amount of coke produced. This projected production of coke-oven gas
is shown in Figure 19.
Coke is produced in an airtight chamber by distillation of coal brought
about by heating the coal. The heat required for this process is supplied
by burning almost all (99%+) in-plant gases. About 3. 8 million Btu is
required for each ton of coke produced. Historically, heat requirements
for coking have been fairly constant, so they were assumed to remain
constant from now to 1985. Based on this assumption, Figure 20 shows
the expected annual consumption of in-plant gases to 1985.
IX-47
-------
tou
450
m
CM
2 440
6"
(£
£430
Ul
CO
* 420
0
^ 410
400
39O
/
/i
iASi:
/
3:TO
/
NS C
7
JOKE
/
PR<
X
DDUC
X
;ED
-^
K 7
X
X 10
X
6 Bl
X
u
^
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-169'
Figure 19. PROJECTED PRODUCTION OF COKE-OVEN-GAS
ENERGY BASED ON REQUIRED COKE PRODUCTION
C3M
= 240
m
CM
o
^ 230
%
Ul
UJ 220
9ir»
X
X
X
X
\o
X
BASIS:
3>1
CO
^
KE F
>£
>ROC
.-
UCE
*v
D K
*^
3.8
^^
X 10
^
6 Bl
.--
u
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-'6S3
Figure 20. PROJECTED DEMAND FOR IN-PLANT-GAS
ENERGY FOR UNDERFIRING COKE OVENS
IX-48
-------
As a final part of the energy use projection for the coking operation,
the coal converted to coke was considered. The amount of coal required
to produce 1 ton of coke varies somewhat, depending upon the heating
value of the coal used in the process. Therefore, the amount of heat is
a more convenient term to use. On this basis, about 36 million Btu of
(coal) energy is needed, to produce 1 ton of coke. (For a typical coking
coal, this is about 1.44 tons.) Figure 21 shows the projected total energy
demand for coking coal to the year 1985, as calculated from Figure 18.
900
1971 '72 '73 '74 '75 176 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-'648
Figure 21. PROJECTION OF TOTAL PURCHASED COKING
COAL AND NATURAL GAS, FUEL OIL, AND LPG ENERGY
IX-49
-------
Blast-Furnace Operation
As described earlier, ore, coke, and fluxes are charged into the
blast furnace^ which is then heated by injection of hot (blast) air and
sometimes hydrocarbon fuels. Two significant historical trends were
found in examining this process relative to its energy requirements.
First, to increase productivity, the industry was tending to use higher
and higher blast-air temperatures and is still increasing the amounts of
hydrocarbon fuel injection.
We first obtained data on the amounts of hydrocarbon injection used
between 1967 and 1971. We plotted these data in terms of percentage of
total iron production and found that the increased usage was nearly linear.
We projected this linear growth (Figure 22) to 100% and concluded that
all iron would be produced in furnaces using hydrocarbon injection by
about 1980. ;
IIU
S5
\f\n
o
T : QA
C 1 yO
t O
&u ao
UlZ
Q _ 70
< Z ' U
50
00 en
Z CE 6O
O<
rcy «?n
Q OU
O 4«i /
£ 40
*n
\\
^
h
s\
)
h
7
^
^4*
^
^
'
^
^
^
i
g
1
g
(/)
8?
8
1967 '68 '69 '70 '71 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-1650
Figure 22. PROJECTION OF PERCENTAGE OF IRON MANUFACTURED
BY USING HYDROCARBON INJECTION INTO THE BLAST FURNACES
This method of prediction compares well with an independent steel in-
dustry projection that 100% use of hydrocarbon injection would be realized
by about 1981.
IX-50
-------
We then combined the projection of total iron to be produced (Figure 14),
the percentage to be produced each year using hydrocarbon injection
(Figure 22), and information that about 1. 5 million Btu of hydrocarbon
fuels (natural gas or oil) is used per ton of iron produced to obtain
Figure 23.
£\J
-------
In examining blast-air temperatures for blast furnaces, we found
that temperatures were steadily increased over the last 10 years, but are
now leveling off. We believe further significant increases are unlikely
because they would require major changes in the blast stoves producing
the hot air, connecting ductwork, and possibly the blast furnaces them-
selves. Consequently, we assumed that the energy to heat the blast air
to current temperature levels would remain essentially constant from
now to 1985. ' .
A typical blast-air stove to produce the hot air uses about 2. 25
million Btu of fuel to generate enough, air to produce 1 ton of p"ig iron.
In general, blast-air stoves use in-plant gases, and some natural gas
and oil are used for load balancing of all plant fuels, for ignition, and
during maintenance of the in-plant gas system on these stoves. Histor-
ically, the industry has used about 90% in-plant gases and 10% purchased
fuels for blast-stove operation. We assumed that this ratio of in-plant
to purchased fuel usage would continue to 1985. From the above informa-
tion, a projection of the amounts of purchased and in-plant fuels (Figure 24)
was prepared based on the projected growth of iron production, again
shown in Figure 14.
An additional factor that must be considered in the energy-use picture
of the blast furnace is the amount of blast-furnace gas that is produced
and reused in other plant areas. Predicting the amount of blast-furnace
gas that will be produced is somewhat difficult because the amount varies,
depending on whether or not hydrocarbon injection is used and on the
temperature of the blast air. Data were not available in sufficient detail
to account for these variables as a function of the growth in use of hy-
drocarbon injection, predicted earlier in Figure 22. However, the liter-
ature indicated that the difference in blast-furnace-gas production was
about 5% between using or not using hydrocarbon injection. We chose
to project blast-furnace-gas production as if hydrocarbon injection was
used for all ironmaking an assumption that is really only valid after
1980. Consequently, our projection for blast-furnace gas production be-
fore 1980 is probably about 5% high. The projection is shown in Figure 25,
which assumes 8. 3 million Btu of blast-furnace gas produced per ton
of coke charged to the blast furnace. Figure 25 was prepared from the
IX-52
-------
CD
CJ
O
300
280
260
240
220
^ 200
Q_
2
O
O
UJ
z
UJ
ISO
160
140
120
100
80
60
40
20
GAS, OIL. LPG
Figure
1971 '72 '73 '74 '75 '76 '77 '78 '79 feO '81 '82 '83 '84 1985
YEAR A-II3-I674
24. ENERGY CONSUMPTION IN BLAST-AIR STOVES
J**J
_2
S 540
CM
2 530
§ 52°
0 510
§ 500
o:
0. 490
0 480
CE
g 470
LU
460
>
^
X
X
X
x^
X
X
X1
x
X
x
X
^
^^
^
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I654
Figure 25. ENERGY PRODUCTION (Blast-Furnace-Gas)
FROM BLAST FURNACES
IX-51
-------
projection of the amount of coke to be used between now and 1985, as
shown in Figure 18.
Energy in the form of coke charged to the blast furnace was not con-
sidered because it was accounted for in the coking operation in terms
of coal usage. In addition, relatively small amounts of oxygen, electric
power, and steam are used in the various steps of ironmaking. The en-
ergy involved in the production and use of each of these is considered
separately in a later section on a plant-wide basis.
Steelmaking Operations
Once pig iron is produced, it is converted to steel in any one of
several processes. These processes are the basic oxygen furnace (BOF),
the electric-arc furnace, and the open-hearth furnace.
Total Projected Steel Production :
Several references that have been found state that the steel industry
expects to produce 185. 5 million tons of steel in 1985. We estimated
the steel production between 1970 and 1985 by linear extension of the
industry's prediction for 1985 and actual production in 1970, as shown in
Figure 26.
Projected Steel Production by Process
The growth in production of steel by the BOF process has been re-
markably linear between 1965 and 1971. We projected future growth
(Figure 26) by linear extension of the past growth rate. Production by
the electric-arc furnace process also exhibited a linear growth between
1965 and 1971. Again, we projected future growth by a linear extension
of past growth.
Projecting the future of the open-hearth furnace process is slightly
more complex. Open hearths are being phased out of use because of a
variety of problems discussed earlier in this report. Information from
the steel industry indicates that the rate of phaseout depends on the num-
ber of furnaces that reach the end of their useful life each year. The
useful life of an open hearth is estimated at about 20 years. Based on
the rate of construction prior to 1965 and the rate of decline or phaseout
after 1965, we determined that open-hearth production will decline at a
IX-54
-------
rniiii
STEEL INDUSTRY
PROJECTION' /
1965 '66 '67 '68 '69 '70 '71 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
B-M3-I659
Figure 26. PROJECTED ANNUAL STEEL
PRODUCTION TO 1985
IX-55
-------
rate of about 9% per year, with the newest furnaces going out of service
about 1985, as shown in Figure 26.
As a means of checking the validity of our projections for each of
the three steelmaking processes, the projected production for each year
was totaled and compared with the projection of total steel production,
which was independently derived. The agreement was satisfactory, as
shown in Table 7.
Table 7. COMPARISON OF TOTAL PROJECTED STEEL
PRODUCTION WITH SUM OF PROCESS PRODUCTION PROJECTIONS
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
EOF
65
72.4
79
86
92.5
99.4
106
113
119.8
126. 5
133
140
146.6
153.5
160
EF*
20. 5
21.5
22. 6
23.7
24. 8
26. 0
27. 0
28. 0
29. 0
30. 2
31.4
32. 5
33. 5
34. 6
35.7
37
32
28
24. 6
21. 8
19.4
17
15
12.8
10.8
8. 6
6.5
4.5
2.4
0
j. uta.i
Total Production
-106 tons/yr-
122. 5
125.9
129. 6
134.3
139.1
144. 8
150
156
161.6
167.5
173. 0
179. 0
184.6
190.5
195.7
120
125
129.5
134.2
139
143.5
148
154
159
163
168
172
177
182
186
Difference
-2.5
-0.9
-0.1
-0.1
-0.1
4-1.3
4-2. 0
4-2. 0
4-2. 6
4-4.5
4-5. 0
4-7. 0
4-7. 6
4-8. 5
4-9.7
*
Electric-arc furnace.
Open-hearth furnace.
Energy Usage in the Open-Hearth Process
The open-hearth process was designed so that it could use a very
wide range of fuels, from in-plant gas, natural gas, and oil to tar and
pitch. As in other processes already discussed, we again found that the
ratio of in-plant gas usage to purchased fuel usage was fairly constant
between 1965 and 1971. The split between fuels for these years was 14%
in-plant gas usage and the remainder purchased fuels. In addition, the
total average energy requirement for open-hearth steelmaking is 3. 55
IX-56
-------
million Btu/ton of steel. Based on these findings and the data of Figure 26,
we projected the decline in energy requirements, as shown in Figure 27.
CD
CM
~O
LU
CO
280
260
240
220
200
ISO
160
140
\
\
O ALL FUEL (INCLUDING COG AND BFG)-
A FUEL OIL, NATURAL GAS,TAR/PITCH
D CQKE-OVEN GAS/BLAST-FURNACE GAS
1965 '66 '67 '68 '69 '70 '71 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-113-1655
Figure 27. PROJECTED DECLINE OF ENERGY USE
IN OPEN-HEARTH STEEL-MELTING FURNACES
Energy Usage in the BOF Process
Very little fossil fuel is used in the BOF process; all the actual
processing heat is supplied by the exothermic reaction of the iron and
oxygen. Fossil fuels generally are used only between charges or heats
to maintain vessel temperature and, maybe some small amounts, during
pouring. An industry average for the amount of fossil fuel used in the
BOF process is 200, 000 Btu/ton of steel, which was calculated from re-
ported fuel use and production data for the years 1965 to 1971. By using
the above heat requirement and the projected production of steel by the
BOF process (Figure 26), fossil-fuel consumption was projected to the
year 1985 and is shown in Figure 28.
IX-57
-------
30
3 25
m
CM
- 20
>"
u
UJ
Ul
15
10
1965'66 '67 '68 '69 '70 '71 '72 "73 '74 '75 '76 '77 '78 '79 "80 '81 '82 '83 '841985
YEAR A-"3-'*58
Figure 28. PROJECTED ENERGY CONSUMPTION OF FOSSIL
FUELS IN PRODUCTION OF STEEL BY THE BOF PROCESS
(Primarily Heat Used Between Charges)
Fossil fuels are indirectly used in the BOF steel process in the form
of the oxygen used to react with the iron. Fossil fuels are converted
to electrical energy, which in turn is used to make the oxygen. Later
in this report, the amount of fossil-fuel energy necessary to make the
oxygen is considered and included in the overall energy picture of the
steel industry.
Energy Usage in the Electric-Arc Furnace Operation
The literature indicates that the electric steelmaking process uses
about 1. 7 million Btu of energy (electrical) per ton of steel produced.
Based on these data and the projection of steel to be produced by elec-
tric furnaces, we projected the energy needs for this process to 1985.
The amounts of fossil fuels required to produce the needed electric power
are considered in a later section. However, the projection of electric
energy required is shown in Table 8,
IX-58
-------
Table 8. PROJECTED ELECTRIC POWER REQUIREMENTS FOR
ELECTRIC FURNACE OPERATION FROM 1973 TO 1985
Projected Steel Projected Power
Production, Required,
Year 106 ton/yr 101* Btu/yr
1973
1975
1977
1979
1981
1983
1985
24
26
29
31,
34
37
40
40.8
44.2
44. 3
52. '7
57. 8
62. 9
68
Finished-Product Steel Processing
Finished-product steel processing, in this analysis, includes all the
steps from the time molten steel is available to producing typical steel
plant products. All these steps are grouped into one or the other of two
basic processes: reheat and form or continuous casting. When steel is
processed by what we call reheat and form steps, it is first cast into an
ingot, allowed to partially or completely cool, reheated in soaking pits
and rolled to smaller dimensions, again allowed to partially or completely
cool, and reheated again and subjected to additional form changes.
This process of reheat and form can occur only once or several
times, depending on the end product desired. Therefore, it would be
extremely complex to calculate energy consumption by considering the
individual furnace processes that may be involved. However, average
energy data are available on all heating operations as one average oper-
ation but including continuous casting. Once the amount of heat used
for continuous casting operations is subtracted, we obtain the average
energy required for the reheat and form process on a per-ton-of-steel-
produced basis.
Continuous Casting and Conventional Reheat Production
In the continuous casting process, the molten steel from the BOF,
open-hearth, or electric-arc furnace is formed directly into a finished
or semifinished product. This process bypasses the normal steps of
cast, cool, reheat, cool, etc. Continuous casting in the U. S. is very
new and consequently represents only a small portion of the steel pro-
cessing currently being done. In 1971 only 4. 0% of all steel processed
IX-59
-------
in the U. S. was continuously cast. However, the steel industry predicts
that 20% of all steel will be continuously cast by 1985. Based on cur-
rent continuous cast production and; this prediction, a linear growth in
the use of continuous casting was assumed in terms of the percentage of
total steel produced (Figure 29).
C %
s/
UBLI
IN 1
^/
SHE
981
^
D
f
STEEL INDUSTRY
PROJECTION FOR/
1985
y
y
/
/
X
x^
r
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I683
Figure 29- LINEAR PROJECTION OF CONTINUOUSLY CAST
STEEL AS A PERCENTAGE OF TOTAL STEEL PRODUCED
Therefore, we were able to project the amount of steel to be continuously
cast between 1971 and 1985 from the projected total steel production given
in Figure 26 and again in Figure 30. The difference between the amount
of continuous cast steel and total steel production gives the amount of
steel that must be processed by the conventional means of reheat and form.
We realize that some continuous cast steel is reheated later for final
finishing. However, the literature suggests that the quantities are rela-
tively small. We are therefore not considering this aspect.
IX- 60
-------
190
160
150
140
w 130
Z
O
to
O
O
O
O
(T
Q.
UJ
UJ
(/>
100
90
30
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I656
Figure 30. PROJECTED GROWTH OF CONTINUOUS CASTING
AND CONVENTIONAL PRODUCTION OF STEEL
IX-61
-------
Continuous Casting and Conventional Reheat
Energy Requirements
Continuous casting requires only moderate amounts of fuel for heating
transfer ladles, spouts, and portions of the casting equipment. Current
equipment requires about 250, 000 Btu/ton of steel cast, and the fuel is
usually natural gas. Table 9 shows the projected amount of natural gas
to be used for continuous casting between 1971 and 1985 based on the
projected production of continuous cast steel.
Table 9. PROJECTED ENERGY REQUIREMENTS
FOR CONTINUOUS CASTING
Year
1971
1973
1975
1977
1979
1981
1983
1985
Production,
106 tons
4.8
8.4
12.2
16.9
21.8
27. 3
33.3
40. 0
Energy,
1012 Btu
1.2
2.1
3.05
4.23
5.45
6.81
8. 31
10. 0
The average amount of energy used to process steel in the conven-
tional reheat and form steps has cycled from a. low of 4. 3 to a high of
4. 8 million Btu/ton over the last 6 years. An average for this time
period was 4. 7 million Btu/ton, which was used as the basis for future
projections shown in Figure 31.
General Energy Uses
Several energy-use areas in the steel industry cannot be tied to
any one processing step. Steam, electric power, and oxygen are inter-
related and used throughout the mill. Therefore, this report deals with
these energy sources in this separate section. None of these items are
energy in the usual sense, but substantial amounts of fossil fuels are
used in their production.
IX-62
-------
ovjw
2 750
m
CM
0
700
0
x
X
X
x
f
X
X
X
X
X
x
/
'
^
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
Figure 31. PROJECTED FUEL ENERGY REQUIREMENTS
FOR STEEL REHEATING AND CONTINUOUS CASTING
Production and Use of Oxygen
The major uses of oxygen in an integrated steel mill are for the BOF
process, the open-hearth furnace, blast-furnace injection, scarfing, electric-
arc melter injection, and general steel cutting with gas-oxygen torches.
The oxygen requirements of each of these processes were discussed
earlier in this report. The amounts of oxygen required by each of these
processes on a production rate basis are given in Table 9.
IX- 63
-------
Table 10. PROCESS OXYGEN REQUIREMENTS
Process
EOF Process
Open-Hearth Furnace
Electric-Arc Furnace
Blast-Furnace Injection
General Plant Usage
Oxygen
Requirement,
ton /ton product
0. 0789
0. 0451
0. 0137
0. 0067
0. 0095
Based on this information and on production predictions, we have pro-
jected the individual demands of each process for oxygen (Figure 32).
17
16
15
14
13
12
co II
I 10
% 9
g" 8
O
6
5
4
3
2
I
1971 '72 '73 '74 '75 '76 77 '78 '79 *80 '81 '82 '83 '84 1985
YEAR A-"3-'663
Figure 32. PROJECTED STEEL INDUSTRY DEMAND FOR
OXYGEN AS A FUNCTION OF PROCESS TYPE
GENERAL USAGE-
BLAST FURNACE-
ELECTRIC MELTING -*
IX- 64
-------
The need for oxygen in the BOF process will increase from about
6.4 million tons in 1973 to 1Z. 6 million tons in 1985; in other words, it
will,almost double. The use of oxygen for open-hearth injection will de-
cline with the decline in open-hearth usage. In 1973, only 1.2 million
tons of oxygen was used by open hearths; this is expected to decline to
0 by 1985. Blast-furnace injection of oxygen and the demand for oxygen
in electric melting will increase slightly, but not as strongly as for the
BOF process. The amount of oxygen required for general plant usage
was derived from consumption data for 1966 to 1971. Published data for
general-plant oxygen usage divided by steel production in each year showed
that requirements in this area are 0. 0095 ± 0. 0003 ton of oxygen/ton of
steel produced.
The individual projections of oxygen demand by process were summed
to obtain a total demand projection (Figure 32). This summation predicted
that the total demand for oxygen will grow from about 9. 2 million tons
in 1971 to 16 million tons in 1985. The calculated demand for oxygen
in 1971 of 9.2 million tons agrees well with published data, which show
an actual usage of 9. 14 million tons.
Only a small portion of the oxygen used by the steel industry is
produced in oxygen plants owned by the steel plant. In 1967, this so-
called "self-produced" oxygen accounted for only about 8. 8% of the total
amount used. However, between 1967 and 1971, the steel industry has
built and operated an increasing number of oxygen plants. Figure 33
shows the amount of self-produced oxygen produced between 1967 and
1971 in terms of the percentage of total oxygen used. The trend toward
increasing amounts of self-produced oxygen was projected to 1985 with a
linear extrapolation. The data of Figure 32 (projected demand) and
Figure 33 (percentage self-produced) were used to calculate the quantities
of oxygen to be purchased and the quantities to be self-produced (Figure 34).
