AND STEEL
               ... ;
Public Health Service

                  Jean J. Schueneman
                     M, D. High
                     W. E. Bye
             Technical Assistance Branch
        Robert A. Taft Sanitary Engineering Center
                 Public Health Service
                Division of Air Pollution

                  Cincinnati 26, Ohio
                      June 1963

The ENVIRONMENTAL HEALTH SERIES of reports was estab-
lished to report the results of scientific and engineering studies
of man's environment:  The community,  whether urban, subur-
ban, or rural, where he lives, works, and plays; the air,  water,
and earth he uses and re-uses; and the wastes he produces and
must  dispose of in a way that preserves these natural resources.
This SERIES of reports provides  for professional users a  central
source of information on the intramural research activities of
Divisions and Centers within  the Public Health Service, and on
their  cooperative activities with  State and local agencies,  re-
search institutions, and industrial organizations. The  general
subject area of each report is indicated by the two letters  that
appear in the publication number; the indicators  are

                AP   Air Pollution
                AH   Arctic Health
                EE   Environmental Engineering
                FP   Food Protection
                OH -  Occupational Health
                RH -  Radiological Health
                WP -  Water Supply
                        and Pollution Control

Triplicate tear-out abstract cards are provided with reports in
the SERIES to facilitate information retrieval.  Space is provided
on the cards for the user's accession number and additional key

Reports  in the SERIES will be distributed to requesters, as sup-
plies  permit.  Requests should be directed to the Division iden-
tified  on the title page  or to the Publications  Office, Robert A.
Taft Sanitary Engineering Center, Cincinnati 26, Ohio.
 Public Health Service Publication No.  999-AP-l

    This review of the air pollution aspects of the iron and steel
industry was prepared to provide part of the basis for considera-
tion of means for control of air pollution problems that may arise
from operation of iron and steel works and activities incident
thereto.  In this report,  published and other information available
as of April 1961 has been collected, summarized, and intrepreted.
Attention is concentrated on air pollution aspects of the industry,
but background information is included to facilitate understanding.

    The report is divided into  seven major sections.  The sum-
mary is followed by a description of iron- and steel-making pro-
cesses and sections on the air  pollutants generated and their
control, effects of pollutants on the community, public health
aspects of emissions, and air pollution laws and regulations.  An
appendix contains certain information  on the size, location, and age
of iron and steel works and some historical data.

    The section dealing with calculated dispersion of pollutants
from open hearth furnaces was prepared by Messrs.  P. A. Hum-
phrey and F. Pooler (U.  S.  Weather Bureau personnel stationed
with the Public Health Service).  Mr.  S.  M. Rogers contributed
materials concerning legislation applicable to emissions of air
pollutants.   Dr.  May Sherman  prepared the discussion concerning
effects of emissions on human  health,  and Dr.  C. S.  Brandt as-
sisted with matters associated with effects of  emissions on vegeta-
tion and animals.

    The authors express their appreciation to  manufacturers  of
air pollution control devices for their  contribution of information
and to iron- and steel-making companies for their review of the



PART I  -  SUMMARY                                     1



    HOW IRON AND STEEL ARE MADE                    10

        Raw Materials                                  10
        Making Pig Iron                                 11
        Coke Production                                 13
        Sintering Operations                             16
        Making Steel in Open Hearth Furnaces             17
        Making Steel in Electric-Arc Furnaces             20
        Making Steel in Bessemer Converters              22
        The Basic  Oxygen Steel-Making Process           24
        Miscellaneous Operations                         26


        Raw Materials and Sintering Plants                27
        Coke Plants                                     28
        Blast  Furnaces                                  29
        Open Hearth Furnaces                           30
        Electric Furnaces                               31
        Bessemer  Converters                            31
        Basic Oxygen Furnaces                           32

             THEIR CONTROL                           33

        Sintering Plants                                 33
        Coke Production                                 35
        Blast  Furnaces                                  40
        Ferromanganese  Blast Furnaces                  43
        Open Hearth Furnaces                           45
        Electric-Arc Steel Furnaces                      57
        Bessemer  Converters                            64
        Basic Oxygen Furnaces                           67
        Miscellaneous Operations                         69


             COMMUNITY                                73

        Neighborhood Air Quality Studies                   73
        Relative Magnitude of Iron and Steel Mill
          Emissions                                      79
        Estimated Effect of Use of Oxygen Lances in
          Open Hearth Furnaces on Community Air
          Quality                                         81
        Effects of Fluorides From Certain Iron and Steel
          Works on Vegetation and Animals                 84

             FROM IRON AND STEEL MILLS               85

             AND THE IRON AND STEEL INDUSTRY        89

        Existing Laws                                    89
        Relationship of Some Regulations to Efficiencies
          of  Control Equipment                            96

REFERENCES                                           103

APPENDIX                                              119

    This report is a summary of published and
other information on the air pollution aspects of the
iron and steel industry, including coke plants inci-
dent thereto.  Processes,  equipment, and raw
materials are briefly described.  The magnitude
and location of plants and process trends are noted.
Air pollutant emissions and means for their control
are discussed in detail, with respect to sintering;
coke production; blast furnaces; open hearth, Bes-
semer,  electric, and basic oxygen steel-making
furnaces;  and other operations.   The effects of
pollutants on community air quality are described,
and knowledge of health aspects of pollutants is
summarized.  Laws regulating pollutant emissions
are given,  and control equipment and measures
needed to  comply with certain laws are listed.

                            Part  I


    This report is a summarization of published and other infor-
mation available as of April, 1961.  As in most fields, new in-
formation is constantly becoming available,  making it necessary
for the reader to consider this report in the light of develop-
ments since its preparation.

    The steel industry is concentrated primarily in about 20
metropolitan areas of the country.  Six states,  Pennsylvania,
Ohio, Indiana, Illinois,  Maryland, and New  York,  contain about
three-fourths of the Nation's iron- and steel-producing capacity.

    The first major step in the conversion of iron ore into steel
takes place in the blast furnace.  Iron ore,  coke, and limestone
are charged in alternate layers at the top and, to promote com-
bustion, a strong draft or blast is supplied by blowing heated
air into the lower part of the furnace.  To produce 1 ton of pig
iron requires, on the average,  1. 7 tons of iron ore,  0.9  ton of
coke, 0. 4 ton of limestone,  0. 2 ton of cinder, scale, and scrap,
and 4.0 to 4. 5 tons of air.   In addition to pig iron, the furnace
yields about 0. 5 ton of slag  and about 6  tons of exhaust gases
per ton  of pig iron produced.

    Blast furnaces produce 400 to 2400 tons of pig iron per day.
The average furnace produces 1000 tons.  At the same time it
generates 100 tons of flue dust (200 pounds per ton of finished
pig iron).  Blast furnace gases contain about 25 percent carbon
monoxide.  The gases are cleaned to prevent plugging of check-
erwork  in the heat-recovering stoves, to render the  gas usable
for heating coke ovens,  etc., and to recover process materials
such as iron and coke.  With only preliminary and primary
cleaners 2 to 3 tons of dust  would  still be emitted each day
whereas with high-efficiency control equipment the emission is
reduced to less than 0. 5 ton per day. Normally, gases are used
for heating or are burned in waste gas flares.  Carbon monoxide
in the gases is oxidized to carbon dioxide, which is of little
concern as an air pollutant.  The amount of  dust produced in
ferromanganese blast furnaces is about 50 percent greater than
that produced in iron blast furnaces, and the dusts are much
more difficult to control.  However, electrostatic precipitators
have been used with some success in controlling ferromanganese

    Two of the most important advances in  blast furnace opera-
tion in recent years are the use of beneficiated iron ore and the
use of sinter in blast furnaces.   This has resulted in a major
reduction in the number of "slips" in the blast furnace burden

                                  IRON AND STEEL INDUSTRY
and a concomitant reduction in emission of air pollutants asso-
ciated with slips.

     Coke, the chief fuel used in blast furnaces, is the residue
after distillation of certain grades of bituminous coal.  It is
usually produced in byproduct-type ovens although at times a
small amount may be produced in beehive-type ovens.  As many
as 100 byproduct-type ovens may be set together in a battery for
ease in charging and discharging the coal and coke.  With this
process,  the volatile matter emitted during coking is piped to
special equipment in which its valuable ingredients are extracted.
The coke oven gas subsequently is used for heating the coking
chambers or in metallurgical processes in the steel plant.

     Most of the coal distillation products generated in byproduct
coke ovens are confined to processing equipment and are not
emitted to the atmosphere.   However, smoke and gases  escape
during oven charging and discharging operations,  during car-
burization, through leaking oven doors,  and during coke quench-
ing operations.  Fundamental features of a coke oven battery
cannot be changed during its lifetime, which amounts to  20 to 30
years.  Finally, it is razed and a new structure erected to re-
place it.  Thus, operating procedures and design  of new ovens
should receive major emphasis for reduction of air pollution.

     In beehive-type ovens,  distillation products of the coking
operation are emitted to the atmosphere.  The emissions may
amount to 25 percent by weight of the coal charged to the oven
and include smoke, dust, sulfur gases,  carbon monoxide, and a
host of organic compounds.

     Sintering plants are used to convert iron ore fines and blast
furnace flue dust into a product more suitable for  charging into
the blast furnace.   This is achieved by burning a mixture of ore-
bearing fines and coke breeze or other fuel on a slow-moving
grate through which combusion air is drawn.  Modern sintering
plants have capacities of 2000 to 6000 tons of finished sinter  per

     The sintering process generates dust from two sources.
The primary source is the combustion waste gas,  which  entrains
dust as it passes through the sinter bed.  The  average-size
machine would discharge about 10 tons of dust daily if un-
controlled (20 pounds per ton of finished  sinter), but only about
1 ton daily if centrifugal separators are installed.  As the
sinter leaves the  moving bed, it is  broken,   screened, and
cooled. Dust generated at this point from an average machine
totals about 11. 2 tons daily (22 pounds per ton of finished sinter)

It may be substantially controlled by use of enclosures exhausted
through centrifugal separators.  In addition to the dust, about
300 pounds of sulfur dioxide and various amounts of other
gaseous combustion products would be emitted daily.

    The steel-refining process reduces the quantity of impur-
ities in the pig iron or steel scrap.  Common impurities which
must be controlled are carbon, manganese,  silicon,  sulfur,
and phosphorous.  The most common steel-making furnace is
the open hearth, which produces 82 percent of the steel in this
country directly, plus another 4 percent that is partially proc-
essed by Bessemer converters.   Electric furnaces, basic oxygen
furnaces, and Bessemer converters are also used.  The most
important modification in the steel-making processes has been
the increasing use of pure oxygen.  Today, 20 to 25 percent of
the open hearth furnaces use oxygen; oxygen-fuel mixtures are
used in electric  furnaces to supplement scrap meltdown; oxygen
enriched air blasts increase the amount of scrap that  may be
used in Bessemer converters; and the basic oxygen process (LD
process) is  gaining wide acceptance in the steel industry.

    In the open hearth process for making steel, a mixture of
scrap iron and steel and pig iron is melted in a shallow rectan-
gular basin  or hearth in which oil, coke oven or natural gas,
tar, or producer gas  provide heat.  Some limestone and other
materials are added.  Impurities are removed in a slag, which
forms in a layer on the molten metal.  Oxygen injection into the
furnace speeds the refining processes, saves fuel, shortens
furnace cycles, and increases steel production rates.  A com-
plete cycle (one heat) takes about 12 hours for  conventional fur-
naces, but with the use of oxygen lances or an oxygen enriched
fuel the heat time may be reduced to 8 hours.

    Open hearth furnaces without oxygen lances  generally gener-
ate about 5. 4 pounds of fume per ton of finished steel or about
1 ton of fume and dust daily.  With oxygen lances and  continuous
operation, the amount of fume produced daily is  about 50 percent
greater.  In terms of the amount of fume per ton of steel pro-
duced, the increase is about 25 percent since more steel is pro-
duced per day when oxygen is used.  Sulfur dioxide and other
pollutants formed by the combustion of fuel in the furnace also
are emitted.  In a very few western plants located where available
iron ore has a high fluoride content, emissions of particulate
and gaseous fluorides have created damage to vegetation and
animals.  Because of the small particle size of the fume from an
open hearth,  only certain types of high-efficiency collectors are
effective in removing a large percentage of the fume from open
hearth waste gas.  Five percent  of existing furnaces (44 of 906)

                                  IRON AND STEEL INDUSTRY
are now (March,  1961) fitted with air pollution control devices;
they represent 8 percent of the open hearth steel-producing
capacity.  Planned installations will soon bring the total to 77
controlled furnaces, or 15 percent  of the Nation's open hearth
capacity.  As a result of advances in steel-making technology,
the basic  oxygen furnace is replacing the open hearth furnace.
This will  influence future use of open hearth furnaces and in-
stallation of air pollution control devices on them.

     The basic oxygen process for making steel is carried out in
furnaces superficially similar to Bessemer converters.  In
1960, about 3 percent of the  Nation's steel was produced by this
relatively new process.  It is also  known as the LD (Linz-
Donawitz) process.  In the basic oxygen furnace, oxygen is
blown onto the surface of the bath at high velocity,  resulting in
violent agitation and intimate mixing of the oxygen with the mol-
ten pig iron.  An average furnace will produce 100 tons of steel
in a heat time of  1 hour.  The basic oxygen furnace generates
over 40 pounds of fume per ton of steel produced or about 50
tons of fume per  day.  However, all 12 basic oxygen furnaces in
use in the United States as of January 1,  1960, were equipped
with high-efficiency controls at the time they were built. Even
though these gas-cleaning installations are very expensive,  some
European operators believe that the value of recovered heat, fume,
and dust helps offset the cost.

     Electric furnaces used primarily to produce special alloy
steels produce 8. 5 percent of the Nation's steel.  Heat is fur-
nished by direct-arc-type electrodes extending through the roof
of the furnace. In recent years oxygen has been used to increase
the rate and uniformity of scrap meltdown and to decrease power
consumption.  An average-size electric furnace will generate
10. 6 pounds of fume per ton  of finished steel or about 1. 3 tons of
fume per  day.  Wide variations are reported.  Oxides of nitrogen
are formed at a rate of 0. 7 to 4.1 pounds per hour per furnace.
The quantity formed is a function of the degree of arcing during
heating and is not related to furnace size.  The characteristically
small particle size of the fume limits the type of control  equip-
ment capable of giving high-efficiency performance.  Fabric
filters are most commonly used for emission control,  but elec-
trostatic precipitators and venturi scrubbers also are used.
Information on the use of air pollution controls is incomplete,
but over 10 percent of the 301 electric furnaces at steel plants
are estimated to have control equipment.

    Bessemer converters may be used to make steel from pig
iron;  however, they are now  essentially obsolete.  Their use has
declined,  and at present, only about 2 percent of the Nation's

steel is produced in these devices.  The Bessemer converter
is a pear-shaped tilting steel vessel lined with refractory bricks
and clay.  The converter is tilted on its side to receive the
charge of molten iron, the air is turned on, the converter is
returned to a vertical position,  and the impurities are oxidized
by air blown through the molten iron for about 15 minutes.  An
average-size Bessemer  converter produces a 25-ton heat of
steel in about 30 minutes.  The  air blast may be enriched with
oxygen to increase production 15 to 20 percent. Bessemer
converters discharge about 17 pounds of fume per ton of product;
an average-size 25-ton-per-heat converter would discharge 10
tons of fume daily.  None of  the Bessemer converters in use in
the United States are equipped with air-cleaning devices.   The
primary difficulty in adapting control equipment to converters
is confinement of the effluent.  Although several control methods
have been suggested and research is continuing, no reasonable
solution to this  gas-cleaning problem has been found.

     Air pollution measurements have been made in an effort to
determine the effect of emission from iron and steel works on
community air quality.  During and after the  1956 steel strike,
studies were made in four iron- and steel-producing communities
to determine changes in  pollution levels from the strike to post-
strike periods.  Since certain other activities that cause pollu-
tion are also curtailed or shut down during steel strikes,  the
changes in pollution levels reflect total pollutant emissions from
all activities, including iron and steel works.  Measurements
were made 0. 125 to 1 mile away from the  steel mills.  When the
mills were operating, suspended particulate levels were 44 to 171
percent higher, soiling levels were as much as 100 percent
higher, and iron content of suspended particulate was 260 to 1080
percent higher than during the strike.   A similar study in  1950
showed soiling levels 50 percent greater during the post-strike
period.  Other studies have shown that pollution levels around
steel mills are  many times greater than average community
levels.  Typical levels of pollution in large cities (population
 100,000 to 1,000,000) are: dustfall 10 to 60 tons per square
mile per month; suspended particulate matter,  100 to 175  micro-
grams per cubic meter;  and  soiling index, 1 to 2 Cohs per 1000
linear feet of air.  Pollution levels several times these values
occur from time to time in some cities. Air samples collected
very close to steel mills in heavily industrialized upper Ohio
River Valley communities showed dustfall of  123 to 556 tons per
square mile per month,  suspended particulate matter of 62 to
 1238 micrograms per cubic meter, and soiling indexes of 5. 3 to
 5. 5 Cohs per 1000 linear feet.   In another study in the same area,
however, at air-sampling stations about 1.75 miles from a steel
mill,  suspended particulate pollution levels were not excessive.

                                  IRON AND STEEL INDUSTRY
Studies showing high pollution levels  near steel mills have been
conducted elsewhere in the United States and in England,  Ger-
many, Belgium, and Russia.

     Iron and steel mill operations may add substantially to the
total quantity of particulate matter emitted into the air over a
community, unless means to prevent pollutant emissions from
the mill are utilized.  For example, the weight of particulate
matter emitted from 13 open hearth furnaces without air pollu-
tion control equipment  is roughly equivalent to particulate
emissions from about 35,000  coal-fired home heating plants.

     No conclusive evidence exists to show whether or not air
pollutant emissions  from the iron and steel industry, by them-
selves, are involved in producing adverse effects upon human
health. Because of  the lack of definitive information concerning
the possible hazard  of iron- and steel-plant emissions, either by
themselves or in combination with other pollutants,  studies should
be conducted in the laboratory and the field to remedy this

     Many communities have adopted air pollution control ordi-
nances.  Such laws should be  tailored to meet the needs of a
particular area.  Many air pollution control ordinances adopted
years  ago embody limitations designed to control emission of
particlate matter from  coal-burning furnaces.  Although these
ordinances are not the most appropriate for use in bringing about
control of emissions from most iron and steel operations, a num-
ber of them include such operations among those to be regulated.
A few jurisdictions have limited particulate emissions on the basis
of the weight of materials introduced into a process  (the process
weight), exclusive of liquid and gaseous fuels and combustion air.
The  persent of the process weight that may be discharged de-
creases as the process  weight increases; thus more-efficient dust-
collection equipment is  required on larger operations than on
smaller ones.  In addition, the Los Angeles County law (for ex-
ample) prohibits emission of more than 40 pounds of particulate
matter per hour from any process,  a severe regulation made
necessary by meteorological conditions  existing in that area and
the large amount of pollutants  that were being emitted.

     Some ordinances such as that of Allegheny County, Pa.
have specific provisions for control of various steel  mill opera-
tions.  The Allegheny County ordinance also prohibits beehive
coke ovens and installation of new Bessemer converters until
control of their effluent is possible.

     Nearly all air pollution control ordinances limit visible
emissions,  particularly dense smoke.  In a few ordinances,
this provision has been adapted to include limitations on visible
emissions,  other than black smoke,  that have an opacity equiva-
lent to a plume of black  smoke.  Since many of the particles
emitted from steel mill  operations are small and cause a visible
plume even when the weight discharged is small, this limitation
is very restrictive.  A few ordinances limit emission of sulfur
dioxide to 0. 2 percent of the exhaust gases, and another is based
on ground level concentrations of the gas off the plant premises.

     To comply with the  more stringent regulations for control
of particulate matter in  Los Angeles and Allegheny Counties and
in the San Francisco Bay Area (for examples), most steel mill
operations would require highly efficient equipment such as
electrostatic precipitators,  fabric filters,  or  venturi scrubbers.
Less-efficient  collectors would be sufficient to insure compli-
ance with the more-lenient standards of certain other cities such
as Lorain,  Ohio.  The Chicago air pollution law presently (April
1961) exempts  certain iron and steel operations from air  pollution
regulations and provides that research shall be conducted to im-
prove means for control of pollutant  emissions.  In any case, local
ordinances  should take into consideration the topographical and
meteorological conditions and other factors affecting pollution
levels in the area of jurisdiction.
  GPO 823-573-2

                            Part  II


         Location  of Iron  and  Steel Works
    The majority of the iron and steel industry in the United
States is concentrated in about 20 metropolitan areas.  The
larger of these industrial complexes consist of integrated steel
operations (blast furnaces and steel-making furnaces, plus
accompanying apparatus  such as sintering plants, coke ovens,
scarfing machines,  rolling mills, etc.).  The  major iron and
steel centers are located in or near the following areas, 1  in
approximate order of decreasing capacity:  Pittsburgh, Penn-
sylvania; Gary-East Chicago, Indiana; Chicago,  Illinois;
Youngstown-Warren, Ohio; Baltimore, Maryland; Buffalo, New
York; Detroit, Michigan; Weirton, • West Virginia-Steubenville,
Ohio;  Cleveland, Ohio; Birmingham, Alabama; Aliquippa-Mid-
land,  Pennsylvania; Bethlehem, Pennsylvania; and Fontana,
California.  Large iron and steel  works are also located in or
near Middletown,  Ohio;  Fairless Hills, Pennsylvania; Lorain,
Ohio;  Canton-Massillon,  Ohio; Johnstown, Pennsylvania; Geneva,
Utah;  Harrisburg, Pennsylvania; Granite City-Alton, Illinois;
and elsewhere.

    Listings of the annual capacities of blast furnaces, coke
ovens, and steel-making furnaces in the U. S.  as of January 1,
1960,  are given in Tables Al through A6 (Appendix).
      Locations of Major Iron- and Steel-Producing Centers.


10                                IRON AND STEEL INDUSTRY

            How  Iron and  Steel  Are  Made


     Iron ore, plus fuel, flux, air, and water are the basic raw
materials required for making iron and steel.   Most iron ore
minerals are oxides of iron, either hematite (Fe^Oo) or
magnetite (Fe^O^), although there may be small proportions of
limonite (2  Fe2O3«3 HgO) or siderite (Fe COn)  present.  In pure
form, these minerals contain from 60 to 70 percent iron.  In
addition to the iron minerals, the ores contain varying amounts
of impurities (gangue), which consist mostly of silicon and
aluminum compounds,  plus moisture. %> ^> ^ In the  United
States, depending upon the location and the type of ore, the iron
content ranges from about 35 to 65 percent. Lake Superior ores,
which supply about 85 percent of the iron ore used in the United
States, average about 51 percent iron.'*

     To supply its heat requirements, the steel  industry depends
upon three major natural fuels: coal, oil,  and natural gas.  Of
these, coal is the most important, since it supplies more than
80 percent of the iron and steel industry's total annual heat and
power requirements.  The majority of the coal  is used in the
form of coke for the blast furnace process.  Oil and  gas are used
principally  in the manufacture  of steel.  About 70 percent of the
fuel oil used by the steel industry is consumed in metal melting,
mostly in open hearth furnaces.  Some natural gas is used in open
hearth furnaces; however, in recent years, more than 60 percent
of the steel  industry's consumption was burned in heat-treating
and annealing furnaces.

    In order to confine the molten metal and also conserve heat,
iron- and steel-making furnaces are lined with refractory brick.
It is necessary that the refractories have a chemical composition
that will not unite undesirable elements with the metal and that
will not react with the slag.  Three types of refractories are
used, acid,  basic, and neutral.  On this fact rests, in part, the
nomenclature of steel-making processes.  When acid slag and
flux are required to purify certain grades or iron and steel,  the
refractory brick must contain similar acids to avoid  the absorp-
tion of foreign elements into the steel.  When acid refractories
are used in  open hearth steel making, the furnace is  called an
acid open hearth.'  Silica brick is a popular acid refractory
since it has  the ability to carry heavy loads at high temperatures.
Basic refractories are those that contain magnesia or lime   such
as magnesite (magnesium ore) and dolomite (lime magnesium
carbonate).   Magnesite, the standard material used to make

Making Iron and Steel                                        11
bottoms of basic open hearth and basic electric furnaces, is
mined chiefly in California and Washington.  The most important
of the neutral refractories is chromite (chromium ore),  which is
used to repair basic open hearths and soaking pits.

    The procedures employed in making iron and steel are re-
fining processes; therefore, impurities that ordinarily would not
melt at operating temperatures must be removed.   Fluxes, such
as limestone (calcium carbonate) and dolomite, are used to com-
bine with the gangue elements and carry them off in a fusible

    Another additive, namely fluorspar (calcium fluoride), is
used in the melting process to make the slag more  fluid.  As a
result of the increased fluidity,  impurities are more quickly
removed from the molten steel and the rate of heat transfer is
increased, also.

    Water is used by the steel industry as a cooling agent,  as a
catalyst, as a conveying medium for transport of materials and
for disposal of waste, as a diluent or dispersive medium,  as a
cleansing agent, and in the production and distribution of heat
and power.  The industry uses nearly 5 billion gallons (21
million tons) of water daily.  A typical blast furnace with a
capacity of 1000 tons of pig iron per day will use about 11 million
gallons of water in 24 hours, primarily for cooling.

    Air is commonly overlooked in a discussion of the materials
of steel making.   However, without air, combustion could not be
supported.  Blast furnaces, open hearth furnaces,  and Bessemer
converters would be useless. *

     The first step in the conversion of iron ore into steel takes
 place in the blast furnace (Figure 1).  The blast furnace itself
 is a large steel cylindrical structure approximately 100 feet high,
 lined with heat-resisting bricks.   Iron ore, coke,  and limestone
 are  charged at the top,  and to promote combustion,  a strong
 draft or blast is supplied by blowing heated air into the lower
 part of the furnace.   The air blast is heated in "stoves, " which
 are  bricklined  regenerators (checker-work) enclosed in a
 circular steel shell with a flat bottom and dome-shaped top.
 Modern stoves, usually three per furnace, are 26 to 28 feet in
 diameter and over 100 feet in height.  The checker-work, which
 consists of a multiplicity of small passageways, contains between
 250,000 and 275, 000  square feet  of heating surface. 3

                                   IRON AND STEEL INDUSTRY
            Figure 1. Blast furnaces and stoves. (Courtesy American
                   Iron and Steel Institute)
     One method of "lighting" a furnace is to fill it with more
 coke and less ore  than the regular charge, with much wood at
 the  bottom for quick ignition.  Once in blast,  the furnace is
 operated continuously, day and night,  for long periods, until the
 lining wears  out or product demand falls off.   The furnace is
 hotter at the  bottom than at the top. As the raw materials melt
 and decrease in volume, the  entire mass of the charge descends.
 The addition  of alternate layers of ore,  coke,  and limestone
 compensates for the decreased bulk, and thus a constantly de-
 scending column of raw materials is maintained within the
 furnace.4  As the  charge  descends through the  increasing heat,
 the  iron oxide of the ore reacts with the hot carbon monoxide
 from the burning coke,  and the ore loses a large part of its

Making Iron and Steel                                        13
oxygen.  This reaction continues while the charge is in the top
half of the furnace.  About in the middle of the furnace, the coke
acts to take out still more of the oxygen  in the ore, and the lime-
stone begins to crumble and react with impurities in the ore  and
coke to form a molten slag.  As the charge enters the zone of
fusion, all the materials but the coke become pasty or fused.
The iron becomes a porous mass.  It then passes through the
melting zone and becomes liquid.  In this zone the ash from  the
burned coke is absorbed by the liquid slag, while the  iron absorbs
silicon from the slag and  carbon from the coke. ^

    The  iron and slag form a molten mass in the hearth, the
slag floating on a pool of iron 4 or 5 feet deep. About every  4
or 5 hours iron and slag are drawn off or "tapped. "  The slag is
tapped more frequently than the iron.  From 100 to 300 tons  of
iron are  drawn off at each tap.   The hot-metal or ladle cars
which receive the iron range in capacity from 40 to 160 tons.
The latter usually is a special type of tank car that makes it
possible  to deliver hotter  iron to the steel works,  even though
it may be 20 miles away.   Most of the  metal  produced in the  blast
furnace is used in molten  form for the manufacture of steel in
open hearth and other types of furnaces.  ^

    To produce one ton of pig iron requires, on the average, 1. 7
tons of iron ore,  0. 9 ton of coke, 0. 4 ton of limestone, 0. 2 ton
of cinder, scale,  and scrap, and 4. 0 to 4. 5 tons of air.  In
addition to the pig iron, the furnace yields about 0. 5 ton of slag
and about 6 tons of gases per ton of pig iron produced.  Air con-
stitutes over one-half of the material entering the furnace,
whereas  gases constitute more than three-quarters of the
materials leaving the furnace.  The difference is due  to the fact
that much of the carbon and oxygen entering as solids, in the
coke and ore,  respectively, emerges as  gases.  These gases,
piped from the top of the furnace, are  rich in carbon monoxide,
which can be burned.  They are used to heat  stoves and generate
power.  About 20 percent  of the gas is required to heat stoves
and the remainder is used for steam generation, underfiring of
coke ovens,  or for the heating of soaking pits.  The heating value
of blast furnace gas is about 100 Btu per cubic foot, which  is
about one-tenth the heating value of natural gas. '

    Coke,  the chief fuel used in blast furnaces, is the residue
after distillation of certain grades of bituminous coal.  It is made
in two types of ovens, the beehive and the recuperative or
byproduct oven.  In either type of oven, the distillation or coking

14                                 IRON AND STEEL INDUSTRY
process consists mainly of driving off certain volatile matter,
leaving in the residue a high percentage of carbon mixed with
relatively small amounts of impurities.