Oxygen usually is produced by liquefaction of air, which requires a
total of about 340 kWhr of electric power per ton of oxygen. If the oxygen
is to be stored, in part, as a liquid, about 750 kWhr of electric power
is required per ton of oxygen. Self-generated oxygen usually is produced
and used without the liquid storage step. Only small amounts of storage
are required to adjust for daily production fluctuations. Purchased oxygen
IX-65
-------
28
26
24
S 22
o
E 20
O
8E l8
3? 16
14
12
10
nS
A
(.
xl
)
t
>^
M
r
X
^
X
x
X
^
x
^
x
/
x
1967 '68 '69 VO '71 '72 '73 '74 '75 "76 '77 '78 '79 '80 '81 £2 £3 '84 1985
YEAR A-M3-I6M
Figure 33. PROJECTED PERCENTAGE OF OXYGEN PRODUCED
BY STEEL INDUSTRY FOR PLANT CONSUMPTION
usually is liquefied for ease of transport by truck or cryogenic pipeline.
Therefore, the calculation of energy usage for making self-produced
oxygen was based on 340 kWhr of power, and that for purchased oxygen
was based on a power requirement of 750 kWhr/ton of oxygen (Figure 34).
The quantity of fossil fuels used to generate the electric power
that, in turn, is used to make the oxygen, is considered in the next
section.
Production and Demand for Electric Power and Steam
Published information for 1962 to 1971 shows that electric consump-
tion for general plant use (excluding electric melting and oxygen produc-
tion) correlates closely with steel production. These data show that
0.98-1. 09 million Btu of electric power is used per ton of steel produced.
Based on the average consumption of 1. 06 million Btu/ton of steel and
our earlier projection of steel production, general plant electric require-
ments were predicted (Figure 35) to 1985. This projection shows that
general plant electric consumption will grow from 128 trillion Btu/yr in
1971 to 212 trillion Btu/yr in 1985. The electric energy requirements
IX-66
-------
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-'662
Figure 34. PROJECTED OXYGEN REQUIREMENTS WITH
RELATIVE AMOUNTS OF PURCHASED AND SELF-PRODUCED OXYGEN
!
and production projections discussed earlier were used to predict demands
for electric melting and self-produced oxygen. The electric demand of
these three areas was summed to estimate total electric power demands.
Electric power is much like oxygen production; that is, a portion of
the demand is satisfied by in-house production and some power is pur-
chased. Historical data on self-produced power (Figure 36) from 1967
to 1971 show a strong trend toward purchasing electric power. The per-
centage of total electric power to be self-produced from 1971 to 1985 was
estimated from a linear extrapolation. This resulted in a calculated
self-produced electric power projection that showed generation capacity
declining from 38 trillion Btu in 1971 to about 0 in 1985 (Figure 37),
assuming 3.23 Btu of fuel energy per Btu of electric energy produced.
IX-67
-------
Q
UJ
V)
o
a:
a.
cc o
o
o
UJ
UJ
30.0
27.5
25.0
22.5
20.0
17.5
15.0
12.5
10.0
7.5
5.0
2.5
0
2.5
0L
w
1967'68'69 '70 '71 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '841985
YEAR '
B-II3-I682
Figure 35. PROJECTION OF ELECTRIC POWER REQUIREMENTS
FOR PRODUCING OXYGEN REQUIRED BY STEEL INDUSTRY
In addition to electric power, process steam is produced in boilers;
this requires, historically, about 1. 7 million Btu of fuel energy per ton
of steel produced. Based on the projected total steel production (Figure Z9)
and the average energy required for producing process steam, Figures 38
and 39 were prepared to show the amounts of fuel required between
1971 and 1985. Figure 40 shows the projected total amount of fuel
energy for use in steel industry boilers, including both electric power
and process steam. The increase in fuel demand for boilers increases
very little from 1971 to 1981 and declines from 1981 to 1985. This is
caused by the combination of growth in the demand for steam and the
decline in self-generated electricity production.
Fuel Types and Demand for Boilers
The steel industry boiler can readily operate on in-plant gases, coal,
oil, or gas. We assumed that all available in-plant gases would be used
first with purchased fuels used to make up any differences to meet the
energy demand.
IX-68
-------
CD
C\J
O
of
LU
O
OL
O
CC
O
LJ
_l
UJ
2.5
1971 '72 '73
'75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I67Z
Figure 36. PROJECTED REQUIREMENTS FOR ELECTRIC POWER
IX-69
-------
OC
UJ
o
.1
UJ
CO
O\J.\J
27.5
25.0
22.5
20.0
17.5
15.0
12.5
10.0
7.5
5.0
2.5
0
191
h
X
c;
K
N
^
X
X
X
X
s.
^
X
X
x,^
X
X
x
X
X
rx
x
B7'68 '69 '70 '71 '72 *73 *74 '75 "76 '77 '78 *79 "80 '81 "82 to '84 1985
YEAR A-"3-'661
Figure 37. RATE OF DECLINE (Linear Projection) OF
STEEL-INDUSTRY-GENERATED ELECTRIC POWER
Earlier in this report, we indicated the amounts of in-plant gases that
would be produced by the coke ovens and blast furnaces in terms of Btu/
ton of iron, with the necessary corrections for changes in coke require-
ments, etc. From these data, we prepared a projection of total in-plant
production of fuel gases (Figure 41). In 1971, steel industry records
show that 854 trillion Btu of in-plant gases was produced. Our projec-
tion indicates that the production of gases will increase, but at a declining
rate. Approximately 1 billion Btu will be produced in 1985. Data pre-
sented earlier in this report also indicated the total amount of energy
needed by each iron and steel process and a typical split between the use
of in-plant gases and purchased fuels. This information was used to
calculate the use of in-plant gases for all processes other than in boilers.
These data are shown in Figures 42, 43, and 44. Figure 42 shows the
projected demand for in-plant gases for steel reheating and continuous
casting based on the tptal energy requirements of these processes as shown
IX-70
-------
iovj
120
2 110
CO
M '°°
0 90
| 80
t 70
rj 60
CO
0 50
O
>-40
CC 30
LJ
U 20
10
n
"^
\
X
x
\
\
\
\
\
\
\
\
\
s
k
\
y
\
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 "83 '84 1985
YEAR A-II3-I664
Figure 38. ENERGY REQUIREMENTS FOR BOILERS TO PRODUCE
PROCESS STEAM FOR THE IRON AND STEEL INDUSTRY
JtW
330
320
3
£ 310
CM 300
O
~ 290
0 280
1 270
W 260
8 25°
& 24°
U 230
Z
LLJ 220
210
onn
/
/
/
/
/
V
J
/
J
/
jt
/
y
/
/
/
/
/
/
/
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 "83 '84 1985
YEAR A-.I3-I665
Figure 39. ENERGY REQUIREMENTS FOR BOILERS TO PRODUCE
ELECTRIC POWER FOR THE IRON AND STEEL INDUSTRY
IX-71
-------
360
3
£ 350
w
O
CONSUMP
OJ OJ
ro S
o o
ENER
300
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-II3-I666
Figure 40. TOTAL FUEL ENERGY REQUIREMENTS FOR
STEEL INDUSTRY BOILERS (Steam and Electric Power)
in Figure 29 and the historical split between fuel types. Figure 43 shows
the total in-plant gases required from ore preparation through pig iron
production. Figure 44 shows the in-plant gases remaining after subtract-
ing ore preparation to iron production and what is used for steel reheat
and continuous casting. Figure 44 also shows the demand for in-plant
gas for steel finishing processes and the resulting supplies for boiler
fuel. Subtracting the availability of in-plant gas from the total projected
demand for boiler fuel uses yields the total amounts of purchased fuels
needed (Figure 45).
The purchased fuel used by boilers can be coal (boiler grade), oil,
or gas. In 1971, the industry used 109 trillion Btu of coal in its
boilers; this matches the total purchased needs for boilers, as shown in
Figure 45. However, there has been a strong trend toward replacing coal
firing with other purchased fuels, probably because of pollution regulations.
We cannot be sure this trend will continue because of the current shortages
of oil and gas. However, for this analysis, the trend was assumed to
IX-72
-------
1020
oo 980
CVJ
O
&
or
UJ
z
UJ
UJ
u.
840
X
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-'679
Figure 41. PROJECTED TOTAL PRODUCTION OF IN-PLANT GASES
(Coke-Oven Gas and Blast-Furnace Gas) FROM IRONMAKING PROCESS
J( i>
550
525
500
475
~ 450
CD
y 425
O
400
I 375
UJ
z 350
Ul
-J 325
UJ
{?. 300
275
250
225
200
175
150
X
.
^X
-
^
X
^^~~
f
1
G>
puJi
-
^
N"^
3&SE
n
r^
-C
* INCLUDES
AMOUNT C
1 'I I
v^G
^
!
~-~~-
^
~
^
^
-
^
j-
REHEATING PROJECTI
DF IMPORTED STEEL
rn
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-U3-I681
Figure 42. PROJECTED DEMAND FOR IN-PLANT GASES AND
HYDROCARBON FUELS (Excluding Coal) FOR STEEL
REHEATING AND CONTINUOUS CASTING OPERATIONS
IX-73
-------
540
520
480
CD
CM
~b
400
360
or
UJ
z
UJ
UJ
u.
200
180
140
X
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-M3-I666
Figure 43. PROJECTED DEMAND FOR OIL, NATURAL GAS,
AND IN-PLANT GASES (Coke-Oven Gas and Blast-Furnace Gas)
FROM ORE PREPARATION TO PIG IRON PRODUCTION
IX-74
-------
CD
ce
Ul
z
UJ
425
400
375
350
325
300
275
IN-PLANT GASES AVAILABLE
FOR STEELMAKING TO
FINISHED-PRODUCT PROCESSES
IPG DEMAND FROM STEELMAKING
TO FINISHED-PRODUCT PROCESSES-
IPG AVAILABLE FOR STEAM,
ELECTRIC (BOILER) GENERATOR
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-H3-I675
Figure 44. BREAKDOWN OF IN-PLANT GAS USE FOR
STEELMAKING TO FINISHED PRODUCT PROCESSES AND FOR USE
TO GENERATE STEAM, ELECTRIC POWER, AND
MISCELLANEOUS PROCESS HEAT
3
s
UJ LU
< LJ
11
130
1971
1973
1975
1977 1979
YEAR
1981
1983 1985
Figure 45. HYDROCARBON FUELS NEEDED
FOR BOILER MAKEUP
IX-75
-------
continue. Historically, the use of boiler-grade coal in the steel industry
has declined at a rate of 9% per year between 1965 and 1971. Project-
ing this decline to the year 1985 yielded the data shown in Figure 46.
IIU
100
90
80
3
m
~ 70
O
z"
2 60
Q_
5
^ 50
8
0 40
o:
UJ
UJ
30
20
10
n
X
N
t
\l
\
\
\
\
I
\
\
1
\
\
\
\
\
\
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 fel '82 '83 "84 1985
YEAR A-|I3H669
Figure 46. PROJECTION OF COAL ENERGY DEMAND
FOR OPERATING STEEL INDUSTRY BOILERS
IX-76
-------
By subtracting the coal usage projected in Figure 46 from the projected
total demand for purchased fuels (Figure 45), we were able to estimate
the demand for oil and gas for boilers, as shown in Figure 47.
ITNJ
130
3 120
55
CM 110
0
~ 100
0 90
H-
!| 80
W 70
I 60
& 50
£ 40
'* 30
20
10
n
^
^
/
/
/
/
/
/
/
/
/
/
/
v
/
/
/
f
A
/
/
/
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-"3-'671
Figure 47. PROJECTION OF OIL AND GAS ENERGY DEMAND
FOR OPERATING STEEL INDUSTRY BOILERS
IX-77
-------
Miscellaneous Uses of Energy
The iron and steel industry uses significant quantities of energy for
such things as flame cutting of steel and direct-fired space heating. The
uses are too numerous to project energy consumption on an individual
basis. However, the total consumption of energy for these so-called
"other uses" historically is about 10.5% of all process energy, excluding
coal for coke and boiler fuel. Based on the assumption that this ratio
would be maintained, Figure 48 was prepared to project the need for
"other uses" based on a summation of projected energy demand by all
processes presented in earlier figures. Almost all this fuel is oil, nat-
ural gas, or liquid propane.
IOU
m
0 150
O
>-
LL!
1 . 1
^
^
X
x
^
x
X
x
,
x
X
^
'
s*
^
^
~ '
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I673
Figure 48. PROJECTED DEMAND FOR HYDROCARBON
FUELS FOR "OTHER" PLANT USES
Summary of Energy Demand Projections
Summing all the energy demand projections of this discussion yields
the following conclusions. Between 1973 and 1985
The demand for steel industry boiler-grade coal will drop from about
108.6 trillion Btu/yr to about 0 (Figure 49).
The demand for fuel energy to make the oxygen purchased by the
steel industry will grow from about 20.4 to 30.2 trillion Btu
(Figure 49).
IX-78
-------
J\J\S
280
260
240
220
200
m 180
CVJ
0 160
e> 140
1|2°!
100
80
60
40
20 '
0
r
U
i x1
^ (
PURCHA
Tl
^
? -
SEO
v-
9Vj
V
£
r
|ap/<
<<
-t:
0^
r
«|
OXYGEN (ENERGY
» A
"" t
^
fi
ft*
^
^
"r,
S^
EQUIVAL
H A i
& I
r
N
ENT
^
s
_r
k,
)
\
i
'
^u
Y
\
^
n
t
^4
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR 4-"s-|67°
Figure 49. PROJECTION OF TOTAL PURCHASED
ELECTRIC POWER, BOILER-GRADE COAL, AND
PURCHASED OXYGEN (Energy Equivalent) ENERGY
The demand for fuel energy to make the electric power purchased
by the steel industry will grow from about 128 to about 292 trillion
Btu (Figure 49).
» The demand for hydrocarbon fuel by the steel industry will grow from
about 3150 to about 3572 trillion Btu (Figure 50), and coal for coking
will be about 66% of this demand, as shown in Figure 49.
Environmental Impact Pattern Projections to 1985
The preceding section presented projections of energy consumption
based on the implementation of new technology, the phasing out of several
processes, and the growth of some existing processes. The combined
effect of the energy consumption and production of each process will de-
termine the ultimate impact on the environment by the iron and steel
industry. The projected emission levels of each of the process areas
are discussed below.
IX-79
-------
OOUU
3500
3
m
^ 3400
k
p 3300
2 3200
O
g 3100
UJ
3000
9QOO
/
/
/
.
X
/
/
X
/
/
^
^
'
- '"
1971 '72 '73 '74 '75 '76 '77 '78 '79 "80 '8! '82 '83 '841985
YEAR A-II3-I640
Figure 50. PROJECTION OF DEMAND FOR
ALL PURCHASED FOSSIL FUELS
Particulate Emissions
Information on participate emissions was obtained for iron and steel
processes both using and not using stack cleanup equipment. The pro-
cesses producing the major portion of all emissions are sintering of ore,
pelletizing of ore, coke-manufacturing, open-hearth furnace, basic oxygen
furnaces, and electric-arc furnaces. Many other processes, such as steel
reheat furnaces, blast air stoves, etc., emit particulate emissions; however,
the steel industry estimates that these total less than 0.5% of all emissions.
Consequently, these emissions werd :not specifically considered in this
analysis. '»>.,
Table 5 showed the typical particulate emission rates of each iron-
making process. Based on these data, the emission projections of Figures 51
and 52 were calculated. Figure 51 shows the annual emission rate for the
steel industry with stack cleanup systems. Several assumptions were
i
necessary to calculate the emission rate of open-hearth furnaces and
IX-80
-------
58OO
5600
5400
5200
5000
4800
4600
4400
4200
.a
4000
u>
O 3800
. 3600
CO
g 3400
to 3200
§ 3000
£ 2800
^ 2600
O
P 2400
QL
£ 2200
$ 2000
UJ
0 ,800
o:
Q- 1600
1400
1200
1000
800
600
400
2OO
X
X
-^
-
/
B^
X
2=1
~ ~
/
=r:
i
^
g^
1
/
.
\
fX
vtf*j
^**
.
of
-- -
f
*&
«$
Q2.
/
COK
&
<
^
<
^^
1$$
E P
^
**,<
-\O^
£^*1
^
^
^
EWT
2U
RODI
7
Eg/\
^\
JCTI
^V
j|5
/
H
^-->
ON
?^^
'/
$e
-
^
/
^1
^^-^.
0^
/,
i^
*^^,
^--
X
/
0ii
^
"
X
/
r^'
p<^-
*^"
13600
13400
13200
13000
12800
12600
12400
12200 -
(O
12000 2
1 1800 -
11600 0
CO
11400 fO
11200 UJ
11000 K
10800 =J
O
10600 h-
10400 $
10200 <
10000 g
9800 O
9600
9400
9200
90OO
8800
8600
8400
8200
1971 '72 '73 "74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 '85
YEAR B-l 13-1690
Figure 51. PROJECTED EMISSIONS OF PARTICULATES FROM THE
IRON AND STEEL INDUSTRY WITHOUT CLEANUP CONTROLS
IX-81
-------
600
O
co
500
£400
CO
UJ
LJ
O
I-
CE
CO
UJ
O
Q.
300
200
100
SINTERING ORE
Ir
"PELLETIZINGORE
1100
(O
O
1000'
co
'00 §
CO
LU
800 UJ
5
O
700 £
2
600
O
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 '85
YEAR
A-113-1667
Figure 52. PROJECTED EMISSIONS OF PARTICULATES FROM THE
IRON AND STEEL, INDUSTRY WITH CLEANUP CONTROLS
electric-arc furnaces. The literature indicates that the particulate emis-
sion rate for open-hearth furnaces is 22 Ib/ton of steel when oxygen lances
are used and 12 Ib/ton when oxygen is not used. Data were not available
on the exact quantity of open-hearth steel produced by using oxygen lances,
but this technique usually is employed only after the furnace begins to
lose capacity due to age. Thus, we assumed that oxygen is used for more
than one-half of the furnace campaign, or that 50% of all open-hearth steel
is produced by using oxygen lancing. Oxygen lancing also is used with
the electric-arc furnace, and it affects the particulate emission rate.
Again, we assumed that 50% of all electric-arc steel is produced by using
oxygen.
Figure 52 shows the projected emissions of particulates when stack
cleanup controls are used. These data were calculated from the data of
Figure 51 by assuming cleanup efficiencies of 91% for ore preparation,
90% for blast-furnace operations, and 90% for steelmaking operations.
IX-82
-------
Particulates also are released from the steel industry boilers that
produce the steam and electric power used in each of the above processes.
The amount of particulates depends upon the relative use of natural gas,
in-plant gases, and coal in the boilers. Earlier in this analysis, the
relative usage of fuels was considered (and presented in Figures 46, 47,
and 48). The only significant quantity of particulates is produced by coal
burning (Figure 46), which was assumed to typically produce 240 pounds
of particulates per ton of coal burned, or 9. 5 pounds of particulates per
million Btu of coal, assuming a 15% fly-ash content of coal. Figure 53
shows the projected emissions from coal-fired boilers. Particulate emis-
sions from boilers operated by electric utility companies that sell electric
power to the steel industry were not considered.
II\JU
1000
- 900
- 800
en
0 70°
600
* 500
UJ
!5 400
_i .
3 300
(-
^ 200
£
100
n
^
^s
\
N
N
\
\
\
\
>
\
\
s
\
\
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR A-II3-I694
Figure 53. PARTICULATE EMISSIONS FROM BOILERS
Carbon Monoxide Emissions
The emission of carbon monoxide from the various steel processes
is a function of both process materials and combustion of fuels. The
major sources of carbon monoxide emissions are the sintering process,
the blast furnace, the coking ovens, and the BOF process. Earlier in
this report, we indicated that carbon monoxide emissions from the blast
IX-83
-------
furnace were 1750 Ib/ton of iron produced. However, most of this pol-
lutant is reclaimed as blast-furnace gas. The literature reports that
only about 1% of the carbon monoxide escapes, or about 17.5 Ib/ton of
pig iron. Other fuel utilization areas that emit carbon monoxide but are
not reported elsewhere are the pelletizing operation, blast-air stoves,
and general steel-processing furnaces, such as soaking pits. The liter-
ature does not report carbon monoxide data for these systems because
they are generally minor. However, for this analysis, we assumed an
average emission rate of 0. 0774 Ib/million Btu consumed. This level of
emissions is typical of hot-air generators and reverberatory furnaces
when operated properly.
Figure 54 shows the total carbon monoxide emission rate projected
to 1985, and Figures 55 and 56 show the projected emission rates by
process. These data were calculated from the emission levels reported
in Table 11 and the production and energy consumption projections pre-
sented earlier.
Q
X =9 5500
%<°r,
O - 5000
2 co
§1450°
CD C/)
o: en 4000
1971 '72 '73 '74 '75 '76 '77 '78 '79 "80 '81 '82 '83 '84 1985
YEAR A-II3-I693
Figure 54. TOTAL CARBON MONOXIDE EMISSIONS
FROM IRON- AND STEELMAKING PROCESSES
IX-84
-------
2500
UJ
k
O ^ 2000
CO~
gl
m co
oc co
<
o
-------
BLAST-AIR STOVES
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I695
Figure 56. CARBON MONOXIDE EMISSIONS FROM
IRON- AND STEELMAKING PROCESSES
-------
350
300
250
co
O
CO
CO 200
5
UJ
x
O
2 150
CO
CO
8
£? 100
50
BLAST-A1R_STOVES
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-II3-I69I
Figure 57. PROJECTED NOX EMISSIONS FOR
IRON- AND STEELMAKING PROCESSES
IX-87
-------
NO Emissions
x
Very little information is published in the literature on the emissions
of NO from steel processes. References on this pollutant lead the reader
to believe that very little has been done to measure emission rates. We
were able to determine emission rates for the coking operation and open-
hearth furnace only. The coking operation typically produces 0. 04 Ib/ton
of coal charged. The open hearth produces typically 0. 70 Ib/ton of steel
produced. However, the NO emissions of many of the other processes
can be estimated based on the temperature, combustion characteristics,
and type of fuel used. For example, blast-air stoves use primarily in-
plant gases, which have a low combustion intensity (generally low NO ),
and operate in the temperature range of 2500°-2800°F (generally inter-
mediate NO levels). Data from similar heating systems show NO levels
X. ' Ji,
in the range of 200-400 ppm. This type of analysis was applied to other
steel processes, with the result given in Table 12.
Table 12. TYPICAL NO EMISSION LEVELS
x
Sintering (100 ppm)
Pelletizing (600-800 ppm)
Blast Furnace Unknown
Coke Ovens (200 ppm) 0. 04 Ib/ton coal
Blast-Air Stoves (300 ppm) 0.23 Ib/million Btu
Open-Hearth Furnace 0. 70 Ib/million Btu
Basic Oxygen Furnace 0. 23 Ib/million Btu
Electric-Arc Furnace Unknown
General Processing Furnaces 0.46 Ib/million Btu
Based on the above information and the energy consumption and pro-
duction rates published earlier in this report, projections of NO emissions
X.
were calculated and are shown in Figure 57.
References Cited
1. American Gas Association, A Study of Process Energy Requirements
in the Iron and Steel Industry^ Catalog No. C-20005. Arlington, Va.,
n. d.
2. American Iron and Steel Institute, Annual Statistical Report 1971.
Washington, B.C., 1972.
3. Bernstein, N. , Reuss, J. L. and Woolf, P. L., "A Cost Comparison:
Production and Smelting of Prereduced U.S. Iron Ore Pellets,"
J. Metals _18, 652-56 (1966) May.
IX-88
-------
4. Brown, J. W. , "Electric Arc Melting A New Era in Steelmaking, "
J. Metals 20, 21-26 (1968) October.
5. Ciotti, J. A., "A New Era in Melting, " J. Metals 23, 30-35 (1971)
April.
6. Dailey, W. H., "Integrated Steel Plants of the Future," Iron Steel
Eng. 49, 87-94 (1972) April.
7. De Witt, C. V., Jr., "Oxy-Gas Roof Burner Experience at Inland's
No. 3 Open Hearth," J. Metals 21, 69-73 (1969) July.
8. Fitch, T. S. and Wilson, L. H. , "The Controlled Pressure Pouring
Process for Producing Slabs and Certain Cast Product, " Iron Steel
Eng. 46, 69-85 (1969) December.
9. Flint, R. V., "Effect of Burden Materials and Practices on Blast-
Furnace Coke Rate, " Reg. Tech. Meetings Am. Iron Steel Inst. 1961,
9-44.
10. Grossi, F. et al., "Experiences and Advantages in Employing Pre-
heated Cold Charge in Electric Furnaces and Their Influence on New
Electric Steel Plant Lay-Out. " Paper presented at the Second
International Symposium on the Iron and Steel Industry, Moscow,
September 1968.
11. Hemon, W. C. L., Ed., "Air Pollution Problems of the Steel Industry
Informative Report," J. Air Pollut. Control Assoc. 10, 208-18, 253
(I960) June.
12. Innes, J. A. and Melouney, H. F., "Metallized Agglomerates: A
Raw Material for Steelmaking," J. Iron Steel Inst. 207, 1437-43
(1969) November.
13. Koros, P. J., "Effect of Oxygen Lances on Thermochemistry of
Open Hearth Processes, " in Proceedings of Open Hearth and Basic
Oxygen Steel Conference 7. New York: AIME, 1963.
14. Leavy, R. I. and Woolf, P. L. , "Blast Furnace Operation With
Natural Gas Injection and Oxygen-Enriched Blast. " U. S. Bureau of
Mines Report of Investigations No. 6977. Washington, D. C. : U. S.
Government Printing Office, T967.
15. McBride, D. L., "Auxiliary Fuels for the Blast Furnace," Iron
Steel Eng. 44, 68-74 (1967) August.
16. McGannon, H. E. , Ed., The Making, Shaping and Treating of Steel,
9th Ed. Pittsburgh: Herbick and Held, 1971.
17. Melaker, N. B. and Fine, M. M., "Prereduced Iron-Ore Pellets
State of the Art," J. Metals 18, 795-802 (1966) July.
IX-89
-------
18. Melcher, N. B. et al., "The Use of Natural Gas and Oxygen in the
Stack and Tuyeres of an Experimental Blast Furnace, " AIME Ironmaking
Proc. 28 (1968).
19. National Air Pollution Control Administration, Environmental Health
Service, Public Health Service, U.S. Department of Health, Education,
and Welfare, "Air Pollutant Emission Factors," Publication No.
PB 206 924, under Contract No. CPA 22-69-119. Washington, D. C. :
U.S. Government Printing Office, April 1970.
20. Peart, J. A. and Pearce, F. J. , "The Operation of a Commercial
Blast Furnace With a Prereduced Burden," J. Metals 17, 1396-1400
(1965) December.
21. "Prereducing Iron Ore Attracts Steelmakers, " Chem. Eng. News 50,
35,36 (1972) April 1?.
22. Rowe, A. D., Jaworski, H. K. and Bassett, B. A., "Waste Gas
Cleaning Systems for Large Capacity Basic Oxygen Furnaces, " Iron
Steel Eng. 47, 74-90 (1970) January.
23. Schwabe, W. E. and Robinson, C. G., "Report on Ultrahigh Power
Operation of Electric Steel Furnaces," J. Metals 19, 67-75 (1967)
April.
24. Speight, G. E. , "Air Pollution Control in Iron and Steel .Industry, "
Steel Times 200, 395-407 (1972) May.
25. Strassburger, J. H., "Blast Furnace Oxygen Operations, " Yearbook
AISI, 175-205 (1956).
26. Tenenbaum, M. and Luerssen, F. W., "Energy and the U.S. Steel
Industry, " in Proceedings of the 5th Annual Conference of the
International Iron and Steel Institute, 86-101, Toronto, Ont. ,
October 1971.
27. Thansky, D. P., Pollution Control in Steelmaking; Fact or Fiction?
Rand Corporation, January 1971.
28. Tihansky, D. P., "A Cost Analysis of Waste Management in the
Steel Industry, " J. Air Pollut. Control Assoc. 22, 335-41 (1972) May.
29. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors." Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C. , April 1973.
30. Woolf, P. L., ''Blast Furnace Operation With Prereduced Burdens,"
J. Metals JUS, 243-47 (1966) February.
31. Woolf, P. L. , "Evaluation of a Formcoke for Blast Furnace Use,"
U.S. Bureau of Mines Report of Investigations No. 6717. Washington,
D. C. : U. S. Government Printing Office, 1966.
IX-90
-------
X. SIC CODE 3331 - PRIMARY COPPER
The primary nonferrous metals industry consumes approximately
3.5% of the total energy consumed by industry annually in the U.S. The
study is being limited to the copper, lead, zinc, and aluminum indus-
tries'* because these industries consume approximately 95% of the total
energy consumed by the primary nonferrous metals industry as a whole.
Based on the typical energy consumption per ton of metal produced for
these primary nonferrous metals, an estimated 516 trillion Btu of energy
was consumed in 1971.
SIC Code 3331 pertains to companies that are involved in the smelt-
ing of copper from raw ore and in the refining of copper by electrolytic
or other processes. This classification does not include companies that
engage in secondary smelting, that is, the smelting of scrap and recycled
copper, nor does it include companies that engage in the rolling, drawing,
or extruding of copper.
During the 1960's, growth of the primary copper industry was min-
imal; the annual growth rate was only 1. 2% (Figure 1). Production in
i960 was about 1,550,000 tons, increasing to 1,775,000 tons in 1971. 3
In spite of this low growth rate, annual production is expected to increase
to 3,250,000 tons by 1985.
Energy consumption by primary copper smelters and refiners in
1971 was 71.4 trillion Btu. At current operating efficiencies, annual
energy consumption will be approximately 130 trillion Btu by 1985, an
increase of more than 80% over 1971. However, the potential for re-
ducing energy consumption is very good. Two new processes that have
been developed for copper production claim a 66% reduction in the energy
required to produce a ton of copper concentrates.
Sections XI, XII, and XIII discuss SIC Codes 3332, 3333, and 3334, the
lead, zinc, and aluminum industries, respectively.
X-l
-------
(A
C
o
10
O
3000
2500
z
o
I-
o
o
Q
O
cr
Q.
? 2000
1500
1000
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 1986
YEAR
A-93-1340
Figure 1. ANNUAL COPPER PRODUCTION
(Projected After 1972)
One significant factor in the growth of the copper industry will be its
ability to control sulfur oxides (SO ) emissions from the smelters. The
most reasonable solution to this problem is the conversion of the SOX in
the flue gases to sulfuric acid or sulfur, which then can be sold commer-
cially. However, the success of this approach is limited by the concen-
tration of SO in the flue gases, which, in turn, affects the economics.
J\.
Primary Copper-Manufacturing Processes
Primary copper production comprises a chain of processes in which
the raw material is concentrated progressively to a high-purity final
product. Figure 2 is a process flow sheet for primary copper produc-
tion. Typical energy consumption figures for each process also are
presented. About 80% of the world's primary copper is produced by
these well-established concentrating, smelting, and converting techniques.8
Although this percentage is likely to decrease, these processes probably
will continue to dominate in the future.
X-Z
-------
REVERBERATORY
SMELTING FURNACE
I
cast & cool
cast blister cake
(for shipment)
FIRE REFINING
FURNACE
molten
copper
cast into
anodes /" 400 KWH
I \~~\ per ton
CAST COPPER INGOTS,
SLABS, WIRE BARS
Figure 2. FLOW DIAGRAM FOR PRIMARY
COPPER PRODUCTION (Source: Ref. 1)
X-3
-------
Concentration
Because of the diminishing availability of rich ores, most of the
copper-bearing materials, primarily low-grade sulfide ores, must be
concentrated to make recovery of the copper profitable. Thus, the in-
troduction of the froth flotation process in the 1920's made practical the
treatment of vast deposits of low-grade sulfide ore. This process is
the most important means of concentrating the copper in ore and is not
likely to change in the near future. Concentrators built during the next
decade will not differ very much from those operating now. Because
much of the ore being processed contains only 2-3% copper, it would be
very uneconomical to ship the ore in this condition. Thus, initial con-
centrating usually is carried out at the mines. Typically, the end product
of froth flotation is 25% copper. No significant amount of energy is
consumed by this process.
Roasting
Roasting, done in the copper plant, is the next step in the concen-
tration of the ore. The purpose is to remove some of the sulfur and
iron in the ore by heating the ore to the temperature necessary for the
oxygen in the air to react, forming gaseous sulfur oxides and solid
metal oxides. Because iron oxide is formed preferentially to copper
oxide, the copper remains as copper sulfide ore, to be converted in a
later step. Roasting is an autogenously heated process, consuming
1.0 million Btu of energy per ton of copper for startup.
Roasting is used optionally to treat ores that are too high in iron
and sulfur concentrations. Consequently, the energy consumed by this
process nationwide cannot readily be determined. However, if roasting
were universally applied, the annual energy consumption would have been
about 1.80 trillion Btu in 1971, or about 3% of the total energy consumed
by the copper industry.
Smelting
Copper smelting has changed little during the last century; it is still
basically matte (a type of copper concentrate) smelting. The changes
that have occurred include the use of different sources of matte from the
blast furnace to the reverberatory furnace because of changes in smelter
feed and the replacement of one type of converter by another.4
X-4
-------
In the typical smelting operation, the copper concentrate as received
from the mine or after roasting is charged into a reverberatory furnace.
Siliceous fluxes are added, and the charge is melted at 2500°F to form a
copper matte (molten copper sulfide) and an iron silicate slag. The
copper matte is periodically tapped from the furnace and transferred to
the converter, where in a two-stage operation it is oxidized with air.10
Molten converter slag, produced in the subsequent converting operation,
is recycled back to the reverberatory furnace as well. When charged
into the furnace, it mixes and interacts with the bath, resulting in the
recovery of a large part of its high copper content. The reverberatory
slag is tapped from the furnace periodically for disposal.
'%% ' .
'""ithe energy requirement for smelting varies considerably depending
on a number of factors, including furnace design, operating practice,
moisture content, and flux additions. The most important factor is feed
grade, that is, the percent copper in the concentrate. Energy consumption
varies from approximately 60 million Btu/ton of copper using a 10%
concentrate feed to 15 million Btu/ton for a 40% concentrate, based on
a fuel requirement of 6 million Btu to smelt a ton of feed. On this basis,
a reasonable average figure for energy consumed is about 18 million Btu/
ton of copper. The majority (90%) of the fuel consumed by this process
is natural gas; the remaining 10% is fuel oil.
Converting
The copper matte product of the reverberatory furnace is reacted in
a two-stage process in oxidizing converters to produce blister copper
with a 98% copper content. In the first stage, iron sulfide in the matte
is oxidized, and the resulting iron oxide is combined with silica flux to
form an iron silicate slag. This slag then is skimmed and recycled
back to the reverberatory furnace. The second-stage conversion is the
oxidation of copper sulfide to metallic copper. The two-stage process is
imposed by the thermodynamic requirement that most of the iron sulfide
X-5
-------
be oxidized before the metallic copper can be produced. No fossil-fuel
energy is consumed directly by this process. Oxidation is promoted by
the blowing of air or oxygen-enriched air through the matte at the temper-
ature of the matte as received from the reverberatory furnace.
Refining
Blister copper is refined in furnaces heated with natural gas. These
furnaces consume about 1. 0 million Btu/ton with molten blister copper
and 4. 0 million Btu/ton if the blister is cold and must be remelted. The
bulk of the copper is refined electrolytically. The copper cathodes, or
product pigs, from the fire refinery are later remelted and cast into
forms suitable for metalworkihg operations. The melting process re-
quires about 4. 0 million Btu/ton of metal. The primary fuel used in
the melting process is natural gas.
Energy Utilization Pattern
The above analysis of the production of primary copper indicates
that approximately 23.0 million Btu is consumed in producing 1 ton of
copper. However, the 1972 Census of Manufactures cites annual energy
consumption as being 66.9 trillion Btu,11 whereas primary copper produc-
tion was 1,775,000 tons. On this basis, the average consumption of
energy per ton of primary copper produced is 37.7 million Btu. This
difference of 14.7 million Btu per ton of finished copper indicates that,
during periods of no production, a significant amount of energy is used
for such things as maintaining furnaces at temperature. Furthermore,
the production of copper is not a continuous process; consequently, much
energy is wasted.
New Technologies in Copper Manufacturing
In recent years, several new processes for producing copper have
been proposed and are being developed: flash smelting, continuous smelt-
ing by means of the Worcra and Noranda Processes, electric furnace
smelting, and various hydrometallurgical techniques as alternatives to
smelting. In all these processes, the energy consumption per ton of
copper produced is less than that of the processes they replace.
X-6
-------
Noranda and Worcra Processes9*10
The Noranda and Worcra Processes are alternatives to the typical
reverberatory and converter smelting of copper. Both processes offer
a means for continuously smelting and converting copper. In addition,
both processes combine the smelting function of the reverberatory furnace
and the converting function of the converter into a single furnace. The
key to developing these processes was to establish conditions such that
the smelting of concentrate at a high rate and the oxidation of iron sul-
fide to iron oxides and copper sulfide to copper could proceed simulta-
. neously in the same furnace. Figure 3 is a cross-sectional view of the
Noranda Process reactor or furnace, and Figure 4 shows two types of
Worcra furnace designs.
,?*=, *"-' "-
__1, j .,*:-, .ti*
, M
ntr
Figure 3. LONGITUDINAL CROSS SECTION OF
NORANDA PROCESS PILOT PLANT REACTOR10
The reactor bath in the Noranda Process consists of slag, high-
grade matte, and metallic copper layers. The copper concentrate and
silica flux are distributed on the bath surface in the smelting and con-
verting zone. This zone is strongly agitated by air or oxygen-enriched
air blown through the tuyeres. Because the bath is dynamic, even in
the presence of substantial quantities of iron sulfide, copper is produced
and settles to the bottom of the reactor. The slag is formed by combi-
nation of the flux with the oxidized iron.
X-7
-------
This process is very efficient in terms of heat utilization because
the heat evolved by the converting reactions is used for smelting. The
additional heat that may be needed is supplied by burning fossil fuels.
Energy consumption by this process in a pilot-plant facility ranged from
5. 5 to 6. 2 million Btu/ton of copper produced,10 which is a 66% reduc-
tion in the energy consumption typical of the reverberatory furnace-
converter process. For a commercial plant, the energy consumption is
estimated to be even lower, approximately 3.4 million Btu/ton of copper.9
The proponents of the Worcra Process claim that energy consumption is
even lower than for the Noranda Process, approximately 2. 5 million Btu/
9
ton of copper. The Noranda Process has been commercialized in Canada,
but operating data are not yet available. An additional feature of both
processes is that the off-gases contain enough sulfur dioxide to make
recovery for conversion to sulfuric acid feasible.
Figure 4. TWO WORCRA FURNACE DESIGNS - (Left)
A PLAN VIEW OF THE U-SHAPED SEMICOMMERClAL
FURNACE AND (Right) PLAN AND ELEVATION VIEWS
OF THE STRAIGHT-LINE COMMERCIAL FURNACE9
Flash Smelting
In the flash smelting process, oxidation reactions are carried out in
a vertical shaft in a downward flowing cocurrent system. Preheated air
and dried concentrate are fed in proper proportions into a burner on the
top of the reaction shaft. The air and concentrate are effectively mixed
in the burner, and the resulting suspension is directed vertically down-
ward into the shaft. When the particles enter the hot shaft, ignition takes
place instantaneously. By exothermic reactions, the temperature of the
particles is raised to the smelting temperature. Ultimately the particles
are collected in the molten bath located horizontally below the reaction
X-8
-------
shaft. Iron oxide and silica from the feed react, forming slag, and at
the same time, molten matte drops are separated and collected on the
furnace hearth. The matte then is charged into a converter as in a typ-
ical smelting and converting operation. 2
In general, flash smelting results in greater thermal efficiency,
higher matte grade, and richer sulfur dioxide gas. The process does
require a feed carrying 25% or more sulfur in order to maintain its
attractiveness to the industry.8
Oxygen Enrichment
Oxygen enrichment was first used as a means of increasing smelting
rates in 1952, and although the benefits appear to technically justify the
process, it is not widely used. Basically the process involves the addi-
tion of raw oxygen to the combustion air of the reverberatory furnace.