     The beehive oven is the older, less used of the two types
of oven.  Their use at iron and steel works has nearly disap-
peared although a few are still used, especially during times of
maximum steel production or when the value of materials re-
covered in byproduct ovens is particularly low. 3  Its domelike
structure is built of refractory or fire-resisting brick.   It has a
flat floor sloping slightly toward the  front.  In the roof is an
opening through which coal is charged and the products  of distil-
lation and combustion escape.   A door in the front permits both
the  regulation of the amount of air admitted during the coking
process and the discharge of the coke after the  process has been
completed.   A typical beehive oven is about 12 feet in diameter
by 8 feet high, and will hold about 6. 5 tons of coal.  It is care-
fully insulated with loam or clay to prevent loss of heat.  As
many as 40 of them may be placed in a row. 4

     To start a cold oven,  wood and coal fires are stoked until
the temperature has reached the intensity needed to start coking.
Thereafter,  enough heat is retained  so that, as one charge is re-
moved through the door at the bottom,  a new charge can be put in
through the top and the fire maintained.  As the heat begins to
work, volatile gases are  ignited in the coal.  The fuel assumes a
pasty, or semi-fused, state during the process, and expands
appreciably.  When the smoke subsides from the hole in the roof
of the oven,  and the flames shorten at the surface of the charge,
coking is finished.  The bricks, which have been placed in the
door, are torn away, and the coke is sprayed with water.  This
causes  cooling and contraction in the charge, which then breaks
into irregular pieces having a column-like structure. This struc-
ture distinguishes beehive coke.  When cooled,  the coke is taken
off on a conveyor that ends at a  screen.  The fuel passes over
this "sifter, " allowing the dust to fall away,  and then slides down
a chute into cars for shipment. 4

     In the byproduct coking process, coal is  heated in the
absence of air.  The volatile matter is not allowed to burn away,
but is piped to special equipment that extracts its  valuable in-
gredients. After the extraction  process, some of the gas
(heating value, 550 Btu per cubic foot) returns to the ovens for
use in heating the coking chambers and for heating in other
processes in the steel plant.   These  ovens are rectangular in
shape.   They may be from  30 to  40 feet long,  6 to  14 feet
high, and 11 to  22  inches wide.   As many as  100 of them
may  be  set  together in a battery for ease  in  charging and

Making Iron and Steel
discharging the coal and coke (Figure 2).  A modern byproduct
oven can receive a charge of 16 to 20 tons of coal through ports
at the top.   The ports are then sealed and coal begins to fuse,
starting at the  walls of the oven, which may generate heat from
16000 to 210QOF.  The fusing works toward the center of the
charge from both walls,  and meets in the center, causing a
crack down the middle of the mass.  This crack  and the porous
structure of the byproduct coke are its distinguishing features.
When coking is finished (18- to 20-hour carburizing period),
doors at the ends of the oven chamber are opened, and the pusher
ram shoves the entire charge of coke into railway cars.  The
load is taken to a quenching station, where it is watered by an
overhead spray.  After this, it is taken to a wharf to cool prior
to screening. *
            Figure 2. Battery of forty chemical recovery coke ovens.
                   (Courtesy of American Iron and Steel Institute)

     The volatile products that have passed out of the ovens are
piped to the chemical plant where they are treated to yield gas,
tar, ammonia liquor, and light oil.  Further  refinement of the
light oil produces benzol, toluol,  and other complex chemical
compounds. ^

     The fundamental features of a coke battery cannot  be  changed
during its lifetime, which amounts to  20 or 30 years.  At  the end
of its life it is completely razed  and a new structure embodying
current technological ideas is erected to  replace  it. 5


    Sintering plants are designed to convert iron ore fines and
blast furnace  flue dust into a product more acceptable for
charging into  the blast furnace.  This is achieved by burning a
mixture of ore-bearing fines plus a fuel consisting of coke
dust, coal,  or wood shavings.   Combustion air is drawn through
the flat porous bed of the mixture (Figure 3).  The principle of
sintering is to supply just enough fuel to the material to be
sintered so that a sticky mass will be produced, but the material
will not be melted sufficiently  to cause it to run.^  The bed is
formed on a slow-moving grate (about 6 feet wide) composed of
receptacle elements having perforated bottoms,  known as pallets.
The assembly of such pallets end to end in a hinged or linked
arrangement comprises an endless  metal  belt with large sprockets
at either end approximately 100 feet apart. 5  The ignition furnace
is either gas or oil fired,  and  its purpose is to bring the fuel in
the charge to  its kindling temperature, after which the down  draft
of air through the bed keeps it burning.
      FigureS. Schematic of a sintering machine. (Courtesy of Mechanical
      Engineering, ASME)

Making Iron and Steel
     The sintered material is dumped from the grate as it passes
over the head sprocket upon a screen, the undersize becoming
the return fines, and the oversize, which is still at a red heat in
the center, passing to a sinter cooler.  The former practice of
spraying water on the hot sinter causes a severe thermal shock
that tends  to crack many of the larger lumps into small pieces.
Now, instead, the cooler is usually a large rotating apron upon
which the sinter is deposited and cool air is blown through
louvers located in the apron.  As the cooler reaches a certain
position, stationary scraper bars push the sinter off the apron
into cars or conveyors. ^

     Modern sintering plants have capacities  ranging from 2000
to more than 6000 tons of sinter per day.  One plant of the latter
capacity has a bed width of 12 feet and a bed length of about
150 feet.   Exhaust fans draw air through the  bed at a rate of
more than 500,000 cfm measured at a temperature of 350°F.
This flue gas volume is subject to cleaning for removal of
dust or fume.

     The exhaust system for control of dust incident to crushing
and screening of the finished sinter involves  air flows in the
vicinity of 150, 000 cfm. 5

     The open hearth furnace is the unit in which some 90 percent
 of the steel made in the country is produced.  In this process,
 steel is made from a mixture of scrap and pig iron in varying
 proportions, depending on the cost and availability.5 The object
 of the operation is to reduce the impurities present  in the scrap
 and pig iron, which consist of carbon, manganese, silicon,
 sulfur, and phosphorus, to the limits specified for the different
 qualities of steel.  The refining operation is carried out by means
 of a slag that forms a continuous layer on the surface of the liquid
 metal. This slag consists essentially of lime combined with the
 oxides of silicon, phosphorus,  manganese,  and iron, which are
 formed or added during the operation. ^

     Open hearth furnaces are of two types, depending on the char-
 acter of the refractory material that forms the basin holding the
 metal. Where the  refractory material is silica sand, the furnace
 is described as "acid, " and where the basin is lined with dolomite
 (or magnesite), it is termed a "basic" furnace.5  The furnace
 proper consists of  a shallow rectangular basin or hearth enclosed
 by walls and roof,  all constructed of refractory brick,  and pro-
 vided with access doors along one wall adjacent to the operating

                                   IRON AND STEEL INDUSTRY
floor (Figure 4).  A tap-hole at the base of the opposite wall
above the pit is provided to drain the finished molten steel into
ladles.   Fuel in the form of oil,  coke oven or natural gas, tar
from coke making, or producer gas  (a gas rich in carbon
monoxide manufactured by blowing a limited quantity of air
through a hot bed of solid fuel) is burned at one end.  The flame
from combustion of the fuel travels the length of the furnace
above the charge resting on the hearth. 5  Upon leaving the fur-
nace, the hot gases are conducted in a flue downward to a
regenerative chamber called checkerwork or checkers.  This
mass of refractory brick is systematically laid to provide a large
number of passageways for the hot gases.  The brick mass
absorbs heat,  cooling the gases to around 1200°F.  All the ele-
ments of the combustion system burners, checkerwork, and flues
are duplicated  at each end of the furnace, which permits frequent
and systematic reversal of flow of the  flame, flue gases,  and  pre-
heated air for  combustion.  A  system of valves in the flue effects
the gas reversal so that the heat stored in checkers is subse-
quently given up to a reverse-direction stream of air flowing to
the burners.  In some plants,  the gases leaving the checkerwork
pass  to a waste heat boiler for further extraction of heat which
reduces the temperature from around  1200°F to an average of
500°  or 600°.  Open hearth furnace capacities span a wide range.
II    II
 Figure 4.  Cross section of an open hearth furnace, showing regenerative checker
 chambers preheating incoming air.  (Courtesy of the American Iron and Steel Institute)

Making Iron and Steel                                        19
The median is between 100 and 200 tons per heat (batch of finished
steel), but there are many of smaller capacity and an increasing
number of larger capacity.  Time required to produce a heat is
commonly between 8 and 12 hours.5

    The open hearth process consists of several stages: tap to
start, charging, meltdown, hot-metal addition, ore and lime
boil, working (refining),  tapping, and delay.  The period between
tap and  start is spent on normal repairs to the hearth and plug-
ging the tap hole used in the previous heat.  During the charging
period,  the solid raw materials (which usually include a combina-
tion of pig iron,  iron ore, limestone, scrap iron, and scrap steel)
are dumped into the furnace by special charging machines.  The
melting period begins when the first scrap has been charged.  The
direction of the flame is reversed every 15 or 20 minutes. When
the solid material has melted, a charge of molten pig iron is
delivered direct from the blast furnace in large ladles and poured
into the open hearth through a spout set temporarily in the furnace
door.   This is the normal sequence for a "hot-metal" furnace, but
in the case of a "cold-metal" furnace, only solid materials (pig
iron and/or steel scrap) are added, usually in two "batch" charges.

    The hot-metal addition is followed by the ore and lime boil,
which is a bubbling action much like the boiling of water and is
caused by the oxidized gases  rising to the surface of the melt.
Carbon  monoxide is generated by oxidation of carbon and is
characterized by a gentle boiling action called ore boil.  When
carbon dioxide is released in the calcination of the limestone, the
more violent turbulence is called the lime boil.

    The aims of the working period are (1) to lower the phosphor-
us and  sulfur content to levels below the maximum level specified,
(2) to eliminate  carbon as rapidly as possible and still allow time
for proper conditioning of slag and attainment of proper process
temperature, and (3) to bring the heat to a condition ready for
final deoxidation in the furnace or for tapping.2  At the end of
the working period the furnace is tapped,  with the temperature of
the melt at approximately 3000°F.

    The delay period includes waiting time during the heat cycle
(e. g.  equipment breakdown,  tapping equipment in use on another
furnace,  etc.) plus repair work not usually done during the tap
to start period.  For normal  operation of  a 10-furnace shop as  a
whole,  the following breakdown  of the heat stages has been made:

20                                IRON AND STEEL INDUSTRY
      Period                             Percent of time in
                                         indicated period
      Tap to start                                6
      Charging                                   12
      Meltdown                                   12
      Hot-metal addition                          3
      Ore and lime boil                           38
      Working (refining)                           19
      Tapping                                    2
      Delay                                      8
     The use of consumable lances to inject gaseous oxygen into
 the bath during the refining period and speed the oxidation re-
 actions, shorten heat time, save fuel, and increase production
 has become more or less standard practice over the last 10 to 12
 years.  Within the last 3 to 4 years, water-cooled lances inserted
 through the furnace  roof have been coming into prominent use. °
 Frequently, oxygen  lances are used throughout the  heat with the
 exception of the charging and hot-metal-addition periods.  By
 use of high oxygen flow rates from hot metal to tap, production
 rates of 90 to 100 tons per hour are  conceivable in  a 300-ton fur-
 nace.  Oxygen consumption under these conditions ranges from
 600 to 1000 cubic feet per ton (900 to 1667 scfm during the period
 oxygen is being added). ^

    One company has been experimenting with oxy-fuel lances,
 i. e.,  the use of oxygen in combination with the fuel.  This pro-
 cedure plus the substitution of burned lime for limestone has
 increased the steel output of a 200-ton furnace from 20 to approx-
 imately 30 tons per hour.   The oxy-fuel lance was developed for
 application in the open hearth furnaces of one  company and is a
 relatively new procedure.  Oxy-fuel  lances currently are being
 constructed for other companies.10

    The function of electric furnaces is in general much more
specialized than that of open hearth, Bessemer, and basic oxygen
furnaces,  in that the former are especially adapted to and are
primarily used for the production of special alloy  steels. 5

Making Iron and Steel
     The furnaces employed in electric-arc melting practices in
the steel industry are refractory-lined cylindrical vessels with
large carbon electrodes passing through the furnace roof (Figure
5).  They are normally of the three-electrode, direct-arc type
with supply currents ranging from 10,000 to 20,000 amperes.
Electric furnaces range in size from about 7 to 22 feet in diam-
eter and produce from 2 to 125 tons  of steel per batch.  Within
the past 4 years, furnaces of 200-ton capacity and with shells
approaching 30 feet in diameter have been  installed. ^> H-
              Figure 5. Electric furnace tor making steel.
                     (Courtesy of the Wheelobrator Corporation)

     Both acid and basic process cycles are  essentially batch
operations requiring 1. 5 to 4 hours.  The cycle consists of the
meltdown, the molten-metal period, the boil,  the reducing or
refining period, and the pour (tap).H  The stages of this cycle
closely approximate those of the "cold-metal" open hearth with
the exception that a shorter over-all time is needed.

     The impetus in the steel industry, because of the development
and  expanding use of higher alloy and stainless steel, has lead to
the increasing use of the basic lined furnace.  This furnace can
employ both high- and low-grade alloy scrap and plain carbon

22                                IRON AND STEEL INDUSTRY
scrap to produce steels that meet the stringent chemical,
mechanical, and purity specifications for straight carbon, and
high- and low-alloy steels. ^

    Electric furnace operations have involved use of oxygen for
more than 15 years.  The process of utilizing oxygen-fuel gas
burners during the scrap meltdown period has been reported to
increase production from 15 to 20 percent and decrease power
consumption 15 to 20 percent.  The use of oxygen has resulted in
uniform scrap-melting rates.  Evaluation tests on a 50-ton fur-
nace were made at an oxygen flow of 860 cubic feet per ton and a
natural gas flow of 545 cubic feet per ton. 13

     Bessemer (Pneumatic) converters serve basically the same
function as open hearth and electric furnaces, i. e. they transform
pig iron to steel by lowering the carbon, silicon, and manganese
content according to the desired quality of the finished steel.
They are essentially obsolete and presently account for only 2
percent of the steel produced in the U. S.  The Bessemer con-
verter receives a charge  of molten pig iron in quantities ranging
from 25 to 30 tons in the case of older converters on up to double
or more that quantity in more recently constructed units.  Con-
verters of 30-ton capacity can effect the complete oxidation re-
action in 10 to 15 minutes; however, this production rate advantage
over open hearth furnaces is reduced somewhat by the limited
extent to which converters can melt scrap  metal.5  Bessemer
converters are cylindrical steel vessels  lined with refractory
and with a spout or nose surmounting the top at an angle with the
main axis and are mounted on trunions on which they can rotate
(Figure 6).  One of the trunions is hollow and serves as an air
duct for passage of air from a blower to  a  chamber, at the bottom
of the vessel,  known as the wind box.  Air passes  upward (when
the  converter  is vertical)  into the molten metal through holes
(tuyers) in the refractory  barrier separating the wind box from
the  molten-metal bath.  Air pressure prevents  metal from
trickling downward through the tuyers into  the wind box.  When
the  converter  is tilted 90 degrees on its side, the surface of the
molten metal is below all tuyers and the  air blast can be  shut off
when the converter is in that position. °

Making Iron and Steel
             Figure 6.  Bessemer converter during charging and
             blowing operations. (Courtesy of the American Iron
             and Steel Institute)

     The operating sequence of a Bessemer converter is outlined
in the following extract:^

             The  vessel is turned on its trunions until it
         assumes an almost horizontal position and scrap,
         scale,  or ore is dumped into the vessel, if de-
         sired.  The molten pig iron is  then poured in.
         The blast is started and the vessel turned to  a
         vertical position and it remains in this position
         throughout the balance of the blowing period,
         unless "side-blowing" is  resorted to for increas-
         ing temperature.   The Bessemer blow is usually
         divided into three  parts, the first period, the
         second period and the after blow.  The first
         period or the silicon blow, begins as the blast
         is turned on and the vessel turned up.  During
         this period of the blow,  a short transparent flame
         extends from the  mouth of the vessel.  As the
         blowing continues  the flame starts to lengthen
         after about four minutes and the second period

24                                IRON AND STEEL INDUSTRY
         or carbon blow begins.  It is during the second
         period that the flame attains its full brilliance
         and length,  extending as much as 30 feet beyond
         the mouth of the converter.   This flame results
         from the evolved carbon monoxide burning to
         carbon dioxide as it comes in contact with the
         air at the mouth of the converter.   The long,
         brilliant flame which is characteristic of the
         carbon blow continues until  the elimination of
         carbon approaches completion, whereupon there
         is a definite change in the appearance of the flame.
         The length of the flame gradually drops and it
         seems to fan out.   When the blow is subsequently
         terminated,  the vessel is turned down and the
         blast turned off.

     In recent years,  it has been reported-^ that the air blast may
be enriched with oxygen or a mixture of oxygen and steam may be
used instead of air to increase production from 15  to 20 percent.

    A smelting process that is superficially similar to the
Bessemer converter is the top-blown oxygen converter or the
basic oxygen furnace.  It is also known by the term Linz-Donawitz
(or simply LD) process,  derived from the names of two Austrian
towns where early developmental work occurred.  The converter
vessel in this process is similar to a Bessemer converter although
considerably larger than most of them. 5  Vessel capacity (since
1954) has  gone from  50 to 250 tons and  oxygen blown from 3, 000
to 20,000  scfm.15 A principal  difference between  basic oxygen
and Bessemer furnaces is in the means for supplying oxygen to
the molten metal.  Instead of bubbling air under pressure upward
through the bath of molten pig iron, a stream of oxygen is supplied
through a water-cooled pipe extending from an overhead position
downward into the converter, the end being positioned at some
distance above the surface of the bath (Figure 7).   The high
velocity of the oxygen results in impingement on the liquid-metal
surface, which results in violent agitation and intimate mixing of
the oxygen with the molten pig iron.  Rapid  oxidation of the dissolved
carbon and silicon (and also of some of  the iron) ensues.

                                                      GPO 822-100-2

Making Iron and Steel
                                      DUST RECLAIMER
                                       OXYGEN LANCE
                                          OXYGEN AT
              Figure 7. Basic-oxygen steel-making furnace.
              (Courtesy of the American Iron and Steel  Institute)

     The following cycle for an 82-ton heat was reported in
January 1960:16

              Sequence                   Time, min.

      Charge scrap (55,000 Ib)               2. 39
      Charge hot metal (130, 000 Ib)           2. 68
      Prepare for oxygen                     1.78
      Oxygen time  (6, 500 cfm)              20. 22
      Temperature and inspection            1.82
      Temperature adjustment               5. 46
      Tapping time                           4. 32
      Slagging off                            1- 78
      Heat time                             40.45
      Delays                                 6. 12
      Tap to tap time                        47.02
      Production rate  106 tons per hour

26                                IRON AND STEEL INDUSTRY
     For this same installation, a 108-ton heat has been proposed,
with a tap-to-tap time of 55. 41 minutes for a production rate of
117 tons per hour. 1"

Heating and Reheating Furnaces

     Following the refining operation in the open hearth, Bessemer
converter, electric, or basic oxygen furnace,  the molten steel is
poured from the furnace into a ladle and subsequently into a mold.
The solidified steel castings are called ingots.  Before these
ingots are rolled, the metal is allowed to solidify throughout.
A heat supply is carefully manipulated to bring the whole body of
metal to a uniform rolling temperature.  In the past,  ingots were
placed in holes in the ground, covered, and allowed to "soak"
until rolling temperature was reached.  Hence the term "soaking
pit" arose. In order to control the soaking operation better and
render it adaptable to varying operating conditions, means for
supplying heat were introduced.  Thus the modern soaking pit is
a kind of heating furnace.  Throughout the steel plant, heating
and reheating furnaces are used to prepare the steel for various
finishing processes.  Fuels for these  furnaces frequently are
coke oven  gas, blast furnace gas, or other byproducts produced
within the  steel mill.
Scarfing Machines

     As the technology of steel fabrication has developed, so has
the requirement for still higher qualities of steel products.  Ever
increasing attention has been devoted to conditioning of semi-
finished products.  A major  element in this area involves the need
for removing surface defects of blooms, billets, and slabs prior
to shaping, as by rolling, into a product for the market.  Such
defects as rolled seams, light scabs, checks, etc.,  generally
retain their identity (although not shape) during  subsequent form-
ing processes and result in products of inferior quality.

     In the earliest  days of the development, pneumatic chisels
were employed to remove such surface defects.   About 25 to 30
years ago the scarfing process developed and today represents an
important operation in the making of high-grade steel products.
It consists essentially of supplying streams of oxygen as jets to
the surface of the steel  product under treatment while maintaining
high surface temperatures  that result in rapid oxidation and local-
ized melting of a thin layer of the metal.  Originally the process

Making Iron and Steel                                       27
was a manual one consisting of the continuous motion of an
oxyacetylene torch along the length of the piece undergoing treat-
ment. In recent years the so-called hot scarfing machine has
come into wide use.   This is a production machine adapted to
remove a thin layer  (one-eighth inch or less) of metal from all
four sides of red-hot steel billets, blooms,  or slabs as they
travel through the machine in a manner analogous to the motion
through rolling mills. 5
 Power Plant Boilers

     Boilers in steel plants generate steam for driving blowers,
 electric-power generators,  service water pumps,  and a multitude
 of miscellaneous equipment, as well as provide steam for the
 heating of buildings, shops,  and offices and for general process
 work.   The efficient utilization of byproduct fuels provides a
 surplus, in well-integrated plants, for the  generation of electric
 power.  A  modern fully integrated plant is  capable of generating
 all of its own power from available surplus byproduct fuel.  Dur-
 ing the  years 1943-1945, 40 to 45 percent of the electrical power
 requirement of the  steel industry was generated by the industry
 itself. 2
   Magnitude of Operations  and Process  Trends


     Probably in no other field of the iron and steel industry has
 so much progress been made as in the improvement of raw ma-
 terials.  In the past, iron ores were charged into blast furnaces
 as direct-shipping ores from the mines. With beneficiation
 (removal of impurities such as sulfur and earthy materials),
 low-grade  ores have been made into high-grade  concentrates with
 iron contents ranging from 55 to 68 percent compared with direct-
 shipping ores of 45 to  53 percent iron content.  The production of
 sinter and  pelletized iron ore is now revolutionizing blast furnace
 practice. 1?

     According to Steel Facts, October  1956, there were 103 sin-
 tering machines under construction at that time. The new'machines
 (annual capacity 25 million tons per year) were scheduled for com-
 pletion before the end  of 1957 and would bring the total capacity
 to 63 million net tons per year.^°  In 1959, it was reported that
 the sintering capacity  of United States plants had doubled in the
 previous 5 years. ^

28                                IRON AND STEEL INDUSTRY
    In addition to an increased number of sintering plants, the
physical dimensions and therefore individual plant capacity have
grown in the last few years.  The first sintering plants,  con-
structed in  1913, were 3. 5 feet wide and 23 feet long.  In 1925 a
6-foot-wide machine was developed and in 1959 there were two
installations that were 12 feet wide.  One was 150 feet long and
the other was 200 feet long. ^0

     Sinter,  which often contains a major part, if not all, of the
flux materials for the furnace burden (self-fluxing sinter), to-
gether with pellets,  has  made possible daily iron production
increases of more than 50 percent; proportional decreases in
coke consumption result in lower iron costs. *'  In a recent test
on an 18-foot-diameter blast furnace,  production was increased
43 percent and the coke rate (pounds per ton of pig iron)  was
reduced 26 percent by using a 100 percent self-fluxing sinter
burden instead of a burden containing 40 percent completely
unfluxed sinter.21  In the future, as ore bodies of heterogeneous
structure are opened up,  new beneficiation methods, including
magnetic roasting and partial direct reduction, will probably be

     The number of slot-type (byproduct) coke ovens in use has
increased steadily since 1900,  except for a slight decline during
the depression years (Table A7, Appendix), while the use of
beehive ovens has been variable.  Coke is used in several
industrial processes in the United States.  However, slot-type
ovens in this report are dealt with only if used in plants connected
with the  steel industry, while the uses of beehive coke are not

     Until the close of World War I beehive coke ovens were the
main source of metallurgical coke in the United  States,  reaching
a maximum in  1910 when there were more than 100, 000 ovens of
this type.    However,  they have been marginal producers of
coke since 1919.  In the depression years of the 1930's the number
of beehive ovens dropped to 10, 816.  The demand for blast furnace
coke in the years preceding the United States entry  into World War
II made it necessary to rehabilitate many ovens that had been idle
for many years.  By the end of 1941,  18, 669 ovens  were available
for producing metallurgical coke.  The number remained approx-
imately the  same during the war years but dropped  to 12,179 at
the end of 1945. Increased demand for coke during the Korean
War again made it necessary to press into service a substantial
number of idle beehive ovens.  In 1951 the number of ovens rose

 Making Iron and Steel                                       29
to 20,458, with an annual coke capacity of 13.9 million tons.
Since that year, however, steel companies have constructed
slot-type coke ovens to support expansions in blast furnace
facilities, and demand for beehive coke has diminished,  re-
sulting in a decline in number and capacity of beehive ovens. ^4
At the end of 1959,  there were 8500 beehive ovens in existence
and the iron and steel industry had 1392 of these. ^3

     The capacity of these 1392 ovens was 877,100 tons of coke
annually, compared to the total byproduct oven capacity from
13,816 ovens of 71, 432, 600 tons per year.1  The primary reason
for increased use of byproduct ovens  is that this process offers
long-range economic advantage from  the production of byproduct

     A byproduct coke oven expansion program was started in
the early 1950's and was virtually completed by the end of 1958
(Table A7, Appendix).   Most of the new ovens (808) constructed
in 1958 were rebuilds or replacements of worn-out ovens; only
117 represented additional capacity.  As a result of the construc-
tion and  modernization program, coke ovens were in the  best
shape in many years and approximately 72 percent of the active
ovens at the end of 1958 were less than 20 years old (Table A8,
Appendix).  Also, the average age of  coke ovens at steel  mills
dropped  from 19 years in 1950 to 13 years in 1958. ^2

     With increasing technological advances in the blast furnace
process,  it is expected that the amount of coke  used to produce
a given amount of pig iron will continue to decline.  The ultimate
average  is expected to reach 1200 to 1300 pounds of coke per ton
of hot metal by 1975.  In 1948,  1947 pounds of coke was needed;
in 1957,  1703 pounds; and in 1959, an estimated 1500 pounds.
The declining coke rate will, however, be offset by a modest
growth in steel demand. ^5

    With reserves of low-sulfur coking coals limited,  the develop-
ment of new continuous  coking and desulfurization methods for the
production of coke from marginal coals will take place.  Our en-
tire concept of blast furnace coke quality, shape, and sizing will
probably change during the next decade.1'

    In 1960, pig-iron- and steel-ingot capacities were the highest
in the history of the industry.  In 1940 there were 232 blast fur-
naces with a total capacity of  55, 724,000 net tons per year,2^
whereas in 1960 these values were 263 and 96, 520, 630 respec-

30                                IRON AND STEEL INDUSTRY
tively16 (Table A9, Appendix).

     In addition to advances in raw material preparation, there
are many recent technological changes in blast furnace practice
that tend to increase production.  Several new techniques,  in-
cluding elevated blast temperatures with gaseous, liquid, or
solid fuel injection, together with moisture (steam) and oxygen
additions to the blast, are being used to achieve both decreased
coke consumption and increased productivity.  Another expected
development is the use of higher top pressures with possible
power recovery, of the energy of the top gases, by gas turbines. ^
The blast furnace operation is also becoming more automated
through the use of automatic stockhouses  and charging equipment,
as well as use of computers for overall process control.  These
developments are expected to bring production to or beyond the
3000-ton-per-day performance aim and will lower coke consump-
tion toward the  1250 pounds per ton figure. ^

     Recent developments in the field of direct reduction have been
reported.27  The first commercial operation of a new smelting
process that could open a new era for the steel industry is owned
by a New York corporation. The process is designed to take low-
grade ores,  contaminated ores, and extremely fine ores, which
would be unsuitable for conventional blast furnace use,  and produce
a high-quality iron that can be turned into steel more rapidly.
The process divides the smelting operation of the conventional
blast furnace into two processes.  The first,  a horizontal,
slowly rotating,  gas-fired kiln,  removes  55 percent of such im-
purities as oxygen and sulfur from ores of such low grade that
blast furnaces cannot handle them and conditions the  ore for the
second step by a specially designed  electric smelting furnace that
finishes the job.  The slag from the electric furnace  can be put
through a series of similar furnaces to draw off other metals
such as  chrome, copper,  zinc, and  manganese.  When developed,
this process is expected to be used in  remote steel-making areas
and not in the large integrated steel operations.

     In 1940 steel ingot capacity was 81, 619, 000 tons26 per year,
 and in 1960 it was  148, 570,970 l (Table A10, Appendix).  During
 the past few years and in years to come, increasing capacity will
 continue to result through enlargement of existing facilities and
 improved technology.

     The trend of production of steel from open hearth furnaces
 generally has been upward since 1920 (Table A10, Appendix).

Making Iron and Steel                                        31
 Virtually all open hearth steel comes from the basic rather than
 the acid process.  Although specific data were not obtained,
 increased capacity in the last few years was probably due to
 enlargement of existing furnaces and process modifications, plus
 construction of a few installations in the large size range (300 to
 400 tons).

     The most important process modification is the increasing
 use of oxygen, not only during the refining period but also during
 the meltdown and the ore and lime boils.  Oxygen lancing with
 consumable lances has become more or less standard practice
 over the last 10 or 12 years.  Within the last 2 or 3 years, oxygen
 injection with roof lances has become prominent.  It has been
 estimated28 that from 20 to 25 percent  of the open hearth furnaces
 in the United States are equipped with roof lances, and this figure
 is rapidly increasing.

     The annual steel production from electric furnaces has in-
 creased rather slowly since 1920 (Table A10, Appendix).  Except
 in areas of low electric power costs and for special applications
 (high-quality steel), the electric furnace is not competitive with
 open hearths.

     Electric furnace operators have been using oxygen for more
 than 15 years.  It is used for  scrap preparation, scrap cutting
 in furnace  doors, tap hole preparation and clean out,  decarboniza-
 tion, temperature control, and to increase flushing action and
 improve alloy recovery in stainless steel production.  "

     A  recent development has involved the addition of oxygen-fuel
 gas burners to accelerate scrap meltdown  and thereby increase
 production and reduce power costs.  When these burners are
 included in the design of new furnace  installations,  their inclusion
 can conceivably reduce capital expenditures while increasing
 production capacity. ^

    Bessemer converters are obsolete.  Their use declined
steadily during the first half of the century (Table A10,  Appendix),
and they presently produce only about 2 percent of the Nation's
steel.  In  1950 the rated converter capacity in the U. S. was
approximately 12,000,000 tons per year, with a considerably
greater potential capacity.  However, over half of this capacity

32                               IRON AND STEEL INDUSTRY
was represented by blown metal for subsequent use in open
hearth furnaces (duplex process). 2 In 1960 there were 21
Bessemer converters with a capacity of 3, 396, 000 tons of ingots
per year plus 11 others that were used in melting the charge for
open hearth furnaces. 1

     In most cases the cost of building a Bessemer plant is appre-
ciably less than the cost of an open hearth plant of equivalent
capacity.  This is offset by the fact that greater blast furnace
capacity is required for operation of a Bessemer plant since
only a small amount of scrap can be processed (10 percent
compared to 35 to 60 percent for open hearth).  Oxygen enrich-
ment of the air blast or an oxygen-steam blast may increase the
use of scrap in Bessemer converters to about 25 percent. 2

     The Bessemer process has certain economic advantages when
scrap is scarce and costly, but there are  periods in our economy
when scrap is plentiful and cheap. Therefore, economic pres-
sure over a period of years is a controlling factor in the  Bessemer
steel-making capacity of the Nation. 2

     As of January 1958, there were 26 basic oxygen furnaces in
the world,  with an ingot capacity of 7, 000, 000 tons.  Eight of
these were in North America, with a capacity of 5, 000, 000 tons.
The largest of these was rated at 65 tons.  As of January 1960,
2 years later, there were 90 furnaces built or being built with an
annual capacity of 28, 000, 000 tons.  Of these, 22 are in North
America, with  a capacity of 12, 000, 000 tons  (for existing capacity
in U. S., see Table All, Appendix).  Aside from three 110-ton
vessels proposed in England, the average vessel used throughout
the world (outside of North America) is rated at about 45 tons.
On the other hand, in North America, the average vessel size is
about 75 tons, with the largest one being  200 tons.15

     It has been estimated that a basic oxygen steel plant, includ-
ing the oxygen-generating plant,  can be built for less than 50
percent of the cost of  an open hearth plant that produces the same
tonnage. The basic oxygen furnace has somewhat lower operating
costs and produces a product of quality equal to that produced^
in the open hearth. 29  The basic oxygen process has one limita-
tion, similar to the Bessemer process, in that a limited amount
of scrap can be charged. Scrap  charged, believed limited to 25
percent originally, is now approaching 35 percent in some shops.  1(

                           Part III


               AND  THEIR CONTROL


Rate of Pollutant Generation

    According to an article published in 1957, there were 103
sintering plants in the United States with a combined annual
capacity of 38 million net tons;30 thus, the average-size facility
at that time would  have produced 1000 tons of sinter per day.
Dust in the sinter combustion gases from such a plant would be
about 10 tons per day.  One investigator in the United States
reports a tremendous variation of dust emission, 2. 7 to 50. 7
tons per day, from two similar-sized (1000 to 1250 ton per day)
sintering machines. 31

    Three  separate British investigators reported that 7. 5,  9,
and 12 pounds of dust are carried in the combustion gas for each
ton of sinter produced. 32,  33, 34 Q'Mar a reports a wide varia-
tion in loading, from 5 to 100 pounds per ton.31  For the U.  S.
plants a value of about 20 pounds of dust per ton of sinter is
likely.  For each ton of sinter produced, there is 200,000 standard
cubic feet (scf) of waste combustion gas to be cleaned. 33 This
effluent has a temperature  of about 160° to 390°F.  About 120 to
160 scfm of waste  gas is evolved per ton of sinter produced per
day with a  dust loading of 0. 5 to 3 grains per scf. 31
    As the sinter leaves the end of the sintering machine, it is
broken, screened, and fed to a cooler.  At one British plant
where this operation is enclosed and vented, dust loadings were
6. 2 grains per scf of discharged gas. 34  The air volume exhausted
at this point would be 15,000 to 20,000 cfm for a 1,000-ton-per-
day sintering plant.  At a 2500-ton-per-day plant,  a volume of
40, 000 cfm was reported, whereas,  at a_10, 700-ton-per-day plant,
a volume of 192,000 cfm was measured. 34, 35 At an assumed dis-
charge of 17, 500 cfm containing 6. 2  grains of  dust per scf, a 1000-
ton-per-day plant would emit 11. 2 tons of dust daily if emissions
were uncontrolled  from the product discharge  end of the machine.

    A few iron and steel works in the western part of the country
use an iron ore mined in southern Utah.  This ore contains an un-
usually large amount of fluoride,  and when it is passed through the
sintering process,  much of the fluorine is driven off. ' No specific
data were found on exit gas loadings, but fluorides are discharged


34                                 IRON AND STEEL INDUSTRY
from the sinter plant and open hearth furnaces in sufficient quantity
to cause a recognized problem in the community surrounding one or
two plants (see elsewhere herein).

Nature of Pollutants

     Most of the particles discharged from a sintering machine
are very large.   About 50 percent by weight of the particles are
larger than 100  microns. "  Particle size distribution has been
reported by a British investigator as follows:^

        > 420 microns                    3. 7 percent

     420 - 178 microns                   22. 6 percent

     178 -  76 microns                   36. 8 percent

        <  76 microns                   36. 9 percent

     A sulfur balance on an English plant shows that the sintering
process is an excellent desulfurizer of raw materials; 71 percent
of the sulfur present in the raw materials is carried up the
stack. 34  Mesabi Range iron ore contains 0. 01 percent sulfur on
a dry basis. ^ This  is equivalent to 0. 2 pound of sulfur per ton
of ore,  and 0.142 pound of sulfur per ton of ore is discharged
out the stack.  Thus a 1000-ton-per-day sintering plant would
emit 284 pounds of sulfur dioxide per day, if all sulfur in the
stack was in that form.

     Gaseous and particulate fluorides are emitted from a few
western sintering plants that are processing high-fluoride-content
ore. f  One review article 36 indicates that hydrofluoric acid (HF)
and silicon tetrafluoride (Si F4) are emitted.  Emission of other
compounds would seem possible.

Control Methods

     Since dust generated in the sintering operation can be returned
to the process, most plants are now constructed with centrifugal
separators installed as an integral part of the plant to clean the
combustion gas.  These  cyclones operate at an efficiency of over
90 percent by weight on the sinter dust,  because of large particle
size. 35, 37  Effluents from these cleaners contain from 0. 2 to
0. 6 grain of dust per cubic foot of gas. 31, 34,  38 Dry-type
cleaners are best suited for this cleaning because of  the sulfur
content of the gas stream. 33 Wet scrubbers would be troubled
by corrosion. The first electrostatic precipitator was installed
in series with a cyclone to clean the sinter combustion gas in

Air Pollutants Generated and Their Control                   35
 about 1952.    The precipitator operates at an efficiency of 95
 percent, and the final discharge  contains only 0. 05 grain per scf.
 Another similar installation cleans the gas to 0. 01 grain per scf. ^8
 One company is planning to construct a baghouse to control emis-
 sions from a sintering operation. These fabric filters should be
 effective in removing particulate matter.  If their service life is
 satisfactory, they may be  used more extensively in the future.

     To eliminate dust from the discharge end of the sinter
 machine, the operation may be enclosed and vented.  Cyclones
 have been used to clean this dust-laden air in both the U. S. and
 England. A British report shows an inlet loading of 6. 2 grains
 per cubic foot, an efficiency of 93 percent, and a final effluent
 containing 0. 43 grain per  scf .34

     Control of both gaseous and particulate fluorides has been
 accomplished at one or more sintering plants using a combination
 of methods.  Limestone is mixed with ore to be used in the
 sintering machine.  This reduces the amount of fluorides liber-
 ated in sintering by about  50 percent. Gas volume is reduced by
 recycling about 40 percent of the gases back over the sinter bed.
 Final clean-up of exit gases is done by feeding powdered limestone
 into the gas  stream to react with gaseous fluorides  and form
 particles that contain the fluoride.   The particles are then passed
 through cyclones and finally an electrostatic precipitator.  At one
 plant,  this system reduced emission of fluorides by 96 percent. '

 Nature of Pollutants and Rate of Generation

     The main pollutants from byproduct coke ovens are smoke,
 dust, hydrogen sulfide, and phenols.  Other contaminants gen-
 erated by destructive distillation of the coal include pyridine,
 cresol, carbon monoxide, ammonia, methane,  ethane, and
 ethylene, in addition to a host of other organic compounds found
 in coal tar.  Some escape to the atmosphere mainly during charg-
 ing and discharging of ovens, and during carburization through
 leaking oven doors. Dust also arises from quenching of finished
 coke. 40

     Gaseous and particulate matter released in the coking opera-
 tion, except that which escapes from ovens to the atmosphere, are
 conveyed in ducts to a coal chemical processing plant. Treatment
 given materials in such plants varies widely.  At one plant, "*! tar
 and water vapor are removed by condensation.  The coal tar is
 dehydrated and sold to refiners or used as fuel in the open hearth.

36                                IRON AND STEEL INDUSTRY
    Ammonia and pyridine homologs are removed by absorption
in sulfuric acid.  The pyridines are sprung from solution and
sold as a crude for further refining.  The crystals of ammonium
sulfate, which form in the acid solution, are removed, dried,
and sold to the fertilizer industry.  The gas is further cooled,
to condense crude naphthalene, which is recovered and sold as
such.  The various light oil constituents are next removed from
the gas by absorption in straw oil.  These oils, benzol, toluol,
xylol,  etc., are stripped from the straw oil, separated, and
refined. ^1  The  coke oven gas remaining after these operations
amounts to about 11, 000 cubic feet per  ton of coal charged in the
ovens.  The gas has a gross heating value of 525 to 535 Btu per
cubic foot.  A portion of the total gas is diverted at this point and
used to underfire some of the ovens.  The remainder of the gas
is pressurized to 15 to 20 psig and distributed throughout the
steel plant.   The portion used in the open hearth furnaces is
treated for hydrogen sulfide removal. *1
    Some pollutants, such as those that escape during the charging
and quenching operations,  are emitted at only one time during each
coking period.  However, when a number of ovens are operated in
a single battery,  they are sequentially operated to effect a nearly
constant gas flow from the ovens.  Pollutants are, therefore,
discharged at a fairly constant rate.  No objective data were
found on emissions from charging or discharging operations, but
it is common to see large clouds of smoke escaping to the atmos-
phere at these times.  These clouds no doubt contain the destruc-
tive distillation products listed above.   Russian investigators
working near plants that may be very different from those in
the United States   report that sufficient phenol escapes to the
atmosphere from coke plants to cause public complaints  be-
cause of its irritating odor.
    Emissions from coke quenching have been investigated to a
limited extent.  One series of tests indicated an average loading
of 0. 073 grain per cubic foot for three samples, indicating that
emissions are minor. 43 Another investigator reports, howeyer,
that emissions are substantial and that a corrosion problem was
caused by droplets of water falling out of the plume from the
quenching tower. 44  A third group is of the opinion that the quan-
tity of emission is small and that since the particles are fairly
large  they tend to settle very close to the quenching operation
and therefore create no neighborhood problem.  This group
indicates the corrosion problem mentioned above is due to use of
contaminated water for quenching. 5

Air Pollutants Generated and Their Control                   37
     No specific data on pollution emissions from byproduct pro-
 cessing plants were found.  One would expect some losses to the
 of the oven because of the distance the gas must travel and the
 partial obstruction caused by piling up of coal as portions of the
 oven become nearly filled. 5  One arrangement for minimizing
 this effect, available for incorporation in new plants, provides
 two gas-collecting mains and a gas ascension pipe with a steam
 jet aspirator at each end of the oven.  Thus the length of gas
 travel to the point of exit is halved and escape of smoke is
 lessened. 5 it should be observed that effectiveness of smoke
 control with this arrangement is also greatly dependent on the
 charging operators, who must manually open and close the steam
 valves supplying the jet aspirators. 5

     Another design feature somewhat similar to the double main
 consists of a gas-pressure-equalizing main on the side opposite
 the collecting main. The, equalizing main serves simply to con-
 nect all ovens to each other through this  main; it is not connected
 to the principal gas main flue.  The equalizing main provides an
 avenue of egress for gas generated during the charging (which
 cannot leave through the ascension pipe into the principal col-
 lecting main) to pass into one of the nearby ovens that,  because
 it is in its coking stage, is at a lower pressure and can thus
 accommodate this gas.  This arrangement is somewhat less
 positive and,  therefore, less effective than the double main. ^

     One oven design element that minimizes the escape of smoke
 from the oven interior during charging is appropriate spacing of
 the charging holes in relation to oven volume.  The objective in
 this respect is to avoid piling up of coal anywhere in the oven in
 a manner that creates a barrier to the free passage of gases from
 any part of the oven toward the gas takeoff flue.  This condition
 is attained by calculated spacing of the charging holes in relation
 to the volume of the oven. 5

     Another feature that has received attention in recent years
 aims to provide an  enclosure between the hopper of the larry car
 and the top of the charging opening to prevent the escape of smoke-
 laden gases from the oven interior during charging.  This takes
 the form of drop sleeves and shear gates.  The operator lowers
 the sleeve to the top of the oven preparatory  to charging the coal
 and closes the gate of each hopper as  soon as it has been emptied,
 the latter to prevent the passage of gases upward through the hop-
 per.  Immediately following this operation the operator trips the
 dropsleeve to raise it and quickly  replace the lid.  Then the leaks
 around the charging-hole lid are sealed either by luting or by dry
 seal. 5

38                                IRON AND STEEL INDUSTRY
    Since a part of the problem of allowing free passage of gas
from one end of the oven to the other, during charging,  is related
atmosphere: materials in the coke oven gas and some of the by-
products produced.

    In beehive coke ovens, all of the materials resulting from
destructive distillation of coal are emitted to the atmosphere
(see composition above).  There are large clouds of smoke and
objectionable odors associated with this process.  No quantitative
data were found.  However,  a material balance would indicate
that about 25 percent of the coal charged to the oven is emitted as
gaseous and particulate pollutants.  This varies widely, depending
on fuel composition and carburization temperature.

Control Methods

    The Technical Coordinating Committee  (TI-6) of the Air
Pollution Control Association has published  a review of  methods
for controlling emissions from coking operations. •* The following
material is taken from the committee's report and includes most
of that published.

    Soon after an oven has been emptied of coke at the end of the
coking cycle, it is refilled with a new charge of pulverized coal,
which may range in size from 3-inch lumps  down to pulverized
material, and since the oven interior is at bright-red temperature,
volatilization of gases from the coal mass begins at once and
escapes to the exterior as visible clouds of yellow-brown smoke. ^

    Considerable progress has been made over the years in
reducing the quantity of such smoke partly in arrangements and
practices for minimizing its escape from the interior and partly
in shortening the time required for charging, since in the latter
respect the sooner this is accomplished and the openings closed,
the less is the total emission.  Some of the positive accomplishment
is ascribable to improvements in coke oven design and some to
improved operating practices. 5

    A number of newer design features have been under discussion
in recent years, on some of which agreement is general and on
others differences of opinion prevail within the  industry.  One  im-
portant improvement over older practice includes facilities for
aspirating gases from the interior of the oven during charging by
means of steam jet aspirators in the ascension pipe elbow, whereas
in former times it was the practice to seal off the oven interior
from the gas-collecting main during charging.  This resulted in
the emission of enormous quantities of smoke during this period,  in
comparison to that under current practice. 5 Even though a steam

Air Pollutants Generated and Their Control                   39
jet aspirator is operated in the gas ascension pipe during charging,
some smoke tends to escape  from the openings at the opposite end
to the volume  of coal charged from a given hopper, attention has
been given to means for better control; this has resulted in the
development of volumetric sleeves on the top of the larry car
hoppers.  These make it possible to adjust the bulk of coal in
each hopper to match the requirements of ovens of different
cubical capacity.  A volumetric sleeve  makes it possible to
charge the correct volume of coal and thus minimize the gas-
passage difficulties discussed above. 5

     Any arrangements that reduce the time required for transfer
of the coal charge from larry car hopper to oven interior reduce
the total escape  of smoke;  several mechanical devices have been
advanced toward this end,  including hopper vibrating mechanisms
in conjunction with smooth stainless steel liners, cylindrical
hoppers and bottom turn-table feeders, and a screw feed mech-

     Control of the bulk density of the coal mass is one of the
factors contributing to coal feed rate as well as to certain pro-
duction aspects.  In those operations where the coal is pulverized
and  blended, it has become increasingly common to  control the
mositure content and, by addition of small quantities of oil to the
coal, to modify the bulk density further.  Oil-sprayed coal moves
out of the larry car hoppers with much  greater facility, thus
reducing time required for charging. 5

     During the oven-charging process, a leveling operation en-
sues that involves operation of a leveling bar, a mechanism that
is an element  of the  coke pusher equipment.  The leveling bar is
operated through the leveling bar opening on the pusher side near
the top of each oven.  Operation of the steam jet aspirator in the
ascension pipe of the oven being serviced tends to prevent the
escape of much smoke from the oven interior, but it is not entirely
successful. To prevent the emission of smoke at this point, each
oven can be equipped with a smoke seal box surrounding the level-
ing bar opening.  In  conjunction with the effect of the steam jet
aspirator,  the seal box greatly reduces the escape of gas from the
opening, and on batteries having a gas-collecting main on the
pusher side, practically eliminates all  escape of smoky gases. 5

     During the 18- to 20-hour carbonization period,  leakage of
smoke from the interior of an oven may occur around the doors
unless measures are adopted to minimize it.  In older designs
the joint between the coke oven door and the jamb is  sealed by
luting, i. e., by hand-troweling a wet mixture of clay and coke
breeze into a channel between the  door and the jamb.  In recent

40                                IRON AND STEEL INDUSTRY
years a number of self-sealing door designs have appeared in
which metal-to-metal contact between a machined surface and
knife edge, together with mechanical arrangements for exerting
pressure, provides the seal and avoids the disadvantages of the
older luting method, which is so dependent for its effectiveness
on the attitude of the workmen having this responsibility.  It is,
however,  necessary that a superior maintenance program be
applied  to this equipment since wear and tear inevitably allows
the development of leakage. ^ An optimum maintenance program
for self-sealing doors,  as proposed by coke oven operators of
Allegheny County, Pennsylvania, includes changing the knife-edge
material to stainless steel,  and adoption of a numerical system
whereby a complete  history of each door is  kept and a system of
having operators tag doors to inform the maintenance force of
the reason a particular door was taken out of service. 5

    As  to luted doors,  it it advocated that tamping of the luting
after completion of charging be practiced to minimize smoke
emission from that source. 5

    At the end of the coking cycle the incandescent coke is trans-
ferred from the oven into a coke-quenching car by pushing equip-
ment.   The quantity of smoke arising from the mass  during the
period required for transport to the quenching station is dependent
on the degree of coking. Incompletely carbonized coke ("green"
coke)  gives rise to considerable quantities of smoke;  and con-
versely, thoroughly carbonized coke gives rise to very little smoke.5

Rate of Pollutant Generation

    The average blast furnace in the United States has a production
capacity of about 1000 tons of pig iron per day. 1  In producing this
quantity of pig iron,  110, 000, 000 to  150, 000, 000 cubic feet of
waste gases containing 100 tons of dust are evolved. 3, 45  Thus,
the blast furnace produces 0. 1 ton of dust for each ton of pig iron

    Bishop described a modern blast furnace producing 1400 tons
of pig iron daily. 46  Using a direct proportion of 1000/1400 to
convert his reported data to a typical (1000 ton) furnace, he re-
ported the daily  input to a typical blast furnace would be:  2000
tons ore, 900 tons coke,  400 tons limestone and dolomite,  and
3570 tons air; the daily output would  be:  1000 tons pig iron, 600
tons slag, 68 tons coarse flue dust, 32 tons fine flue dust,  and
5100 tons blast furnace gas.  Expressed another way, the untreated

                                                     GPO 823-573-4

 Air Pollutants Generated and Their Control                  41
gases from a blast furnace contain from 7 to 10 grains of dust per
scf of gas (7000 grains equals one pound).  It is common practice
to clean these gases before discharging them to the atmosphere.

    Carbon monoxide is formed in the blast furnace operation at
a rate of about 2200 to 3000 pounds per ton of pig iron produced.
Gases leaving the furnace contain about 25 percent carbon monoxide.
For a 1000-ton-per-day furnace, Carbon monoxide would be formed
at a rate of about 55 tons per hour. ^  Practically all of this carbon
monoxide is burned for heating purposes or in waste gas flares.

    Most of the  sulfur entering the blast furnace is contained
in the coke, although some is contained in the iron ore and lime-
stone.  One might expect this sulfur to burn and leave the furnace
as gaseous sulfur dioxide, but it does not, to any large degree.
Most of the sulfur goes into the slag,  but a little ends up in the
pig iron. 2

    Other common pollutants resulting from combustion are
probably emitted from blast furnaces, including a relatively small
amount of nitrogen oxides, but no data on them were found.
Gaseous  pollutants associated with combustion also arise  from
burning blast furnace gases for heating purposes or in waste gas
Operations Causing Pollutant Generation

    "Slips" are the principal factor in pollution arising from
modern blast furnaces having typical air pollution control equip-
ment. 34, 45 -phis problem is much less severe today than it was
a few years ago.  A slip is caused by arching of the furnace charge.
The arch finally breaks and the burden slips into the void.  There
is a rush of gas to the top of the furnace, which develops abnorm-
ally high pressures, much greater than can be  handled through
the gas-cleaning equipment.  When this occurs, bleeders or
safety valves open to  release the pressure; the result is a dense
black or red cloud of  dust discharged to the atmosphere.  Slips
not only create air pollution,  but also reduce efficiency of the
blast furnace and may cause damage to the interior of the furnace.

    Blast furnace, operators are constantly striving to reduce the
incidence of slips, thus increasing efficiency and production and
reducing air pollutant emissions.  Furthermore,  with increased
understanding of their cause, further steps are being taken to
utilize practices and procedures to eliminate slips.  The use of
sinter, with a reduction in the amount of fines fed to blast fur-
naces, has resulted in smoother operations.  In recent years the

42                                 IRON AND STEEL INDUSTRY
practice of regular "checking" has been introduced in some
British mills.  This involves easing the blast every 15 minutes
or so to cause a controlled minor slip.  Since these controlled
slips are normally not severe enough to open the safety valves,
the release of uncleaned gas is virtually eliminated.  Another
procedure that has been used is the differential pressure blowing
technique. 34  The blower volume  is automatically controlled
according to  the difference in pressure between the top and bottom
of the furnace.  The blast is automatically reduced if the furnace
"tightens up," thus preventing the bleeders from opening and
allowing excessive dust discharge.

Nature  of Pollutants
     Particulate matter from blast furnaces reportedly varies in
diameter from 0. 25 inch to 0. 1 micron, 47 but no detailed particle
size distribution was available.  From a  1400-ton-per-day furnace,
Bishop reported 68 percent coarse flue dust and 32 percent fine
flue dust, but again,  no detailed values were provided on size
distribution.  The coarse dust referred to was removed by greatly
reducing the velocity of the gas and at the same time suddenly
changing the direction of the gas stream.  This would indicate
that most of the particles were greater than 50 microns in size.
The dust contains around 30 percent iron, 15 percent carbon, 10
percent  silicon dioxide,  and small amounts of aluminum oxide,
manganese  oxide, calcium oxide,  and other materials.

     The principal gaseous pollutant arising in a blast furnace is
carbon monoxide.   However, at nearly all plants, blast furnace
gas is used  for heating, thus the carbon monoxide is burned to
carbon dioxide, which is of little concern as an air pollutant.
That carbon monoxide not burned for fuel is burned in waste gas
flares.  No  data were found on other gaseous  combustion products
that are  probably formed.

     Although a few iron ores contain relatively large quantities
of fluorides, no reports were  found to indicate that fluorides are
present in important amounts  in blast furnace exit stacks. One
set of tests  indicated emission of 0. 7 pound per day from two
blast furnaces  designed to produce 1450 tons of pig iron per day. 48
However, the furnaces were not using an  ore noted for its high
fluoride  content.

Control Methods

     Blast furnace gas is cleaned in three stages; the first two,  at
least,  are used almost universally throughout the industry.  The
majority of  furnaces have secondary cleaning facilities as well.

Air Pollutants Generated and Their Control                   43
 The three stages and the equipment used in each are:

    1.  Preliminary cleaning - settling chambers or dry-type

    2.  Primary cleaning - gas washers or wet scrubbers.

    3;  Secondary cleaning - electrostatic precipitators or high-
                            energy washers.

    Dust-laden blast furnace gas,  with a dust concentration of 7
to 10 grains per scf is first passed through the preliminary cleaner;
upon leaving this unit, it contains about 3  to 6 grains per scf. 4°,  46
The gas then passes through the wet scrubbers; effluent from this
type of cleaner contains on the order of 0. 05 to 0. 7 grain per scf
of gas. 4,  45, 46  if a secondary cleaner,  such as an electrostatic
pi*ecipitator or venturi scrubber, is employed, the  exit gas con-
tains only about 0. 004 to 0. 08 grain  per scf. 40> 45> 49

    Aside from the need for air pollution  control,  one of the main
reasons that the industry cleans blast furnace gas is to render it
sufficiently clean for heating coke  ovens,  boilers,  stoves,  soaking
pits, and  gas engines.  The blast furnace  gas  has a heating value
of about 100 Btu per cubic  foot, but must be cleaned before it can
be successfully used. As an example, it is known that checkerwork
in the stoves is most efficient for heat recovery if the openings
can be kept small.  However, the size opening that can be used is
dictated by gas cleanliness. 4^  For a gas  containing 0. 2 grain
per scf, the minimum opening that can be used without troublesome
plugging is 4. 5 inches square.  If the gas  contains only 0. 02 grain
per scf,   the minimum opening can be reduced to 2  inches square.

    Blast furnace flue dust is returned to the  iron-making process.
Since the  dust contains about 15 percent carbon, it will help
support combustion in sintering machines.  The dust also contains
about 30 percent iron, which can be  put back in the process after

Rate of Pollutant Generation

    The number of furnaces producing ferromanganese is small.
However, effluent from a ferromanganese blast furnace (a 75 to
80 percent manganese iron is produced) is reported to be a greater
air pollution porblem than that of an iron blast furnace. 4^  In fact,
this effluent, if uncontrolled, is said to be the most prolific pol-

lution producer of any of the metallurgical processes. 50 studies
by Bishop in 1951 established a fume loading in the waste gas of
4. 5 to 8. 5 grains per cubic foot,  with an average of 7. 5. ^5  The
emissions from two 350-ton ferromanganese furnaces averaged
9 grains per scf in a gas volume  of 135, 000 scfm. ^9  These two
furnaces produced about 125 tons of dust per day.  Another
investigator reported that electrostatic precipitators collected
100 tons of dust per day from ferromanganese furnaces producing
670 tons per day. 50  This is equivalent to nearly 150 tons of dust
per 1000 net tons of metal produced compared to 100 tons per
1000 for iron blast furnaces.
Nature of Pollutants

     Fume from ferromanganese blast furnaces contains 15 to 25
percent manganese and 8 to 15 percent alkali. 47  The fume is
extremely small, with 80 percent  ranging in size from 0.1 to 1.0
micron. 45, 51  properties of ferromanganese blast furnace fume
reported by Bishop are shown in Table 1.  The waste gas has a
slightly higher heating value (120 to 135  Btu/ft3) than iron blast
furnace gas, which is around 100 Btu per cubic foot. 46, 50

        	(Reference 51)	


         Total alkali (as Na20 and K20)

         Silicon dioxide (Si 02)

         Aluminum oxide (Al^ Og)

         Calcium oxide (Ca 0)

         Magnesium oxide (Mg 0)

         Total sulfur (as S04)

        15 to 25

       0. 3 to 0. 5

         8 to 15

         9 to 19

         3 to 11

         8 to 15

         4 to 6

         5 to 7

         1 to 2
         Apparent density, 12 lb/ft3

         Particle size (average), 0. 3 micron

Air Pollutants Generated and Their Control                  45
Control Methods

    Cleaning ferromanganese blast furnace gas is more difficult
than cleaning the usual blast furnace gas.   This is due primarily
to the high percentage of fine fume. 4°  in addition, the tempera-
ture of the waste gas is higher, 40 and moisture variations are
extremely wide, complicating control by electrostatic precipitators.
However,  in this country electrostatic precipitators have been
used with  some success to control ferromanganese fumes.  The
collected particulate matter weighs only 12 pounds per cubic foot,
and there  is at present no economical way to make use of collected

Rate of Pollutant Generation

    Concentration of dust and fume in the effluent gas from open
hearth furnaces after it is passed through the checkerwork has
been reported by several investigators (Table 2).  The most
striking feature of the data is the wide variation in results re-
ported.  Some investigators have reported emission rates as
related to the several stages involved in producing a heat of steel
(Table 3).  Again a wide range of values is reported for the same
stage of operations; especially high rates were  reported during
oxygen injection.   However, the average emission rate from a
hot-metal open-hearth furnace appears to be about 0. 4 grain per
scf for the conventional furnace and 0. 6 for  the oxygen-lanced
furnace.   These values were about the average  of emission rates
shown in Table 2,  and they can be substantiated somewhat by using
the 10. furnace shop cycles given by Purvance (see "Open Hearths"
under "How Iron and Steel are Made") for one furnace and applying
emission rates from Table 3 for each cycle.  Stack gas loadings,
process cycle times, and  stack gas volumes may be converted into
total daily emissions from the furnaces. Some results of such
calculations are shown in  Table 4.