At one facility that installed such a system, a 25-30% increase in re-
verberatory furnace throughput was achieved by enriching the combustion
air to 27% oxygen. Although reliable data are not available, the use of
oxygen enrichment is believed to decrease energy consumption per ton of
copper produced.
Hydrometallurgy
During the past 10 years, the amount of copper chemically extracted
from concentrates, dumps, and on-site ores has increased considerably.
However, most of these hydrometallurgical processes apply only to copper
oxide ores or to ores consisting primarily of oxides with only small
amounts of sulfides present. To date, no hydrometallurgical process has
been successfully commercialized for the treatment of copper sulfide.
Typically, the metal is leached from the ore by a solvent and then re-
covered from the resulting solutions in a relatively pure form. One ad-
vantage of hydrometallurgy is the elimination of the sulfur dioxide pollution.
problems that normally bother smelters. The capital investment is
relatively low, and the processes are economical even on a small scale.
Thus, such a process can be set up at the mouth of a mine to turn out
refined copper instead of ore that must be shipped. In addition, the
working temperatures are relatively low, and as a result, energy require-
ments are minimal. In spite of all its advantages, hydrometallurgy is
not expected to replace any existing smelters, although it may be useful
as an adjunct to smelters.5
X-9
-------
Cathode Melting
Cathodes are usually melted in reverberatory, direct-arc, or shaft
furnaces. Until recently, most cathodes were melted in reverberatory
furnaces. One problem with the reverberatory furnace is the exposure
of the copper to contamination from refractories and from sulfur in the
fuel. Direct-arc melting has been going on for the last 20 years.
Direct-arc furnaces are more efficient melting units than reverberatory
furJiaces and less prone to contamination problems. However, with the
advent of the shaft furnace, even arc melting has become obsolete.8
The shaft furnace is the most notable development in copper refining in
many years. Its primary advantage is the ability to produce a pure
product. However, it requires the use of low-sulfur fuels and consistently
good-quality feed.
Effect of New Technology on Energy Usage
Traditionally, about 80% of the total energy consumed by the primary
copper industry has been natural gas. Thus, based on an estimated total
energy consumption of 71.4 trillion Btu in 1971, approximately 57 trillion
Btu of natural gas was consumed. By 1985, annual copper production is
estimated to be about 3. 25 million tons, and based on current levels,
energy consumption will be 130 trillion Btu. However, the probable an-
i :
nual energy consumption in 1985 will be considerably lower than current
estimates if such new technology as the Noranda or Worcra Processes
gains in usage. Currently, only 16 smelters exist in the U.S. Thus, the
conversion of one or two of these companies to a continuous smelting
process would have a marked effect on the energy'consumed by this in-
dustry. For example, if four of the 32 smelting furnaces currently in
existence were replaced by the Noranda furnace and if the smelters being
replaced produced 12. 5% of the total annual copper production, total
energy consumption by 1985 would decrease by 5%. (See Figure 5.) If
one-half of the annual copper production were to occur via the Noranda
Process, total energy consumption by 1985 would decrease by 18.5%.
We cannot predict whether any of these furnaces will be in operation by
1985. The first production-scale Noranda. furnace was scheduled for com-
pletion by early 1973 in Canada. Plans for the construction of a Noranda
furnace in the U. S. have just been announced.
X-10
-------
UJ .
si
I- H
O Q.
a:
o.
en
z
o
o
IOV
120
110
100
90
80
yn
<**
«
^
PRESENT CONDITION
2 NORANDA FURNAC
4 NORANDA FURNAC
8 NORANDA FURNAC
^2
^- "
^
^
2^
^
-^
S
ES
ES
ES
. ^
^<
-
^
..*
y^-
<^
* *
^
s^
^
f
^
^
^
1972 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
A-93-1342
Figure 5. EFFECT OF USAGE OF NORANDA FURNACE
ON PROJECTED ENERGY CONSUMPTION IN 1985
In addition, new facilities are being planned that will utilize
flash smelting, electric smelting, and hydrometallurgy. However, none
of these facilities are expected to be in operation before 1980 .and,
consequently, will not have a great impact on the industry's energy
consumption pattern between now and 1985.
Air; Pollutant Emissions in Copper Manufacturing
The major air pollutant emission from copper -manufacturing processes
is sulfur dioxide. In 1966, according to one source, a total of 2, 830, 000
tons of sulfur dioxide, or about 1. 6 tons per ton of copper concentrate
produced, was emitted by copper producers,6 primarily from roasters
(where used), reverberatory furnaces, and converters. On this basis,
approximately 2, 850, 000 tons of sulfur dioxide was emitted in 1971.
According to a second source, total SO emissions are only 1250 Ib/ton
of copper concentrate produced. On this basis, only 1,110,000 tons of
SO was emitted in 1971. The breakdown of SO emissions by process
-*t JC
is given in Table 1.
Present air pollution standards call for a 90% reduction in current
emissions. However, the establishment of this standard appears to have
been based on a study that used mathematical models. The study suggested
that the combination of sulfuric acid plants and limestone slurry scrubbing
could achieve this level. Copper producers contend that such reductions
are not possible because a commercial limestone slurry process has not
X-ll
-------
Table 1. EMISSION FACTORS FOR PRIMARY
COPPER SMELTERS WITHOUT CONTROLS*
(Source: Ref. 12) ,
Particulates" SO.
Type of Operation lb/tort-
Roasting 45 60
Smelting (reverberatory furnace) 20 320
Converting 60 870
Refining 10
Total Uncontrolled 135 1250
»
Emission factors expressed as units per unit weight of
concentrated ore produced.
t
Electrostatic precipitators reduce emissions by 99.7%.
as yet been successfully developed. Furthermore, even if the technical
problems relating to limestone slurry can be solved, a new problem would
be created by the amount of solid waste produced by this method of sul-
fur control. Copper producers contend that a 50-60% sulfur recovery
is feasible.13
The other major air pollutant from the production of copper is par-
ticulates. Table 1 also shows the particulate emissions by process from a
conventional uncontrolled copper-producing facility. The literature has
reported that 99. 7% of the particulates can be recovered by electrostatic
precipitators.
Carbon monoxide and nitrogen oxides (NO ) probably are also emitted
X
by the processes for producing copper. However, no quantitative data
have been reported in the literature.
Effect of New Technologies on Air Pollutant Emissions
Because SO emissions are of prime concern at present, the major
portion of this discussion deals only with the effects of the new techno-
logies previously discussed on sulfur dioxide emissions. The obvious
solution to the sulfur dioxide emissions problem faced by the copper pro-
ducers is to convert these emissions to either elemental sulfur or sulfuric
acid. This conversion is technically possible and, in fact, is practiced
by many of the copper producers. However, in order to make the con-
version economically feasible, the percentage of sulfur dioxide in the
X-12
-------
off-gases must be fairly high, at least 8-12%. Typically, the off-gases
from reverberatory furnaces contain only 1-2% sulfur dioxide, the roaster
gases (when used) contain about 7% sulfur dioxide, and the converter
off-gases contain an average of 4-5% sulfur dioxide.7 Of these off-gases,
only those from the roaster have been used successfully to prodiice sulfuric
acid. Thus, if new technologies are to have a beneficial effect on sulfur
dioxide emissions, they must ultimately produce off-gases with higher
percentages of SO .
Ji
As previously stated, roasters, which have declined in usage over
the years, are being revived. The main reasons for this revival are
the interests in sulfuric acid production and in air pollution control. The
new fluid-bed roasters are ideal because they operate autogenously and
the off-gases contain up to about 15% sulfur dioxide. Thus, these off-
gases can be readily converted to sulfuric acid.
For the smelting process, sulfur dioxide in the off-gases probably
cannot be increased from the 1-2% level. However, new processes that
have been developed to increase production rates and reduce fuel con-
sumption also produce greater concentrations of sulfur dioxide in the
off-gases. Flash-smelting furnaces using preheated air produce off-gases
from the furnace containing 11-14% sulfur dioxide, and these gases are
used successfully in manufacturing sulfuric acid. Such furnaces have
been in operation in Finland and Japan for many years but are not used
in the U.S.
A second method for producing high percentages of sulfur dioxide in
smelting off-gases is electric smelting. The off-gases from electric
smelting consist primarily of the gases generated from the charge, water
vapor, and air leaking into the furnace. If air leakage is minimized,
off-gas volume is minimized and sulfur dioxide concentrations in the off-
gas range from 10 to 20%. At present, electric smelting is not used in
the U.S. because of the availability and relative cheapness of other fuels.
Also, this type of smelter eliminates the generation of dust-laden off-
gases and, consequently, should decrease particulate emissions.
X-13
-------
The use of oxygen-enriched air in converters was first introduced
commercially in Japan; it has been in limited use in the U.S. for sev-
eral years. Theoretically, the concentration of sulfur dioxide in the
converter off-gases increases in proportion to the percentage of oxygen
in the air used in blowing the converter. Sulfur dioxide concentrations
in the off-gases of converters using oxygen-enriched air are reported to
range from 10-20%. If oxygen enrichment of the blast to the converter
is combined with a gas collection system comparable with that devised
for the basic oxygen furnace 'in the steel industry, gas streams containing
15-25% sulfur dioxide or even higher concentrations should be obtained. 7
A process for continuous smelting and converting of copper makes
control of SO emissions easier than in the current smelting and con-
X.
verting processes. These continuous processes include both the Noranda
and Worcra Processes. Thus far, both processes have been tested only
on a pilot-plant scale. However, the Worcra Process is reported to
produce a gas stream containing 9-12% sulfur dioxide. Presumably, if
oxygen-enriched air is used, even higher sulfur dioxide concentrations in
the off-gases seem likely. At present, no information is available on
sulfur dioxide concentrations in the off-gases from the Noranda Process,
but the results probably will be similar to those from the Worcra Process.
Finally, a number of processes that are being developed in the field
of hydrometallurgy would permit reduction of copper while eliminating
part or all of the pyrometallurgical processes that produce sulfur dioxide.
One advantage of these processes is that the by-product is elemental sul-
fur, which is very attractive in areas where markets for sulfuric acid are
limited. However, much work remains to be done on these processes.
There are several alternatives to the sulfur dioxide emission prob-
lem, and their relative attractiveness depends on several factors. Con-
tinuous smelting and converting is the most attractive solution in terms
of production and pollution control. Yet, neither this process nor any
of the others previously discussed will gain widespread usage by 1985,
primarily for economic reasons. However, if all of the sulfur dioxide
being emitted into the atmosphere by the copper industry were converted
to sulfuric acid, sulfur dioxide emissions would be reduced by 99%. This
estimate includes the sulfur dioxide emissions that are given off in the
X-14
-------
production of sulfuric acid. Figure 6 shows that if the copper industry
were to convert 7% of its production to processes producing high-sulfur-
dioxide-containing flue gases and convert this sulfur dioxide to sulfuric
acid, a 99% reduction in sulfur dioxide emissions could be achieved by
1985. However, such a conversion rate is not likely with current fuel
costs, the depreciation rate of current equipment, and the capital required
for conversion.
O.O
5.0
4.5
4.0
(A
o 3.5
"2 ,0
to"
i 25
8 2.0
2
U 1.5
CM
0
CO 1.0
0.5
n
<
^
^
^
\
^-*"
^
PRO,
AT C
RAT
^
"*»»
JECTE
:URRE
ES
^
^
ID SO
INT t
^-^
-^
2 EM
:MISS
^
ISSIO
ION
^N
^
NS
^-'
^
^
^^
^
PROJECTED S02 EMISSIONS
ASSUMING 7% AVERAGE
/^"INCREASE IN CONVERSION
S02 TO H2S04 PER YEAR
^\
^^
^^
^^
OF
\
1970 '71 '72 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-93-I34I
Figure 6. PROJECTED SULFUR DIOXIDE
EMISSIONS BY COPPER INDUSTRY
X-15
-------
References Cited
1. American Gas Association, A Study of Frocess Energy Requirements in the
Non-Ferrous Metals Industry, Catalog No. CZ0010. Arlington, Va., n.d.
2. Bryk, P. et al., "Flash Smelting Copper Concentrates, " J. Met. 10,
395-400 (1958) June.
3. "Copper Profile," in Metal Statistics 1972, pp. 79-138. New York:
Fairchild Publications, 1972. ,
4. Milliken, C. L. , "What is the Future of the Copper Smelter?"
J. Met. 2£, 51-54 (1970) August.
5. "New Boost for Hydr©metallurgy, " Chem. Week 113, 44 (1973)
April 25.
6. Rohrman, F. A. and Ludwig, J. H. , "Sulfur Oxides Emissions by
Smelters," J. Met. 2,0, 46 (1968) December.
7. Semrau, K. T. , "Control of Sulfur Oxide Emissions From Primary
Copper, Lead and Zinc Smelters A Critical Review, " J. Air Pollut.
Control Assoc. 21, 185-93 (1971) April.
8. Smith, G. A. , "Primary Copper: A Review of Methods of Production
and Quality Control," Met. Mater. 4, 461-65 (1970) November.
9. Stevens, B., "Continuous Copper Smelting," Chem. Process Eng.
j>3_, 28-30 (1972) January.
10. Themelis, N. J. et al., "The Noranda Process," J. Met. 24, 25-32
(1972) April.
11. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D.C.: U.S. Government
Printing Office, July 1973.
12. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors," Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C., April 1973.
X-16
-------
XL SIC CODE 3332 - PRIMARY LEAD
SIC Code 3332 pertains to those establishments primarily engaged in
smelting lead from the ore and in refining lead by any process. Establish-
ments engaged in the rolling, drawing, or extruding of lead are classified
elsewhere.
During the 1960's, the primary lead industry was very stable, exhibiting
minimal growth until 1968, when production began to increase at a record-
breaking pace. By 1972, total annual production of primary lead in the
U.S. had increased by nearly 50% (over 1968 production) to about 695,000
tons. 1 However, this growth rate is not expected to continue; rather, it
is expected to increase as it has during the last three decades at between
1 and 2% annually. 6 Consequently, annual production of primary lead will
be about 860,000 tons by 1985, as shown in Figure 1.
Energy consumption by primary lead producers is estimated to have
been 18. 5 trillion Btu in 1972. By 1985, total energy consumption is
expected to increase to nearly 23. 0 trillion Btu, a 24% increase over 1972.
These estimates are based on an average energy consumption bf 26. 7 million
Btu/ton of primary lead produced. 10 However, the newly published figures
for 1972 indicate that energy consumption was only 13. 3 trillion Btu, 11 or
a unit consumption rate of about 19.2 million Btu/ton of lead produced.
This indicates that a considerable improvement in the efficiency of fuel
utilization has been achieved by lead producers since 1967. Further
reductions in unit energy consumption are not likely by 1985.
Primary Lead-Manufacturing Processes
Primary lead production comprises a chain of processes in which the
ore concentrate is beneficiated, smelted, and purified to produce 97%
pure lead bullion. A schematic process flow sheet for primary lead pro-
duction is presented in Figure 2.
Sintering
The first step in the manufacture of primary lead is sintering, in
which the lead concentrates, lead ores, and roasted dusts from electro-
static precipitators are combined with about 1% coke fines by weight and
roasted in a Dwight-Lloyd sintering machine. Sintering is a universal
XI-1
-------
V)
o
*^
8
O
O
O
O
-------
ORE
CONCENTRATE
DWIGHT LLOYD
SINTERING MACHINE
DE-COPPERIZED BULLION
ZINC-
; DROSS CONTAINING
'.ARSENIC, ANTIMONY,
COPPER
LEAD
DE-SILVERING
KETTLE
-»-SILVER RICH ZINC
DE-SILVERIZED LEAD
DE-ZINCING
KETTLE
--ZINC RICH DROSS
PURE REFINED LEAD
(to Manufacturing Processes)
Figure 2. FLOW SHEET FOR PRIMARY AND SECONDARY
LEAD PRODUCTION 2 '
XI-3
-------
although oil can also be used. On this basis, the sintering of lead concentrates
'i ' '
consumed about 1. 56 trillion Btu of energy in 1972.
Smelting
The smelting of lead is generally a two-phase operation in which the
feed material, sinter, is charged into a blast furnace and the blast-furnace
output is refined in dross reverberatory furnaces and lead-refining kettles.
Blast-Furnace Operation
The purpose °f the blast furnace in the manufacture of lead is to reduce
the lead oxide in the sinter to-metallic lead. For this process, blast-furnace
operation is similar to that used in the steel industry, except that furnace
temperatures rarely exceed 1700°F, which is considerably lower than that
used in the steel industry.
The furnace is principally a shaft furnace with water pockets for
cooling. The tuyeres through which the combustion air is introduced are
located near the furnace bottom. The furnace charge consists of a mixture
of sinter and metallurgical coke, limestone, silica, and slag-forming
materials. Coke makes up 8-15% of the charge, and sinter makes up 80-90%
of the charge. 7 The furnace uses carbon monoxide derived from the coke
feed to reduce the sinter to bullion. Most of the impurities are eliminated
in the slag.
; The use of oxygen in the blast furnace is beginning to increase fairly
rapidly. Oxygen enrichment of the blast air results in significant reductions
in coke consumption and large increases in smelting rate. As early as
1949, tests were conducted with oxygen enrichment. The results of these
tests showed a 10% reduction in coke consumption and a 15-20% increase
in smelting rate at an oxygen enrichment level of 2. 5%. 3 Other studies
conducted since then have shown coke consumption reductions as high as
16% and smelting rate increases as high as 53% for a 5% enrichment
level. Figure 3 shows the effect of oxygen enrichment on coke rate and
smelter rate as determined by a number of investigators over a period of
years. However, data on the number of lead producers currently using
oxygen enrichment are not available, and consequently, neither are data
on average energy consumption. Without oxygen enrichment, this process
consumes an estimated 7. 5 million Btu/ton of lead produced. 8 Thus, on this
XI-4
-------
01234
OXYGEN ENRICHMENT LEVEL, %
A-44-575
Figure 3. EFFECT OF OXYGEN ENRICHMENT OF BLAST AIR
ON SMELTER RATE AND COKE RATE OF LEAD SMELTER
basis, approximately 5. 2 trillion Btu of energy was consumed by this
process in 1972.
Furnace products discharged from the bottom are lead metal, matte,
speiss, and slag, all in liquid form. In some plants, the first three prod-
ucts are sent to lead refining without separation from each other. In other
plants, the lead is separated and sent to refining, and the speiss and matte
go to a dross furnace. Slag is removed separately and conveyed to a fuming
furnace for recovery of lead and zinc. Some slag may be gound up and re-
cycled into the sintering process for flux.
Refining
The lead-refining process in a primary lead smelter consists of a dross
reverberatory furnace and several lead-refining kettles. During the course
of this process , impurities such as arsenic, autimony, zinc, silver, and
copper (if it has not already been separated) are removed. Assuming that
the copper has already been removed, the bullion is softened in a rever-
beratory furnace in which the arsenic, antimony, and residual copper are
XE-5
-------
removed. In addition, small amounts of sulfur are oxidized. The lead then
is transferred to a heated kettle to which zinc is added to combine with any
silver that may be present in the lead. The silver-rich zinc is removed,
and the desilverized lead is transferred to another heated kettle, in which
the remaining zinc is oxidized and removed. The resultant product is
virtually pure lead.
Typically, the refining process consumes about 5. 0 million Btu/ton
of lead. 8 On this basis, about 3. 5 trillion Btu of energy was consumed for
refining of lead in 1972. Of this amount, approximately 90% was consumed
as natural gas; the remaining 10% was fuel oil.
Energy Utilization Pattern
According to the above analysis of process energy consumption, approxi-
mately 13. 25 million Btu of energy is required to manufacture 1 ton of primary
lead. As in other industries, the actual energy consumed was considerably
higher than the energy required. In 1972, 13. 2 trillion Btu of energy was
consumed in the manufacture of primary lead. u This is about 45% higher
than the amount of energy that is calculated by summing typical processes
consumption, which is indicative of the amount of energy that is wasted
and used for miscellaneous process segments. Note however, that according
to Bureau of Census statistics, 10» n this industry seemingly has made sig-
nificant strides in a short time to eliminate the causes of inefficient fuel
uses, as evidenced by the decrease in unit energy consumption to 19. 2 million
Btu/ton in 1972 from 26. 7 million Btu/ton in 1967. Of the total amount of
fuel consumed, coke, coal, and natural gas account for more than 90%.