    By calculation, data in Tables 2 and 3 may also be converted
into terms of dust and fume emission per ton of steel produced.
British investigators54 report emissions of  1 percent or 20 pounds
of dust and fume per ton of steel. According to rates given in
Table 4,  a general average of 4500 pounds of dust and fume per
day is emitted during production of 500 tons of steel.  This is
equivalent to 9  pounds of dust and fume per ton  of steel produced.

Furnace type
(hot or cold
02 lanced
©2 lanced


02 lanced
(net), tons
n. a.
n. a.
n. a.

n. a.




Gas volume,
14, 400
n. a.
n. a.
20, 000
n. a.
n. a.

n. a.
40, 000

33, 000
18, 000 to
60, 000
n. a.
33, 500 to

Dust loading, gr/scf
0. 1 to 2. 0
0. 01 to 0. 08
0. 04 to 0. 18
0. 02 to 0. 07
0.1 to 1.4
0. 07 to 0. 4

0.11 to 1.26
0. 1 to 1. 2

0. 5 to 2. 5 c
0. 1 to 2. 0

0. 11 to 0.34
0.10 toO. 31

0. 8 to 2. 5

n. a.
n. a.
0. 25 b

0. 5b
0. 6b


n. a.
n. a.

n. a.






  an. a.  indicates that data were not available.

  Estimated by the reviewers.
  cRange data are from an earlier reference than the average data; therefore,
   the average figure is probably more accurate.

Air Pollutants Generated and Their Control
Reference Number
Furnace size (net), tons
Effluent gas volume,
Stage of Heat
Melt down
Hot metal addition
Ore and lime boil
Working and refining
52 a
14, 400
33, 500 -
40, 000
Dust loading at 60° F and 29. 9 in. Hg, gr/ft3
only cold
0. 82 d
0. 87 d
only cold
    Tests conducted by Menardi & Co. and staff of Columbia Steel.

   bThe investigator was DeVries; it is assumed that grain loadings reported 45 were at
    a gas temperature of 550°F.

   c Three values were obtained during lime boil; 0.43 is the median.

    High-purity oxygen was injected during this stage by roof lances.
(net), tons

Cold metal
Hot metal
Hot metal
Hot metal
Hot metal

Emission rate,
8, 000 to 18,000
2, 000 to 6,000


    Gaseous emissions from open hearth furnaces include those
usually associated with combustion of fuels.  Open hearth fuel
may be producer, coke oven, or natural  gas; residual fuel oil;
crude coal tar; or pitch creosote.  Characteristics  of these fuels
and partial data on the composition of stack gases resulting from
their use are given in Table 5. ® These are British data.  Values
given by Smith58 for an American  plant indicate 8 to 9 percent
CO2, 8 to 9 percent O2,  2 to 5 ppm SO2,  100 to 200 ppm SOs,
7 to 8 percent H2O,  and the approximate  balance as N2-  Nitrogen
oxide emissions from four  50-ton furnaces is about 2. 6 tons per
day. 168 A mixture  of about 70 percent gas and 30 percent fuel
oil was being used for fuel.  Stack gas concentration varied from
500 to 800 ppm at stack conditions and averaged 700 ppm.

     Sulfur dioxide is one gas formed by burning most fuels  except
natural gas.  According to  the British data in Table 5, a 200-ton
open hearth furnace using  fuel at a rate  that is equivalent to 4
million Btu per ton of steel would emit 136 pounds of sulfur
dioxide per hour if the fuel oil contained  1. 6 percent sulfur.  In
the U. S., oil with a lower sulfur content is usually used and
therefore sulfur dioxide emissions would be less. Use of coal tar
and pitch creosote would produce slightly lesser amounts.   Use of
coke oven gas would produce more than twice as  much sulfur
 Table 5.  OPEN HEARTH FURNACE FUELS (Reference 6)a

Calorific value (Gases,
Btu/ft3; others, Btu/lb
Specific gravity
Sulfur content (Gases,
gr/ft3; other, %)
Flue gas, with 40% excess
H20, %
N2, %
02, %
S02, %
Specific gravity
Dew point, °F
Type of fuel
Coke oven
6. 5 to 7. 5
18, 890
16, 740
16, 780
   a British data; American fuels are somewhat different.

Air Pollutants Generated and Their Control                   49
dioxide as use of fuel oil. 6  Not all of this sulfur dioxide would be
emitted.  Some is absorbed in the steel and in the slag.  One
reportl69 states that,  on the basis of tests, most of the sulfur
emitted from open hearths equipped with waste-heat boilers is in
the form of sulfur trioxide,  with only a small fraction being
present as sulfur dioxide.  This was not believed to be unusual in
view of the oxidizing atmosphere and the presence of a finely
divided catalyst,  Fe2O3,  in the regenerative system.

    Fluorides may be  emitted from open hearth furnaces both as
gaseous and particulate matter.   In most instances, when the
source of the fluoride is fluorspar used as a flux during the final
stage of the heat, the amount emitted is small and does not cause
specific air pollution problems.   Emission data from one study
indicate that 39 pounds of fluorides per day were emitted from an
open hearth shop  consisting of 13 110-ton furnaces. 48 However,
the report is based on limited data.  British reports indicate
that about half of the fluorine added as fluorspar is emitted. 63, 64
The amount of fluorspar added to furnaces varies widely but
averages about 6. 5 pounds per ton of steel produced.   Iron ore
mined in southern Utah contains more fluorides  than others, and
when this ore is used in a furnace,  it can cause  higher fluoride
emissions. Although no emission data were found, it is known
that at least one western plant has been troubled by fluoride
emissions resulting from the use of fluorspar fluxes and high-
fluoride ore.  In the neighborhood of the plant, problems due to
fluorides were indicated. 7 The fluorides  are  reported to be
emitted as hydrofluoric acid (HF) or silicon tetrafluoride
(SiF4).63, ei
Effect of Oxygen Lancing on Fume Generation

    As a means of developing a concept of the effect of oxygen
roof lances on generation of particulate matter, calculations have
been made for a hypothetical furnace.  These are presented in the
following paragraphs.

    The average-size open hearth in the United States has a
capacity of about 175 net tons of steel per heat.  A heat takes
about  11 hours when oxygen is not injected into the furnace.  If
oxygen is  injected with roof lances at a rate of  about 600 cfm for
about  3 hours after  hot metal has been added and the bath has been
melted, the total length of the heat may be reduced to about 9 hours.
The average furnace discharges gaseous effluents at a rate of
about  35, 000 cfm with or without lances.

    Fume loading in the exit gases averaged over the heat are

50                               IRON AND STEEL INDUSTRY
about 0. 4 grain per scf without oxygen injection and around 0. 6
with oxygen injection as indicated above. 55,  57  Daily emission
of dust and fume from this average 175-ton furnace is about 2900
pounds for the standard furnace and about 4300 pounds for a fur-
nace with oxygen injection, if no air pollution controls are
operating.  These values vary a great deal depending on furnace
practice and many other factors as indicated by Tables 2 and 3.

    The amount of dust and fume emitted per ton of steel pro-
duced, based on the assumptions above, would be  7. 5 pounds per
ton of steel without oxygen injection and about 9. 3 with oxygen in-
jection.  Thus, on the basis of emissions per unit time,  oxygen
injection would increase emissions by 50 percent, but on the basis
of emissions per unit of production, oxygen injection  would in-
crease emissions about 25 percent.  These values must  be con-
sidered only rough general estimates in view of the variables in-
volved and the few data available.
Operations and Variables Affecting Rate of Fume Generation

    Although several factors are believed to affect the rate of
generation of dust and fume from an open hearth furnace, not
many of these have been studied thoroughly.  Cold-metal furnaces
generally produce less particulates than the-hot metal type. ^5
However, if the scrap iron contains oil,  oxides, dirt, wood, etc.,
or if a large amount of galvanized scrap iron is used in the  charge,
particulates from a cold-metal furnace may be very high. 52 Work
by the British has shown that use of pitch creosote  and coke oven
gas produces three times as much particulate matter as use of
coke oven gas alone. 65 in those tests, a V-notched probe was
used to measure deposition between the waste-heat boilers and
the stack.  When oxygen was added to combustion air,  particulate
matter was reduced by a factor of 2. 5.  Many tests on a 100-ton
cold-metal open hearth and a 300-ton hot-metal open hearth
furnace showed that the peek rate of fume formation occurs in
the cold-metal furnace at the end of charging and in the hot-metal
furnace just after the hot-metal addition.

    Bishop has noted that the peak rate of particulates leaving a
hot-metal furnace occurs  during the ore  and lime boil and during
oxygen lancing. 57 He attributed the added fuming to the increased
agitation of the molten bath and vaporization of the iron resulting
from higher localized metal temperatures (5000° to 6000°FJ
in the vicinity of the lance. Oxygen lancing may increase the rate
of fume formation,  mostly ferric oxide, as much as 3-fold over
that formed during normal operation. 54  Kohlmeyer,  investi-
gating the effect of blowing air oxygen on iron carbon melts,

Air Pollutants Generated and Their Control                   51
found that fuming did not start before the surface was at a
temperature of 2138°F, i. e.,  just above the melting point of the
melt, and was at a maximum at 2552°F.  When the temperature
of the melt exceeded 2732op, fuming ceased. 65  Holden and other
investigators have found that only alloys containing carbon pro-
duce dense brown fumes.  The British Iron and Steel Research
Association has also investigated the influence of metalloids and
found that molten iron or molten alloys without carbon produced
no fuming when held in a stream of oxygen, whereas with molten
iron-carbon alloys fumes were evolved. *>5  Other investigators
have substantiated that the amount of fume evolved increases
with the carbon content of the  melt and with the oxygen input
rate. ^6  Further tests suggest that the depth of submersion of
the oxygen lance tip may be an important factor affecting fuming
Nature of Particulate Pollutants from Open Hearths

     Most (up to 90%) of the particulates from open hearth
furnaces are usually iron oxide, predominantly Fe20s. 55  However
during the lime boil period,  the percent iron oxide may be less. ^
In cold-metal furnaces where large amounts of scrap are used in
the charge,  zinc oxides may be predominant during part of the
heat. The high percentage of zinc oxide would be due to use of a
large amount of galvanized scrap in the charge,  a rather unusual
circumstance.  StraussSO studying a 200-ton oxygen-lanced open
hearth furnace found that iron oxide averaged about 50 percent of
the total particulates over most cycles but reached above 60
percent during working and ore boiling and about 80 percent during
oxygen lancing.

     Composite dust samples taken over an entire heat show that
a great number of the particles, about 50 percent, are less than
5 microns in size; however, the size distribution varies con-
siderably during the heat.  One investigator found that whereas 46
percent of the particles in a composite sample (over one heal) were
less than 5 microns,  about 77 percent of the particles collected
during the lime boil were less than 5 microns. 5? For particle
technology work a mean particle size of 0. 5 micron and a particle
density of 5.2 grams per cubic  centimeter has been used. 53,  68

     A recent analysis5? of the  composition of an oxygen-lanced
open hearth furnace fume gave the following percentages: Fe203,
89. 1; FeO, 1.9;  Si02, 0.9; A1203, 0. 5; CaO, 0.9; MnO, 0.6;
alkalies,  1. 4; P2C>5, 0. 5; and S, 0. 4.  The particle size analysis
(composite over entire heat) was as follows:

52                                IRON AND STEEL INDUSTRY
         Diameter,                         Percent less
          micron                        than (by weight)
             2                                 20
             5                                 46
            10                                 68
            20                                 85
            40                                 93

    In number,  most particles are below 0. 1 micron in diam-
eter. 67  Vajda remarked "Experiments ... indicate that the out-
let dust loading must be down to 0. 05 grains per cubic foot to
give a clean (visually) stack.  Above this a bystander cannot tell
whether there is a cleaning unit on the stack or not." 69 However,
another investigator found that during decarb lancing 0. 05 grain
per cubic foot does cause a visible plume. '^  This suggests that
the particle size produced during oxygen lancing is smaller than
during other stages of the heat.
Air Pollution Control Methods

    The iron oxide fumes from open hearth furnaces are hard to
collect economically because of four factors:  the small particle
size (down to 0. 03^),  the large volume of gas, the high gas
temperature, and the low value of the  recovered material.

    Some 10 years ago Vajda°" conducted tests of certain devices
for control of open  hearth dust and found that  the devices tested
were not highly efficient. The efficiencies reported on selected
control  equipment are listed:

         Unit Tested                      Efficiency, %

Trion precipitator (commercial type)          80. 7
Impingo (Pebble  filter) 6 in. H2O pressure  drop 77. 8
Type N  Rotoclone (with gravel filter)          65. 7
Type N  Rotoclone                            49. 6
Multiclone                                   43. 4
Type D  Rotoclone                            22. 8

    The inlet dust loading on these tests was  generally in the
range of 0. 10 to 0. 40 grain per cubic foot at 60°F.  Today, iron
oxide dust and fume may be removed efficiently by use of three
types of collectors: electrostatic precipitators,  high-efficiency
wet scrubbers (venturi and Theissen Disintegrator), and fabric
filters.   A review of present cleaning devices for open hearth

Air Pollutants Generated and Their Control                    53
waste gases  has been published. 60

     The first electrostatic precipitator installed on an open
hearth furnace in this country was in California in 1950 and
1951. 39  The precipitator cleaned effluent from a 58-ton cold-
metal furnace.  The precipitator efficiencies were reported to
be 98. 5 percent, and the dust concentration leaving the precipi-
tators  averaged 0. 016 grain per scf. Other descriptions of the
operation of electrostatic precipitators in open hearths are avail-
able in the literature. 59,  61

     A venturi scrubber has been installed on a 220-ton oxygen-
lanced open hearth furnace.  Operated at a pressure drop of 14
inches of water across the venturi throat, it removed 98 per-
cent of the fume.    The effluent averaged 0. 04 grain per scf.
These data indicate better efficiencies at relatively low pressure
drops than do more recent data presented by Bishop. ^  He
describes a venturi scrubber system installed on a 200-ton oxygen-
lanced open hearth.   Efficiencies  range from 86 to 98 percent
with inlet dust loadings of 0. 35 to 0. 45  grain per scf and pressure
drops of 26 to 40 inches of water. During oxygen lancing with
grain loadings of 0. 82 to 0. 87 grain per scf the efficiency varied
from 89 to 99 percent with respective pressure drops of 26 to 40
inches of water.  Effluent dust loadings were consistently below
0.10 grain per scf, and with the scrubber was operated at high
pressure drops,  exit-gas loadings were below 0. 05 grain per scf.
The higher pressure  drops needed to get  better efficiency increase
operating power requirements  considerably.

     It was suggested by Pring^ in 1954, after some pilot plant
experiments, that fabric bag filters would be commercially
feasible on open hearth furnaces.  In his  experiments the gases
were cooled to 220°F and passed through "Orion" filter bags. The
efficiency was virtually 100 percent. Thfe first installation of a
large fabric bag filter plant on a 300-ton  tilting furnace using
oxygen refining has been reported in England. < 1  The capital cost
of the bag filter unit is lower than that of an electrostatic pre-
cipitator, but the space requirements are great, and the mainten-
ance costs, particularly with the replacement of bags, might be
higher.  An order has been placed for the first full-scale glass
fiber bag filter on an open hearth (oxygen-lanced) furnace in the
United States. ?2  At another eastern plant, a 10-compartment unit
will be equipped with sonic generators to remove the dust cake from
the filter fabric.

     The extent of existing and near future use of air pollution
control equipment is  summarized in Tables 6 and 7.   Only 44
(4. 9 percent) of the 906 open hearth furnaces in the United States

in March 1961 were using control equipment.   These furnaces
represented about 7. 8 percent of the annual open  hearth steel-
producing capacity.  There are air pollution control installations
presently under construction or to be installed shortly for an
additional 33 furnaces, which represent about 7. 2 percent of the
nation's annual open hearth steel-producing capacity and 3. 6
percent of the number of furnaces.  When these installations are
completed, about 15 percent of the  steel made in  open  hearths
will be produced in furnaces with air pollution controls.  As a
result of advances in steel-making technology, the basic oxygen
furnace is replacing the open hearth furnace.   This will influ-
ence future use of open hearth furnaces and installation of air
pollution control devices on them.
       (Reference 73)
United States
United States
United States
United States
United States
Plant location
Fontana, Calif.
Fairless Hills, Pa.
Homestead, Pa.
Braddock, Pa.
Geneva, Utah
Torrance, Calif.

Number of
with controls
(net), tons
Total annual
capacity (net),
million tons
1. 5
in roof
Type of
E. P.
E. P.
E. P.
V. S.
E. P.
E. P.
 a E. P. - electrostatic precipitation; V. S. - venturi scrubber.

 b Two furnaces have oxygen lances in furnace roof; four have oxygen-enriched combustion air; three do
  not use either.
     Control of gaseous and particulate fluoride emissions from
open hearth furnaces has been achieved.  As previously mentioned,
potential fluoride emissions are of concern at only a few western
plants where ores containing unusually large amounts of fluorides
are used.  One installation  in operation in Utah7 has been described.
Gases from 10 310-ton oxygen lanced open hearth furnaces are
collected in a large duct and conveyed to  a battery of eight electro-

Air Pollutants Generated and Their Control
static precipitators.  Average particulate loading is 0. 5 grain
per scf for the shop.   Pulverized hydrated lime (Ca(OH)2) is
mixed with the furnace gases to react with gaseous fluorides and
form particulate matter containing the fluoride.  The particles
are then removed from the gas stream by cyclone collectors and
electrostatic precipitators in series.

    Aside from preferential use of low-sulfur fuels,  which
results in lower sulfur dioxide emissions,  no reports of measures
for control of gaseous emissions (except fluorides) were found in
the literature.
         CONSTRUCTION OR TO BE INSTALLED March 1961 (Reference 73)

Jones & Laughlin
Jones & Laughlin

Plant location
Cleveland, O.
Pittsburgh, Pa.
Sparrows Ft. , Md.
Lackawanna, N. Y.
San Francisco, Cal.
Cleveland, O.
Sparrows Pt. , Md.

Number of
that will be

(net), tons

Total annual
capacity (net),
million tons
0. 79b

in roof

Type of
E. P.
E. P.
E. P.
E. P.
E. P.
F. F.
E. P.
F. F.
a E. P. - electrostatic precipitator; F. F. - fabric filter.

b Estimated by authors.
Research and Development

    The most promising future developments known of today are
effective slag-wool filters, beds of granulated refractory material,
and better high-temperature filter fabrics.  All have been shown
to be  capable of high efficiencies.

    The American Iron and Steel Institute is sponsoring research
at Harvard University on new means of controlling potential
emissions from steel-producing operations.   A  1-inch layer of

56                                IRON AND STEEL INDUSTRY
slag wool was used as a filter medium in studies that showed 90
percent of the iron oxide fume is removed from a gas stream at
1022°F at a filter face velocity of 1. 67 fps. 53> 67  However,  in
pilot plant investigations the over-all efficiency was only 60 per-
cent.  In this test the slag wool,  in the form of a slurry,  was fed
to a moving-chain belt conveyor.  The hot gases dried the slurry
and formed  the filter bed.  Difficulty arose through poor  mechan-
ical sealing of the bed. A modified unit using a circular  conveyor
had even lower efficiency, averaging about 44 percent. Decreased
efficiency was due to edge leakage during rotation.  Research
now is centered on developing an economical  cleaning method to
remove the  trapped fume  particles.  Most promising is the use
of sonic shock waves to remove the particles from the in-place
filter bed.

     Straus and Thring have investigated the fume-collecting effi-
ciencies of beds of different depths of high-temperature insulating
brick, crushed and graded between 7 mesh and 5/16 inch. A
9-inch filter bed attained  efficiencies of up to 95 percent with a
pressure  drop of less than 3 inches water  gage and gas velocity
of 1. 7 fps.  The average filter efficiency was 80 percent with an
average gas velocity  of 1. 2 fps.  The pressure drop increased
from 2 to 4  inches water gage during four  melting cycles  over a
48-hour operation period,  when the bed was used regeneratively. 60

     Research investigations seeking means to increase the size
of open hearth fume and thus make it easier to collect have been
reviewed. ^0 Vacek and Schertler have worked on magnetic
agglomeration of fume from a basic oxygen furnace.  Vajda,
using sonic  agglomeration,  improved the performance of  a type
W Rotoclone.  Silverman has reported work with a screw-type

    A study of formation and suppression of smoke in top-blowing
processes was initiated at Battelle Memorial  Institute as  part of
the American Iron and Steel Institute's program late in 1960.  95
These studies are satisfactorily explaining some of the sharp
differences in experimental results of various other investigations
into iron-smoke-forming processes.  The  studies are still in
progress, but they already give confidence to the possibility of
effectively reducing the fume from top-blowing processes.  Be-
cause of the more general applicability of the results,  the
primary emphasis will continue to be  placed on the determination
of  the mechanism of formation of the fumes.  This knowledge  will
be useful to  the entire industry in devising methods of suppressing
the fumes produced.  Along with the direct study of mechanisms,

Air Pollutants Generated and Their Control                  57
however, information will be developed that may be of direct
immediate use to the steel industry.

    The characteristics, quantities, and means for control of
pollutants from electric-arc steel furnaces were summarized by
Brief,  Rose, and Stephan in their paper "Properties and Control
of Electric-Arc  Steel Furnace Fumes. " *•*•  Much of the following
is taken from that paper.  For convenience of the reader,
original references given by those reviewers are  given herein.
Rate of Pollutant Generation

    Data on particulates in effluents from a number of electric-
arc furnaces are presented in Table 8.  A mean quantity of 10. 6
pounds of fume is generated per ton of metal in electric-arc
steel-melting furnaces, the range being from 4. 5 to 29. 4 pounds
per ton.  Values as high as 37. 8 pounds per ton have been
reported. ^ An assumed "typical"  furnace of  50-ton capacity
producing five heats per day would generate a mean amount of
fume of 2650 pounds per day.  Tests have also been reported 82
that show that  fume concentrations frequently reach 3 grains per
cubic foot,  with a maximum of 6. The wide variations are
attributable to the following factors: type of furnace  process,
furnace size, formulation of charge, quality of scrap,  cleanliness
of scrap, sequence  of charge additions, melt-down rate, metal-
refining procedure, and pouring temperature.

    Emissions of nitrogen oxides from electric-arc furnaces
range from 0. 7 to 4. 1 pounds per hour per furnace.  Tests made
on large (75-ton) and small (2-ton) capacity furnaces indicate
that the size of furnace is not a measure of nitrogen-oxide-pro-
ducing capability.  It has been concluded that the amount formed
is a function of the degree of arcing during the heating. 83  NO
data were found on the amount of ozone, sulfur dioxide, and other
pollutant gases discharged.
Variables Affecting Pollutant Generation

    Any effects of furnace process or furnace size on fume
generation are not obvious in the data presented.  No significant
difference in fume formation in acid and basic furnace processes
is apparent.  It has been stated 52' 8^ that the quantity of fume
per ton of metal processed increases with furnace size, but this

58                                IRON AND STEEL INDUSTRY
 relationship also is not evident here.

     The quality of scrap charged,  however, is of extreme im-
 portance, since the inclusion of large quantities of lower-boiling
 nonferrous  metallic impurities in the melt inevitably leads to high
 concentrations of  oxides of these metals in the fume. 74,  77 The
 cleanliness of the  scrap likewise is an important factor because
 most impurities (paint,  oil, grease, wood, fabrics, etc.) are
 volatilized at the  high temperatures within the furnaces.  75

     Fume generation may be affected further by the sequence of
 charge additions to the furnace.  Metal oxide fumes from the
 melt are normally decreased after slag additions,  because of the
 slag blanket formed. 85, 86  Impurities may then be included
 within the slag rather than  vaporized to the furnace atmosphere.
 Partially offsetting this  reduction is the volatilization of the newly
 added slag components.

     The method employed for metal refining also has a pronounced
 effect on fume  generation.  An oxygen lance leads to higher fume
 release 12   because of the breaking of the slag film by effluent
 bubbles and because of the  extremely  high temperatures 87
 reached during this operation. These temperatures lead to in-
 creased volatilization of the melt constituents.  A mill-scale boil
 or a spontaneous non-initiated boil has a similar but less pro-
 nounced effect.  Another recognized factor affecting quantity of
 fume is the pouring temperature.   High pouring temperatures
 required for small or thin-walled castings lead to greater fume
 formation because of the rapid increase of the vapor pressure of
 the molten components with temperature.

     Although quantative  effects of all of the above factors are not
 apparent from  data presented in Table  8,  the significance of these
 factors is clearly  shown in  either the previous literature  or the
 original data analyzed for this paper.   Coulter^, 75 performed
 several tests (Table 8,  Case A) under identical operating conditions
 except for differences in cleanliness and quality of the scrap
 charged.  The amount of fume per ton of metal melted increased
 100 percent when dirty,  subquality  scrap was used.  Substantiation
 of these effects is  reported  by Kane and Sloan. 77  Their tests
 (not  given on Table 8) show  an increase of over 40 percent in the
amount of fume released per ton of metal processed when poorer
 scrap was charged to the furnace.   (The actual fume increase was
over 40 percent since fumes evolved during a full quarter of the
melting cycle were lost because of a defective collecting thimble.)

   The rate of  fume release appears to reach a peak during the boil
 and  refining periods.  When an oxygen lance is employed,  dense fumes

Air Pollutants Generated and Their Control





per ton
q oa
9. O
18. 6b

Furnace process

Basic, single slag

Basic, single slag

Acid, oxygen blow
Acid, oxygen blow

Basic, oxygen blow
Basic, oxygen blow


Acid, single slag
Acid, single slag
74, 75




79, 80
  a Average for one 50-ton and two 75-ton furnaces processing normal scrap.

  b Average for one 50-ton and two 75-ton furnaces processing dirty, sub-
    quality scrap.

  c Refer to same furnaces as Case A.

  "* Two 2-ton furnaces operating in parallel.

60                               IRON AND STEEL INDUSTRY
of hematite and magnetite are produced.12  As previously noted,
this peak fume release is less pronounced when a mill-scale or a
spontaneous boil is employed.  It can be said in general that
during refining,  fume rates are high, reaching a final peak as
pour temperature is approached.

    In addition to the above factors, the effects of several other
variables on total fume generation may be hypothesized,  although
not substantiated by experiment.  The formulation of the charge
may be a factor, the higher the proportion of subquality scrap
contained in the  charge, the greater the amount of impurities in
the furnace.  The impurities are,  in part, subject to volatilization;
hence, fume formation increases.  Also,  since the rate of heat
flux to the bath increases in proportion to the  rate of melt-down,
higher temperatures exist in the molten pool when the higher melt-
down rates are employed.  The resulting  increased temperature
should cause greater over-all fume generation.
 Nature of Pollutants

     The effects on fume generation rate and chemical composition
 of the fume appear to be a function of the melting-cycle phase and
 the charge formulation and quality.  EricksonSS and Assel84 have
 reported that fumes may analyze 40 to 50 percent iron oxide during
 melt-down.  Erickson further states that when the scrap is com-
 pletely melted the iron oxide  content of the fume drops to well
 below 20 percent, with the remainder being composed of oxides of
 calcium, silicon,  phosphorous, manganese,  and sulfur.  Assel
 also reports that after the slag is made up the iron oxide content
 may fall as low as 5 percent,  with the calcium oxide content of
 the fume increasing to the 45  to 50 percent range.

     In addition to the variations in fume composition resulting
 from melting-cycle phase, the effects of charge formulation and
 quality are significant.   The inclusion in the charge of nonferrous
 metallic impurities results in the oxides x>f these  metals being
 present in the fumes.  Coulter1^ has reported the following average
 composition of fume produced during an entire melting cycle.  It
 is likely that the charge contained an unusually high percentage of
 galvanized steel scrap.

Air Pollutants Generated and Their Control                  61
            Component                   Weight percent

         Zinc oxide (ZnO)                      37
         Iron  (Fe)                             25
         Calcium oxide (CaO)                   6
         Manganese oxide (MnO)                4
         Aluminum oxide (A^C^)                3
         Sulfur trioxide (SO3)                   3
         Silicon dioxide (SiC-2)                   2
         Magnesium oxide (MgO)                2
         Cupric oxide (CuO)                     0. 2
         Phosphorus pentoxide (Pz°5)           °- 2
     As noted, the composition includes a high percentage of
 nonferrous metallic oxides in addition to those constituents nor-
 mally expected in the fume.  Further substantiation that metallic
 impurities in the charge produce nonferrous oxides in the fume
 is found in the work by O'Mara, ^  who reports approximately 25
 percent zinc oxide present in one such fume.

     Certain physical  properties are characteristic of the par-
 ticulates from electric-arc steel-melting processes.  Specifically,
 these particles have a strong tendency to adhere to both natural
 and syntheticJabric surfaces; 85 they possess a high angle of
 repose; 85 they are difficult to wet; 88,  89 and they posses a
 high electrical resistivity. ^»  85  These properties limit the
 applicability of control equipment, and specific consideration for
 offsetting these inherent properties of the fume must be provided
 for in satisfactory control installations.

     Size distribution  measurements by the Los Angeles County.
 Air Pollution Control District ?6 and by Erickson 8^ (Table 9,
 Cases A and B) show  that electric-arc steel furnace fume parti-
 cles are predominantly (approximately 70% by weight) in the
 range below 5 microns.  Electron photomicrographs ^2 have shown
 that 95 percent by number of the particles in such fumes are less
 than 0. 5 micron in diameter.   The presence of appreciable per-
 centages by weight ol larger particles,  as determined by sedimenta-
 tion procedures, ">, 85 is undoubtedly due to the high agglomerating
 tendency of this type  of fume.

     Nitrogen oxides are emitted in small amounts.  Ozone formed
 by the electric -arc is thought to be emitted, but no specific data
 were found.  Sulfur gases are  also probably emitted in  fairly small

            Table 9.  ELECTRIC-ARC STEEL FURNACE
                    IN WEIGHT PERCENT (Reference 11)
Size range,
0 to 3
0 to 5
3 to 11
5 to 10
11 to 25
10 to 20












Control Methods

    The type of hooding provided in each melting furnace installa-
tion to collect fumes from the furnace and  convey them into a duct
determines the temperature and volume  relationship of the effluent
gas to be processed.  Three types of control hoods are in general
use:  (1) canopy-type hoods, (2) enclosing or roof-ring hoods, and
(3) direct furnace taps.  Canopy hoods located above the crane-way
offer  the least interference with operating  procedure; but to capture
furnace effluent efficiently, large volumes  of indraft air are re-
quired, resulting in lowered gas temperatures.  Roof-ring hoods
closely fit around the charging door,  pouring spout, and electrodes
and have a disconnecting flange in the exhaust duct  to allow for
furnace movement. A modification of this  type of hood is the
direct furnace tap in which the furnace itself serves as a refractory-
lined  hood.  These last two types of hood require smaller volumes
of indraft air for efficient collection, and higher effluent gas
temperatures result.

    The characteristically small particle size of electric-arc
steel  furnace fume precludes the use of dry centrifugal collectors,
settling chambers,  etc., when high efficiencies are required.

Air Pollutants Generated and Their Control                   63
Furthermore, those types of control equipment capable of giving
high-efficiency performance must operate within the following
specific limitations: wet scrubbers require high power inputs,
baghouse collectors need special provisions for increasing the
effectiveness of the shaking mechanisms and for controlling the
gas temperature and humidity, and electrical precipitators must
have adequately conditioned gas.

    High-efficiency scrubbing systems can be used for the con-
trol of electric-arc steel furnace fumes.  However, high power
requirements tend to negate their use.  In one installation the
fume collection system consists of a spark box, a washer, a
disintegrator (a high-efficiency water-fume contactor), and a
moisture eliminator. ^0  A gas volume of 50, 000 cfm at 60°F
is handled with an inlet fume loading of 1. 12 grains per cubic foot.
The material, which is collected at 98 percent efficiency,
analyzes just over 50 percent
     Use of an electrostatic precipitator to control fumes on one
 50-ton and two 75-ton furnaces has been reported. ^^> ^  The
 fumes are removed through direct furnace taps to a two-stage
 evaporative cooler-conditioner.  The precipitator handles 105,000
 cfm at 12 7° F with an inlet fume concentration varying between
 0. 68 and 1. 35 grains per  cubic foot.  Collection efficiencies of
 over 97 percent are  realized, and in the particulate concentration
 range concerned,  should be relatively independent of inlet loadings.

     A second electrostatic-precipitator installation although
 designed to employ gas conditioning actually does not because of
 a reduction in plant furnace capacity. ?8 A ring-type hood directs
 the furnace effluent through a radiation cooler after which temper-
 ing air is added to reduce the gas stream temperature to 80°F.
 Approximately 33, 500  cfm then enters the precipitator at this
 temperature with a fume loading of 0. 115 'grain per cubic foot.
 In this case,  a collection  efficiency of 92 percent is obtained,
 which as in the case above should also be relatively independent
 of inlet loading.

     Fabric filters (baghouses) are used to control fumes from
 electric-arc  steel furnaces.   One such installation has been
 described by Dok. **8  it has Orion fabric bags and handles 50, 000
 cfm of 124°F gas from one 1-ton, one 3-ton,  and two 6-ton electric-
 arc steel furnaces.  The total of roughly 40, 000 cfm from the four
 furnaces is normally tempered with 10, 000 cfm of  cooling air to
 reduce its temperature to that required for baghouse operation.
 Efficiencies of greater than 98 percent are readily obtained.

64                               IRON AND STEEL INDUSTRY
    Another baghouse handles a volume of 16,000 cfm at l30° to
18QOF for a 6-ton acid furnace. '79' &°  The furnace employs
single-slag operation with an oxygen lance during the boil.   No
specifically designed gas  cooler is in use, but a long duct run
serves as an effective radiation cooler.   Actual efficiencies
have not been measured but are assumed  to be 98 percent or better.
Cotton bags rotted after only short usage  and have been replaced
with Orion bags.  The deterioration of cotton is probably caused by
the presence of ozone and oxides of nitrogen,  in addition to the
acidic nature of the gas. 91

    A third baghouse operation processes the effluents from a 3-
ton furnace in which an acid, single-slag  melting cycle is used in
combination with a mill-scale boil. 81» 92  >phe gas passes through
an enclosed roof hood and then is cooled by radiation and air
tempering to 142°F before it enters the baghouse. The bags are
of Orion fabric. About 12, 000 cfm are cleaned at an average
inlet loading of 0. 22 grain per cubic foot.   Collection efficiencies
of over 99 percent are reached.

    Although data collected are  incomplete, indications are that
approximately 40 electric-arc furnaces at iron and steel mills
have been equipped with air pollution control devices. ^  The
largest number are using baghouses.   Electrostatic precipitators,
venturi scrubbers, and other high-energy dynamic scrubbers are
also used.  There are many electric-arc  steel furnaces that are
not located at iron and steel mills.  There are about 301 furnaces
at iron and steel works.   The 40 with air  pollution controls com-
prise about 13 percent of  the total.

    Bessemer converters are not extensively used.  Only about
2 percent of the Nation's total steel is produced in these units.
Rate of Pollutant Generation

    An average Bessemer converter is rated at 20 to 35 ton per
heat with two or three cycles per hour. 31 The rate of emission
from a 25-ton converter with a 10- to 15-minute blow has been
estimated to be 10 grains  per cubic foot, or between 15 and 20
pounds per ton of steel produced. 40> 45  A 25-ton converter
operating two  cycles per hour and emitting 20 pounds per ton of
steel produced would emit a total of 12 tons per day.  Lower  grain
loadings of 0. 15 to 0. 4 grain per cubic foot have been reported in
the case of an oxygen-blown 20-ton basic converter, with sampling

Air Pollutants Generated and Their Control                  65
                                                       Q ^
being carried out at some distance from the vessel throat.
Other test work indicates dust loadings in the order of 0. 8 grain
per cubic foot. 31

    Data from  German experiments have been  reported for a
30-ton converter. °»^  Dust measurements indicated an average
emission of solids of  132 pounds per minute or 1320 pounds for a
10-minute melt (44 Ib/ton of steel).  These emission  figures  in-
clude all the very coarse materials, a major portion  of which
would would settle out rapidly on the premises  immediately
surrounding the installation. 5

    During  the operating cycle the volume of gas emitted varies
from near zero during the non-blow periods to  about 2 million
cubic feet at 2700° to 3500°F during a 10- to 15-minute blow. 40> 45
The rate of  dust emission would also vary considerably throughout
the cycle.  It has been stated that fume emissions increase very
considerably when an oxygen enriched blast is used,4^, ^3 but the
use of an oxygen-steam blow is said to reduce fume production to
about the same level as that experienced with air blowing.4^
Nature of Pollutants

    The Bessemer flame,  which may reach 30 to 40 feet in length,
contains carbon monoxide that is rapidly burning to carbon dioxide,
plus two kinds of particulate matter.  Pellets of molten metal and
slag are mechanically ejected by the violence of the air blast
bubbling through the molten bath.  These are known as spittings,
composed, as they are, of relatively coarse particles that tend to
settle out on the premises  close to the source.   The other type
is the visible orange-colored fume of iron oxide, resulting from
volatilization in the converter of some of the iron and its sub-
sequent oxidation in the open air.  These small particles are
suspended in the hot gases, which because of their buoyancy rise
to great heights, transporting the visible fume with them. *>  It has
been reported5? 94 that about three-fourths of the participates,
by weight, are larger than 100 microns.

    The pellets, which are spit  out, rise to a height of  50 or 60
feet.  When they impinge upon a solid object, such as a wall,
girder,  or roof, they flatten and stick much like a metal spray.
In time,  heavy masses of iron and steel are formed and unless
removed, they drop off and endanger those who are working below.

    Measurement  of air velocities and qf temperatures in the
plume at a distance of 55 feet from the c'onverter mouth indicated
that the total flow of flue gas (including the air that had become

66                               IRON AND STEEL INDUSTRY
turbulently entrained in the 55-ft distance) amounted to 250,000
to 300, 000 cfm at temperatures between 450° and 550°F.  This is
the gas volume that must be subjected to gas cleaning. ^
Control Methods

    None of the Bessemer converters presently in use in the
United States is known to be controlled.  The primary difficulty
in adapting control devices to converters is confinement of the
effluent.  Normal operating procedure,  plus the spitting action
of the process, precludes the installation of a control device near
the mouth of the converter.
Research and Development

     The American Iron and Steel Institute (AISI), during the last
several years,  has been sponsoring a research program at
Battelle Memorial Institute aimed at preventing the formation of
iron smoke at its source,  thus obviating the need of collecting it.

     The initial specific objective of the AISI research program
was to discover practical  methods of reducing,  to unobjectionable
levels, the emission of smoke and dust from steel conversion by
the Bessemer process. ^^ It was believed from the outset that if
the emissions could be eliminated from this process the knowledge
could be extended to suppression of smoke emissions from all
steel-making processes.  In this research, operating converters
were studied.   Thereafter, a 300-pound experimental converter
was operated to simulate the commercial Bessemer converter.
With this converter,  hydrogen,  methane, or ammonia added to the
blast gas suppressed smoke.  From a practical standpoint, it
appears that the possibility of adding methane to suppress smoke
in Bessemer converters has merit and seems worth trying in a
field test.

     More recently, the AISI research program was modified
to direct the objective at the determination of the mechanism by
which iron smoke is formed.  Once the mechanism is  known, it
should be possible to devise methods of suppressing smoke or at
least controlling the physical characteristic of the smoke so it
can be removed readily.

     The first step toward  understanding the fundamentals of iron
smoke formation was to establish the separate effects of tempera-
ture, metal composition, blast gas composition,  and other external
variables on the formation of smoke.  Simultaneously, details of

Air Pollutants Generated and Their Control                  67
possible smoke-forming reactions were analyzed thermochemically.
Combining these approaches has led to a much improved under-
standing of the mechanism of smoke formation and possible prac-
tical methods of reducing the amount of particulate matter from
both top-blown and bottom-blown steel-making processes.


Rate of Pollutant Formation

    The average basic oxygen furnace in the United States in
April 1961 had a capacity of about  100 tons per heat.  Only a year
or so earlier the average would have been about 60 tons per heat,
but at the rate that large furnaces  are being installed, ^ 100 tons
per heat may soon be outdated. If this furnace completed  one heat
cycle per hour, it would generate on the order of 1 to 2 tons of
dust and fume per hour  (25 to 50 tons per day). ^, ^9,96 Thug
represents 1 or 2 percent of the steel produced.   At the Donawitz
plant in Austria, a basic oxygen furnace rated at 30 tons per heat
generates 35 to 60 pounds of dust per minute during the blow. ^0> 70,
This is equivalent to 13  to 23 tons  per day, if blowing oxygen takes
one-half of the heat cycle,  or equivalent to 2 or 3 percent  of the
steel produced.

    Dust loadings in the waste gas are reported to be about 7 to 8
grains per scf at a plant in Canada. 29> 49  Control equipment
designers in this country have used a value of 5 to 5. 2 grains
per scf. ^

    The major operation causing dust and fume generation is
oxygen blowing.  Other  steps in the heat cycle are relatively
minor contributors.  The waste volume produced from oxygen
blowing is not necessarily related to the nominal tonnage rating
of a vessel.  Instead it is proportional to the blowing rate.  By
plotting design oxygen blowing rates versus exhaust gas volumes
for systems both presently in service and proposed,  Gaw9"  ar-
rived at a design volume rating of  25 scfm of waste gases  per
cfm of oxygen blown.  He found that the ratio for venturi cleaning
systems often was slightly less at 20:1.  These ratios take into
account the  excess dilution air drawn into the system.  Dilution
air is desirable to reduce the explosion hazard and to allow
proper control of the temperature of combustion  products.

Nature of Pollutants *

    The composition of  gases emitted from a basic oxygen con-
verter,  after ignition above the furnace, are reported as follows,9^
in percent:

 68                                IRON AND STEEL INDUSTRY
         Carbon dioxide (CO2)             0. 7 to 13. 5
         Oxygen (O2)                  •   11. 1 to 20.0
         Nitrogen (N2)                    74. 5 to 78.9
         Carbon monoxide (CO)            0. 0 to  0. 3
         Illuminants                       0. 0 to  0.2
         Hydrogen (H2)                    0. 0 to  0.4
 Chemical analysis of dust and fume arising from a basic oxygen
 furnace is as follows, ^6 in percent:

         Iron oxide (Fe2O3)               90.0
         Manganous oxide (Mr^C^)         4. 4
         Ferrous oxide (FeO)              1. 5
         Silicon dioxide  (SiO2)             1. 3
         Calcium peroxide (CaO2)          0.4
         Phosphorous pentoxide (P2Os)     0. 3
         Aluminum oxide (Al2Os)          0. 2
     During large-scale testing of various gas-cleaning systems on
the Donawitz plant in Austria, several dust samples revealed that
the Fe2O3 content increased with decreasing particle size.^  At
Donawitz, approximately 40 percent of the dust particles, by
weight, were less than 5  microns with the 1- to 2-micron size
predominating. 97  An investigator in the United States reports
particle size to be somewhat smaller, as follows:^

                    <0. 5 micron - 20 percent
              0. 5 to  1.0 micron - 65 percent
              1.0 to 15.0 micron  15 percent
Control Methods

    All 12 basic oxygen furnaces operating in this country as of
January 1, 1960,  were equipped with high-efficiency control
devices.   The furnaces are all relatively new and were equipped
with air pollution control equipment when built.  This practice is
also common in foreign countries.  The gas-cleaning systems are
divided between high-pressure-drop venturi scrubbers and
electrostatic precipitators.  The wet system has the disadvantages
of high operating horsepower requirements and need for water
clarification.  Electrostatic precipitators have a potential explosion
hazard and a dust disposal problem.  Still,  these are the only
systems that have been found satisfactory.  This was determined at
the Donawitz plant in Austria through 509 experiments on eight

Air Pollutants Generated and Their Control                  69
different dust-collecting systems.97  Gaw has reported on the
advantages, costs,  and maintenance of electrostatic-precipitator
and venturi cleaning systems. 96

    Dust and fume control facilities are very expensive to install
and operate.  Depending on a number of factors pertaining to
specific establishments, some operators,  particularly those in
Europe,  believe it may be possible to defray a part of the costs
through use of waste-heat boilers for cooling of gases in the
dust-collecting system and by returning the collected dust to the
process, after suitable preparation.  73, 96,  97,  170

    Final effluent from these control devices will contain around
0. 05 grain per scf.   A venturi system is reported to discharge to
the atmosphere about 0.03 to 0.09 grain per scf,  with a maximum
of 0.12,4*  varying with furnace operating conditions and pressure
drop across the venturi throat.


    There are several miscellaneous sources of  air pollution in
the steel-making industry,  such as heating and reheating furnaces,
scarfing machines,  power plant boilers, slag dumps and crushers,
and rail locomotives.  Although their individual significance as
air pollution sources may be relatively minor in comparison with
some of those discussed previously,  when taken collectively they
may prove important if suitable attention is not given to air
pollution control.

Heating and Reheating Furnaces

    With the almost universal use of instruments and automatic
combustion controls,  modern furnaces seldom emit black smoke.
However, in the earlier days of the industry when the technology
of furnace design was less well understood, there were frequent
instances of smoking chimneys due to lack of automatic fuel-air-
ratio controls and to insufficient combustion space, with the result
that the fuel gases were incompletely consumed before coming in
contact with the relatively cold steel, and  black smoke resulted.
Some of the older heating and reheating furnaces  of poor design
are still in operation but are rapidly becoming obsolete and being
replaced with modern units. ^  The principal pollutants emitted
from properly designed and operated furnaces are the normal
gaseous products of combustion and perhaps particulate matter,
the amount and characteristics depending on the type of fuel used.
There also are times when too much fuel is fed into a furnace
containing cold  metal,  resulting in emission of smoke.

70                                IRON AND STEEL INDUSTRY

    Information is very meager on pollutants generated in scarf-
ing operations.  Spot tests on a 45-inch slabbing mill and a 40-inch
blooming mill indicated dust loadings of around 0. 4 grain per
scf. 3$  One investigator found a stack loading of about 0. 22 grain
per cubic foot in an effluent volume of 36, 840 cfm  from a 32-inch
slabbing mill and from 0. 42 to 4.4 grains per cubic foot in an
effluent volume of 87,000 cfm from a 53-inch slabbing mill.43
An effluent of 50, 000 cfm containing 0.  4 grain of dust per scf,
if uncontrolled, would discharge  over 2 tons of dust per day.
This is probably conservative since the fume loading from scarf-
ing machines has been accepted in some cases as 0. 8 grain per
scf. 70  Gas volume ranges from 20,000 cfm for the lightest two-
edge table to 150,000 cfm for the heaviest four-side  machine.

    Fine material produced in scarfing operations amounts to
approximately  36 grains per pound of metal removed, and in
typically heavy duty scarfing operations, this may amount to
50 to  75 pounds of solids per hour. ^

    The use of air pollution control equipment on this process
would be desirable.   The extent of use  in the industry was not
determined.  Many plants  use a baffled  settling chamber that
collects large particulates.  One such installation  pollects 7000
to 9500 pounds of iron oxide and other dust per week. 98  The
cleaners collect approximately 2 pounds of dust per ton of steel
Power Plant Boilers

     Emissions from boilers might include smoke and flyash,
plus varying amounts of gaseous pollutants (sulfur dioxide, nitro-
gen dioxide,  etc.).  In general, particulates may be controlled by
standard techniques (cyclones, precipitators, etc.).   Control of
gaseous pollutants from such operations is not practiced at steel
mills nor at any fuel-burning installations, except for use of low-
sulfur fuels at some plants,  as an air pollution control measure.
Slag Dumps and Crushers

     Dust generated at slag dumps and in crushers,  as well as in
other materials-handling operations, are commonly suppressed
by ordinary methods.  However, if the dust from these operations
is not controlled,  it may cause severe local nuisance problems.

Air Pollutants Generated and Their Control                  71
Rail Locomotives

    Nearly all mills have diesel-powered railroad engines to
handle material transfers and finished-product movement.  A
few coal-fired engines may still be in use.  Rail transportation
is used to move raw materials such as coal, iron ore, and lime-
stone  from storage points, where processing begins; to move
intermediate products such as coke and pig iron to subsequent
processing points; and to move finished products from the manu-
facturing area to warehouse facilities.  The primary emissions
from coal-burning locomotives are smoke and flyash.  The usual
gaseous products of combustion are also emitted.   Diesel
locomotives emit some small-size particulate matter and gaseous
pollutants, but they  are insignificant in comparison to other steel
mill emissions.

                           Part  IV


            IRON  AND  STEEL WORKS

               ON  THE  COMMUNITY

    Pollutants in a community atmosphere may arise from many
sources including industrial  plants,  automobiles, burning dumps,
home heating plants,  commercial operations, and backyard trash
burners.  Larger particles (larger than say  5 microns in size)
settle from the  atmosphere and cause soiling of automobiles,
porches, window sills, laundry, vegetation,  furniture, rugs,
goods on shelves in stores, etc. These larger particles tend to
settle out of the atmosphere  close to their point of emission.
Smaller particles (smaller than say 5 microns in size) attach
themselves to all  sorts of objects  and cause soiling of house paint,
curtains and drapes,  building stone, windows,  merchants'  stocks,
etc.  They also decrease beneficial sunlight and cause a  gloomi-
ness in the community atmosphere.  They tend to travel  great
distances from their point  of emission. Gaseous pollutants may
cause offensive odors, deterioration of materials, and damage to
vegetation.  Beyond these effects, air pollution, in general, in
certain areas (especially in major cities) is a hazard to man's health.

    As background information for use in judging significance of
pollution levels reported later herein, some typical values will be
useful.  Pollution levels several times higher than the "typical"
values occur from time to  time in some cities.  Lower pollution
levels also occur.  Dustfall in suburban areas of large cities are
often around 10 or 15 tons  per square  mile per month. Central
parts of large cities may experience 30 to 60 tons per square mile
per month.  Values above 25 are considered undesirable  from a
community welfare standpoint by some authorities.  Suspended
particulate matter (on a weight basis)  ranges from an average of
40 micrograms per cubic meter of air in non-urban areas to
more than 2500 in highly polluted areas, on particular days.
Long-term average values for cities having 700,000 to 1,000,000
population, at locations in  the central  part of town, are around
144 micrograms per cubic meter of-air. Cities having 100,000
to 400,000 population experience an  average of ^about 120 micro-
grams per cubic meter of air.  Values for the soiling power of
suspended particulate  matter range from near zero with  a clean
atmosphere to more than 10  Cohs per 1000 linear feet for a dis-
agreeably smoky day. 99 One authority classifies values  from
zero to 0. 9 Coh per 1000 linear feet as light smoke, from 1. 0 to
1.9 as moderate,  from 2. 0 to 2.9 as heavy, from  3. 0 to  3. 9  as
very heavy,  and 4.0 or more as extremely heavy. 100 Average

74                                IRON AND STEEL INDUSTRY
values for a reasonably clean city would be around 1. 0 or less on
an annual basis.  Sulfur dioxide values range from near zero in
small communities to more than 5 ppm in the vicinity of major
sources for short periods.  General average values for the
central parts of large cities (1,000,000 population or more) might
be around 0. 03 to 0. 06 in the  summer and 0. 03 to 0. 25 in the
winter, depending on fuels burned in the area, industrial activity,
pollution dispersion, etc.  The minimum detectable by the sense
of smell is 3. 0 ppm and by sense of taste, 0. 3 ppm.  Some species
of vegetation are damaged by  concentrations of 0. 3 ppm,  if the
concentration persists for a number of hours.   Sulfur dioxide
values, as indicated by lead peroxide candles, may exceed 4. 0
milligrams 863 per 100 square centimeters per day  during the
winter in the central parts of  fairly heavily polluted cities.  In
the summer values are around 1.0 milligram SO3 per 100 square
centimeters per day.  In the residential suburbs of such a city,
these values are around 0. 4 in the summer and 1. 0 in the winter.
Higher values would be expected near large sources  of sulfur
dioxide pollution.

    All values reported later herein must be considered  in the
light of changes that occur in  meteorological conditions.   The
wind seldom blows in a certain direction continuously.  Thus, for
a fixed sampling station,  emissions from a given source  may be
carried to the station for only a small part of an hour,  a  day, or
a month.  The values reported represent average conditions at
a point during the sampling period.  No doubt large variations
occur that cause instantaneous pollution concentrations to be many
times greater or many times  less than the average.

Studies Made in the  United States

    During the 1956 steel strike, pollution measurements were
made in four steel-producing  areas, for comparison with similar
measurements  made after the steel mills resumed operation. lO*
Tests were made during July  and August, at which time pollution
from space heating plants was not involved.  Certain other activ-
ities that may cause pollution also are curtailed or shut down
during steel strikes. Thus, the change in pollution levels from
the strike period to the post-strike period was due to the total
effect of rate of pollutant emissions  from all activities.   Sampling
stations were located from 1/8 to 1 mile away from steel mills.
Suspended particulate pollution during the post-strike period was
higher by 44 to 171 percent than during the strike, varying from
one city to another.  Average  suspended particulate levels ranged
from 183 to 279 micrograms per cubic meter in the four  areas
during the post-strike period.  Sulfur dioxide pollution measured
in one  area did not vary, from the strike period to the post-strike
period, indicating that the steel mills did not cause major pollution

Effects of Air Pollutants on the Community                   75
by this gas.  The soiling power of suspended particulate matter
measured in one community was twice as great during the post-
strike period as it was during the  strike period, indicating that
the steel mills (and other activities discontinued during the strike)
are a significant source of this type of pollutant.  Average post-
strike values were 1.2 Cohs per 1000 linear feet, with 8 percent
of the values being greater than 2.  The concentration of iron in
suspended particulate pollutants during the post-strike period
was 2. 6 to 10. 8 times greater than during the strike period,
indicating that the steel mills are  an important  source of this

    A study was made in  1950 in a northern Ohio city  during
periods when the'steel mills were shut  down because of a labor
strike and for a comparable period after operations were
resumed. 102 Mean sulfur dioxide concentration during the strike
period was 0.034 ppm, with one sample in the range 0. 29 to 0. 39
ppm.  During the post-strike period the average was 0.050 ppm,
with one sample in the 0. 39 to 0. 49 ppm range;  however, this
increase was not considered statistically significant.  Measure-
ments of fluorides were practically the same during the  strike
and post-strike periods.   The soiling value of airborne dust, as
indicated by shade or darkness of  deposits on filter papers, was
50 percent greater during the post-strike period than during the
strike.  These data indicate that operation of steel mills  in this city
considerably increased suspended particulate matter,  which causes
soiling, and had a minor effect on sulfur dioxide concentrations,
but no demonstrable effect on fluoride concentrations.

    A 1-month study has been made of  air pollution in the upper
Ohio River Valley area. 103 There are two major steel mills,
two large coke plants,  steam-powered electric-generating
plants,  and other industries in the area. Sampling stations were
located in the immediate neighborhood of major spurces of
pollution.  Dustfall was 566 tons per square mile per month near
a steel mill,  and 123 at another site.  Total suspended particulate
matter averaged 383 micrograms per cubic meter of air in one
city and 186 in another.  A maximum of 1238 micrograms .per
cubic meter was observed in one community, and one  suspended
particulate matter sample from that city contained 30. 8 micro-
grams of iron per cubic meter  of air.  Soiling values averaged 5. 5
and 5. 3 Cohs per 1000 linear feet in two cities.   Although other
sources contribute to pollution  in this valley area,  the data dem-
onstrate the high pollution levels that may be found in the  imme-
diate area of some iron and steel works.

76                               IRON AND STEEL INDUSTRY
    As part of a generalized study of the occurrence and sources
of benzo(a)pyrene as an air pollutant, a short-term study was
done in an upper Ohio River community in late March and early
April 1961.  166  Forty-eight-hour (approximate) samples were
collected in high-volume filter paper samplers at four locations
for a period of 12 days.  Sampling sites were about 1-3/4 miles
from a large, integrated steel mill with many coke ovens.  Three
of the four sampling stations were on hills, at an elevation of
about 1000 to 1100 feet above sea level.  One was located in a
valley, at about 680 feet above sea level, the approximate ground
elevation at the steel mill.   Total weight of suspended particulate
matter in the air ranged from 23 to 189 micrograms per cubic
meter and averaged between 69 and 142 for the four stations; the
highest station average (142) was at the station in the valley.
None of these data indicated grossly  excessive suspended particu-
late pollution. Benzo(a)pyrene  concentrations for composite
samples for the 12-day period at each station ranged from 0.006
to 0.01 microgram per cubic meter of air. Similar concentrations
of benzo(a)pyrene are also found in many cities  of the country. 112
The amount of benzene soluble organic matter in the samples
ranged from 2.0 to 11. 5 micrograms per cubic  meter and averaged
6.7.  Such values are in the same range as those reported for the
air over many cities. 167 Concentration of certain metallic ions
were determined in each of the  four samples.  Although little
significance can be attached to this small number of analyses, it
was found that concentrations of iron, manganese, titanium,
chromium, cadmium, and antimony were above the average found
in many cities. 167  Concentrations of lead, copper, vanadium,
zinc, and nickel were below the averages found  in many cities.