New Technologies in Lead Manufacturing
Beginning with the initial processing of the concentrate as received by the
plant, the major new technology revolves around the development of an up-
draft sintering machine. Updraft sintering has been known since 1898, when
it was used in sinter pots. However, with the advent of continuous sintering,
updraft sintering was found difficult. Consequently, the downdraft process
became common practice. In the downdraft process, a draft of air is blown
down through the blended charge in the sintering machine after it has been
ignited on the moving grate. In the updraft process, the draft of air is
blown up through the charge.
XI-6
-------
The updraft process offers several advantages over the downdraft
process: higher output, less equipment wear, and lower power consumption. 13
The reduction in power consumption is achieved by increased air output at
lower pressures, resulting in a decrease in power consumption by the sinter
fans. The reduction in power consumption is estimated to be about 20%.
Actual consumption figures are not available. Updraft sintering offers other
advantages such as greater product homogeneity,and higher lead content.
Such a sinter would promote the smelting operation in the blast furnace.
Updraft sintering also results in increased sulfur removal, thus resulting in
the production of off-gases suitable for direct conversion into sulfuric acid.
In recent years, updraft sintering has begun to gain usage in the U. S. to the
point that most companies currently use it.
Direct Smelting
The idea of single-step smelting has been around for many years, but it
is only recently that research has made such a process feasible. Advantages
of the process include the elimination of the sinter plant, a decrease in fuel
consumed for heating and reduction, the elimination of the recirculation of
diluent (such as blast-furnace slag), the increase of sulfur dioxides in the
off-gases to make recovery as sulfuric acid economically attractive, and
the use of concentrates as received without special treatment except for drying. 5
The basic principle of the process is'the oxidation of lead sulfide to
form lead and sulfur dioxide. For this process, a special reaction vessel
was designed. At the start of a compaign, hot refined lead is charged into
the vessel. At this point, the air/gas burner is lit and the vessel rotated
such that the lead covers the tuyeres. Feeding of the lead concentrates is
initiated, and air is'Mown through the bath and concentrates by means of
the tuyeres. The result is the formation of lead and slag, which are then
withdrawn from the converter.
Energy requirements for this process are minimal. On a theoretical
basis, the energy required by this process is 0. 14 million {itu/ton of
concentrates, assuming no heat losses. A more realistic estimate of energy
requirements would be 0. 4 million Btu/ton of concentrates, assuming no
heat losses.
This process has been tested on an experimental converter, but has not
been completely developed as yet. The main problem is the unusually high
XI-7
-------
refractory wear, expecially in the tuyere area of the reaction vessel. Some
work to develop suitable refractories is being performed; however, no
solution has yet been found. ;
Flash Smelting4
The use of flash smelting of lead concentrates is relatively new,
although flash smelting is used in the copper and nickel industries. (For
an explanation of flash smelting, see the section entitled "Flash Smelting"
under SIC Code 3331 Copper.) Flash smelting of lead has been accomplished
on a pilot-plant scale, demonstrating that lead smelting can be performed
continuously without sintering. Although no operating figures are available,
the indications are that total fuel consumption is relatively low. Another
advantage of this process is that very high grade lead concentrates can
be used without dilution by return slag. The only fluxes needed are those
required to make a suitable slag composition with a gangue of the concentrate.
This results in a much smaller slag fall and lower lead losses in the slag
than in conventional melters.
As in flash smelting in the copper industry, thermal efficiency is increased
and richer sulfur dioxide gas is given off. Thus, treatment of the sulfur dioxide
in acid plants is more attractive. In spite of these advantages, flash smelting
of lead is hot in use in the U. S. at present.
Boliden Process3
The main production unit in the Boliden Process is an electric hearth
furnace. In this process, electrical energy supplies the heat necessary to
carry out the smelting of lead-rich concentrates. The only material prep-
aration required is drying of the concentrates.
In the Boliden furnace, electrodes are submerged in a molten bed of
lead and slag. High-voltage current is forced through the electrical circuit
formed by the electrodes and slag. The resistance to the current flow by the
slag layer provides the necessary heat of reaction.
Compared with conventional processes, the Boliden process offers the
advantages of low capital cost, avoidance of metallurgical coke, recovery
of waste heat in the form of steam, and production of a gas rich enough in
sulfur dioxide to make sulfuric acid production feasible. Reportedly, energy
consumed in this process is about 2. 75 million Btu/ton of lead refined.
XI-8
-------
If fossil fuel consumption for generation of the electricity is considered,
total energy consumption is about 9. 2 million Btu/ton of lead refined.
Hydrometallurgy
As in the copper industry, hydrometallurgical processes for lead con-
centrating and refining have been developed. However, no commercial plants
exist nor are any expected to be built, at least until the design and con-
struction problems surrounding the leaching step are solved. In addition,
recovery and separation of minor components are anticipated to be more
difficult than in pyrometallurgical processes. In spite of the existing
problems, there is no doubt that commercialization will eventually occur
because the incentives in terms of reduced energy utilization and air pol-
lutant emissions are strong.
Effect of New Technologies on Energy Utilization Pattern
In spite of the availability of new processes for lead manufacturing,
no significant changes are expected in the foreseeable future. Consequently,
the pattern of energy utilisation is not likely to change either. However,
at present, either the Boliden process or a direct smelting process would
be acceptable to the industry if a change were to be made. From the
industry's point of view, widespread usage of either process would result
in considerable reductions in energy consumption. Figure 4, for example,
shows the potential effect of a direct smelting process on total energy con-
sumption as a function of the utilization of such a process by the industry
and compares this projected energy consumption with the projected
annual consumption in 1985. The Boliden process also would result in
significant reductions in energy consumption from the industry's point of
view if it were to be widely accepted. However, if the energy consumed for
electricity generation also is considered, the reduction in energy consumption
from current practice would be insignificant.
Air Pollutant Emissions From Lead-Manufacturing Processes
The primary air pollutant emissions from lead-manufacturing processes
are sulfur dioxide and particulates. In 1966, an estimated 146,000 tons of
sulfur dioxide was emitted by lead producers, or 0. 33 ton/ton of lead pro-
duced. On this basis, an estimated 229,000 tons was emitted in 1972.
These data are in good agreement with emission data previously published
by the EPA (Table 1).
XE-9
-------
OTAL ANNUAL ENERtf
:ONSUMPTION, I012 Bti
._ _ iv> rv> f\
> o> oo o ro 4
k
^
^
._
' E
L_
NERG
1
-^
»ROJ_
Y CO
k
|CT§
NSUN
»^^
^v
D_rqi
IPTIO
k^
PAL
TIN
^^>
L
1985
1
% LEAD PRODUCED BY DIRECT SMELTING
A-44-574
Figure 4. EFFECT OF UTILIZATION OF DIRECT SMELTING
ON TOTAL ENERGY CONSUMPTION
Table 1. AIR POLLUTANT EMISSION FACTORS
FROM PRIMARY LEAD SMELTERS12
Particulates'
Sulfur Oxides
Process
Sintering and Sinter
Crushing
Blast Furnace
Reverberatory
Furnace
IH/f nn
50°
75
660
d
12
Electrostatic precipitators collection efficiency, 96%;
baghouse collection efficiency, 99%.
Emission factors expressed as units per unit weight of
concentrated ore produced.
Pounds per ton of sinter. ;
Overall plant emissions for SO , 660 Ib/ton of ore produced.
XI-10
-------
The main source of sulfur dioxide emissions is the sulfur in the raw
ore. Compared with that in copper and zinc ores, the concentration of sul-
fur in lead ore is relatively low, about 0. 155 Ib/lb of metal. 6 The primary
method of control, as in the copper industry, is the recovery and conversion
of the sulfur dioxide to sulfuric acid. To make this conversion feasible,
concentrations of 6-7% sulfur dioxide in the effluent gases are required.
Typically, the sintering operation is the major source of sulfur dioxide
emissions. In this process, 85% of the total sulfur is eliminated from
the feed as sulfur dioxide. Off-gases from a smelter using updraft sintering
machines can contain 4-6% sulfur dioxide. However, flue gas recirculation
is required to achieve these concentrations because most machines currently
in use have large air leaks that dilute the sulfur dioxide to concentrations
of 1-3%.
The lead blast furnace is a relatively minor, but quite variabiej source
of SO emissions. Concentrations from 0. 005 to 0. 018% have been reported
on one furnace, whereas concentrations as high as 6. 05% have been reported
on others.9 In general, SO emissions from the blast furnace vary in accor-
Jt
dance with the percentage of sulfur in the feed, which is usually less than
15% of the sulfur originally in the ore. The SO emissions from .refining
Jf,
operations are negligible.
The sintering operation and the lead blast furnace are the primary
sources of particulate emissions, as shown in Table 2. Particulates from
the sintering operation are primarily lead, zinc, and sulfur. Particulates
emitted from the blast furnace are a combination of sulfates, oxides, lead
sulfide, and coke dust. 6 Particulates from the sintering operation are re-
covered from the effluent gas stream by settling in large flues and by
electrostatic precipitators and baghouse filters (Table 3). Collection
efficiencies are up to 96% for precipitators and 99- 5% for baghouses.
After cooling, the blast-furnace effluent is cleaned in baghouses by
using wool or fiber-glass bags. The collected dust usually contains a
high concentration of lead as well as quantities of cadmium and arsenic.
Consequently, the collected dust usually is recycled to the sintering
machine.
XI-H
-------
Table 2. PARTICULATE EMISSIONS FROM
LEAD-MANUFACTURING PROCESSES6
Process
Emission
Factor
Collection
Efficiency
Emissions,
tons/yr
Ore Crushing
Sintering
Blast Furnace
Dressing Kettle
Softening Furnace
Desilvering Kettles
Cupeling Furnaces
Refining Kettles
Dross Reverberatory
Furnace
Materials Handling
Total
2 Ib/ton ore
520 Ib/ton lead
250 Ib/ton lead
20 Ib/ton lead
5 Ib/ton lead
0.0
0.86
0. 83
0. 50
0.32
6,950
25,298
14,769
3,475
1, 182
51,674
Table 3. LEAD SMELTER CONTROL EQUIPMENT6
Efficiency
Process
Sinter Machine
Blast Furnace
Reverberatory
Furnace
Control Device
Secondary
Electrostatic preci-
pitate r, bag filter
Electrostatic preci-
pitate r
Waste-heat Electrostatic preci-
boilers, tubular pitator, bag filter
coolers
Primary
Centrifugal
Centrifugal
Primary Secondary
80-90
80-90
70-80
95-99
95-99
95-99
Effect of New Technology on Air Pollutant Emissions
Sulfur dioxide emissions will reach an estimated 284,000 tons/yr
by 1985 if current emission rates continue unchecked. However, conversion
of these emissions to sulfuric acid would reduce this figure by 99% to a mere
2840 tons/yr including sulfur dioxide emissions from the acid plant itself.
However, the emission levels required to meet this figure are not expected
to be reached in the foreseeable future.
XI-12
-------
In the more distant future, the development and satisfactory commer-
cialization of the direct smelting of lead sulfide concentrates will drastically
reduce sulfur dioxide emissions. Such a process would eliminate the roasting
and sintering process, the major source of sulfur dioxide. Other new pro-
cesses include converting the concentrates to metal by air blowing, which
produces off-gases containing 13-14% sulfur dioxide; flash smelting, which
produces off-gases containing 10-12% sulfur dioxide; and the Boliden process
for direct smelting of concentrates in an electric furnace, which produces
off-gases containing 6% sulfur dioxide. In all these processes, the gases
would be fed into an acid-manufacturing process. As previously stated,
none of these new technologies is likely to be in widespread usage before the
year 2000; thus, their contribution to air pollutant emissions reduction in
the interim should be minimal.
References Cited
1. American Bureau of Metal Statistics, Year Book for 1972. New York,
June 1973.
2. American Gas Association, A Study of Process Energy Requirements
in the Non-Ferrous Metals "industry^ Catalog No. C200iO]
Arlington, Va. , n. d.
3. Anderson, J. N. and Queneau, P. E. , Ed., Pyrometallurgical Processes
in Nonferrous Metallurgy. New York: Gordon and Breach, 1967.
4. Bryk, P. , Malmstrom, R. and Nyhohn, E. , "Flash Smelting of Lead
Concentrates, " J. Met. 1_8, 1298-1302 (1966) December.
5. Fuller, F. T. , "Process for Direct Smelting of Lead Concentrates, "
J. Met. 20, 26-30 (1968) December.
6. Jones, H. R. , "Pollution Control in the Nonferrous Metals Industry,"
Pollution Control Review No. 13, Park Ridge, N. J. : Noyes Data
Corporation, 1972.
7. McKay, J. E., "Lead, " in Kirk-Othmer Encyclopedia of Chemical
Technology, Vol. \2_, 2nd Ed. , 207-47. New York: John Wiley , 1967.
8. Rosenberg, R. B. , "Energy Use for Industrial Heat and Power and New
Process Developments for Conservation, " Final Report, IGT Project
8938. Chicago: Institute of Gas Technology, August 1972, Revised
October. 1972. ,
9. Semrau, K. T. , "Control of Sulfur Oxide Emissions From Primary
Copper, Lead and Zinc Smelters A Critical Review, " J. Air Pollut.
Control Assoc. 2±, 185-93 (1971) April.
XI-13
-------
10. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1967 Cenaua of Manufactures, Special
Report No. MC67(S)-4. Washington, D. C. : U. S. Government
Printing Office, June 1971.
11. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C. : U. S. Government
Printing Office, July 1973.
12. U. S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-42, 2nd Ed. Research
Triangle Park, N.C., April 1973.
13. Wundelrom, H. B. , Peucker, M. O. , and Massion, W. P. , "Updraft
Sintering of Lead Concentrates," J. Met. 11, 748-51 (1959) November.
XI-14
-------
XII. SIC CODE 3333 - PRIMARY ZINC
SIC Code 3333 pertains to those establishments engaged in smelting
zinc from ore and in refining zinc by any process. This classification
does not include establishments engaged in secondary smelting, nor does
it include establishments engaged in the rolling, drawing, or extruding
of zinc.
Prior to 1969, the zinc industry exhibited a moderate growth rate
of about 4% annually. However, after 1969, production declined sharply
because of the closing of seven smelters that accounted for about 30%
of the' 1969 U.S. smelter capacity.1 (See Figure 1.)
1000
-. 900
g 800
O
E 7°°
o
£ 600
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 1986
YEAR
A-44-569
Figure 1. ANNUAL PRIMARY ZINC PRODUCTION
WITH PROJECTION TO 1985 .
Most of these smelters were shut down because of outmoded equipment
and operations, in addition to the burden of environmental cleanup. An
additional 20% of the 1969 smelter capacity is currently in jeopardy of
being shut down for the same reasons. In spite of these setbacks, the
future of the industry is bright, as the domestic demand for zinc is
expected to increase at the rate of about 3. 5% annually. This increase
in demand will help support the domestic price of zinc and thus tend to
motivate new entries into the industry. Consequently, by 1985, domestic
production of primary zinc slabs should approach 1. 050 million tons/yr,
a 53% increase over 1972 production.
XII-1
-------
Energy consumption by primary zinc smelters and refiners is
estimated to have been about 38. 0 trillion Btu in 1972. This is based
on an average unit energy consumption rate of 55. 2 million Btu/ton of
zinc produced.12 In spite of the expressed optimism about the future of
the industry, zinc technology is not expected to offer any significant im-
provements in production practices. As a result, the unit energy con-
sumption rate is not likely to change from current practice. However,
the figure for the energy required to produce 1 ton of zinc given above
is an average figure representing four current major zinc-smelting pro-
cesses that have widely varying energy requirements. Two of these
processes are expected to disappear altogether, and utilization of the two
remaining processes, being more efficient from an energy standpoint,
should result in a decrease in average unit energy and consumption and
pollutant emission.
Primary Zinc-Manufacturing Processes
Four major extraction processes are available for producing zinc,
all of which are currently used in the U. S: the horizontal retort, the
vertical retort, the electrothermic process all of which are pyrometal-
lurgical and the electrolytic process, which is hydrometallurgical.
Up to a point, all four systems contain three similar processing steps:
1) the initial sulfur removal is carried out in roasters, 2) further im-
purities are removed by sintering in the pyrometallurgical methods and
by leaching in the electrolytic process, and 3) the prepared zinc-bearing
material is fed to final, and divergent, metal-producing units for which
slab processes are named. (See Figure 2.)
Ore Concentrate Preparation
Prior to the extraction of zinc, the ore concentrates are prepared
by any of several different combinations of processes. The choice of
method depends on the physical and chemical properties of the raw material;
and on the extraction process that is to be used. Zinc extraction requires
that the raw material, primarily zinc sulfide, first be processed to con-
vert the zinc to zinc oxide. This conversion is accomplished by roasting,
sintering, and/or calcining operations. Some plants both roast and sinter
the zine sulfide concentrates before extraction, whereas other plants use
only sinter machines. The latter plants, in effect, are roasting and
XII-2
-------
DISTILLATION
PROCESSES
ORE
CONCENTRATE ELECTROLYTIC
PROCESS
I
ROASTING
FURNACE
,
CHEMICAL
LEACHING
ELECTROLYTIC
CELLS
CATHOD
ZINC
MELTING
FURNACE
ULTRA-PURE SLAB
ZINC
LEAD BLAST
FURNACE SLAG
Figure 2. FLOW SHEET FOR PRIMARY AND
SECONDARY SLAB ZINC PRODUCTION2
XII-3
-------
sintering in one step. Calcining is performed only on oxide ores or on
material that has previously been oxidized by roasting.
Roasting and Sintering
The first step in the preparation of the zinc sulfide ore is the
roasting operation, in which the ore is mixed with coke and heated
autogenously. Operating conditions for roasting vary from plant to
plant according to the composition of the raw material and the specific
use of the roasting calcines. Table 1 summarizes typical zinc roasting
operations currently being used and shows the various ranges for all
the main operating parameters.
Table 1. TYPICAL ZINC ROASTING OPERATIONS3
Operating Feed Capacity, Dust in Off-Gas, Off-Gas
Type of Roaster* Temp, °F tons/day % of feed SO2, %
'0
Multiple Hearth 1200-1350 50-120 5-15 4.5-6.5
Multiple Hearthb 1600-1650 250 5-15 4.5-6.5
Roppc 1200 40-50 5 0.7-1.0
Fluid Bedd
(Dorr-Oliver) 1640 140-225 70-80 7-8
Fluid Bedc
(Dorr-Oliver) 1650 240-350 75-85 10-12
Fluid Bed
(Lurgi) 1700 240 50 9-10
Suspension 1800 120-350 50 8-12
Fluid Column 1900 225 17-18 11-12
Dead roast except where otherwise noted.
First stage is a partial roast in multiple hearth, and second stage is a
dry-feed dead roast in Dorr-Oliver fluid bed.
Partial roast.
d Slurry feed.
Of the roasters listed, the Ropp hearth roaster is the oldest type
still in use. It is a long, narrow, mechanically rabbled reverberatory
furnace divided into two parallel hearths. Operating conditions are gen-
erally on a small scale, that is, low hearth temperatures and low feed
rates. One disadvantage of this type of roaster, from a pollution point
of view, is the low sulfur dioxide concentration in the off-gases.
.XII-4
-------
Multiple-hearth roasters are the next oldest type currently in use.
These units are 20-25 feet in diameter and have from 7 to 16 hearths;
their feed capacity rates range from 100 to 250 tons/day. In this
roaster, the required heat is obtained from the combustion of the sulfur
with the oxygen in the air. However, when concentrations of less than
1.0% of sulfide sulfur in the calcine are required, some auxiliary fuel,
about 4 million Btu/ton of feed, must be used. Because of the relatively
low operating temperatures and the large volume of off-gases, waste-heat
recovery is not economically practical.
The suspension, or flash, roasting process is similar to the combus-
tion of powdered coal in furnaces, in that the finely ground particles are
sprayed into the combustion chamber into a stream of combustion air.
As in the multiple-hearth roasters, the sulfur content of the ore acts as
the* primary fuel source. Typically, a suspension roaster is a refractory-
lined, cylindrical steel shell enclosing a large combustion chamber in
its upper portion. The lower part contains two to four hearths. Rabble
arms on a rotating shaft move the material on the hearths. The lower
one or two hearths dry the ore before it is ground and sprayed into the
upper combustion chamber, where calcination takes place. Process
temperatures approach 1900°F, making waste heat recovery economically
practical.