    In one Pennsylvania town, industrial development was
dominated by a steel and wire plant,  and  a zinc  plant.   The  steel
mill,  in 1949, consisted of two blastfurnaces, eleven 110-ton
open hearth furnaces, a sintering plant,  and other equipment.
There  were also a coke plant and two steel plants'at another city
a few miles  away.  Measurements made between February 16 and
April 27, 1949,104  gave the distribution of values for suspended
particulate matter shown in  Table 10. Station 9 Was 300 yards
from the open hearth shop; station 10 was 800 yards from the open
hearth Shop  and about 2 miles from the steel mill in the neighbor-
ing city; station 11 was 600 yards from the blast furnaces and 500
yards from the open hearth shop; and station 12  was 70 yards from
the open hearth shop and 800 yards from the blast furnaces.
Although average values are not given,  the number of samples
containing more than 500 micrograms per cubic meter (22 out of
53) indicate  very high pollution levels.  Other sources contributed
to pollution present,  but there is little doubt that the steel mill
made a sizeable contribution.  High levels were found in April when

Effects of Air Pollutants on the Community
use of fuel for space heating by domestic and commercial
establishments was at a relatively low rate.
              (Reference 104)

0 to 400
500 to 900
1000 to 1400
1500 to 1900
2000 to 2400
2500 or more
Number of measurements
in the range indicated
      a Two-hour samples.
    An extensive study has been made of air pollution in a large
Michigan community. In 1951,105 suspended particulate matter
was measured at 31 stations.  At one station near two coke plants,
iron and steel foundries, a blast furnace,  and many other industries,
suspended particulate matter concentrations in the  air averaged
343 micrograms per cubic meter over a 5-week period.  Iron
content of suspended particulate matter averaged 5. 8 micrograms
per cubic meter. During the same period,  in a residential area,
total suspended particulate matter averaged 122 and iron content
of suspended particulate matter averaged  0. 6 micrograms per
cubic meter of air.   Dustfall measured from 1951 through 1953106
averaged 116 tons per square mile per  month in the industrial
area while in a residential area it averaged 34.  The high pollu-
tion levels were not due to iron and steel operations only,  but
there is little doubt that these played an important part in causing
the high levels.

    Processing of blast furnace slag may cause pollution in the
surrounding neighborhood.   The operation usually consists of
moving slag from the blast furnace to a dump.  After cooling, the
slag is segregated,  crushed, classified by screening, and loaded

78                               IRON AND STEEL INDUSTRY
for shipment.  Dustfall in and around a small community (500
population), resulting from operation of one such plant where some
steps had been taken to control emissions, has been reported. HO
Even with some controls in use, it was estimated on the basis of
analyses of collected dust that from 5 to  109 tons of dustfall per
square mile per month at stations located about 0.4 miles from
the plant resulted from blast furnace  slag processing.   The
variation was  due to relative prevalence  of wind directions,
specific station location, and other factors.  Suspended particulate
samples were also collected at one station located 0.4 miles from
the plant, and often downwind from it. Loadings ranged from 124
to 1980 micrograms per cubic meter  of air, and averaged 411.
At another station located 0.8 miles from the plant, and in the
same direction from the plant as the other station,  concentrations
averaged 195, with a maximum of 477.  While other pollution
sources contributed to these pollution levels, chemical analyses
indicated that  35 to nearly 100 percent of the suspended dust at
stations 0.4 miles downwind from the plant came from blast
furnace slag processing.
Studies Made in Foreign Countries

    Studies made in foreign countries may not be applicable to
the United States situation, since different equipment,  materials,
and processes may be used.  Also, equipment may be in a dif-
ferent state of repair, and operating practices may be different.

    A detailed study of dustfall rates in the vicinity of an inte-
grated iron and steel works was made in England in 1955, 1956,
and 1957. Measurements were made at 64 stations both on the
grounds of the mill and in the surrounding neighborhood. 10?  On
the plant premises, dustfall rates from 250 to 500 tons per
square mile per month were found.   Values decreased to about 50
tons per square  mile per month at distances of 1/2 to 3/4 mile
from the plant.   In the same study,  sulfur dioxide was measured
by the lead peroxide candle method at 30 stations.   Measurements
were made during the summer months when space heating units
were not in operation.   Values were about 1.0 milligram of 803
per 100 square  centimeters per day at a distance of about 1/2
mile from the plant and increased to about 2.0 on the plant
premises. The  data indicate that emissions from the plant cause
excessive dustfall in the neighborhood of the plant and make some
contribution to sulfur dioxide pollution levels.

    Measurements  of dustfall in the Ruhr district of Germany in
the vicinity of a  steel milll°8 indicated dustfall rates of 42 to 133
tons per square  mile per month at locations within 1/2 mile of the

Effects of Air Pollutants on the Community                    79
mill, varying with specific sampling location, wind direction, etc.

     Suspended atmospheric dust was measured at a number of
locations in Belgium by exposing vaseline-coated aluminum
spheres 12 centimeters in diameter. 109 Average values for the
amount of dust collected were 1. 5 milligrams per day in the
country,  6 in the central parts of cities, and 20 in the neighbor-
hood of a coke plant.  These data indicate that the air near a coke
plant contains more than 3 times as much dust (of the type
collected by this instrument) as the air in the central parts of

     Measurements of benzo(a)pyrene were made  in the neighbor-
hood of a Russian pitch-coke and coke-chemical plant of obsolete
construction.  Concentrations of 0. 002 to 0. 4 milligram per cubic
meter were found on the plant grounds. HI  The plant was
modernized and pollution emissions brought under a measure of
control.  Benzo(a)pyrene values dropped to 0. 0004 milligram per
cubic meter. There was not sufficient information given in the
paper to evaluate significance of the data; however, a value of
0. 00005 milligram per cubic meter is  considered relatively high
for the general atmosphere of polluted cities,112 indicating that
the plant investigated was a substantial source of this pollutant.
It is generally accepted that destructive distillation of coal results
in formation of benzo(a)pyrene and a number of other similar

     The concentration of phenol in the air has been measured in
the area surrounding a large coke-chemical plant in a Russian
community.  The "characteristic odor of coke-chemical plants"
was clearly perceptible to all points within 1. 86 miles of the
plant. 42  Maximum concentrations of 0. 8 ppm of  phenol were
found within 0. 6 mile of the plant and 0. 3 ppm at  a distance of
1.86 miles.  The odor threshold of phenol is less than 0. 9 ppm
and may be as low as 0. 05 ppm.  (There are conflicting reports
on the odor threshold of phenol.)   Thus, people living in an area
of 1 to 3 square  miles may be, from time to time, bothered by
odors of phenol from the plant under study.   The  "characteristic
odor of coke-chemical plants" would, from time to time,  be
detectable by people residing in an area of about  3 square miles
surrounding this plant.

     As a means of developing a concept or feeling as to the magni-
 tude of emissions from various operations at iron and steel works
 and to relate these to emission of pollutants  from other sources,

                   IRON AND STEEL INDUSTRY
     Name of equipment
     and its capacity
Air pollution control
equipment; exit gas grain
loading; and other condi-
tions assumed
Emission from
described oper-
tlon, Ib/day
 emission with
 best available
control, Ib/day
   250 ovens.
   5,100 tons of coal used
   per day.
   3, 560 tons of coke produced
   per day.

   1,000 tons of sinter pro-
   duced per day in one
   Four units each pro-
   ducing 1000 tons per day.
   Thirteen furnaces,  each
   with 175-ton rated capac-
   ity. No oxygen roof lances.
   Two furnaces with 50-ton
   rated capacity each. Five
   heats per day.
   One 45-inch, four-sided
No special equipment.
0.1 % of coal used emitted
as solids (estimated).
Reasonably good practice.
 Cyclones clean 140,000
 scfm of combustion gases
 to 0. 2 gr/scf.

 Cyclones clean 15,000
 scfm of gases from belt
 discharge enclosure to
 0. 4 gr/scf.
 Preliminary cleaners and
 wet scrubbers handle
 87,000 scfm per furnace.
 Exit loading 0.3 gr/scf.
 No control equipment.
 25,000 scfm per furnace.
 Average exit loading is
 0.4 gr/scf.
No control equipment.
Loading 0. 43 gr/scf in
30,000 scfm for each

No control equipment.
Loading 0. 4 gr/scf in
85,000 scfm.
                             10, 200
   21, 500
   26, 800
                         Pollution source
                                    Partlculate pollutants
                                      emitted,  Ib/day
  5, 000 homes burning coal; average winter day; northern city.               3, 780
 5r, 000 homes burning fuel oil; average winter day; northern city.            6, 000
 50, 000 homes burning refuse in open fires.                                4, 400
Refuse from 100, 000 homes burned in municipal incinerators.                 3, 520
300, 000-kw coal-fired electric power station;  good dust collectors.          11, 000
300, 000-kw coal-fired electric power station;  fair dust collectors.           40, 000

a Values estimated by present authors based on data in this report
  and on published emission data concerning sources of pollution
  other than iron- and steel-making processes.
   No estimate of probable reduction available.

Effects of Air Pollutants on the Community                   81
a tabulation has been prepared (Table 11).  The nature and
magnitude of the individual iron- and steel-making operations
selected for comparison may be found at some mills,  although
wide variations exist.  Emissions of particulate matter only are
tabulated since these are of greatest concern in the case  of iron
and steel works.  Application of available means for control of
emissions from iron and steel mill operations would bring about
a major reduction even though the equipment selected and listed
in Table 11 was taken to include certain air pollution control

    It can be seen that particulate emissions from the selected
iron and steel mill operations  are sizeable in comparison with
certain other sources of particulate pollutants.  For example,
emissions from the 13 open hearth furnaces without air pollution
control equipment are roughly equivalent to particulate emissions
from about 35, 500 coal-fired home-heating plants.  Particulate
matter emissions from the sintering machine having only cyclones
for air pollution control would be roughly equivalent to particulate
matter that would be released to the atmosphere if about 66,000
home owners burned their refuse in open fires. Table 11 can be
used to make other comparisons.  It is apparent that an iron and
•steel mill could be responsible for a sizeable part of the total
particulate matter discharged to the atmosphere of a community.
As indicated by emissions from a large pulverized-coal-fired,
steam-powered, electric-generating station, however, other
major sources may also play a substantial role.

    It is generally agreed that injection of oxygen into open hearth
furnaces through lances installed in the furnace roof increases the
amount of dust and fume discharged from the furnace into the
atmosphere.  The question arises as to the effect this increased
emission has on air pollution levels in the surrounding neighbor-
hood. Since no field measurement data are available to help
answer this question, estimates  have been based on calculations in
which atmospheric dispersion theory and reasonable assumptions
of meteorological and plant-operating conditions were used.

    In these estimates,  it was assumed that there was one open
hearth with a stack 150 feet high and 5 feet in diameter at the top.
Furnace discharge rate was taken as 79,800 cfm at 1200°F (25,000
cfm at 60°F).  Particulate loading in the exit gases was taken as
0.4 grain per scf for the regular open hearth and 0. 6 for the fur-
nace with oxygen lances.  The dust and fume was considered to

82                                IRON AND STEEL INDUSTRY
have a mean particle diameter of 0. 5 micron and a particle dens-
ity of 5. 2 grams per cubic centimeter.  The particle size distri-
bution given by Bishop et al. 57 (see earlier herein) was used.
Meteorological conditions assumed were:  wind speed 11.2 mph;
air temperature 59°F; air pressure 29. 2  inches of mercury; air
density 1. 1945 kilograms per cubic meter; atmospheric diffusion
coefficient 0.12 mn/2; and a turbulence parameter of 0. 25.  The
height of rise  of the plume above the stack, calculated by Holland's
equation, H3 was 88. 5 feet; thus, the particles were  carried to
about 240 feet above ground before being dispersed.  This effec-
tive stack height was used in the calculations.

    Deposition of open hearth dust in the  surrounding neighbor-
hood (dustfall) results from both fallout and diffusion, the latter
being the most significant for small particles.   For both fallout
and diffusion,  the deposition rate at various distances from the
stack was determined along an imaginary line downwind from the
stack.  Total deposition  rate was obtained by adding fallout and
diffusion.  Deposition at certain distances downwind from the
stack and at various distances away from the center line of the
plume were also  computed by use of the crosswind diffusion
component of Button's equation. 113 Airborne particulate matter
concentrations at ground level and the distance from  the stack to
the point of maximum concentration were computed from atmos-
pheric diffusion equations. H3

    Based on  results of  these calculations, several conclusions
were drawn:

         1.  Diffusion is  especially  important in dispersion
            of the very fine particles emitted by an open
            hearth furnace, whether oxygen lances are used
            or not.   Only about 6 percent of the total  emis-
            sion falls to  the ground within a mile of the stack.
            The bulk of the material is still airborne at that

         2.  The most important effect of using oxygen lances
           in the furnace is that ground level concentrations
           are increased at all points in  the neighborhood of
           the plant.  The increase in community pollution
           levels is in direct proportion  to the increase in
           emissions.   Therefore, use of oxygen lances
           increases average pollution concentrations 50
           percent, over those from furnaces  operated
           without lances, providing both types of furnace
           are operated continuously.  At the point of
           maximum concentration of airborne particulate

Effects of Air Pollutants on the Community                   83
            matter from the furnace under consideration,
            concentrations would be 95 micrograms per
            cubic meter of air for a regular furnace and
            142 for a furnace with oxygen lances.  The
            point of maximum concentration occurs about
            5000 feet from the stack under the conditions
            given above.

         3.  Dustfall in the community surrounding the
            furnace are appreciable, even though most
            of the particles tend to remain airborne.
            One open hearth stack causes a dustfall rate
            of 10 tons per square mile per month or more
            for a distance of 2 miles downwind from the
            stack if oxygen is  not used,  and for a dis-
            tance of nearly T miles if oxygen lances are
            used.  For purposes of comparison, dustfall
            rates in suburban areas from all  sources of
            pollution is often about 10 tons per square
            mile per month.

    The average steel mill has about 12 open hearth furnaces.
The foregoing was based on a single furnace.  Operation of more
than one furnace of course increases pollution in the surrounding
community;  however,  the relative increase at ground level would
be somewhat less than the relative increase in emissions from
additional stacks.

    In the air pollution levels  calculated above, a constant wind
direction was assumed.   The pollution level estimates are correct
for a specific moment, but they would not be representative of
longer time  periods,  such as 30 days for dustfall and 24 hours for
suspended particulate matter.  The wind direction that will carry
pollution from a source to a certain location in the neighborhood
may occur about 15 percent of the time over a long period.   Thus,
the pollution levels estimated above would be 6 or 7 times too
high.  However,  since they are based on a single furnace whereas
the average  open hearth shop has 12 furnaces, the estimates are
probably representative of average neighborhood pollution from
the operation of an open hearth shop.

    Pollution levels were estimated on the basis of "average" at-
mospheric dispersion conditions.  Actually, during the daytime
stack plumes spread laterally  and vertically with a decrease in
average concentrations,  whereas night conditions often  stretch
out the plume and cause concentrations higher than those given.
Generally, either daytime or nighttime conditions prevail rather
than "average" conditions.

84                               IRON AND STEEL INDUSTRY
    Iron and steel mills usually include many sources of pollution
other than open hearth furnaces.  Thus, the percentage increase
of emissions from the plant as a whole that would occur  if oxygen
roof lances were to be added to the open hearth furnaces would be
reasonably small,  perhaps in the neighborhood of 10 to 15 percent.

    Two steel mills in the western part of the country have been
associated with neighborhood air pollution problems caused by
the fluorides emitted.  The fluoride emission was largely due to
the higher-than-average fluoride content of the iron ore processed
and to a limited extent to fluorides in fluxes used.  These high-
fluoride ores are used by very few mills.  Fluxes are used by
most  mills,  but have not caused a recognized problem.   The
problem has been accumulation of fluorides in certain species of
vegetation.  This has resulted in damage to certain vegetation,
and cattle have been injured from eating contaminated forage.

    In the case of one of the mills, claims have been made and
settled for damage to cattle and for some damage to small fruits
(apricot and plum).  There have also been some limited nuisance
claims for ornamental and other  crops.  In the case of the second
mill,  the claims were chiefly for damage to grape and citrus with
limited claims for damage to other agriculture products.

                           Part V



    Knowledge about the health effects of air pollutant emissions
from the iron and steel industry is seriously deficient,  and
studies specifically directed at the problem are almost nonexist-
ent.  Lacking such definitive information,  we can attempt to pull
together bits of several types of more or less pertinent informa-
tion: (1) Air sampling study areas adjacent to steel mills may
give indications of the extent that the industry contributes to the
general atmospheric pollution  - this aspect was reviewed in a
previous section, (2) Observations on the health of community
populations in the vicinity of steel mills may indicate whether
residence in such areas  is proving hazardous, 114-125 (3) Data
from the annals of occupational medicine should offer clues to the
possible effects of long exposure to high doses of specific
agents,126-159 (4) animal experiments under controlled condi-
tions may give us a firm base line of specific effect ijrom speci-
fic cause."9, 160-163 ^s promising as this outline may sound,
the yield obtainable from it is far less than is desirable.

    The evidence from these and other sources  presents  a com-
plex and uncertain picture concerning operations of the iron and
steel industry.  Emissions might cause detrimental effects on
man's health if present for sufficient periods of time at high
enough concentrations.  For some emissions, the evidence of
potential toxicity is conflicting, one investigator describing ad-
verse effects and another finding none under comparable  condi-

    From air sampling studies comes evidence  that steel mills
do contribute gases and dusts to the atmosphere and that this
contribution may significantly increase the  levels  of some com-
ponents in the vicinity within a varying geographic radius.  The
proportions_of this contribution depends upon the type of opera-
tion, the adequacy of controls, meteorologic and topographic
factors, and the quantities of the components contributed to that
particular atmosphere by other contaminating sources.

    From observations of community effects,  we know of acute
episodes to which steel operation emissions certainly contributed
and are reminded that polluted agricultural crops  may contribute
a hazard.  From Russia, long-term studies suggest that given
sufficient time and the proper  approach other hazards may be
brought to light.

    From studies of industrial health we know that  many of the
dusts—even those usually accepted as "inert" — and at least some

86                                IRON AND STEEL INDUSTRY
 of the gases present in steel industry emissions may, in suf-
 ficient concentrations over long enough periods of time, produce
 changes in human tissues.  Some of these changes are incapaci-
 tating, and  may ultimately be fatal in themselves or may provide
 the nidus for secondary diseases.

     Experiments in which animals are exposed to pollutants.
 provide information useful in considering possible effects of
 pollutants on man. A vast amount  of work of this type has been
 done.  Among the many studies is one in which blast furnace
 stack  dust was found to be inert in  rats under the test conditions
 used, •""

     We do not know the significance of increases in atmospheric
 pollution levels  indicated by some of the air sampling studies,
 because we have yet to establish permissible community levels
 for most pollutants.  Acute episodes like that in Donora are
 always possible given the proper set of circumstances,  but this
 does not mean that the steel industry is any more responsible
 than many others,  nor does it mean that the long-term health of
 the populace is necessarily suffering.  Studies—still too embry-
 onic to be meaningful—of comparative mortality rates have not
 brought forth any convincing evidence that a heavily polluted area
 like Donora has a seasonal pattern of deaths different from that
 of the rest of the country,  or that human mortality rates are
 lower during a steel strike than while the mills are operating.
 Since acute  pollution  episodes are,  or soon should be, completely
 avoidable, it may be  possible to ignore the effects upon  subse-
 quent health and survival of those who became ill from exposure
 to such episodes. ^^1

     Attempts to apply observations based on occupational  diseases
 to air  pollution health effects are apt to distort the point of view
 in two directions:  First,  they describe effects produced by ex-
 posure 8 hours a day 5 days a week to one or a few agents; and
 second, these effects have occurred predominantly in a healthy
 adult working population whose members may be less suscep-
 tible to injury than the young, the aged, and the ailing.  The
 Russians  have chosen to set many of their industrial threshold
 limits considerably lower than our own.  It may be that they are
 crying wolf  or that we are wearing blinders.

     Threshold limits for single agents have been established for
 industrial exposures, and the manifestations of acute toxicity are
well described.  For  some pollutants, cumulative effects are also
 known; for others,  long-term exposure to industrial concentra-
tions produces pathologic lesions that are usually considered
 benign; for still others, the chronic effects are controversial.

Public Health Aspects of Emissions                           87
    Existing information on either the toxicity or absence of
toxicity of these individual substances cannot be clearly applied
to the concentrations or the composite mixtures in which  they
exist in community situations.  With few exceptions,  such as
the Los Angeles alert levels for some gases and the California
ambient air standards,  atmospheric threshold limits have not
been adopted even for the single agents.

    In conclusion, evidence  is adequate to show that industry
contributes (to a greater or lesser degree, depending on the
plant involved) to the total pollution of the surrounding commun-
ity, and that this contribution includes,  in some degree, potentially
toxix and hazardous  substances such as sulfur dioxide,  particulates,
and^probably certain potentially carcinogenic materials.  Further
research that will better define the possible effects of air. pollutants
on man's health must be done before positive cause and effect rela-
tionships can be firmly  established.  Although it cannot be singled
out as the contributor of a specific toxicant resulting in any special
disease condition, parts of the iron and steel industry and many
other pollution sources  do add to the steadily increasing burden of
known and unknown agents that are potentially a hazard to man's
health and welfare.  The diminution of these in the community air
is desirable, not only by this industry, but by the other contributors
as well.

                             Part  VI




    It is not intended here to present standard or model legisla-
tion or to suggest desirable words,  phrases,  paragraphs or
provisions to be incorporated into laws.  The purpose of the
following discussion is to point out the action that would be
required at iron and steel works to bring any uncontrolled
emissions into compliance with certain kinds of laws existing
today.  It is always wise to determine the nature and extent of
pollution in the  community, the number and kind of pollution
sources, and the probable impact of legislation on the community
before actually  adopting many of the possible provisions of a law.
The legislation  must be tailored to meet the needs of each par-
ticular area.

Regulations of Particulate Matter Emissions (Weight Limitations)

    Air pollution  control laws and regulations adopted prior to
the middle 1920's were designed to bring about control of black
smoke emitted from fuel-burning installations.  Hand firing of
coal  was a common practice,  and this, and other practices, re-
sulted in emission of excessive amounts of readily visible black
carbonaceous particles.  As control of this problem moved for-
ward, the public and government officials turned their attention
to ash emitted from fuel-burning units, an emission that was in-
creasing because  of more general use of  mechanical  stokers,
higher firing rates, pulverized-coal-fired units, and other
factors.  Thus, laws and regulations adopted after the 'late 1930's
were often designed to  bring about control of emission of ash as
well  as smoke. *64  ^sh emission regulations, expressed in
various language to limit the weight of material discharged, were
generally such that no collectors would be required on domestic
stokers or commercial underfeed or traveling-grate  stokers
operating at reasonably low firing rates.  For pulverized-fuel-
burning installations and for some types of stokers operating at
high  firing rates,  a stack gas cleaning efficiency of 85 percent
or more (by weight) would be required to comply with the limits
established.  These  regulations were based on judgments of the
amount of emission that would create a nuisance, efficiency of
available control  equipment, and the economic impact of the
regulations on parties  responsible for emissions and, perhaps,
their control.

    The majority of the laws and regulations desi

90                               IRON AND STEEL INDUSTRY
about control of ash emissions embody a provision that the limi-
tations shall apply only to emissions from combustion equipment.
Some do not,  however, and the limitations have been applied to
all types of sources of particulate matter, including certain
operations at iron and steel works, even though, in many situa-
tions, the limitations are grossly inadequate to provide for the
necessary improvement in air quality and are otherwise  inappro-

    Probably the most commonly used limit prohibits emissions
in excess of 0. 85 pound of dust per 1000 pounds of carrying gases,
adjusted to 50 percent excess air for products of combustion.
This value was used in the "Example Sections for a Smoke Regu-
lation Ordinance" published by the American Society of Mechanical
Engineers in 1949. This is approximately equivalent to 0. 45 grain
per cubic foot at  68°F or 0. 25 grain per cubic foot at 500°F,
limitations that appear in a number of ordinances.  Other laws
permit emissions as high as  2. 5  pounds per 1000 pounds of gases
(roughly 1.33  gr/ft ^ at 68°F), while other  more  restrictive
laws prohibit emissions in excess of 0. 57 pounds per 1000 pounds
of gases (roughly 0. 30 gr/ft 3 at  68°F).  Most of these laws con-
tain a stipulation concerning  excess air for  products of combustion,
in terms of percent excess air per se or indirectly in terms of
carbon dioxide content of the gases ^see "circumvention" later

    A more recent development in regulation of emission of par-
ticulate matter to the atmosphere has grown from efforts to control
air pollution in Los Angeles County, California.  Regulations
used in other areas in the past were considered inappropriate
because of the nature and extent of pollution sources and the ad-
verse meteorological and topographical conditions for dispersion
of pollutants in the area.  Therefore, during 1948 and 1949, the
Los Angeles County Air Pollution Control District set out to
develop new standards.   Measurements were made of particulate
emissions from a variety of operations,  including many types of
metallurgical processes.  Initially, attempts were made to
develop emission standards based on visual  appearance of the
stack effluent or on the total process enthalpy. These were not
found entirely satisfactory.  Emission limits were finally
developed that specified a certain maximum allowable emission
in pounds per hour, depending on the total process weight per
hour.  These limits have been in  use since March 1949 in Los
Angeles County,165 and similar limits have been adopted by a
number of other jurisdictions.  Regulations  concerning visual
appearance of stack effluents were also adopted (see later herein).
In determining allowable emission rates, measured emissions
were reduced by factors  based on known efficiency of dust- and

Air Pollution Laws and Regulations                          91
fume-collecting equipment.  The final table of allowable emis-
sions,  in terms of pounds per hour,  in general,  requires a
collection efficiency of about 80 percent for small operations and
about 90 percent or more for large operations.  These collection
efficiencies were considered necessary to provide for needed air
quality improvement in Los  Angeles, were obtainable with avail-
able control equipment,  and were within the economic reach of
the industries that would have to provide  control equipment.

    Regulations based on process weight apply to operations con-
ducted at iron and steel works, except combustion operations con-
ducted for space and water heating, steam raising,  incineration,
or salvage purposes.  The maximum amount of particulate  matter
that may be discharged (in pounds per hour) ranges from about 0. 5
percent of the process weight (0. 5  Ib/hr)  for plants handling 100
pounds of process materials per hour to 0. 067 percent (40 Ib/hr)
for plants handling 60, 000 pounds per hour of process materials.
The Los Angeles  rule prohibits emission of more than 40 pounds
per hour from any one source, whereas the San Francisco Bay
Area regulation permits emission of  more than 40 pounds per
hour from very large plants  but allows only a very small percent
of the process weight to be emitted, e. g., 0. 0069 percent of the
process weight  for a 1,000,000-pound-per-hour operation.  Some
other jurisdictions using this type regulation permit a higher
percent of the process weight to be emitted in the case of large
operations.  Riverside County, California,  has a rule of this type.
The Bay Area regulation has a further stipulation concerning the
minimum concentration  of particulate matter that can be required,
in terms of grains per scf in exit gases.  Values  range from 0. 1
grain per scf for  an emission of 7000 scfm to 0. 02 grain per scf
for an  emission of 1,000,000 scfm or more.

    The Allegheny County, Pennsylvania, Health Department on
July 5, 1960, enacted regulations dealing specifically with weight
of particulate matter discharged from iron and steel mill opera-
tions.  Emissions from blast furnaces, after burning, are limited
to not more than 0. 35 pound per 1000 pounds of gases.  Particulate
matter in excess  blast furnace gas being  bled to the atmosphere is
limited to 0. 50  pound per 1000 pounds of  gases.   The limitations
do not apply during "slips" of the furnace burden.  A device for
recording occurrence of  "slips" is required.  Persons responsible
for operation of blast furnaces must  conduct research to determine
methods of controlling emission of pollutants during "slips. "
Emissions  from open hearth and electric furnaces and sintering
plants built after  July 5,  1960, are not to exceed 0. 2 pound per
1000 pounds of  exhaust gases.  A program that will require con-
tinued installation of air pollution control devices (or replacement
of facilities) to reduce emissions from existing installations to this

92                                IRON AND STEEL INDUSTRY
level is to be recommended by an advisory committee for approval
of the Board of Health; a reasonable period of time for compliance
will be allowed.  Persons responsible for these units are also
required to  do research on methods of further controlling emis-
sions.  Basic oxygen steel furnaces are limited to an emission of
0. 2 pound or less per 1000 pounds of exhaust gases.  Particulate
emissions from heating and reheating furnaces are limited to not
more than 0. 30 pound per 1000 pounds of gases, and emissions
from air furnaces are not permitted to exceed 0. 50 pound per
1000 pounds of gases.

     Steubenville, Ohio, has an ordinance similar to that of
Allegheny County but has slightly more lenient limitations on a
few processes.  Lorain, Ohio, prohibits emissions of more than
1. 0 pound per 1000 pounds of exit gas from blast furnaces,
heating and reheating, and air furnaces.   That city further pro-
vides that coke plants, open hearth furnaces, sintering plants,
and Bessemer converters "shall incorporate those means of
controlling  particulate matter emission which have been proven
economically practical."

     Enactment of laws or regulations providing for control of
emissions of particulate matter based on technical standards
(such as Ib/hr or lb/1000 Ib of exhaust gases) impose a respon-
sibility to measure emissions.  This is a formidable job, re-
quiring trained personnel,  laboratory and field equipment,
process information, etc.  The  control agency must be provided
with resources to do this work,  or  as an alternative, the law may
require persons responsible for emissions to conduct necessary
tests. Even then the control agency must have personnel who
can insure that tests are properly conducted.
Regulation of Visible Emissions

     Nearly all air pollution control ordinances or regulations
embody a limitation on visible emissions, particularly dense
smoke.  The Ringelmann Chart is used almost universally as the
standard for determining smoke density, although some laws
permit use of other equivalent means.  These regulations were
designed primarily for use in controlling black smoke from fuel-
burning devices used for space and water heating, steam raising,
and  refuse incineration.  In some areas, however, the same
regulation used to control smoke from fuel-burning operations is
used in an effort to control emissions from iron and steel works,
even though,  in many cases,  such regulations are grossly inade-
quate to provide for the necessary improvement in air quality.

Air Pollution Laws and Regulations                           93
     The most common fuel-burning regulation prohibits smoke,
the shade or appearance of which is equal to or darker than No. 2
of the Ringelmann Chart. Some darker smoke is usually permitted
for a few minutes (usually 4) in any half-hour and for a few min-
utes (usually 3) in a 15-miriute period, when a fire is being
cleaned or built.  A wide variety of limitations,  more and less
severe, exists.

     A few communities have adapted use of the Ringelmann Chart
to a procedure for measuring visible emissions other than  black
smoke.   Emissions having an opacity that obscures an observer's
view to a degree equal to or greater than smoke of a stated
Ringelmann number are prohibited.  This "equivalent opacity"
regulation would be applicable to certain iron and steel mill
emissions, particularly open hearth, electric, and basic oxygen
steel furnaces.  For these sources,  the equivalent opacity limita-
tion may be more difficult to comply with than a relatively severe
limitation on weight of particulate emissions.  This situation is
due to the great light-scattering power of the very fine particles
emitted from these  furnaces.  They would not constitute  a great
amount of weight and are more difficult to collect than large
particles.  As little as 0. 05 grain per scf in an exhaust stack may
cause a visible plume.

     The Allegheny  County, Pennsylvania, Health Department
regulations include  special provisions for dense smoke from
byproduct coke ovens.  A generally applicable clause prohibits
smoke darker than  Ringelmann No.  2.  This applies to coke plants
and includes smoke leaking out of oven doors. However, more
dense smoke is permitted for 12 minutes in any 60-minute period
when coke is charged to a battery of ovens and when coke is pushed
from a battery of ovens.  Emission of smoke from heating,  re-
heating,  and air furnaces may not be darker than Ringelmann No.
2,  but during an aggregate of 6 minutes in any 60-minute period,
darker smoke is permitted.   Blast furnaces, open hearth and
electric furnaces, sintering machines, basic oxygen furnaces,
and Bessemer converters are exempted from the dense smoke
 Miscellaneous Regulations

     The great majority of air pollution laws and regulations con-
 tain a clause relative to nuisances caused by air pollutants.  Fur-
 thermore,  nearly all jurisdictions have a general nuisance law
 applicable to unsatisfactory conditions caused by air pollutants.
 Such regulations  typically prohibit emission of pollutants in a
 manner or in amounts that cause injury, detriment, nuisance,  or

94                                IRON AND STEEL INDUSTRY
annoyance to the public,  that endanger the public comfort, wel-
fare, health, or safety,  or that cause injury to business or
property.  These anti-nuisance laws can be used as a basis for
requiring control of air pollution problems caused by iron and
steel works, problems such as odors, excessive dustfall,  and
soiling of property.  The legal and administrative procedures are
cumbersome, however,  and it is often difficult to demonstrate
conclusively that a certain problem or specific portion of a
problem is caused by a given pollution source.  This is especially
true in large urban areas.

     Allegheny County, in addition to other regulations described
above, has prohibited  operation of  beehive coke ovens.  Also,
construction of new Bessemer converters is not permitted until
equipment has been developed that  will prevent emissions in ex-
cess of 0. 65 pound of dust per 1000 pounds of gases at 50 percent
excess air for products of combustion.

     Most of the limitations on weight of emission of particulate
matter do not have special provisions  concerning size of particles
emitted.  A few regulations (usually those designed primarily
for control of emissions from coal-fired boilers) limit  the part
of the total particulate emission that may consist of particles
greater than 44 microns in diameter or that  are retained on a
U. S. Standard 325-mesh screen.  The most  common regulation
of this type prohibits emission of more than  0. 2 pound of particu-
late matter,  larger than 44 microns,  per 1000 pounds of gases.
Some steel mill operations might be affected by such regulations.

     Regulations applicable to emission of gaseous air pollutants
have been adopted by only a few jurisdictions and with respect to
only a few pollutants.  The usual anti-nuisance laws can, however,
be used to regulate gaseous emissions that create odor problems;
cause some sort of demonstrable injury,  detriment, or annoyance
to man; or cause injury to vegetation,  animals,  or property.
Sulfur dioxide and fluorine compounds (the latter may be in gaseous
or particulate form) are pollutants  that arise at steel mills and
have been subject to specific regulations.  A few jurisdictions
(perhaps six) limit sulfur dioxide concentration in exhaust gases
to 0. 2 percent.  One regulation (San Francisco Bay Area) further
limits emission on the basis  of concentration of sulfur dioxide
caused at locations off the plant premises, at ground level.  This,
in effect,  limits rate of emission.  One regulation (San Bernardino
County, Calif.) prohibits emission  of "that amount of fluorine
compounds which causes injury to the property of others. "

     Since many regulations embody limitations on emission of

Air Pollution Laws and Regulations                           9 5
particulate matter in terms of grains per cubic foot or pounds of
dust per 1000 pounds of carrying gases, it becomes necessary to
arrive at some reasonable volume of gases,  going through the
equipment and being discharged,  that is needed to carry on the
operation properly.  Otherwise,  before discharged,  the effluent
could be diluted with air to the extent necessary to meet the
emission limitation without actually reducing the total weight of
particulate matter emitted. Since pollution levels in the ambient
air are a function of the amount of pollutant discharged per unit
time, such dilution would  have nearly no beneficial effect on
ambient air quality.  In the case of emission of particulate
material from combustion operations, the matter is handled by
providing that stack concentration of particulate matter be ad-
justed to the basis of 50 percent excess air for products  of
combustion; however, this procedure is not applicable to many
iron- and steel-making processes.  A few regulations contain a
specific clause  to control  actions that would make it possible for
an operation to  comply with regulations and yet not actually reduce
emission of pollutants.  The Los Angeles County  regulation on
this matter (circumvention) states that "a person shall not build,
erect, install, or use any  article,  machine, equipment,  or other
contrivance,  the use of which, without resulting in a reduction in
the total release of air contaminants to the atmosphere,  reduces
or conceals an emission which would otherwise constitute a
violation... "  Regulations that limit emissions in terms of pounds
per hour are  much less subject to problems associated with
determination of a reasonable gas volume discharge since, in
such regulations, it makes no difference whether the pollutants
are more concentrated in  a small volume of gases or more
diluted in a large volume.

     The Chicago ordinance of May 1, 1959, exempts from air
pollution regulations all blast furnaces and auxiliaries, byproduct
coke plants, open hearth furnaces,  sintering plants, Bessemer
converters, pneumatic steel-making furnaces, electric furnaces,
and a few other installations.  The ordinance provides that re-
search shall be conducted to determine feasible methods for con-
trolling smoke and particulate matter from this equipment. Until
such time as  definitive standards are established for these in-
stallations, emissions of  smoke and particulate matter are not to
exceed those  associated with "normal good operating practice. "

     Some ordinances and  regulations stipulate the maximum effi-
ciency to be required of dust-collecting equipment.  Commonly
used maximum required efficiencies are 85 percent for installa-
tions built subsequent to adoption of the regulation and 75 percent
for installations built prior to the effective date of the regulation.
These provisions were designed primarily for fuel-burning equip-

96                                IRON AND STEEL INDUSTRY
ment and represent an economic compromise based on collection
equipment and equipment costs existing 10 to 15 years ago.  These
values are based on consideration of ash content of coal,  effect
of boiler load changes on collector efficiency, space available to
install collectors, expected life of plants,  etc.

     A few regulations contain specific provisions concerning
control of wind-borne dust from handling or storing materials or
from dusty roadways or parking lots, building or wrecking opera-
tions,  etc., The Allegheny County regulations require that rea-
sonable precautions be taken to minimize air pollution in such

     In this section,  consideration is given to the types of control
 equipment needed to meet various emission limitations.  The
 discussion must be considered only  in generalities because of
 the wide variation that exists between  equipment,  raw materials,
 fuels, and operating procedures.  An evaluation is necessary for
 each situation in actual practice.

     Equipment necessary to bring iron- and steel-making
 processes into compliance with regulations will be discussed in
 terms of the following: (1)  Los Angeles County process-weight
 regulations,  (2) San Francisco Bay Area regulations based on
 process weight and minimum grain loadings to be required,
 (3) Allegheny County regulations, (4) visible emission limitations,
 and (5) other regulations such as those concerning odors, sulfur
 dioxide,  and fluorides.
Sintering Machines

    Use of well-designed inertial separators (cyclones and de-
vices of similar efficiency) might make it possible to comply
with Allegheny County regualtions.  In other cases and in order
to meet Los Angeles and San Francisco regulations, high-efficiency
collectors (electrostatic precipitators, high-efficiency scrubbers,
and bag houses) possibly in series with an inertial separator will
be required.  Visible emissions probably will not present problems
nor will odors or sulfur dioxide.  Fluoride emissions may cause
neighborhood problems when certain western United States ores
are used.  Equipment for control of fluorides in such cases is
described in detail elsewhere herein.

Air Pollution Laws and Regulations                           97
 Coke Plants

     Data reviewed were not adequate to determine whether
 emission of particulate matter from coke plants would be in
 violation of the various regulations concerned with weight of
 particulate matter emitted.  Control of emissions to the extent
 necessary to comply with visible emission regulations of
 Allegheny County would require application of operating and
 oven-door maintenance practices described elsewhere herein.
 Emission of materials causing odors in the surrounding com-
 munity would also be reduced by such control practices.  Emis-
 sion of sulfur dioxide would not be expected to exceed  amounts
 allowed by regulations except, perhaps, in the event that a
 sulfuric acid plant were part of the byproduct processing plant.
 Emission of fluorides would not be of major concern.
 Blast Furnaces

     Blast furnace gases, after passing through a preliminary
 cleaner such as a settling chamber or cyclone and a primary
 cleaner such as a wet scrubber, would in many cases contain
 particulate matter in excess of that allowed by Allegheny County
 regulations and in nearly all cases would exceed those allowed by
 Los Angeles and San Francisco regulations.  Compliance with
 these regulations  would require secondary collectors (in series
 with primary collectors) with efficiencies of around 90, 80, and
 70 percent, respectively, in Los Angeles,  San Francisco,  and
 Allegheny County.  Since most of the larger particles are removed
 by the first two stages of the gas-cleaning system,  the secondary
 collector would have to be effective in collecting fine particles
 to achieve the required efficiency.  Such devices would be elec-
 trostatic precipitators, high-energy scrubbers, or perhaps fabric
 filters.  No particular measures would be necessary to comply
 with regulations concerned with visible or other pollutant emis-
 sions; however,  the gases should be burned either in heat-exchange
 equipment or in waste-gas burners to prevent release of large
 volumes of carbon monoxide.
Open Hearth Furnaces

     For an average-size furnace (175-ton capacity) a collector
with an efficiency of 70 to 75 percent would be required to meet
emission-weight regulations of Allegheny County,  San Francisco,
and Los Angeles on a furnace not using roof lances.  An efficiency
of around 80 percent would be required if oxygen were used.
Collectors that can provide  this efficiency include electrostatic

98                                IRON AND STEEL INDUSTRY
precipitators, high-energy scrubbers,  and fabric filters.  Higher
efficiencies would be required for larger furnaces in order to
comply with regulations of Los Angeles and San Francisco.
Emission control necessary to meet equivalent opacity regulations
requires high-efficiency collectors.  Exit-gas loadings would
have to be about 0. 05 grain per scf (and perhaps lower) to eliminate
the visible plume from an open hearth.  This would indicate that
collectors on furnaces using oxygen would have to remove about
95 percent of the particulate matter.  The corresponding figure
for regular open hearths is about 92 percent.  Such efficiencies
require excellent collection equipment.  Sulfur dioxide emissions
from open hearths will generally meet regulations without special
control measures,  since process considerations require low
sulfur input to the furnace. Emission of fluorides may cause a
neighborhood situation in violation of regulations at the few plants
where ores containing unusually large amounts of fluorine are used.
The causes, nature, and means for control of this problem are
described elsewhere herein.
 Electric Furnaces

     Although the appearance of emissions from electric furnaces
 gives the impression that a tremendous amount of pollution is
 being emitted,  the actual weight of material discharged is not
 particularly large, since the small particle size of the material
 emitted enhances  its light-scattering properties,  forming a dense
 visible plume.   To comply with emission-weight regulations of
 San  Francisco  and Los Angeles, collection efficiencies  of more
 than 80 percent would be required.  Compliance with Allegheny
 County regulations would require a slightly less efficient collector.
 Compliance with equivalent opacity regulations is difficult for
 electric  furnaces.  Exit-gas grain loadings to meet such regulations
 would probably be lower than loadings permitted by emission-weight
 restrictions.   Collector efficiencies needed to meet visible emis-
 sion limitations may be greater than 85 percent.  Such efficiencies
 can  be achieved only with collectors highly effective  in collecting
 small particles, collectors such as electrostatic precipitators,
 high-energy scrubbers, and fabric filters.   Emissions of odorous
 materials,  sulfur  dioxide, and fluorides are of little concern with
 respect to air pollution from electric furnaces  and no problem in
 complying with regulations would be expected.
 Bessemer Converters

     Emissions from Bessemer converters exceed virtually all
 limitations on weight of particulate matter emitted and on visible

Air Pollution Laws and Regulations                           99
emissions.  No feasible means are available for bringing these
furnaces into compliance with limitations applicable to other
pollution sources.  In Allegheny County, existing furnaces are
exempted from compliance with regulations and construction of
new furnaces is prohibited until suitable controls are available
and used.  In other areas,  violations are allowed to continue by
issuance of a variance to the regulations, by ordinance pro-
visions exempting these furnaces from application of emission
limitations, or by simply not enforcing regulations.
Basic Oxygen Furnaces

    Basic oxygen furnaces produce a large amount of fine
particulate matter.   Emissions far exceed all regulations on
weight of emissions.  To comply with emission-weight regulations
of Los Angeles County,  the San Francisco Bay Area,  and Allegheny
County, efficiencies of 98 percent  or better would be required.
Equivalent opacity regulations will be of particular concern,  and
to produce an effluent in compliance with strict regulations of
this type will require excellent control equipment.  In all cases,
only the most efficient type of collector, properly designed and
operated,  can provide the necessary high performance.  Sulfur
dioxide, fluoride, and odorous material emissions are unlikely
to violate air pollution regulations.
Heating and Reheating Furnaces

    Inadequate firing or combustion-control equipment or improper
operation may lead to smoke emissions from heating and reheating
furnaces in excess of those allowed by regulations.  The use of
instruments and automatic combustion controls and modern com-
bustion equipment will prevent most excessive smoke emissions.
Allegheny County regulations provide a special modest exception
to general smoke emission limits so that some smoke darker than
No. 2 Ringelmann Chart may be legally emitted.  Most of these
furnaces use liquid or gaseous fuels that would not be expected to
cause emission of particulate  matter in excess of that permitted
by regulations dealing wtih the weight of the particulate matter
emitted. If a fuel containing a high percentage of sulfur is burned,
there might be a violation of sulfur dioxide emission limits.  Control,
if needed, would probably be accomplished by changing to a fuel of
lower sulfur content.  Emission of fluorides or odorous materials
would not be expected to exceed that allowed by regulations.

100                               IRON AND STEEL INDUSTRY
 Scarfing Machines

     Data concerning emissions from scarfing machines are
 adequate only for tentative indications of control equipment that
 might be required to comply with air pollution regulations.  That
 control equipment is necessary is indicated by exit-gas grain
 loadings in the range of 0. 2 to more than 0. 8 grain per scf and
 daily effluent rates ranging from 1000 to more than 7000 pounds,
 according to various reports.  Collection efficiencies required to
 meet equivalent opacity regulations would probably be around
 90 percent.  Efficiencies of around 75 percent would probably
 provide sufficient control to comply with the majority of the
 emission-weight regulations.  Emission of sulfur dioxide and
 fluorides would not be expected to exceed legal limits.
 Power Plant Boilers
     Power plant boilers are subject to emission-weight regula-
tions applicable to fuel-burning installations.  Powdered-coal-
burning plants would usually have to use collectors of about 90
percent efficiency to meet these standards.  Electrostatic pre-
cipitators, perhaps  in series with high-efficiency inertial col-
lectors,  are commonly used to  control emissions from powdered-
coal-burning plants.  Spreader  stokers would need control equip-
ment with efficiencies in the 75 to 90 percent range,  depending on
fuel-burning rate and other factors, in order to meet emission-
weight limitations.  In many cases, high-efficiency inertial
separators would be adequate,  but  in others, electrostatic pre-
cipitators may be necessary. Most other types of furnaces using
coal would not need  collectors to comply with regulations unless
extremely  high firing rates were used. Oil- and gas-fired fur-
naces do not emit particulate matter at a rate in excess of
emission-weight limits.

     Power plant boilers are also subject to regulations concerned
with visible smoke emissions.   Excessive smoke may often be pre-
vented by giving suitable attention to fuel specifications, furnace
maintenance,  and firing practices.  Installation of overfire-air,
steam, or steam-air jets would assist in preventing smoke in
some cases.   Smoke from  over-loaded plants can be eliminated
by reducing the load on the plant. Coal- and oil-fired plants may
emit excessive smoke, but gas-fired plants rarely have this prob-
lem.  Emissions from certain oil-fired plants sometimes exceed
equivalent opacity regulations even though no black smoke is em-
emitted.  The visible plume is due  at least in part to  sulfur tri-
oxide and can  be controlled by use of electrostatic precipitators
and, if available, low-surfur-content oil.

Air Pollution Laws and Regulations                          101
     Plants using coal or oil with a high sulfur content may emit
sulfur dioxide in excess of legal limits. No economically
attractive means of controlling this emission is available.
Obviously, use of lower-sulfur-content fuel would eliminate the
problem. Emission of  fluorides and odorous  materials is not
expected to exceed legal limits.

 1. American Iron and Steel Institute.  Directory of Iron and Steel
    Works of the United States and Canada.  29th ed. American
    Iron and Steel Institute, New York, N.  Y. ,  I960.  519pp.

 2. Camp, J. M.  and Francis, C.  B.   The Making,  Shaping, and
    Treating of Steel.  6th ed.  United States Steel Co. , Pittsburgh,
    Pa. ,  1951.  1584 pp.

 3. Hemeon,  W. C.  L. ,  ed.  Air pollution  problems of the steel
    industry.   Technical Coordinating Committee T-6, steel re-
    port, Section I.   JAPCA.  7:62-67.  May 1957.

 4. American Iron and Steel Institute.  The Making of Steel.
    American Iron and Steel Institute, New York, N. Y. ,  1954.
    95 pp.

 5. Hemeon, W. C.  L. ,  ed.  Air pollution  problems of the steel
    industry.   Technical  Coordinating Committee TI-6, Sections
    II through VII.  JAPCA.  10:208-18,253. June I960.

 6. Colclough, T. P. Sulfur in iron and steelmaking.  In:  Prob-
    lems  and Control of Air Pollution, F. S. Mallette, ed.
    Reinhold Publishing Corp.  , New York,  N. Y. , 1955.  pp.

 7. Purvance,  W. T. Atmospheric pollution control.  Chem.
    Eng.  Progr. 55:49-53. July 1959.

 8. Linde roof jet information. Linde Company, Development
    Laboratory, Newark, N. J.   11 pp.

 9. Hodge, A.  L.  and Arnold,  C. S.  How oxygen input affects
    open-hearth steelmaking.  J.  Metals.   12:325-30.  I960.

10. Ford  reveals new technique in OH steelmaking with oxygen.
    Iron Steel Engr.   37:145.  June I960.

11. Brief, R.  S. ,  Rose,  A. H. , and Stephan, D. G.  Properties
    and control of electric-arc steel furnace fumes.  JAPCA.
    6:220-24.   Feb.   1957.

12. Fedock, M. P.  Melting practice and refractories performance
    in basic electric-arc furnaces.   Ind. Heating.  20:135-36,  138,
    140.  Jan.  1953.

104                               IRON AND STEEL INDUSTRY
13.   Hind,  G. W.  and Hodge, A. L.  The use of oxygen-fuel gas
      burners for scrap meltdown in electric furnaces.  Presented
      at the Electric Furnace Conference,  Cleveland, Ohio.  Linde
      Company, Development Laboratory, Newark,  N.  J.

14.   Yocurn, G. and Xidis, L.  Blowing methods in steelmaking.
      Iron Steel Engr.  37:101-07.   Sept. I960.

15.   Vajda, S.  Symposium on basic oxygen furnaces.  Equipment
      layout.  Iron  Steel Engr.  37:73-78.  Oct.  I960.

16.   Smith, D. W. Symposium on basic oxygen furnaces.  Opera-
      tions.  Iron Steel Engr.  37:85-89. • Oct.  I960.

17.   The promise  of the sixties.  AJSE Forum.  Steel.  147:154,  157,
      160,  162,  166, 168.  Sept. 19,  I960.

18.   Jaklitsch, J.  J. , Jr. , ed. Briefing the record   Sintering.
      Mech.  Eng.   79:35-36.  Jan.  1957.

19.   New 5000  tons per day sintering plant at U. S. Steel's Ohio
      works.  Ind.  Heating. 26:86-94.  Jan. 1959.

20.   Morgan, M.  F.  and Collison, W.  H.   Symposium on sinter
      plants-discussion.  Iron Steel Engr.   36:101-22.  June 1959.

21.   McMahan, J.  S.  The use of self-fluxing sinter.   Blast Fur-
      nace Steel Plant.  47:51-54.  Jan. 1959.

22.   DeCarlo,  J.  A.  and Otero, M. M.  Coke plants in the
      United States on December 31,  1958.  Information Circular
      7934,  U. S. Bureau Mines, 1959.  20pp.

23.   Metal Market Bureau.  Beehive coke oven use up; revival
      won't last. Reprinted in Smog News (ASME).  No. 160:19.
      May 15, I960.

24.   DeCarlo,  J.   A. and Ryan, E. E.  Beehive-coke plants in
      the United States that reported to Bureau of Mines in 1956.
      Information Circular 7820, U.  S.  Bureau Mines,  Jan.
      1958.   17 pp.

25.   Raleigh, W.   A. , Jr.  Outlook - steel and coal. Coal Age.
      65:76-79.  Sept. I960.

References                                                 105
26.   U.  S.  Bureau of the Census.  Statistical Abstract of the
      United States,  1958.   79th ed.  U. S. Dept. Commerce,
      Washington, D. C. ,  1958.  1040 pp.

27.   Industry - new era for steel.   Time,  April 7, 1961.  pp.
      90,  92, 94.

28.   Parker, C. M.   American Iron and Steel Institute.
      Personal communication with J. J. Schueneman, U.  S.
      Public Health Service.  Feb.  I960.

29.   Davis, D. O.  The oxygen steelmaking process.  Blast
      Furnace Steel Plant.  44:44-48, 108.  Jan. 1956.

30.   Jaklitsch, J. J. ,  Jr.  loc.  cit.

31.   O'Mara, R.  F.   Dust and fume problems in the steel
      industry.  Air Pollution Symposium.  Iron Steel Engr.
      30:100-06.  Oct.  1953.

32.   Meadley, A. H. ,  Colviii, J. G. ,  and Gamble, H.  J.
      Mitigation of air pollution in sinter plants.  In:  Air and
      Water Pollution in the Iron and Steel Industry,  Special
      Report No. 61.   Iron  and Steel  Institute, London,  Eng-
      land,  1958.  pp.  39-48.

33.   Granville, R. A.  The capital costs of some waste-gas
      cleaning plants for use in iron and steel  works. In:  Air
      and Water Pollution in the Iron and Steel Industry, Special
      Report No. 61.   Iron  and Steel  Institute, London,  England,
      1958.  pp. 23-30.

34.   Brooks, S. H. and Calvert, W. J.  External pollution from
      an iron and steel works and measures towards itv reduction.
      In:  Air and Water Pollution in the Iron and Steel Industry,
      Special Report No. 61.  Iron and Steel Institute,  London,
      England, 1958.   pp.  5-15.

35.   Corzilius, W.  R.  Sintering plant expands blast furnace and
      open hearth capacity.  Blast  Furnace Steel Plant.  47:44-50.
      Jan. 1959.

36.   Semrau,  K. T.  Emission of fluorides from industrial pro-
      cesses - a review.  JAPCA.  7:92-108.  Aug.  1957.

106                                IRON AND STEEL INDUSTRY
37.   Gallear,  C. A.  Reduced turbulence boosts dust collector
      efficiency.   Iron Age.   176:98-100.  Sept. 1, 1955.

38.   Brandt, A. D.  Air pollution control in the Bethlehem Steel
      Company.  Air Repair.  3:167-69.  Feb. 1954.

39.   Specht, S.  E.  and Sickles, R.  W.  New uses of electrical
      precipitation for control of atmospheric pollution.  Air
      Repair.  4:137-40, 170.  Nov. 1954.

40.   Thring, M. W. and Sarjant,  R. J.  Dust problems  of the
      iron and  steel industry.   Iron and Coal Trades  Rev.  174:
      731-34.  Mar.  29,  1957.

41.   Kurtz, J. K.  Recovery and utilization of sulfur from coke
      oven gas.  In:  Problems and Control of Air Pollution, F. S.
      Mallette, ed.  Reinhold Publishing Corp. , New York,  N. Y. ,
      1955.   pp.  215-21.

42.   Beryushev, K. G.  Limits of allowable concentration of
      phenol in the atmospheric air of inhabited localities.  In:
      Limits of Allowable Concentrations of Atmospheric Pollu-
      tants.  Book 2-1955.  V.  A.  Ryazanov, editor, and B. S.
      Levine, translator.  (Publication No. 59-21174).  U. S.
      Dept.  of  Commerce,  Office of Technical Services,  Wash-
      ington, D.  C.   pp.  65-68.

43.   Radcliffe,  J. C. and Delhey,  W. F.  An industrial  air pol-
      lution program.  In:  Proc. 45th Annual Meeting, Air Pol-
      lution and Smoke Prevention Assoc. of America, Cleve-
      land,  Ohio, June 9-12, 1952.   pp.  40-45.

44.   Kemp, W.  E.   Product fallout - a serious corrosion problem.
      Ind. Eng. Chem.  51:75A-76A. July 1959.

45.   Mallette, F. S.  A new frontier: air pollution control.
      (James Clayton Paper).   Inst.  Mech.  Engrs. (London),
      Proc.  l68(No.  22):595-628.  Apr.  1954.

46.   Bishop, C. A.  Metallurgical furnace stacks.  AIHA Quart.
      11:34-39.  Mar.  1950.

47.   Howell, G. A.  Air pollution control in the steel industry.
      Air Repair.  3:163-66.  Feb.  1954.

References                                                  107
48.  Hosey, A. D. and Neviiis,  F.  Evaluation of plant effluents.
     In: Air Pollution in Donor a, Pa. - Preliminary Report.
     Public Health Bull.  No. 306,  1949.  pp.  86-109.

49.  Basse, B.  Gases cleaned by use of scrubbers.  Blast
     Furnace Steel Plant.  44:1307-10.  Nov.  1956.

50.  Wurts, T. C.  Industrial sources of air pollution - metal-
     lurgical.  In: Proc. National  Conference on Air Pollution,
     Washington, D.  C. , Nov.  18-20, 1958.  PHS Publ. 654,
     1959.  pp. 161-64.

51.  Bishop, C.  A.  et al.  Cleaning ferromanganese blast fur-
     nace gas.  Iron Steel Engr.  28:134-36.  Aug. 1951.

52.  Allen, G. L. , Viets, F. H. , and Me Cab e, L. C.  Control
     of metallurgical and mineral dusts and fumes  in Los
     Angeles County, Calif.  Information Circular 7627, U. _S.
     Bureau Mines,  Apr.  1952.  79pp.

53.  Billings,  C. E.  , Small, W. D. ,  and Silverman, L.  Pilot-
     plant studies of a continuous slag-wool filter for an open-
     hearth fume.  JAPCA.   5:159-66.  Nov.  1955.

54.  Open hearth furnace fumes  - their formation and resulting
     problems.  Iron and Coal Trades Rev.  173:1311-17.  Nov.
     30, 1956.

55.  Bishop, C.  A.   Some experiences with air pollution abate-
     ment in the steel industry.  In: Proc.  45th Annual Meeting,
     Air Pollution and Smoke Prevention Assoc.  of America,
     Cleveland,  Ohio, June  9-12, 1952.  pp. 32-37.

56.  Pring, R.  T. Filtration of hot gases.  Air Repair.  4:40-45.
     May 1954.

57.  Bishop, C.  A. ,  Campbell,  W. W. , Hunter,  D. L. , and
     Lightner, M. W.  Successful cleaning  of open-hearth ex-
     haust gas with a high-energy venturi scrubber.  JAPCA.
     11:83-87.  Feb.  1961.

58.  Smith, J. H. , Rounds, G.  L. , and Matoi, H. J.  Some
     problems encountered in sampling open hearth stacks.
     Air Repair.  3:35-40.  Aug. 1953.

108                               IRON AND STEEL INDUSTRY
59.   Akerlow,  E.  V.  Modification to the Fontana open hearth
      precipitators.  JAPCA.  7:39-43.  May 1957.

60.   Strauss, W.  Cleaning waste gases  from open-hearth
      steel processes.  Chem. and Proc.  Eng.  41:339-43,
      351.  Aug. 1960.

61.   Speer, E. B.  Operation of electrostatic precipitators on
      O.H.  furnaces  at Fairless Works.  In:  Air and Water
      Pollution in the Iron and Steel Industry, Special Report
      No. 61.  Iron and Steel Institute,  London, England,  1958.
      pp. 67-74.