The fluid-bed roaster is the newest development in feinc sulfide
roasting; several types exist. The primary difference among these
roaster types is the method used to charge the feed. In some, a wet
slurry is charged, whereas in others, a dry charge is fed to the com-
bustion chamber. Operating temperatures range from 1625° to 1750°F,
and temperature control is achieved by varying the fee'd rate. A newer
type of fluidization roaster recently placed in operation is described as
a fluid-column roaster that approaches true fluidization. This is accom-
plished by pelletizing and close sizing of the feed. The resulting product
does not require sintering.
In all of the roasters used, combustion of sulfur in the feed is the
source of heat; consequently, the process is self-sustaining. The only
additional fuel required is about 1. 0 million Btu/ton of ore for ignition;
the primary fuel used is natural gas. As stated above, some multiple-
XII-5
-------
hearth roasters require auxiliary fuel to achieve the desired sulfur level
in the calcine. Based on this information, approximately 1.0 trillion
Btu of energy was consumed for roasting in 1972.
Sintering in the zinc industry is used primarily to agglomerate the
i
roaster calcine for further processing. If the zinc is to be processed
electrolytically, sintering is not used. Only if the zinc is to be processed
pyrometallurgically is sintering necessary. In this event, Dwight-Lloyd
sintering machines similar to those used in other nonferrous metal in-
dustries are used. Table 2 summarizes typical zinc sintering operations.
Table 2. TYPICAL ZINC SINTERING OPERATIONS3
Case 1 Case 2 Case 3
New Feed Material Calcine Calcine Concentrate
Total Charge Capacity,
tons/day 240-300 400-450 550-600
Machine Size, ft 3.4 X 45 6 X 97 12 X 168
Fuel Added to Feed, % 6^7 10-11 0-2
Total Sulfur in New Feed, % 8 2 31
Recycle, % of new feed 35-75 40-70 80
Dust in Off-Gas, % of feed 5 5-7 5-10
Off-Gas SO2 Content, % 1.5-2.0 0.1 1.7-2.4
The sintering machine comprises moving pellets upon which the feed is
placed. The feed is a mixture of calcines or concentrates, recycled
ground sinter, and coke or coke dust, which have been sized and pellet-
ized. The loaded pellets move under a small gas- or oil-fired muffle
where the bed is ignited. Air is forced down through the bed, resulting
in rapid combustion of the ignited material and formation of a porous,
relatively hard clinker. The product is discharged from the pellets as
they burn over at the end of the machine for return to the feed end.
Fuel consumption is estimated to be about 1. 0 million Btu/ton of con-
centrate processed.8 In 1970, 57% of the zinc produced was processed
pyrometallurgically; consequently, sintering was used on only 57% of the
calcine and concentrates processed. On this basis, less than 0. 5 trillion
Btu of energy was consumed for sintering in 1972.
XII-6
-------
Calcining
Calcining is a heat-treating process that is used exclusively for
oxidized materials, such as oxide ore concentrates or the product mate-
rials from the roasting of sulfide ore concentrates. Calcining is accom-
plished in a rotary kiln, and the product is hard, randomly sized nodules
that subsequently are treated for zinc extraction. This process is not
very widely used and, consequently, is not a large consumer of energy.
Zinc Extraction
Roasting, sintering, and calcining are preliminary steps to extraction:
pyr or eduction, or leaching, and electrolysis. In pyroreduction, or pyro-
metallurgical reduction, three systems are in use: the Belgian, or hori-
zontal, retort; the vertical retort; and the electrothermic furnace. All
three processes are based on the action of carbon monoxide and solid
carbon in sealed retorts at temperatures of 2ZOO°F. On the other hand,
the leaching and electrolytic process is considered to be a hydrometallurgical
process.
Belgian Retort
The Belgian, or horizontal, retort is the oldest zinc smelting method.
The basic retort is a batch system, and its charge is fired indirectly.
The major aim of this and the other pyrometallurgical processes is closely
controlled distillation condensing and collection of liquid zinc with a
minimum amount of oxidation or loss, or both. The basic retort is
approximately 10 inches in diameter and 5-6 feet long. Each production
unit within a horizontal-retort plant turns out only 100 pounds of zinc
every 48 hours. Thus, to produce large amounts of zinc, a horizontal
retort smelter must have many production units. For example, 10, 000
retorts, with 5, 000 charged per day, would be required to produce
90, 000 tons/yr of zinc.
Horizontal retorting does not lend itself to effective automation;
consequently, its efficiency suffers. Typically, a horizontal retort
operates at fuel efficiencies of 5% or less because most of the fuel
energy is lost up the flues. In addition, its zinc recovery is less than
90%.n Energy consumption for this process is about 60.0 million Btu/
ton of zinc produced. The primary fuel used is natural gas.2 Because
XII-7
-------
of its inefficiencies of operation, excessive operating costs, low produc-
tivity, and difficulty of controlling particulate emissions, this process is
not likely to be used in the future.
Vertical Retorts
The vertical retort process for zinc recovery was developed in an
effort to overcome the major problems of the horizontal process. Like
the horizontal process, the vertical retort uses indirect heating to reduce
the zinc calcines. Unlike the horizontal process, the vertical retort is
continuous and can be mechanized. To be continuous, the retort is
vertical so that the charge can pass through it by gravity. The charge
must be briquetted to give free movement through the retort, to permit
heat transfer to the center of the retort, and to provide sufficient porosity
for the escape of gases. Because of these improvements, energy effi-
ciency is about 10%; the typical energy consumption of a vertical retort
is 18.0 million Btu/ton of zinc produced.2 What makes this process
possible is the use of briquettes as a charge; however, briquette produc-
tion requires a major effort and reportedly consumes large almounts of
energy. One company that uses vertical retorts reports that 36 million
Btu of coal and coke per ton of distilled zinc is required for the briquetting
process. However, this company claims an energy consumption by the
retorts of only 10 million Btu/ton of distilled zinc for a total energy
consumption of 46 million Btu/ton.10
Electrothermic Smelting
The electrothermic smelting process was developed in;the U.S. and
put into commercial operation in 1936. Unlike either the vertical or
horizontal retorts, the electrothermic furnace is heated internally by
electrical resistance. In the electrothermic furnace column, the charge
of sinter, briquetted material, and coke is heated electrically to reaction
zone temperatures of 1400°-2500°F. The charge is the current carrier.
The most significant factor concerning electrothermic smelting is that it
has an energy efficiency of 25-30%. u This process consumes about 10
million Btu/ton of zinc as electricity. An additional 21 million Btu/ton
of zinc is consumed as coke, for a total of 31 million Btu/ton of zinc.
However, if the energy for electric generation also is considered (assum-
ing 33% efficiency), total energy consumption increases to 51 million
Btu/ton.
XII-8
-------
Electrolytic Process
Electrolytic recovery is the fourth process used in the U.S. for
producing zinc. In this process, the zinc sulfide concentrate that has
been preroasted to produce calcine is transferred to a leaching operation,
in which the calcine and dilute sulfuric acid are introduced into a series
of tanks. Zinc oxide dust is added to precipitate the impurities. The
resulting solution is filtered and subjected to electrolysis, in which the
metallic zinc is deposited at the aluminum cathodes. The zinc cathodes
then are melted in electric or reverberatory furnaces and subsequently
cast into slabs. The basic electrolytic process is reasonably efficient,
with an estimated energy efficiency of 25%. n The energy consumption
per ton of zinc produced is estimated at 18. 0 million Btu of natural gas
for melting and 33.8 million Btu for electricity including electric generation,
for a total of 51.8 million Btu.ri)
Refining
The zinc produced in the retorting process is transferred in molten
form to fuel-fired reverberatory furnaces for refining. During refining,
impurities, such as lead, iron, and cadmium, are removed. Typical
energy requirements for this process are approximately 3 million Btu/ton
of zinc produced. Upon completion of the refining process, the zinc is
formed into slabs.
Energy Utilization Pattern
Depending upon the combination of processes used, the energy required
to produce 1 ton of zinc slab varies from about 35 to 65 million Btu.
In 197Z, an estimated 38.0 trillion Btu of energy was consumed in the
manufacture of primary zinc slabs, or an average of 55.2 million Btu/ton
of zinc slab. In 1967, the industry consumed an average of 60.3 million
Btu/ton of zinc produced.13 These data show an 8% improvement in
energy utilization efficiency by this industry over a 5-year period. The
main reason for this improvement is the closing of two horizontal retort
plants, the most inefficient of the four processes used. These two plants
had an annual production capacity totaling 113,000 tons.11
XII-9
-------
Of the four zinc-smelting processes in use, the industry seems to
be moving in the direction of the electrolytic process. Both the retort
processes have shown significant decreases in total annual output in
recent years, while the electrothermic process has just been holding
its own. Although the vertical retort process and the electrothermic
process will survive for a while, the electrolytic process is expected to
ultimately prevail.11 Of all the processes available, the electrolytic.
process is better able to conform to automation, customer quality de-
mands, and air pollution regulations. As a result, the energy consump-
tion rate per ton of zinc will remain virtually unchanged in the future as
the process energy requirements, including energy consumed for electric
generation, are about the same as the average energy consumption rate
of all the processes currently in use.
New Technologies in Zinc Manufacturing
Because of the present state of affairs in the zinc industry, research
and development of new technologies in the U.S. is at a virtual standstill.
However, a new process for zinc manufacturing that is being used in
other countries produces more than 10% of the total zinc production in
the free world. This process, the Imperial Smelting Process (I. S.P. ),4"*
is strongly supported by the smelters that use it, but has not been put
into operation in the U. S. In addition to producing zinc, the I. S. P. also
produces lead bullion.
Basically, the I. S. P. furnace is similar to a standard lead blast
furnace except that, in this case, the tuyeres project inside the line of
the water jackets for better heat economy and the air blast is preheated
in alloy-tube preheaters, which are fired by furnace gas. The charge
to the furnace consists of sinter and hot coke, added alternately. The
coke is preheated to 1500°F in a shaft-type furnace, also fired by furnace
gas. The gases leaving the furnace are passed through condensers, where
they are rapidly chilled to below 1100°F. The zinc vapors condense and
go into solution in the molten lead. This solution is removed from the
condensers and then cooled even further, causing the zinc and lead to
separate into two layers,according to their respective solubilities. These
layers are separated easily.
XII-10
-------
The energy requirements for this process are relatively low compared
with those for other processes. Coke consumption is approximately 22. 6
million Btu/ton of zinc produced. An additional 1. 0 million Btu of elec-
tricity per ton and about 1.4 million Btu of furnace gas per ton are con-
sumed. Thus, the total energy consumed by this process is about 25
million Btu/ton of zinc produced.
The current state of the I. S. P. is not ideal, but research to improve
this process is continuing. These improvements include increasing furnace
size, developing computer control, implementing oxygen enrichment, and
using vacuum dezincing. All these improvements will make the process
more economical by reducing costs and improving efficiencies. However,
no data are available on the quantitative effects of these improvements.
Effect of New Technology on Energy Consumption
It is not clear at present just what effect the I. S. P. will have on the
amounts of energy consumed by the zinc industry in the U.S. Currently,
it is not used in this country; moreover, there is great opposition to it.
Supporters of the process believe that it is a revolutionary process that
will eventually smelt most of the world's zinc. Critics think that 1) the
process is not flexible, requiring a carefully balanced lead-zinc feed;
2) it consumes great quantities of expensive metallurgical coke; and 3) it
requires work-force procedures, namely, regular furnace shutdowns, for
cleaning, requiring utilization of maintenance and operating crews on an
irregular basis, which the unions in the U.S. would oppose;
In spite of these objections, the potential for reducing energy con-
sumption by this industry is there, assuming acceptance of the process
by U.S. manufacturers. Figure 3 shows the potential effect of this
process on energy consumption, assuming that 50% of the zinc produced
in the U.S. is produced by this process by 1985. These data also assume
a 3. 5% annual growth rate by this industry.
XII-11
-------
60
50
UJ
_l -
§2 40
Z 9:
30
20
PROJECTED INDUSTRY ENERGY
CONSUMPTION ASSUMING NO CHANGES
IN CURRENT ENERGY
UTILIZATION PATTERNS
PROJECTED ENERGY CONSUMPTION
ASSUMING LINEAR GROWTH OF
I.S.P TO 50% OF TOTAL PRODUCTION
BY 1985
1972 '73 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 1985
YEAR
A-44-570
Figure 3. EFFECT OF IMPERIAL SMELTING PROCESS
ON ENERGY CONSUMPTION BY DOMESTIC ZINC SMELTERS
Air Pollutant Emissions From Zinc Smelters
The primary air pollution emissions from zinc smelters are partic-
ulates and SO ; the latter are emitted primarily from plants that emit
J\.
.dilute off-gases from roasters and sinter machines. In 1972, an estimated
342, 000 tons of sulfur dioxide was emitted by zinc smelters in the U. S. ,
based on an emission rate of about 993 Ib/ton of zinc produced.7 As in
the other nonferrous metals industries, the primary method of control is
recovery of the SO and conversion to sulfuric acid. Of the 15 zinc
JC
smelters currently in operation in this country, nine recover the SO
X.
from the off-gases, two smelt material very low in sulfur content, and
four have no recovery units at all. 3 At present, SOX is recovered only
from the roaster off-gases, because 93-97% of the sulfur in the feed to
a zinc smelter is eliminated in roasting. Off-gases from the newer
fluidized roasters contain from 7 to 12% sulfur dioxide, making them
ideal, economically, for recovery and conversion to acid. The sulfur
XII-12
-------
dioxide concentrations in the off-gases from this and other types of
roasters are shown in Table 3.
Table 3. CONCENTRATION OF SULFUR DIOXIDE IN
OFF-GASES FROM ROASTERS IN ZINC SMELTERS
Roaster Type SO2 Concentration, %
Ropp Roaster 0.7-1.0
Multiple Hearth 4.5-6.5
Suspension 10-13
Fluid Bed 7-1Z
Fluid Column 11-12
With the exception of the Ropp roaster, the other roasters generate off-
gases with high enough sulfur dioxide concentrations to make recovery
and conversion to acid economical, although multiple-hearth roasters are
marginal. Sintering machines, where used, are also sources of SO
X.
emissions, but the emission rate, which depends upon the sulfur content
of the feed, is very low. The best way to reduce sulfur dioxide emis-
sions from this process is to use feed with the lowest possible sulfur
content. Because the concentration of sulfur dioxide in the off-gases from
sintering machines is so low, no attempts are made to recover it. (See
Table 4.)
Table 4. SULFUR OXIDE GENERATION AND
RECOVERY AT U.S. ZINC SMELTERS IN 19699
Sulfur Equivalent
Generated Recovered
Process tons/yr » % Recovered
Roasters 697,300 461,600 66.2
Sinter-Roast Machines 60,100 0 0
Sintering Machines 22,700 0 0
Cokers and Retorts 2, 600 0 0
Total 782,700 461,600 59.0
The SO emissions from the zinc extraction processes are very low and
thus offer no real possibilities for recovery.
XII-13
-------
The other major air pollutant emission is particulate matter. Of
all the processes used in zinc smelting, the fluid-bed roasters and the
retorts are responsible for the largest percentage of the total particulate
matter emitted. The data presented in Table 5 indicate that particulate
emissions from these two sources alone accounted for 60% of the total
emissions in 1972 in spite of the controls used.
Table 5. PARTICULATE EMISSIONS FROM
ZINC SMELTERS IN 1972
Emission Control Emissions,
Process Factor5 Efficiency tons/year
Ore Crushing 2 Ib/ton ore 0. 0 12,400
Roasting
Fluid Bed, Suspension 2000 Ib/ton zinc 0.98 10, 300
Ropp, Multiple Hearth 333 Ib/ton zinc 0. 85 2, 580
Sintering 180 Ib/ton zinc 0.95 1,860
Retorts 49 Ib/ton zinc 0.0 16,860
Materials Handling 71 Ib/ton zinc 0. 32 1, 640
Total 45, 640
The other major source was the ore-crushing process. Particulate
emissions can be controlled by normal methods, such as electrostatic
precipitators and baghouse filters, regardless of the source. Some
difficulty is encountered in collecting the sinter fumes electrically because
of their inherently high electrical resistivity. However, with proper con-
trol of gas temperatures and moisture content, efficient collection can
be maintained. As things currently stand within this industry, the elec-
trolytic process will assume the major part of the zinc production load
,in the future. Particulate emissions from this process are only 5 Ib/ton
of zinc without controls. On this basis, if 100% of the zinc had been
manufactured by this process in 1972, a total of 1032 tons of particulates
would have been emitted, resulting in a total reduction from all zinc
smelters of 35%. Apparently, such a conversion to this process by the
industry would have a significant impact on the particulate emission
problem.
XII-14
-------
Effect of New Technology on Air Pollutant Emissions
The emergence of the I. S. P. in the U. S., should it occur, is not
likely to have a significant effect on air pollutant emissions because the
primary sources of emissions, namely, the roasters and sintering mach-
ines, are still required in this process. In addition, there is no other
new technology that will affect pollutant levels in the near future.
References Cited
1. American Bureau of Metal Statistics, Year Book for 1972. New
York, June 1973.
2. American Gas Association, A Study of Process Energy Requirements
in the Non-Ferrous Metals Industry, Catalog No. C20010.
Arlington,. Va. , n. d.
3. Jones, H. R., "Pollution Control in the Nonferrous Metals Industry, "
Pollution Control Review No. 13. Park Ridge, N. J. : Noyes Data
Corporation, 1972.
4. Morgan, S. W. K., "Recent Developments in Zinc Blast-Furnace
Technology," J. Met. 16, 33-36 (1964) January.
5. Morgan, S. W. K. and Lumsden, J., "Zinc Blast-Furnace Operation, "
J. Met. H_, 270-75 (1959) April.
6. Morgan, S. W. K. and Temple, D. A;, "The Place of the Imperial
Smelting Process in Nonferrous Metallurgy, " J. Met. 19, 23-29
(1967) August.
7. Rohrman, F. A. and Ludwig, J. H., "Sulfur Oxide Emissions by
Smelters," J. Met. 20, 46 (1968) December.
8. Rosenberg, R. B. , "Energy Use for Industrial Heat and Power and
New Process Developments for Conservation," Final Report, IGT
Project 8938. Chicago: Institute of Gas Technology, August 1972,
Revised October 1972.
' 9. Semrau, K. T., "Sulfur Oxides Control and Metallurgical Technology,"
J. Met. 2/3, 41-47 (1971) March.
10. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President, Washington, D. C. : U.S. Government Printing
Office, January 1972.
11. "The Crisis in U.S. Zinc Smelting Spells Trouble for the Mining
Industry, " Eng. Min. J. 173, 69-74 (1972) February.
XII-15
-------
12. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1967 Census.pf Manufactures, Special
Report No. MC67(S)-4. Washington, D. C. ; U. S. Government
Printing Office, June 1971.
13. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C. : U.S. Government
Printing Office, July 1973.
XII-16
-------
XIII. SIC CODE 3334 - PRIMARY ALUMINUM
SIC Code 3334 pertains to those establishments primarily engaged
in producing aluminum from alumina and in refining aluminum by any
process. Establishments engaged in rolling, drawing, or extruding alum-
inum are classified elsewhere.
The primary aluminum industry is growing at a greater rate than
' . !"S- '
any of the other nonferrous metal industries. Historically, production
of aluminum in the U.S. doubles every 10 years (Figure 1), and this .:
growth rate is expected to continue.
IC.VVJV
1 1,000
§10,000
8 9,000
O
2" 8,000
O
k 7,000
3
0 6,000
o:
^ 5,000
g 4,000
< 3,000
2,000
1,000
^
^
^
/*
x
/
X
/
'
^
/
/
y
/X
^'
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-54- 719
Figure 1. ANNUAL PRODUCTION OF PRIMARY
ALUMINUM WITH PROJECTION TO 1985
Between i960 and 1970, the annual production of primary aluminum
increased from 2 million to almost 4 million tons.1 By 1985, produc-
tion is expected to increase to 12 million tons/yr,4 an increase of 200%
over current production rates, in spite of the depressed growth rate in
the early 1970's due to an oversupply situation.
XIII-1
-------
Energy consumption by the primary aluminum industry was about
295 trillion Btu in 1971. 7 This figure does not include the fuel consumed
by the electricity-generating stations, nor does it include fuel consumed
for use as carbon anodes in the electrolysis process. By 1985, the total
annual energy consumption of this industry will be about 900 trillion Btu,
assuming no changes in energy utilization efficiency (Figure 2).