62.   Silverman, L.  Cleaning of open-hearth stack gases.
      Blast Furnace Steel Plant.  43:735-38.  July 1955.

63.   Great Britain.  Ministry of Housing and Local Government.
      85th Annual Report on Alkali etc.  Works by the Chief In-
      spectors,  1948.  H. M. Stationery Office,  London, Eng-
      land,  1949.  43 pp.

64.   Ibid.  86th Annual Report,  1949.  41 pp.

65.   Holden, C.  Factors affecting fuming in open hearth furnaces,
      J.  Iron Steel  Inst. (London).  193:93-102.  Oct. 1959.

66.   Knaggs, K. and Slater, J.  M.  Some factors affecting fume
      evolution from molten steel during oxygen injection.  J.
      Iron Steel Inst. (London).  193:211-16.  Nov. 1959.

67.   Silverman, L.  Technical aspects of high temperature gas
      cleaning for steel making processes.  Air Repair.  4:189-
      96, 231.  Feb.  1955.

68.   Lindstrom, C.  A.  U. S. Public Health Service, Division
      of  Air Pollution.  Personal communication •with M. D.
      High,  U. S. Public Health Service,  Feb. 1961.

69.   Vajda, S.  and Dreher, G. M.  Open hearth dust control.
      In:  Proc. 45th Annual Meeting, Air Pollution and  Smoke
      Prevention Assoc.  of America,  Cleveland, Ohio, June 9-
      12, 1952.  pp. 26-32.

70.   Richardson, H.  L.  Scope of the furnace fume control
      problem.  Iron Steel Engr.  33:105-11.  Jan. 1956.

References                                                 109
71.   Turner, N. H.  Use of oxygen in modified tilting furnace -
      dust removal plant.  J.  Iron Steel Inst. (London).  190:20-
      22.  Sept. 1958.

72.   Dracco receives order.  JAPCA.  10:474.  Dec. I960.

73.   Steel Companies and Air Pollution Control Equipment
      Manufacturers.  Personal communications with J.  J.
      Schueneman.  Feb. and Mar. 1961.

74.   Coulter, R. S.  Bethlehem Pacific  Coast Steel Corporation.
      Personal communication.  Apr.  24, 1956.

75.   Coulter, R. S.  Smoke, dust, fumes closely controlled
      in electric furnaces.  Iron Age.   173:107-10.  Jan.  14,

76.   Los Angeles County Air Pollution Control District.
      Unpublished data.  Los Angeles,  California.  1950-1951.

77.   Kane, J.  M. and Sloan, R. V.  Fume control-electric
      melting furnaces.  Am. Foundryman.  18:33-35. Nov.

78.   Pier,  H.  M. and Baumgardner,  H. S.  Research-Cottrell,
      Inc.  Personal communication.   Apr. 17, 1956.

79.   Faist, C. A.  Remarks.  Electric Furnace Steel, Proc.
      (Am. Inst. Mining  Met. Engrs.).  11:160-61.  1953.

80.   Faist, C. A.  Burnside Steel Foundry Co.  Personal
      communication.  Apr. 25,  1956.

81.   Anderson, E. F.  There are indirect benefits from the
      furnace fume  collector.   Foundry.  83:152-53.  Sept.  1955.

82.   Everling, W.  O. Extent to which available control technique
      have been utilized by industry - iron and steel.  In:  Proc.
      National Conference on Air Pollution, Washington,  D. C. ,
      Nov.  18-20,  1958.   PHS Publ. 654.  1959.  pp.  339-43.

110                               IRON AND STEEL INDUSTRY
83.   Los Angeles Air Pollution Control District.  Emissions
      of oxides of nitrogen from stationary sources in Los
      Angeles County.  Report No.  2, Oxides of nitrogen
      emitted by small sources.  A Joint District,  Federal,
      State, and Industry Project.  Los Angeles County Air
      Pollution Control District,  Los Angeles, California,
      Sept.  1960.  73 pp.

84.   Lewis,  W.  E.  Remarks.  Electric Furnace  Steel,  Proc.
      (Am.  Inst.  Mining Met. Engrs.).   10:  Dec.  1952.

85.   Erickson, E.  O.  Dust control of electric foundries in
      Los Angeles area.   Electric Furnace Steel, Proc.
      (Am.  Inst.  Mining Met. Engrs.).   11:156-60.  1953.

86.   Assel, W.  J.  Remarks.  Electric Furnace Steel, Proc.
      (Am.  Inst.  Mining Met. Engrs.).   10:47.  Dec.  1952.

87.   Kane, J. M.   The application of local exhaust ventilation
      to electric  melting furnaces.  Trans.  Am.  Foundrymen's
      Soc.   52:1351-56.   1944.

88.   Dok, H.  Smog control in the steel industry.  JAPCA.
      5:23-26. May 1955.

89.   Dok, H.  Smog control in the foundry.  Am.  Foundryman.
      26:46-49.   Dec.  1954.

90.   Bloomfield, B. D.   An appraisal of air pollution control
      installations.  AIHA Quart.  17:434-44.  Dec. 1956.

91.   American Foundrymen's Society.   Control of emissions from
      metal melting operations.  American Foundrymen's Society,
      Des Plaines,  111., 1955.   26pp.

92.   Anderson, E.  F.  Wheelabrater Corp.  Personal communi-
      cation.  Apr.  30, 1956.

93.   Sarjant, R. J.  Steelmaking  processes in relation to at-
      mospheric pollution.  Iron and Steel (London).  32:185-90.
      May 1959.

94.   Dehne, W.  Possibilities of removing dust from brown
      converter waste gas.  Iron and Coal Trades Rev.  175:477-
      82.  Aug. 30,  1957.

References                                                  111
 95.   Parker,  C. M.  American Iron and Steel Institute.
      Personal communication with V. G.  MacKenzie, U. S.
      Public Health Service.   Apr.  11,  1961.

 96.   Gaw, R.  G.  Symposium on basic oxygen furnaces.  Gas
      cleaning.  Iron Steel Engr.  37:81-85.  Oct.  I960.

 97.   Vacek, A.  and Schertler,  A.  Waste-gas cleaning systems
      at oxygen steel plants.   In: Air and Water Pollution in the
      Iron and Steel Industry, Special Report No. 61.  Iron and
      Steel Institute, London, England, 1958.  pp. 82-89.

 98.   Dust control: dust settling chambers keep Canton's air
      clean. Iron Age.   175:119-20.  June 23, 1955.

 99.   Hemeon, W. C. L. ,  Haines,  G.  F. ,  and Ide, H. M.
      Determination of haze and smoke concentrations by filter
      paper samplers.   Air Repair.  3:22-28.  Aug.  1953.

100.   Munroe, W. A.  Statewide air pollution survey  - smoke
      index. Public Health News, New Jersey State Depart-
      ment of Health,  Trenton, N. J.  39:227-42.  Aug.  1958.

101.   Tabor, E.  C.  and Meeker, J.  E.  Effects  of the 1956
      steel strike on air pollution levels in several communities.
      In:  Proc.  51st Annual Meeting APCA,  Philadelphia, Pa. ,
      May 25-29, 1958.  58-24.  20pp.

102.   Dyktor,  H. G.  and Goldston,  L.  N.  The effects of the
      steel industry on atmospheric pollution in the Cleveland area.
      Industrial Hygiene Newsletter.  11:22-23.  Feb. 1951.

103.   Consumers Union of U.  S. , Inc.   Pollution in the air we
      breathe.  Consumer  Reports.  25:400-07.  Aug. I960.

104.   Paulus,  H. J. , Hosey,  A. D. , Crothers, R. B. , and
      Byers, D.  H.  Investigation of atmospheric contaminants.
      In: Air Pollution in Donora, Pa.  - Preliminary Report.
      Public Health Bull. No. 306, 1949.  pp. 81-125.

105.   International Joint Commission.  Technical Advisory Board
      on Air Pollution,  United States Section.  Report of results
      of sampling the atmosphere in the Detroit River area dur-
      ing 1951.  Mar.  1,  1953.  55 pp.

112                               IRON AND STEEL INDUSTRY
106.  International Joint Commission.  Technical Advisory
      Board on Air Pollution, United States Section.  Report
      on  1953 environmental studies in the Detroit River area.
      Robert A. Taft Sanitary Engineering Center,  Cincinnati,
      Ohio,  Apr. 1,  1955.   56pp.

107.  Jones, H. G.  andDavies, J. T.  External dust deposition
      and sulfur emission.  In:  Air and Water Pollution in the
      Iron and Steel Industry, Special Report No. 61.  Iron and
      Steel Institute, London, England,  1958.  pp.  16-23.

108.  Kirste, H.  Dust measurements to determine the sources
      of emissions damaging crops.  In:  Proc. International
      Clean Air Conference, London,  Oct. 20-23,  1959.
      National Society for Clean Air, London, England, I960.
      pp. 154-56.

109.  Leclerc,  E.  On the determination of the degree of atmos-
      pheric pollution of city atmosphere.  Rev.  Universelle
      Mines (Belgium).   9(Ser. 9):801-08.  Dec. 1953.

110.  Cholak, J. , Schafer,  L.  J. ,  and Yeager, D.  Air pollution
      near an iron slag  processing plant.  AIHA Quart.  15:220-25,
      Sept. 1954.

111.  Dickun, P. P  and Nikberg, I. I.  A study of air pollution
      with 3, 4-benzpyrene in the vicinity of an old pitch-coke
      plant.  In: U.S.S.R.  Literature on Air Pollution and Re-
      lated Occupational Diseases   Vol. 2.  A survey and trans-
      lation by  B. S. Levine.  (Publication No.  60-21188). U. S.
      Dept. of Commerce, Office of Technical Services, Wash-
      ington 25, D.  C. , Mar. I960.  pp. 135-40.

112.  Sawicki,  E. et al.  Benzo(a)pyrene content of the air of
      American communities.  AIHA J.  21:443-51.  Dec.  I960.

113.  U.  S. Weather Bureau.  Meteorology and Atomic Energy.
      AECU-3066. U.  S. Government Printing Office, Wash-
      ington, D. C. ,  July 1955.   169 pp.

114.  Firket, J. et al.   The causes of the symptoms found in the
      Meuse Valley during the fogs of December  1930. Bull.
      acad. roy. med. Belg.  11:683-741.   1931.   (Abstract 18
      in reference 115)

References                                                 113
115.   Cincinnati.  University.  Kettering Laboratory of Applied
       Physiology.  Annotated Bibliography.  The Effects of At-
       mospheric Pollution on the Health of Man.  University of
       Cincinnati, Cincinnati, Ohio,  1957.  481 pp.

116.   Heimann, H., David, W. D. ,  and Sitgreaves, R.  The
       acute Illness.  In:  Air Pollution in Donora, Pa.  - Pre-
       liminary Report.  Public Health Bull.  No. 306, 1949.
       pp.  11-54.

117.   U.  S. Public Health Service.  Air  pollution research con-
       ducted and supported by the Public Health Service, fiscal
       year I960.  Project No. APM-542.2, 542.3, 542.4  - Air
       pollution and community health in Nashville - a prototype
       study.  Investigators:  Zeidberg,  L. D. et al.  U. S.
       Public Health Service^ Washington 25,  D. C.  Aug. I960.

118.   Zeidberg,  L. D. , Prindle, R. A.,  and Landau, E.  The
       Nashville air pollution  study.  I. Sulfur dioxide and bronchial
       asthma.  A preliminary report.  Amer. Rev. Respirat.
       Diseases.  84:489-503.  Oct.  1961.

119.   Dohan,  F.  C. and Taylor, E.  W.  Air pollution and
       respiratory disease. A preliminary report. Am. J. Med.
       Sci. 240:337-39.  Sept. I960.

120.   Spicer, W.  et al.  Pulmonary function studies in Baltimore.
       To  be presented at the American Thoracic Society Annual
       Meeting, Cincinnati, Ohio.  May 22, 1961.

121.   Ciocco,  A.  and Thompson, D. J.  A follow-up of Donora
       ten years after: methodology and findings.  Am. J. Public
      Health.   51:155-64.   Feb.  1961.

122.   Skvortsova, N.  N.   Pollution of atmospheric air with carbon
       monoxide in the vicinity of ferro-metallurgical plants.  In:
       U.S.S. R. Literature on Air Pollution and Related Occupa-
       tional Diseases - Vol.  2.  pp. 204-12.  (See reference 111).

114                               IRON AND STEEL INDUSTRY
123.   U.  S.  Public Health Service.  Air pollution research con-
       ducted and supported by the Public Health Service, fiscal
       year I960.  Project No. APM-653. 1 - Relationship of
       sudden changes in steel production to changes in mortality.
       U.  S.  Public Health Service, Washington 25, D.  C.  Aug.

124.   Hallgren, W. , Karlsson, N. ,  and Wramby, G.   Molybdenum
       poisoning ("molybdenosis")  in cattle in Sweden.  Nord. Vet.
       Med.  6:469-80.  1954.  (Abstract 155 in reference 115).

125.   Murray, M.  M.  and Wilson, D. C.  Fluorine hazards with
       special reference to some social consequences of indus-
       trial processes.  Lancet.  251:821-24.  1946.  (Abstract 48
       in reference 115).

126.   American Conference of Governmental Industrial Hygienists.
       Threshold limit values for  I960.   Arch.  Environ. Health.
       1:140-44.  Aug.  I960.

127.   Elkins, H. B.   Maximum acceptable concentrations, a com-
       parison in Russia and the United States.   Arch. Environ.
       Health.  2:45-49.  Jan.  1961.

128.   The maximum allowable concentrations of an air pollutant
       to be used for hygienic evaluation of the  air in populated
       areas.  In:  U.S.S.R. Literature on Air Pollution and Re-
       lated Occupational Diseases   Vol. 2.  p. 260.  (See refer-
       ence 111).

129.   Evans, E.  E.  An X-ray study of industrial gases on the
       human lung.  Radiology.  34:411-24.   1940.  (Abstract 468
       in reference 115).

130.   Kehoe, R.  A. et al.  Effects of prolonged exposure to sulfur
       dioxide.  J. Ind. Hyg.  14:159-73.  1932.  (Abstract 446 in
       reference 115).

131.   Myers, J.  A., ed. Diseases of the Chest Including the
       Heart.  Charles C. Thomas  Co. ,  Springfield, 111. , 1959.
       pp.  520-29.

References                                                115
132.   Anderson,  W. A. D.  Pathology.  C. V. Mosby Co. , St.
      Louis, Mo.,  1948.  pp.  164,  167.

13.    lleason, M.  N. , Gosselin, R. E. , and Hodge, H. C.
      Clinical Toxicology of Commercial Products.  Williams
      and Wilkins Co. , Baltimore, Md. , 1957.  pp.  160,  169.

134.   Patty, F.  A.   Industrial Hygiene and Toxicology.  Vol.  2.
      Interscience Publishers, Inc., New York, N. Y. , 1949.
      pp. 607-09.

135.   Vigliani, E. C. and Zurlo, N.  Experiences of the Clinica
      del Lavoro with some maximum allowable concentrations
      (MAK) of industrial poisons.   Arch. Gewerbepathol.
      Gewerbehyg.   13:528-34.  1955.  (Abstract 594 in refer-
      ence 115).

136.   Vigdorchik, N. A. et al.  The symptomatology of chronic
      poisoning with oxides of nitrogen.  J. Ind. Hyg. Toxicol.
      19:469-73.  Nov.  1937.

137.   Grollman,  A.  Pharmacology and Therapeutics.  4th ed.
      Lea and Febiger, Philadelphia, Pa., I960.  pp. 709, 978.

138.   Novikov, Yu.  V.  Effect of small benzene concentrations on
      higher nervous activity  of animals in chronic experiments.
      In:  U.S.S.R.  Literature on Air Pollution and Related Occu-
      pational Diseases.  Vol. 2.  pp.  185-91.  (See reference 111).

139.   Beck, H. G.   The clinical manifestations of chronic carbon
      monoxide poisoning.  Ann. Clin.  Med.  5:1088-96.  1927.
      (Abstract 431 in reference li5).

140.   Goldsmith, J. R.  Recurrent carbon monoxide exposure.
      Presented at Western Industrial Hygiene Conference, San
      Francisco, Calif.  Oct.  7, I960.

141.   U. S. Public  Health Service.  Air pollution research conducted
      and supported by the Public Health Service,  fiscal year  1960.
      Project No. APM-433. 1 - Epidemiologic studies on the  car-
      bon monoxide content of blood from specific population groups.
      U. S. Public  Health Service, Washington 25, D. C.  Aug. I960.

116                               IRON AND STEEL INDUSTRY
142.   Roebber,  H. M.  In:  Fourth Saranac Laboratory Symposium
      on Silicosis.  The Edward C.  Trudeau Foundation, Saranac
      Lake,  N.  Y., 1939.  p.  241.

143.   Ivanova, M.  G.  and Ostrovskaya, I. S.  The effect of
      aluminum dust on the animal organism.  In: U.S.S.R.
      Literature on Air Pollution and Related Occupational
      Diseases   Vol. 2.  pp.  167-74. (See reference 111).

144.   King, E. J. et al.  The effect of aluminum and of aluminum
      containing  5 percent of quartz in the lungs of rats.  J.
      Pathol. Bacteriol.  (London).  75:429-34.  Apr.  1958.

145.   Mitchell,  J. ,  Manning, G.  B. ,  Molyneux, M. , and Lane,
      R. E.  Pulmonary fibrosis in workers exposed to finely
      powdered aluminum.  Brit. J.  Ind. Med.  18:10-20.  Jan.

146.   Schepers,  G.  W. H. and Delahant, A. B.  Prevention of
      "silica shock" by aluminum.  Arch. Environ. Health.
      2:9-15. Jan.  1961.

147.   Sander, O.  A. Foundry workers' pneumoconiosis.  Arch.
      Ind.  Hyg.  and Occupational Med.  10:512-21.   Dec. 1954.

148.   Dechoux,  J.  Pneumoconiosis in iron ore miners in the
      Lorraine Basin.  Nancy,  Imprimerie Georges Thomas,
      1954.  170 pp. (Abstract 570 in reference 115).

149.   Gross, P., Westrick, M.  L. ,  and McNerney, J. M.  The
      pulmonary response to blast furnace stack gas.  ATHA  J.
      20:197-204.  June 1959.

150.   Sadoul, P. et al.  Pneumoconioses of iron miners and  their
      functional  consequences.  Minerva Med. (Turin).  49:4689-
      93.  Dec.   12, 1958.

151.   Hueper, W. C.  Role of  occupational and environmental air
      pollutants in production of respiratory cancer.  Arch.  Pathol.
      63:427-50.  May  1957.

152.   Levin, M.  L.  , Kraus, A.  S. , Goldberg,  I. D. ,  and Ger-
      hardt,  P.  R.  Problems  in the study of occupations and
      smoking in relation to lung cancer.  Cancer.  8:932-36.
      Sept. -Oct. 1955.

References                                                 117
153.   McLaughlin, A. I.  G. and Harding, H. E.  Pneumoconiosis
      and other causes of death in iron and steel foundry -workers.
      A. M. A.  Arch. Ind. Health.   14:350-78.  Sept. 1956.

154.   Guthred, K.  G. and Flynn, M.  J.  Some aspects of chronic
      respiratory diseases in coal miners in New South Wales,
      Australia.  Diseases of the Chest.  37:390-99.  Apr. I960.

155.   Belknap, E.  L. Saranac Laboratory Symposium, National
      Tuberculosis Association.  The Edward C. Trudeau Founda-
      tion,  Saranac Lake,  N. Y. , 1941.  p.  123.

156.   U. S. Public Health Service.  Air pollution research con-
      ducted and supported by the Public Health Service, fiscal
      year  I960.  Project No. APM-432.2 - Geographic and
      temporal variations in lead levels of blood and urine.
      U. S. Public Health Service,  Washington 25,  D.  C.  Aug.

157.   Elstad, D.  Manganese-containing dust as a predisposing
      factor for pneumonia epidemics in an industrial  region.
      Nord. Med.  3:2527-33.  1939.   (Abstract 36 in  reference

158.   McLaughlin,  A. I.  G. and Harding,  H. E.  The  causes  of
      death in iron and steel workers  (non-foundry).  Brit. J.
      Ind.  Med.  18:33-40.  Jan. 1961.

159.   Sepke, G.  Silicosis following sublimial exposure to dust.
      Tuberkulosearzt (Stuttgart).  12:638-42.  Oct. 1958.

160.   Amdur, M. O.  The physiological response of guinea pigs
      to atmospheric pollutants.  Intern. J. Air Pollution.  1:170-
      83.  Jan.  1959.

161.   U. S. Public Health Service.  Air pollution research conducted
      and supported by the Public Health Service, fiscal year I960.
      Project No.  APM-411. 1 -  Measurement of pulmonary physiological
      responses  to air pollutants.  Investigators:  Whittenberger, J. L.
      and Roberts, A.  U.  S. Public Health Service, Washington 25,
      D. C. Aug.  I960.

162.   Hueper,  W.  C. et al.  Carcinogenic bioassays on air pollu-
      tants. Arch. Pathol.  74:89-116.  Aug.  1962.

118                               IRON AND STEEL INDUSTRY
163.  Sawicki, E.  et al.  loc.  cit.

164.  Barkley,  J.  F.  Fundamentals of smoke abatement.  Infor-
      mation Circular 7588,  U. S. Bureau Mines, Dec.  1950.

165.  McCabe,  L. C. , Rose, A. H. , Hamming, W.  J. ,  and
      Viets, F.  H.  Dust and fume standards.  Ind. Eng. Chem.
      41:2388-90.  Nov.  1949.

166.  U.  S.  Dept.  of Health, Education and Welfare, Public Health
      Service.   Unpublished  data.  Robert A. Taft Sanitary Engineer-
      ing Center,  Cincinnati, Ohio.

167.  Robert A. Taft Sanitary Engineering Center.  Air  pollution
      measurements of the national air sampling network, 1953-
      1957.  PHS Publ. 637.  1958.  259 pp.

168.  Mills, J.  L. et al.  Emissions of oxides  of nitrogen from
      stationary sources in Los Angeles  County.  Report No.  3,
      Oxides of nitrogen emitted by medium and large sources.
      Los Angeles County Air Pollution Control District, Los
      Angeles,  California, Apr. 1961.   51 pp.

169.  Dagan, B. N.  The cleaning of open hearth waste gases.
      Unpublished paper, Kaiser Steel Corporation,  Fontana,

170.  Dust removal in oxygen steelmaking.  J.  Metals.   12:554-57.
      July I960.

Size, Location,  and Age of Iron and Steel Works in the United
           States, and Other Historical Data.
GPO 8E3-573-9

Table Al.  BLAST FURNACE CAPACITY January 1,  1960 (Reference 1)
New York
West Virginia
Number of
Number of
Annual capacity
(net), tons
18, 734, 500
7, 955, 200
5, 947, 000
5, 817, 440
5, 480, 000
5, 290; 250
2, 646, 000
1, 997, 800
1, 804, 200
1, 058, 000
925, 000
922, 400
696, 000
217, 740
195, 000
128, 000
96, 520, 630a
Percent of
total capacity
2. 1
a Includes 877, 500 tons ferroalloys capacity.

          January 1, 1960 (Reference 1)
West Virginia
New York
Number of
Number of
5, 365a
15, 208a
Annual capacity
(net), tons
20, 672, 100a
9, 819, 000
6, 149,900
4, 174, 000
3, 856, 000
3, 629, 000
3, 460, 000
2, 638, 000
1, 502, 000
1, 434, 600
985, 500
893, 600
832, 000
664, 000
251, 500
72, 309, 700a
Percent of
total capacity
28. 5
8. 5
a Includes four coke plants with
  877, 100 net tons per year.
1, 392 beehive ovens and a capacity of

Table A3.  ANNUAL OPEN HEARTH CAPACITY January 1, I960 (Reference 1)
New York
West Virginia
New Jersey
Rhode Island
Number of
Number or
Annual Capacity
(net), tons
22, 688, 280
18, 339, 000
9, 842, 000
7, 864, 000
5, 420, 000
4, 786, 000
3, 300, 000
2, 727, 500
2, 300, 000
1, 825, 000
1, 800, 000
1, 363, 000
973, 000
506, 500
420, 000
235, 000
93, 000
126, 621, 630
Percent of
total capacity
1. 1
0. 1

Table A4.  ELECTRIC FURNACE CAPACITY  January 1, 1960 (Reference 1)
New York
West Virginia
New Jersey
Number of
Number of
Annual capacity
(net), tons
2, 400, 400
1, 178, 600
699, 080
670, 020
628, 000
466, 190
420, 000
140, 000
84, 000
60, 000
51, 000
45, 000
40, 000
38, 000
Percent of
total capacity
20. 1
16. 6
8. 1
1. 6
0. 3
0. 1
100. 0

Table A5.  BESSEMER FURNACE CAPACITY January 1, 1960 (Reference 1)
West Virginia
Number of
Number of
Annual capacity
(net), tons
2, 552, 000
408, 000
336, 000
100, 000
3, 396, 000
Percent of
total capacity
75. 1

a Bessemer converters used in melting charge for open hearth furnaces.
         (Reference 1)
Number of
Number of
Annual capacity
(net), tons
1, 440, 000
880, 000
452, 000
4, 157, 400
Percent of
total capacity
21. 1

          ABANDONED IN THE UNITED STATES,  1900-1960 (Reference 22)
In Existence end of year
Number of

Number of
13, 816a
Annual capacity
(net), tons

67, 909, 300
Number of


Number of



aFrom reference 16.

           December 31, 1958 (Reference 22)
Age, yr
Under 5
From 5 to 10
From 10 to 15
From 15 to 20
From 20 to 25
From 25 to 30
From 30 to 35
From 35 to 40
40 and over
Number of
Percent of
Annual capacity
(net), tons
12, 453, 300
19, 650, 000
9, 838, 500
5, 973, 900
750, 600
1, 590, 200
6, 180, 700
Percent of
27. 5
         1920 to 1957 (Reference 26)
Number of
(Jan. 1)
1, OOO's of
short tons
56, 249
57, 855
55, 724
71, 560a
82, 001
83, 971
85, 485
86, 818
91, 000
96, 521
Production, 1, OOO's of short tons
41, 179
35, 338
46, 979
65, 440
77, 790
75, 960
79, 339
Pig iron
40, 593
34, 743
46, 072
64, 587
57, 966
76, 857
75, 068
78, 375


percent of

a Average annual capacity as of January 1 and July 1.

b From 27th edition of Reference 16.

c From Reference 16.

           TYPE  OF FURNACE,  1920 - 1957 (Reference 26)
(Jan. 1)
1,000's of
short tons
60, 220
65, 962
94, 233
99, 983C
104, 230
140, 743
148, 571e
Production, 1, OOO's of short tons
46, 183
44, 591
38, 184
66, 983
66, 603
88, 640
77, 978
96, 836
105, 200
Open Hearth
38, 587
34, 401
61, 573
76, 874
70, 249
86, 263
82, 846
100, 474
101, 658
34, 005
60, 883
82, 143
9, 14Sd
8, 582d
percent of
a Less than 500 short tons.

  Include with electric.

c Average of January 1 and July 1.

  Includes steel made by the basic oxygen process.

** From reference 16.

                    (Reference 15)
Company and plant location
Dominion Foundry & Steel
Hamilton, Ontario
Trenton, Mich.

Jones & Laughlin
Aliquippa, Pa.
Sault Ste-Marie, Ontario
Fontana, Cal.
Chicago, 111.
Colorado Fuel & Iron
Pueblo, Colo.
Jones & Laughlin
Cleveland, Ohio
Startup date
Oct. 1954
Dec. 1956
Dec. 1954
Apr. 1958
Mar. 1960

Nov. 1957

Nov. 1958

Dec. 1958

Jan. 1959

1st qtr. 1961

2ndqtr. 1961






(net), tons








81 to 108





6,000 to 8,000


3, 000 to 4, 000

12, 000 to 20, 000
GPO  823—573—1 1

BIBLIOGRAPHIC:  Schueneman, Jean J.,  High, M. D.,
      1963.  129 pp.  (limited distribution).

ABSTRACT:  This report is a summary of published and
      other information on the air pollution aspects of the
      iron and steel industry, including coke plants inci-
      dent thereto.  Processes, equipment,  and raw
      materials are briefly described.   The  magnitude
      and location of plants and process  trends are noted.
      Air pollutant emissions and means for  their control
      are discussed in detail, with respect to sintering;
      coke production; blast furnaces; open hearth, Bes-
      semer, electric, and basic oxygen steel-making
      furnaces; and other operations.  The effects of
      pollutants on community air quality are described,
      and knowledge of health aspects of pollutants is
      summarized.  Laws regulating pollutant emissions
      are given, and control equipment and measures
      needed to comply with certain laws are listed.


    Air Pollution






BIBLIOGRAPHIC:  Schueneman, Jean J., High, M. D.,
      19fi3.  129 pp.   (limited distribution).

ABSTRACT:  This report is a summary of published and
      other information on the air pollution aspects of the
      iron and steel industry, including coke plants inci-
      dent thereto.  Processes, equipment, and raw
      materials are briefly described.  The magnitude
      and location of plants and process trends are noted.
      Air pollutant emissions and means for their control
      are discussed in detail, with respect to sintering;
      coke production; blast furnaces; open hearth,  Bes-
      semer, electric, and basic oxygen steel-making
      furnaces; and other operations.  The effects of
      pollutants on community air quality are described,
      and knowledge of health aspects of pollutants is
      summarized.  Laws regulating pollutant emissions
      are given,  and control equipment and measures
      needed to comply with certain  laws are listed.


    Air Pollution