ENERGY CONSUMPTION, I012 Btu
rooJAoicD-Niopu
ooooooooc
o o ooo oooc
»-^-
^
s-
S
/
/
/
/
/
/
/
y
r
/
/"
X"
f
I960 62 T64 n66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-54-721
Figure 2. ANNUAL ENERGY CONSUMPTION FOR
POWER AND HEAT BY THE ALUMINUM
INDUSTRY WITH PROJECTIONS TO 1985
This is more than 3 times the current rate of consumption. However,
there are significant inefficiencies in the energy consumption of this
industry, and several new processes that are available will probably
reduce the energy consumed by this industry. The most important new
process, which was just developed, according to its developers, con-
sumes 30% less energy than the process currently used. Details on
this process are not available, but, in time, it is expected to revolutionize
the industry.
XLII-Z
-------
Air pollutant emissions from primary aluminum plants are composed
primarily of particulates and fluorides, the most serious emissions
occurring from the potlines. However, control methods that exist remove
up to 98% of these emissions from the effluent gas streams, although
there is no inventory of methods in use to indicate the real emission
rates for this industry.
Primary Aluminum-Manufacturing Processes
In general, aluminum production is divided into three operations:
mining the bauxite, refining the bauxite into alumina, and smelting the
alumina into aluminum. (See Figure 3.)
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 consumed 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 Z. 3 trillion Btu, less than 1% of the industry's total annual
energy consumption, was used for drying.
R ef ining
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 solution 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
XIII-3
-------
S.ODIUM
HYDROXIDE
BAUXITE
TO CONTROL DEVICE
I
SETTLING
CHAMBER
DILUTION
WATER
(RED MUD
(IMPURITIES)
DILUTE
SODIUM
HYDROXIDE
TO CONTROL
DEVICE
AQUEOUS SODIUM
ALUMINATE
TO CONTROL DEVICE
BAKING
FURNACE
BAKED
ANODES
, TO CONTROL DEVICE
PREBAKE
REDUCTION
CELL
ANODE PASTE
TO CONTROL DEVICE
HORIZONTAL
OR VERTICAL
SODERBERG
REDUCTION CELL
MOLTEN
ALUMINUM
Figure 3. SCHEMATIC DIAGRAM OF PRIMARY
ALUMINUM PRODUCTION PROCESS8
alumina in the bauxite is dissolved in the caustic soda, forming a sodium
aluminate solution. The impurities, which are insoluble, remain as
solids. The mixture is passed through a series of pressure-reducing
tanks and filter presses. Cloth filters hold back the solids known as
"red mud" but allow the liquid, which contains the dissolved alumina,
to pass.
XIII-4
-------
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 ithrough 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. Eventually, a precipitate of hydrated alumina settles out of
solution. This precipitate 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 commercially 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. To produce 1 ton of aluminum in the reduction process,
approximately 2 tons of alumina is required. Thus,' the 7. 8 million tons
of alumina produced in 1971 consumed approximately 115 trillion Btu of
natural gas.
Smelting
Smelting is the process that breaks alumina down into its two com-
ponents, 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 dissolved in a bath of molten cryolite sodium aluminum
fluoride in large electric 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 pot, which serves as the cathode. The molten alum-
inum 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
XIII-5
-------
dioxide. The layer of molten aluminum 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 reduction process. Heat
generated by the electrolysis maintains the cryolite bath in a molten
state, so that the additional alumina charges are dissolved. Periodically,
molten aluminum is siphoned and cast into ingots or alloyed. The re-
sulting aluminum is at least 99. 5% pure.
The primary source of energy for this process is electricity, which
is consumed at the rate of 55-60 million Btu/ton of aluminum produced.
This does not include the fuel consumed for electricity generation.
The three types of po'ts 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 gradually consumed. The rate of con-
sumption is 0.45-0.55 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 car-
bon 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
aluminum sheeting and perforated steel channels, through which electrode
connections, called studs, are inserted in the anode paste (the pitch and
carbon aggregate mixture). As the baking anode is lowered, the lower
row of studs and the bottom channel are removed, and the flexible elec-
trical 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.
XIII-6
-------
Energy Utilization Pattern
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 jprimary form of energy consumed, fossil-fuel consumption for elec-
tricity generation also should be considered.
The electrolytic process for converting alumina to aluminum consumes
about 0. 6 ton of carbon electrodes per ton of aluminum produced. Because
40, 000 Btu of fossil fuel including petroleum, coke, and pitch) is required
to produce 1 pound of carbon, the fuel value of the carbon amounts to
about 48 million Btu/ton of aluminum produced.
According to the 1972 Census of Manufactures report7 on fuels and
electricity consumed, almost 52% of the energy consumed directly by
the aluminum producers was electricity, a total of slightly more than
150 trillion Btu. Because of the low efficiency of electricity generation,
however, the total fuel value for electricity consumed is about 3 times
higher. Assuming a 33% efficiency for electricity generation and trans-
mission, the actual fuel value of the electricity consumed is 455 trillion
Btu, or almost 120 million Btu/ton of aluminum produced. Note that
almost one-half of the electricity consumed by aluminum producers is
generated hydroelectrically. 6 Consequently, aluminum plants tend to be
located near the sources of this low-cost electricity, primarily in the
TVA area> the Ohio Valley, and the Pacific Northwest. Furthermore,
fossil fuels consumed for electricity generation account for only one-half
of the electricity consumed by the industry. As a result, only 230
trillion Btu of fossil fuel was consumed for the generation of electricity
by aluminum producers in 1971. Of the electricity consumed by aluminum
producers in 1971, 75 trillion Btu was generated hydroelectrically. The
actual total fuel value of the electricity consumed by aluminum producers
was 305 trillion Btu, or 78 million Btu/ton of aluminum produced. Given
these considerations, the total amount of fuel consumed by aluminum
producers in 1971, including electricity generation and the production of
carbon anodes, was 635 trillion Btu, or 163 million Btu/ton of aluminum
produced.
xni-7
-------
For the energy consumed by the aluminum industry, Figure 4 com-
pares the case in which energy consumed for electricity generation is
included with the case in which it is excluded. However, there has been
a trend in recent years toward hydroelectric power. From I960 to 1969,
the percentage of electricity from hydroelectric sources consumed by
aluminum producers increased from 39. 1 to 48. 3% . If this trend con-
tinues, the total fuel value of energy consumed by the aluminum industry
will be lower than that projected by the top curve in Figure 4 as a result
of a decrease in fossil fuels consumed for electricity generation.
New Technologies in the Manufacture of Aluminum
The Hall-Heroult Process for converting alumina to aluminum is the
greatest energy-consuming process in the manufacture of aluminum. As
a result, much work has been done to develop means for lowering its
energy consumption. The most significant development to date is a new
process that reportedly consumes 30% less electrical energy than the
traditional Hall-Heroult Process, which has been in use since 1886.
However, details of this process, developed by Alcoa, are not available
at present.
If energy consumption is going to be reduced, the efficiency of the
electrolytic process will have to be increased. Currently there is great
interest in the concept of increasing the efficiency by, adding lithium salts
to the cryolite baths. Lithium salts are particularly attractive because
of their effectiveness in lowering the freezing point and increasing the
electrical conductivity of the electrolyte. Several tests with lithium
fluoride on aluminum potlines have resulted in a 10-12% increase in
production and a 2-3% savings in power consumption.2 In addition, the
net carbon consumption was about 5% lower than in a normal cell.
Thus, additional energy is saved by the reduced carbon consumption.
During the past 30 years, much research has been done to develop
methods for processing clay into alumina suitable for aluminum produc-
tion. 3 This research resulted in the development of several processes
for treating the clay; in addition, many variations of each process have
been developed. In general, these processes utilize chemical means
for reaching the desired product, and in some cases, high-temperature
heating (2000°F) in fuel-fired furnaces and kilns is required. Although
XIII-8
-------
Q.
§
CO
>
CO
u
U
£OUU
2200
2100
2000
1900
1800
1700
1600
.1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
^
- ^-
/
^
X
S^
INC
C
'
^-
LUD
ONS
ELE
GEN
/
/
^
ESE
UME
:CTR
ERA1
i
^/
.
NER
DFO
ICIT1
riON
/
'
-'
SY /
_ L-
R /
r /
/
/
/
f
/
X
f
(
X
X
f
1
1
t
'
/
x EXCLUDES ENERGY
X CONSUMED FOR
" ELECTRICITY
GENERATION
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
, A-54-720
Figure 4. ANNUAL ENERGY CONSUMPTION BY ALUMINUM
INDUSTRY WITH AND WITHOUT ELECTRICITY GENERATION
XIII-9
-------
there are many proposed processes, they can be categorized into three
major classifications: acid-leaching, alum intermediate, and alkaline
sinter. Energy consumption by these processes is variable, and no
operating data are available to allow even an estimate of the range to
be made. At present, none of these processes has been commercialized
nor are they likely to be, until costs are lower, or until the prices for
the conventional route rise sufficiently. Such a rise would be due pri-
marily to increases in fuel costs or increases in the cost of bauxite.
At present, only fuel costs seem to be rising rapidly. Increases in
bauxite costs have been held to a minimum because of the current over-
supply situation. Currently, the Bayer Process is 20% less costly than
the least costly of these new processes. As a result, the research
objective is to lower costs by improving existing processes or developing
improved versions of these processes. Eventually, the process with
the lowest cost probably will be commercialized.
Effect of New Technologies on Energy Consumption
At the present rate of consumption, the total amount of energy
consumed by the aluminum industry annually is expected to increase
threefold by 1985. However, the potential for reducing energy consumption
is good, especially in light of the recent announcement of Alcoa's new
process. It is difficult to assess the impact of this process on the in-
dustry without details on the time required for implementation. Figure 5
shows that if full implementation were achieved by 1985, the projected
energy consumption would be reduced by nearly 25%.
On the other hand, the addition of lithium salts to the cryolite bath,
as previously described, reduces total energy consumption by about 5%.
Implementation of this technology apparently does not require extensive
changes in operation, large increases in capital costs, or other such
prohibitive actions on the part of the aluminum producer. Consequently,
short-range implementation to achieve the reduction in energy consump-
tion seems feasible.
: As previously stated, there are no commercialized processes for
converting clay to alumina; consequently, no operating data are available.
Furthermore, estimates of expected energy consumption also are not
available. As a result, the effects of these processes on the total energy
consumption of the industry is indeterminable at present.
XIII-10
-------
m
(M
~b
o.
(f)
8
>
a:
UJ
LJ
3WV
800
700
600
500
400
300
200
100
~^^
^
^s
x-"~
'
^
CURRENT PROJECTION
BASED ON CURRENT x|
USAGE X
««^
jt
/ .
's
/
*
/
/
/
^'
/
fS
/
/
s
/
^*
s
ASSUMING 100%
IMPLEMENTATION
BY 1985
I960 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86
YEAR
A-54-7Z2
Figure 5. EFFECT OF IMPLEMENTATION OF NEW ALCOA
PROCESS ON ENERGY CONSUMPTION OF ALUMINUM INDUSTRY
(Excluding Energy Consumed for Electricity Generation)
Air Pollutant^ Emissions From Aluminum-
Manufacturing Processes
Air pollutants originate from several processes in an aluminum-
manufacturing plant, but 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 alumina with the carbon anode, emissions from the potlines
include sulfur dioxide, derived from the sulfur in the petroleum coke and
tar used to make the carbon anodes; particulates of vaporized bath
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 other pollutants. 5 Other sources of
emissions include the bauxite-grinding operation, the calcination operation,
materials handling, and anode preparation.
XIII-11
-------
From the point of view of potential damage to the areas surrounding
an aluminum plant, the most important effluents are gaseous and par-
ticulate fluorides released from the molten bath electrolyte of the potlines.
In general, total fluoride concentrations in the waste gases from the
various types of cells are approximately equal. However, the relative
amounts of gaseous and particulate fluorides emitted depend upon the
mechanisms for removing and treating the gases. Table 1 summarizes
the emission rates of total particulates and gaseous and particulate fluorides
from the various types of smelters. Application of the best demonstrated
pollution abatement equipment to collect the primary potline effluents
would result in estimated total fluoride emissions of 1.2-4.7 lb/1000 Ib
of aluminum produced, depending upon the type of potline, compared with
18.5-23 lb/1000 Ib of aluminum produced5 without such equipment. The
addition of scrubbing equipment on the secondary streams would reduce
the total fluoride emissions even further to 0.8-2.0 lb/1000 Ib of alum-
inum. Application of this technology by the entire industry would raise
the overall control efficiency of the industry from the present 74 to 92%.
However, invested industry expenditures for pollution control would in-
crease more than threefold. 5
Table 2 summarizes the emissions from primary aluminum production
processes other than the potlines. Also included in this table are the
effects of various types of control equipment. Emission factors for other
pollutants such as sulfur dioxide and hydrocarbons from prebake anode
furnaces are not available.
Effect of New_ Technologies on Air Pollutant Emissions
The effects, if any, of the new Alcoa process on air pollutant
emission rates from potlines are not known at present. However, the
.other new processes discussed above are not expected to affect air
pollutant emissions from potlines. On the other hand, the processes for
converting clay to alumina are of interest in the more distant future
(beyond 1985). because no waste products are given off. (The various
products aluminum hydroxide, cement, and combustion gases are
all utilized.) Even so, the only emissions affected by the advent of such
a process would be particulate s emitted during the grinding of the bauxite
in the currently used process. Based on an estimated production of 12
million tons of aluminum in-^ 1985 and assuming 100% implementation of
xm-12
-------
Table 1. ALUMINUM SMELTER EFFLUENTS5
New Prebake6
Old Prebake6
VS Soderberg
HS Soderberg
a-
Component
Solid Fluorides"
Quantity, Ib/Mlb Al
Loading, mg/m3
HF
Quantity, Ib/Mlb Al
Loading, mg/m3
Total Fluoride1"
Quantity, Ib/Mlb Al
Loading, mg/m3
Alumina"
Quantity, Ib/Mlb Al
Loading, mg/m3
Total Solids
Quantity, Ib/Mlb Al
Loading, mg/m3
Sulfur Oxides0
Quantity, Ib/Mlb Al
Loading, mg/m3
DiluentAird
106 cu ft/Mlb Al
Collection Efficiency
Solid F. %»
HF, %
Total F, %
Total
10
13
23
20
48
15-50
27.5
Prim.
9.5
61
12.6
81
22.1
142
19.0
122
46.1
295
14^8
90-300
2.5
95
97
96.0
Sec.
0.5
0.32
0.4
0.25
0.9
0.57
1.0
0.64
1.9
1.2
1-2
0.6-1.2
25
Total
10
13
23
20
48
15-50
27.5
Prim.
8.0
51
11.7
75
19.7
126
16.0
103
41.3
265
14-45
90-300
2.5
80
90
85.6
Sec.
2.0
1.3
1.3
0.8
3.3
2.1
4.0
2.6
6.7
4.4
1-5
0.6-3.2
25
Total
3
20
23
3
39
15-50
35.5
Prim.
1.5
47
17.0
543
18.5
590
1.5
47
25.9
826
13-43
420-1380
0.5
50
85
80.4
Sec.
1.5
0.7
3.0
1.4
4.5
2.1
1.5
0.7
13.1
6.1
2-7
0.9-3.2
35
Total
10
.
13
23
20
49
15-50
18.5
Prim.
8.0
37
11.7
53
19.7
90
16.0
73
38.2
174
14-45
64-206
3.5
80
90
85.6
Sec.
2.0
0.9
1.3
0.6
3.3
1.5
4.0
1.8
10.8
18.4
1-5
0.4-2.3
35
» Includes fugitive dusts from cell room.
>> Reported range 12.8 to 33.0 lb/1000 Ib Al.
«10 Ib SOi/1000 Ib Al/percent S in a node coke.
d Represents total cell room air.
elt is easier to control effluents from
new prebakes than from old prebakes.
-------
Table 2. AIR POLLUTANT EMISSIONS FROM PRIMARY
ALUMINUM PRODUCTION PROCESSES EXCLUDING POT LINES8
Type of Operation
Total Particulates
Gaseous Fluorides (HF)
Ib/ton
Participate
Fluorides (F)
Bauxite Grinding
Uncontrolled
Spray Tower
Floating-Bed Scrubber
Quench Tower and Spray Screen
Electrostatic Precipitator
6. 0
1.8C
1. 7
1. 0
0. 12
Negl
Negl
Negl
Negl
Negl
NA*
NA
NA
NA
NA
Calcining of Aluminum Hydroxide
Uncontrolled
Spray Tower
Floating-Bed Scrubber
Quench Tower and Spray Screen
E^ctrostatic Precipitator
ZOO. 0
60. 0
56. 0
34. 0
4. 0
Negl
Negl
Negl
Negl
Negl
NA
NA
NA
NA
NA
Anode Baking Furnace
Uncontrolled
Spray Tower
Dry Electrostatic Precipitator
Self-Induced Spray
3. 0 (1. 0-5. 0)d
NA
1. 13
0. 06
0. 93
0. 0372
0. 93
0. 0372
Negl
Negl
Negl
Negl
Emission factors for bauxite-grinding expressed as pounds per ton of bauxite processed. Factors for
calcining of aluminum hydroxide expressed as pounds per ton of alumina produced. All other factors
in terms of tons of molten aluminum produced.
Includes particulate fluorides.
Controlled emission factors are based on average uncontrolled factors and on average observed
collection efficiencies.
Numbers in parentheses are ranges of uncontrolled values observed.
No information available.
A-54-734
-------
such a process, particulate emissions from aluminum smelters would
be lower by about 400 tons/yr. This estimate is based on current
emission levels of 0. 12 Ib/ton of bauxite processed with electrostatic
precipitators. Beyond this process, emission reductions will come
about primarily through improvements in control equipment.
References Cited
1. Bureau of Mines, 1971 Minerals Yearbook. I. Metals, Minerals,
and Fuels. Washington, D. C. : U.S. Government Printing Office, 1973.
2. Lewis, R. A. , "Aluminum Reduction: Evaluating 5% LdF-Modified
Hall Bath in 10-KA Experimental Reduction Cells, " J. Met. 19,
30-36 (1969) May.
3. Peters, F. A., Kirby, R. C. and Higbie, K. B. , "Methods for
Producing Alumina From Clay An Evaluation," J. Met. 19, 26-34
(1967) October.
4. Predicasts, Issue No. 44. Cleveland: Predicasts, Inc., July 30, 1971.
5. Rush, D. , Russel, J. C. (Singmaster and Breyer) and Iverson, R. E.
(EPA), "Air Pollution Abatement oh Primary Aluminum Potlines:
Effectiveness and Cost," J. Air Pollut. Control. Assoc. 23, 98-104
(1973) February.
6. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, D. C. : U.S. Government Printing
Office, January 1972.
7. U. S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, D. C. : U.S. Government
Printing Office, July 1973.
8. U.S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors," Publication No. AP-42, 2nd Ed. Research
Triangle Park, N. C. , April 1973.
XIII-15
-------
XIV. ELECTRICITY AND STEAM GENERATION
Approximately 43% of the fossil fuels consumed by industry for power
and heat in 1971 was used in boilers.2'7 In addition, more than 13% of the
total amount of energy consumed by industry for power and heat in 1971 was
electricity. 7 Consequently, this last section of the report discusses elec-
tricity and steam generation.
Industrial Boilers
The use of fossil fuels to generate steam varies considerably from one
industry to another. Approximately 10% of the fossil fuels consumed by in-
dustries in SIC Code 32 (stone, clay, and glass) is used to generate steam.2
By comparison, about 95% of the fossil fuels consumed by industries in
SIC Code 26 (paper and allied products) is used to generate steam.2
All the boilers used in industry are classified as intermediate size,
that is, larger in capacity than residential boilers (2.00, 000 Btu/hr) and
smaller than that equivalent to 50 megawatts electrical (500, 000 pounds of
steam per hour).2 The three basic types of boilers used are water-tube, fire-
tube, and cast iron. The efficiency of fuel utilization of these boilers is
usually 80% or higher. Consequently, reductions in total energy consumption
by boilers can be achieved by reducing the steam load by l) switching from
steam to direct-fired processes when feasible and 2) properly maintaining
all the steam plant facilities, including properly insulating all steam lines
and repairing leaks when they occur. For example, a recent survey of the
operations of a division of a large chemical company indicated that up to
8% of the existing steam traps were blowing steam.4 Worn steam jets
with excessive steam pressure were found, accounting for millions of Btu's
lost every hour, as well^as .innumerable steam-line leaks. Under these
circumstances, much energy undoubtedly is being wasted daily by industry,
energy that without too much difficulty could be saved.
Table 1 summarizes estimates of the percentage of fuel that is used in
manufacturing to fire boilers. Note that other fuels also are burned under
boilers, depending on the industry and the availability of such fuels. For
example, in the paper industry bark is used as a boiler fuel and in the steel
industry blast-furnace gas and coke-oven gas, when available, are used
as boiler fuels. This report does not include these fuels because they are
not considered primary fuels. The data in Table 1 and data on fuel consumption
XIV-1 -
-------
by each industry indicate that the paper, food, and chemical industries are
prime candidates for significant reductions in energy consumption. Estimated
reductions of 10-20% of the fuel consumed under boilers could be achieved
by/these industries. Such reductions would represent substantial savings in
the total energy consumption of industry. Such reductions also would result
in substantial reductions in industry expenditures for fuel without increasing
operating costs significantly.
In 1971, approximately 4900 trillion Btu of fossil-fuel energy was con-
sumed by industry in boilers. 2' 7 Of this amount, 1700 trillion Btu was
consumed as coal, 2100 trillion as natural gas, and 1100 trillion as fuel
oil. It is estimated that the consumption of fossil fuels under industrial
Table 1. PERCENTAGE OF FUEL IN
MANUFACTURING USED IN BOILERS2
Percent Burned under Boilers
Industry
Food
Tobacco
Textile
Apparel
Lumber
Furni ture
Printing
Chemicals
Petroleum vV
Rubber, -
Leather
Stone, Cloy, Glass
Fabricated Metal
Machinery
Elec. Machinery
Transp. Equipment
Instruments
Primary Metals
Paper
Miscellaneous
XIV-2
Coal
100
100
TOO
100
100
100
100
75
90
s 100
100
ass 10
1 90
90
90
nt 90
90
90
100
70
Oil
100
85
90
90
90
90
90
40
40
65
85
10
70
70
70
80
80
10
95
50
Gas
90
50
80
80
70
50
85
30
30
40
50
10
20
20
2.0
20
45
10
93
50
-------
boilers will nearly double by 1985.2 The future distribution-of fuels consumed
under boilers is difficult to estimate, but coal and fuel oil consumption is likely
to increase in proportion to natural gas consumption because of the reduced
availability of natural gas.
Air Pollutant Emissions From Industrial Boilers
Air pollutant emissions from industrial boilers consist primarily of SO ,
.X
NO , and particulates. In some cases, hydrocarbons and carbon monoxide
JC
also are emitted, but these are controllable by maintaining proper combustion.
The emission rates of these pollutants vary considerably and depend on the
type of fuel consumed. Table 2 shows emissions typical of industrial boilers
fired by each type of fossil fuel.
Particulates
In 1971, an estimated 2. 3 million tons of particulate matter was emitted
by industrial boilers. This estimate is based on an average emission rate of
0. 95 Ib/million Btu of fuel consumed,2 assuming that the relative ratios of fuel
consumed under industrial boilers have remained unchanged since 1967. At
the present rate, particulate emissions will nearly double by 1985. (See
Figure 1.) However, any shifts in fuel utilization patterns will have a sig-
nificant effect. If the projected shift toward coal and fuel oil occurs because
of a shortage of natural gas, particulate emissions will increase. In general,
industrial-size boilers, many of which are used for space heating and thus are
only used seasonally, are not controlled. By imposing control measures on this
equipment, immediate reductions in particulate emissions would result.
sox
In 1971, an estimated 4.2 million tons of SO was emitted by industrial
3C
boilers, assuming an estimated average emission rate of 1. 7 Ib/million Btu
consumed.2 As in the case of particulates, SO emissions are sensitive to
JC
the type of fuel consumed. Consequently, a future shift toward coal and fuel
oil will result in an increase in SO emissions. Current attempts to alleviate
.x
this problem emphasize the use of low-sulfur fuels and waste-gas treatment,
although the latter approach is generally too expensive to be practical for
industry. Unfortunately, the availability of low-sulfur fuels also is diminish-
ing rapidly. Consequently, SO emissions can be expected to increase in
the future.
XIV-3
-------
Table 2.
EMISSION FACTORS FROM INDUSTRIAL BOILERS8
Fuel
Natural
Gas,
lb/106 Btuf
0. 018
0.0006
0. 120-0. 230a
0. 017
0. 003
« _
Fuel
Residual,
lb/106 Btuf
0.15
1.06Sb
0.27-0. 53C
0. 027
0. 02
0. 007
Oil
Distillate,
lb/106 Btuf
0. 1 0
0. 96Sb
0.27-0. 53C
0. 027
0. 02
0.013
Coal, d
lb/106 Btuf
coal burned
0.50A-0.62A6
14. 8Sb
0. 58-0. 69
0. 04-0. 08
0. 01-0. 04
0. 0002
Pollutant
Particulates
SO
x
NO
x
Carbon
Monoxide
Hydrocarbons
Aldehydes
a Use 120 for smaller industrial boilers (< 500 boiler horsepower) and
230 for larger industrial boilers (>7500 boiler horsepower).
The letter S is weight percentage of sulfur in fuel, which should be
multiplied by value given to obtain factor.
° Use 40 for tangentially fired units and 80 for horizontally fired units.
First value is for boilers consuming 10 to 100 X 106 Btu/hr; second
value is for boilers consuming greater than 100 X 106 Btu/hr.
o
The letter A is weight percentage of ash in coal, which should be
multiplied by value given to obtain factor.
f Pounds per 106 Btu at 1000 Btu/CF, 150,000 Btu/gal, 13,000 Btu/lb.
NO
In 1971, an estimated 1.08 million tons of NO was emitted from
industrial boilers based on the emission factors in Table 2. A number
of factors are involved in the formation of NO in the combustion of
x
fossil fuels, including flame temperature and the nitrogen content of the
fuel. Consequently, the type of fuel as well as the boiler size affect the
rate of NO emissions.
x
More than 40% of the NO emissions from coal-
x
and oil-fired boilers is due to the nitrogen content in the fuel. The nitrogen
content of natural gas is much lower than that of coal or oil, and con-
sequently, NO emissions from natural-gas-fired boilers due to fuel
X.
nitrogen are relatively low. As a result, shifts toward coal and oil
will result in higher NO emissions. Several studies on ways to reduce
2£
NO emissions are being conducted. Some of the proposals include the
use of flue-gas re circulation and water injection.
/OV-4
-------
8
I6
CO
V)
CO
£
o
I
1971 '72 '73 '74 '75 '76 '77 '78 '79 '80 '.81 '82 '83 '84 '85
YEAR A-64-934
Figure 1. PROJECTED ANNUAL EMISSIONS OF SELECTED
POLLUTANTS FROM INDUSTRIAL BOILERS
But because the mechanism of NO formation is not clearly understood,
X
solutions to this problem appear to be in the distant future. Conse-
quently, NO emissions can be expected to continue to increase in the
foreseeable future.
Other Emissions
Compared to particulates,, NO , and SO , the emission rates of
.X -?C
carbon monoxide and hydrocarbons from industrial boilers are low. In
general, these emissions are controllable by maintaining proper com-
bustion. This includes careful monitoring of the air/fuel ratio to ensure
the presence of enough combustion air to completely burn the fuel.
Proper mixing of the air and fuel in the combustion chamber also is
important for maintaining these emissions at a minimum. In some in-
stances, it may be necessary to purchase control equipment to achieve
this end, but in general, continuous observation and maintenance are all
that are required.
XIV-5
-------
Electricity Generation
In 1971, nearly 1800 trillion Btu of electrical energy was consumed
by industry, about 13% of the total energy consumed.7 This is based on
a conversion rate of 1 kWhr = 3413 Btu. However, this conversion rate
does not consider the average efficiency of electricity generation, which
is about 33%. At this generation efficiency, the conversion rate becomes
1 kWhr = 10,400 Btu. That is, 10,400 Btu of fossil fuel, or its equiva-
lent, is required to generate 1 kWhr of electricity, which when consumed
will only do the work equivalent to 3413 Btu.
There are four sources of energy for the generation of electricity.
By far, the largest percentage of electricity is generated by the combus-
tion of fossil fuel in boilers to drive steam engines and turbines.
Hydroelectric power is second, with nuclear power a distant third. (See
Figure 2. ) A fourth source, not shown in Figure 2, is generation by
internal combustion engines, but this source is so small by comparison
(0.4% of the total) that it is not considered important. Unfortunately,
the electric energy consumed by industry cannot be broken down by its
source; thus the full impact of increases in electric energy consumption
by industry on fuel consumption and on air pollutant emissions is difficult
to assess. The one exception is the aluminum industry, where nearly
one-half of the electricity consumed is generated hydroelectrically. 6
Because of the relatively low cost of this electricity, aluminum plants
tend to be close to hydroelectric plants. Because of the relative im-
portance of conventional steam generation of electricity, this discussion
considers only this source.
Historically, electric utilities have relied on coal as a source of
energy, but in recent years, the use of natural gas and fuel oil has in-
creased significantly. (See Figure 3. ) Fuel oil consumption alone in-
creased threefold between 1968 and 1972. This increase in the amount
of electricity generated by using oil represents 10-18% of the total amount
of electricity generated by fossil fuels. This switch, in large part, is
a result of the restrictions on the quality of effluents from the generating
station. Similarly, this also is the reason for the increase in natural
gas consumption. However, whereas the data indicate a continued increase
in fuel oil consumption, natural gas consumption is declining. This
XIV- 6
-------
1-8
1.7
1.6
1.5
I L4
d 1.3
O
z"l.2
g
5 ''
oc
^ 1.0
UJ
OQ.9
y
£0.8
yo.7
LU
_|0.6
<
0.4
0.3
0.2
O.I
TOTAL/
HYDROELECTRI
'CONVENTIONAL
STEAM
I960 '61 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72
YEAR
A-64-935
Figure 2. TOTAL ELECTRICITY GENERATION BY
TYPE OF PRIME MOVER DRIVING GENERATOR1
decline is a result of the extreme shortage of this fuel, a shortage that
is likely to continue for some time. As a result, its use for electricity
generation is likely to continue to decline.
As previously stated, the average efficiency of electricity generation
is about 33%. The efficiency could possibly be increased, but not without
great expense and probably not more than 40%. The matter of efficiency
is very important in considering alternative processes in industry. What
appears to be a reduction in energy consumption by using an electric
process may in fact result in a net increase in total energy consumption.
XIV-7
-------
8
CD
O
Q_
^ 6
CO
O
O
_l
UJ
CO
CO
O
NATURAL
GAS,
COAL
OIL
I960 '61 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72
YEAR
Figure 3. FOSSIL FUELS CONSUMED
FOR ELECTRICITY GENERATION
A-64-936
XIV-8
-------
A case in point is the use of electric melting in the glass industry,
which reportedly consumes about 2. 9 million Btu of electrical energy per
ton of glass melted. However, when the energy consumed for electric
generation is included, the total energy consumption for melting increases
to more than 8. 7 million Btu/ton of glass melted. Compared to a fossil-
fuel-fired unit that can melt the same type of glass at a fuel consumption
rate of 6. 0 million Btu/ton of glass, the net result of switching to elec-
tric melting is an increase in energy consumption. And where a glass
plant might switch to electric melting to reduce air pollutant emissions,
the net result might be an increase in emissions due to the increase in
combustion of fossil fuels for electricity generation. Thus, trends by
industry toward the use of electric processes, as opposed to direct-fired
processes, apparently will result in increases in fossil-fuel consumption
and, in fact, may result in increased air pollutant emissions.
Air Pollutant Emissionsi From Electricity Generation
The electric utilities industry comprises about 675 fossil-fuel plants,
which together produce more than 80% of all U.S. electric power; these
plants also are one of the nation's worst sources of air pollution. 5 In
1968, the most recent year for which data are available, these plants
emitted 5.6 million tons of particulate matter, 16.8 million tons of sulfur
dioxide, and 4.0 million tons of NO ,3 or 20%, 50%, and 20%, respectively,
JC
of these pollutants produced by all U.S. sources.5 The amount of emis-
sions from any plant is determined by the type of fuel burned and the
degree of control practiced. Table 3 shows the emissions from power-
plant boilers according to the type of fuel being burned. This table also
shows that the type of combustion system employed can affect emissions-*,
As shown by the data, tangentially fired units produce lower NO emis-
Jt
sions than horizontally fired units.
Particulates
Coal is by far the largest source of particulate matter, producing
100 times as much as oil and 500 times as much as natural gas when
burned without the use of pollution control equipment. Included in the
particulate matter resulting from the combustion of coal and, to a much
lesser extent, fuel oil, are fly ash and carbon particles resulting from
the incomplete combustion of the fuel.
XIV-9
-------
Coal,
lb/106 Btu
coal burned
0. 62A*
14. 8S*
0. 04
0. 01
0. 69
0. 0002
Fuel Oil,
lb/106 Btu
0. 05
1. 06S*
0. 02
0. 01
0. 70f
0. 007
Natural
Gas,
lb/106 Btu
0. 015
0. 0006
0. 017
0. 001
0. 600*
_ _
Table 3. , EMISSION FACTORS FROM
POWER UTILITY BOILERS8
Fuel
Pollutant
Particulates
SO
x
Carbon
Monoxide
Hydrocarbons
NO
x
Aldehydes
*
A and S are the weight percent of ash and sulfur, respectively, in
the fuel being burned and are to be multiplied by the numbers shown
to obtain corrected values.
Use 50 for tangentially fired units.
Use 300 for tangentially fired units.
The technology for particulate control for coal combustion is well
developed. Electrostatic precipitators, baghouses, and wet scrubbers
that are available prevent the emission of 99. 5% or more of all soot
and ash produced in the combustion of this fuel. 5 The efficiencies of
these controls are reduced considerably when applied to oil-fired units.
Still, all these controls can remove at least 90% of the particulate emis-
sions from oil-fired utility boilers when properly adapted.
SO
x
The air pollutant of greatest concern in the SO group is sulfur
X.
dioxide. The amount of sulfur in the fuel is the determining factor in
the sulfur dioxide emission rate. Of the fossil fuels, coal's sulfur con-
tent is the highest, ranging from 0. 5 to 5% , whereas that of oil varies
from only 0.1 to 3%. Natural gas contains virtually no sulfur.
The state-of-the-art for controlling sulfur dioxide emissions from
power plants is in a transitional period. The two alternatives for control
are 1) to burn low-sulfur fuels or 2) to remove the sulfur dioxide from
the flue gases before they escape into the atmosphere. The use of low-
XIV-10
-------
sulfur fuels is a costly and temporary solution because their availability
is limited. In addition, mining, refining, and transportation facilities
do not exist in sufficient capacity to permit immediate industry-wide
changeover. Similarly, several systems have been proposed and developed
to remove sulfur dioxide from the flue gases, but information on their
costs, efficiencies, and technical problems is sketchy. These uncertainties
have caused delays both in the perfection of available techniques and in
the commitment by utilities to implement currently available controls.
However, changes are beginning to occur. Utilities are switching to low-
sulfur fuels, resulting in a reduction of sulfur dioxide emissions of 50-85%
below previous levels. In addition, control techniques are beginning to
be implemented, generally aiming for reductions up to 90% by 1980. 5
The most promising methods are limestone scrubbing, magnesia scrubbing,
and catalytic oxidation. All these methods are expensive and control
efficiencies conservatively range from 70 to 80%, although higher effi-
ciencies have been reported with some experimental systems. 5
NO
x . \
These pollutants are produced in the combustion of fossil fuels at
high temperatures. Unlike SO and particulates, which result from im-
purities in the fuel, NO are formed by the reaction of nitrogen and
oxygen present in the combustion air, as well as by the oxidation of
nitrogen in the fuel. The rate of formation depends on a number of
combustion characteristics, many of which are not completely understood.
As a result, NO formation varies among different boilers as well as
ji
different fuels.
Much research is currently being conducted to develop methods for
controlling NO emissions. The three most successful methods involve
modification of combustion methods and tend to apply only to natural gas
and oil combustion:
XIV-11
-------
1. Reducing the amount of excess air to the lowest level without causing
incomplete combustion
2. The use of two-stage combustion, in which combustion air insufficient
for complete combustion is fed directly into the burner and the re-
mainder is fed in at a point downstream of the burner to burn the
remaining fuel
3. Recirculating some of the flue gases back into the boiler, reducing
flame temperature and oxygen concentrations.
None of these methods appears to work well with coal-fired systems.
Implementation of these control methods by power plants is occurring only
at low rates, except in southern California, where implementation in some
plants has resulted in a 70% reduction in emissions from oil- and gas-fired
units.
Much can be done to reduce power-plant emissions in the future. Mean-
while, industry is converting many processes to electrical energy. In many
cases, the processes themselves consume less energy and are cleaner than
the processes that they replace. But if the energy requirements and emis-
sion rates of the power plant also are considered, some of these processes,
from a national point of view, will become unattractive. However> evalua-
tions of these processes cannot be presented here; rather, they should be
made on an individual basis with particular attention to the source of the
electricity being consumed.
References Cited
1. Edison Electric Institute, Statistical Year Book of the Electric Utility
Industry for 1972, Publication No. 73-13. New York, November 1973.
2. Ehrenfeld, J. R. et al-, Systematic Study of Air Pollution From Intermediate-
Size Fossil-Fuel Combustion Equipment. Cambridge, Mass.: Walden
Research Corp. , July 1971.
3. Environmental Health Service, Public Health Service, U.S. Department
of Health, Education, and Welfare, "Nationwide Inventory of Air Pollutant
Emissions, 1968, " Publication No. AP-73. Raleigh, N. C. , August 1970.
4. Kline, P. E. , "Technical Task Force Approach to Energy Conservation, "
Chem. Eng. ProR. 70. 23-27 (1974) February.
5. Komanoff, C. et al. , The Price of Power. Electric Utilities and the
Environment. New York: Council on Economic Priorities, 1972.
XIV-12
-------
6. Stanford Research Institute, Patterns of Energy Consumption in the
United States, for Office of Science and Technology, Executive Office
of the President. Washington, B.C.: U.S. Government Printing
Office, January 1972.
7. U.S. Department of Commerce, Bureau of the Census, "Fuels and
Electric Energy Consumed, " 1972 Census of Manufactures, Special
Report No. MC72(SR)-6. Washington, B.C.: U.S. Government
Printing Office, July 1973.
8. U. S. Environmental Protection Agency, "Compilation of Air Pollutant
Emission Factors, " Publication No. AP-4Z, 2nd Ed. Research
Triangle Park, N. C. , April 1973.
XIV-13
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-74-Q44
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Study of Industrial Uses of Energy Relative to
Environmental Effects
5. REPORT DATE
July 1974
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
M. E. Fejer and D. H. Larson
9. PERFORMING ORGANIZATION. NAME AND ADDRESS
Institute of Gas Technology
I IT Center
Chicago, Illinois 60616
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-0643
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
OAQPS, SASD, Cost Analysis Branch
Research Triangle Park, North Carolina
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
27711
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The energy utlization patterns and air pollutant emissions of the 10 largest
energy-consuming industries in the U.S. are presented in this report. Each
industry is described in terms of basic energy-consuming processes, and the
amount and types of energy consumed and the air pollutant emissions for each
process (especially those related to combustion) are presented. The energy
utilization efficiency of each process is discussed with a view toward increasing
efficiency either by improvement of the existing process or by replacement with
a new process. In addition, the effects of such changes on the air pollutant
emissions are determined.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Metal Industry
Chemical industry
Petroleum refining
Paper industry
Glass industry
Portland cements
Structural clay
Electric power
generation
Furnaces
Thermal efficiency
Cost engineering
Air pollution
None
0701
1001
1102
1106
1301
1302
1308
1401
2013
18. DrsTRYmjTTON STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
N/A
21. NO. OF PAGES
320
20. SECURITY CLASS (This page)
N/A
22. PRICE
EPA Form 3220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA reporl number as it appeals on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, it used, in snmllt-r
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, sulil volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/G R ANT NUMBE R
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not-included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COS ATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. & 20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but'exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |