United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600 2-80-036
January 1980
Research and Development
Direct Reduction:
A Review of
Commercial Processes
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-036
January 1980
Direct Reduction:
A Review of
Commercial Processes
by
Larry G. Twidwell
Program Element No. 1AB604
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Direct reduction commercial processes for ironmaking have been reviewed.
The potential for environmental degradation appears to be minimal. A
detailed environmental assessment does not appear to be warranted. It is
recommended that samples of scrubber water and sludge material be collected
and characterized from several gas reductant reactor systems. It is also
recommended that rotary kiln solid reductant reactors be sampled to ensure
that the final gas effluent does not contain harmful concentrations of
organic species, sulfur oxides and trace metal contaminants; and that such
systems be monitored for harmful rates of fugitive emissions.
-------
CONTENTS
Abstract ii
Figures iv
Tables iv
1. Introduction 1
Definition 1
Statistics of Growth 1
Reasons for Growth 4
Terms and Nomenclature 8
2. Direct Reduction Processes 10
Introduction 10
Systems 10
Gas Reductant Systems 13
Midrex 13
HyL 18
Others 22
Solid Reductant Systems 22
SL/RN 26
Others 28
Direct Reduction in United States 30
3. Potential Pollution Problems 31
Gas Reductant Systems 31
Solid Reductant Systems 32
Recommendations 35
4. Steelmaking Using Metallized Product 37
DR-EAF Steelmaking 37
Other Uses 39
References 40
-------
FIGURES
Number Page
1 Growth in production of Direct Reduction sponge iron 2
2 Types of Direct Reduction processes 11
3 Feed material size distribution 14
4 Midrex standard flowsheet 17
5 The HyL process 21
6 The SL/RN process 29
TABLES
Number Page
1 Regional Distribution of Direct Reduction Plants 3
2 Annual Capacity of Direct Reduction Plants, 1977,
1980, 1985 5
3 Sponge Iron Production by Process Types (1977) 12
4 Midrex Process 16
5 Energy Requirements for the Gas Reductant Systems 19
6 HyL Process 20
7 Gas Reductant Processes Excluding Midrex and HyL (1978) .... 23
8 Solid Reductant Systems (1978) 24
9 Rotary Kiln Processes for Treating Steel Plant Wastes 25
10 Energy Requirements for Solid Reductant Systems 27
IV
-------
SECTION 1
INTRODUCTION
Definition
For the purpose of this report "direct reduction" is defined as the
reduction of iron oxides by the use of either solid or gaseous reductants
to produce a solid iron product.
It would be more appropriate to designate the "direct reduction"
processes as alternative processes, i.e., alternatives to the production
of iron in the conventional blast furnace. The reactions that occur in the
"direct reduction" processes are no more "direct" in producing iron than
those reactions occurring in the blast furnace. The difference between the
process types is that the direct reduction processes operate at much lower
temperatures than does the blast furnace and therefore produce a solid iron
product instead of a liquid product.
Statistics of Growth
Present day large scale commercialization of the direct reduction
processes dates back only about 25 years (however, the first iron produced
was probably via direct reduction processes practiced over 3000 years ago).
Growth and projects of growth are given in two graphs (1-3) presented in
Figure 1. The most recent projections (2,4,5) suggest that production
may achieve about 50 million tonnes of direct reduced product by 1985
with the possibility that the production could reach 100 million tonnes
(3,5). This could mean that about 5-10 percent of the world's iron may be
produced by direct reduction processes in 1985. In 1978 about 3 percent
of the world's iron was produced by direct reduction processes.
Miller (4) has tabulated the regional distribution of direct reduction
plants by production and by number of plant installations (Table 1). In
1978 almost three-fourths of the world's sponge iron was produced by the
developing world: Latin America (30.9%), Middle East (19.7%), Africa (8.4%),
and Asia (13.9%); about one-sixth was produced in North America. The
projected production distribution in 1985 is about two-thirds for the
developing world, one-eighth for North America. The number of direct
reduction plants in the world at the end of 1978 was 55. This number is
anticipated to double by 1985 (4).
Miller (4) has also tabulated the individual plants presently operating,
those scheduled for completion by 1980, and those planned for operation by
-------
I960
1965
1970
1975
1980
1985
Greenwalt, R. and J. Stephensen (2)
(Reproduced with permission of AIME.)
IIOO
ooo
I960
OPEN HEAffTW STEEL
l»70 ISC I99O
YM, A'PMEREOUCCD IRON TOR IRON
B*PNEREOUCED IRON FOR STEEL
Miller, R. (3)
Figure 1. Growth in production of direct reduction sponge iron,
2
-------
TABLE 1. REGIONAL DISTRIBUTION OF DIRECT REDUCTION PLANTS (4)
Distribution by Production
production;
co
Region
1S78
% world
kt/year total
1979
% world
kt/year total
1980
% world
kt/year total
1981-85
% world
kt/year total
North America
Latin America
Western Europe
Eastern Europe
Middle East
Africa
Asia
Oceania
Total
Region
North America
Latin America
Western Europe
Eastern Europe
Middle East
Africa
Asia
Oceania
Total
©American Society
2935 16.9
5345 30.9
1650 9.5
3415 19.7
1450 8.4
2409 13.9
120 0.7
17324 100
Distribution by
Installations:
1978
% of
No. total
11 19.6
16 28.6
9 16.1
0
4 7.1
3 5.4
12 21.4
1 1.8
56 100
for Metals and
2935
8645
1650
2500
4415
1450
4134
120
25849
Number of
1979
No.
11
18
9
1
5
3
13
1
61
the Metals
11.3
33.4
6.4
9.7
17.1
5.6
16.0
0.4
2935
9065
3650
5000
5215
1700
4134
120
100 31819
Installations
% of
total
18.1
29.5
14.8
1.6
8.2
4.9
21.3
1.6
100
Society (London)
1980
No.
11
19
11
2
6
4
13
1
67
1977.
9.2
28.5
11.5
15.7
18.4
5.3
13.0
0.4
100
% of
total
16.7
28.8
16.7
3.0
7.6
6.0
19.7
1.5
100
6455
15035
5700
5000
7915
5100
4194
1520
50920
1981-85
No.
16
33
15
2
9
9
14
3
101
12.7
29.5
11.2
9.8
15.6
10.0
8.2
3.0
100
% of
total
15.8
32.7
14.8
2.0
8.9
8.9
13.9
3.0
100
-------
1985. His results are presented as Table 2. (It is known (6) that the Steel
Company of Canada has not started up their facility after shutting down in
late 1976. Also Hecla Mining has closed down its mining operation in Casa
Grande, Arizona.) Note that the United States' direct reduction capacity is
indeed small; e.g., 11.2% of the world's capacity in 1977.
Reasons for Growth
The reasons for the optimism in projected growth of direct reduction
produced iron are:
1. A high purity product can be produced. The conversion of iron
oxide to iron is accomplished in the solid state; i.e., a liquid
iron product is not produced as in the blast furnace. This is
because the reaction temperatures are much lower in direct
reduction furnaces. Less reduction of gangue material occurs and
the solubility of solid iron for impurities is much less in the
solid state than in the liquid state. Secondly the product is
relatively pure iron because the feed materials are chosen to
have low impurity levels, especially sulfur and phosphorus.
2. A wide variety of reducing agents are candidates for use in the
direct reduction process; e.g., coal, coke, charcoal, reformed
natural gas, products of coal gasification, oil, coke oven gas,
and coke breeze. The blast furnace requires coke as the primary
reductant.
3. Refractory life is longer and less expensive refractories are
required for the direct reduction furnaces than are required in
the blast furnace because of the lower furnace operating tempera-
tures.
4. Developing nations can enter the steel business on a moderate
scale with a smaller investment in a direct reduction-electric
arc steelmaking furnace (DR-EAF) combination as compared to the
coke, blast furnace, basic oxygen furnace combination; i.e., the
capital cost is only about 60 percent of the conventional process
capital cost (7-9).
5. Some developing nations have rich iron ore deposits and natural gas
or oil supplies but do not have metallurgical grade coals
available.
6. Developing nations need to develop their basic industries and
therefore prefer to pretreat their ores before shipping to other
nations, i.e.; keep as much industry home as possible.
7. Direct reduction products can be used as feedstock for supplement-
ing scrap in an electric furnace, as a coolant feed material for
the basic oxygen furnace operations, and as an iron value added
to increase the productivity of a blast furnace (4).
-------
TABLE 2. ANNUAL CAPACITY OF DIRECT REDUCTION PLANTS, 1977, 1980, 1985 (4)
Plants in Operation on 1 January 1977*
Plant
No.
1-2
3
4-6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
(cont
Year
of Start
1954
1954
1954
1957
1957
1900
1964
1967
1968
1969
1969
1969
1970
1970
1971
1971
1972
1972
1972
1973
1973
1973
1973
1974
1974
1974
1975
1975
1975
1975
1975
1976
1976
d>1976
Company
Hoganas Grangesberg
Hoganas Corporation
SKF etc.
Tohoku-Satetsu
HYLSA-Monterrey I
HYLSA-Monterrey II
Hitachi Metals
TAMSA
Anglo-American Corp.
HYLSA-Puebla I
Oregon Steel Mills
Kawasaki Steel Co.
New Zealand Steel
Thyssen-Purofer
Georgetown Steel Co.
Nippon Steel Co.
Kawasaki Steel Co.
Hamburger Stahlwerke
Armco Steel Corp.
Acos Finos Piratini
Dunswart Iron & Steel
SIDBEC-DOSCO
MINORCA
USIBA
HYLSA-Monterey III
Nippon Kokan KK
Steel Co. of Canada
Heel a Mining Co.
Sumitomo Metals Co.
Sumitomo Metals Co.
All is Chalmers Co.
Dalmine-Siderca
Sudbury Metals Co.
Fior de Venez.S.A.
Country Process
Sweden Hoganas
USA Hoganas
Sweden Wiberg
Japan Rotary kiln
Mexico HyL
Mexico HyL
Japan Wiberg
Mexico HyL
South Africa High veld kiln
Mexico HyL
USA Midrex
Japan Kawasaki
New Zealand SL-RN
West Germany Purofer
USA Midrex
Japan Koho
Japan Kawasaki
West Germany Midrex
USA Armco
Brazil SL-RN
South Africa Krupp
Canada Midrex
Venezuela HIB
Brazil HyL
Mexico HyL
Japan SL-RN
Canada SL-RN
USA SL-RN
Japan Sumitomo
Japan Kabota
Canada ACCAR
Argentina Midrex
Canada ACCAR
Venezuela FIOR
Reductant
Coke breeze
Coke breeze
Coke breeze
Coal
Natural gas
Natural gas
Coke breeze
Natural gas
Coal
Natural gas
Natural gas
Coke breeze
Coal
Natural gas/CO gas
Natural gas
Coke breeze
Coke breeze
Natural gas
Natural gas
Coal
Coal
Natural gas
Natural gas
Natural gas
Natural gas
Coal
Coal
Coal
Coal
Coal
Coal ,oil ,gas
Natural gas
Natural gas, oil
Natural gas
Annual rated capac-
city of DRI, kt
170
70
90
24
100
270
10
280
1000
315
300
72
120
150
400
48
240
400
330
60
150
400
650
250
475
350
360
60
240
210
50
330
240
400
-------
Table 2 (Continued)*
38
39
1976
1976
Total
Nippon Steel Co. Japan
Ferriere di Arvedi Italy
NSC Oil
Kinglor-Metor Coal
150
40
8804
Plant Year
No. of Start
Plants Under Contract at Start of 1977 and Scheduled for Completion by 1980
Anmia
Company
SIDBEC-DOSCO II
SIDOR III
DIDOR III
COSIQUA
NISIC
HYLSA-Puebla II
Iraq Iron & Steel Co.
Anglo-American Corp.
Consol i dated Gol d Fi el ds
Kawasaki Steel Co.
Nippon Steel Co.
NISIC
SIDERPERU
Acindar
BSC-Hunterston
Qatar Steel Co.
PT Krakatau
NISIC
PT Krakatau
SIDOR IV
SIDOR IV
USSR-Kurak
North Sea Iron Co.
Nord. Ferrowerke
USSR-Kurak
ISCOTT
TIKA
Saudi Arabia-Jubail
Country
Process
Annua
acity
I rated cap-
of DRI. kt
Reductant
Natural gas 625
Natural gas 360
Natural gas 360
Gasified oil 350
Natural gas 330
Natural gas 625
Natural gas 1485
Coal 300
Coal 100
Coke breeze 250
Oi 1 240
Natural gas 1200
Coal 100
Natural gas 420
Natural gas 800
Natural gas 400
Natural gas 575
Natural gas 1000
Natural gas 1725
Natural gas 1200
Natural gas 2100
Natural gas 2500
Natural gas 800
Natural gas 1200
Natural gas 2500
Natural gas 420
Gasified naphtha 250
Natural gas 800
23015
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
1977
1977
1977
1977
1977
1977
1977
1977
1977
1977
1977
1978
1978
1978
1978
1978
1978
1979
1979
1979
1979
1979
1980
1980
1980
1980
1980
1980
Canada
Venezuela
Venezuela
Brazil
Iran
Mexico
Iraq
South Africa
USA
Japan
Japan
Iran
Peru
Argentina
UK
Qatar
Indonesia
Iran
Indonesia
Venezuela
Venezuela
USSR
UK
West Germany
USSR
Trinidad-Tobago
Zambia
Saudi Arabia
Midrex
Midrex
HyL
Purofer
Purofer
HyL
HyL
Highveld kiln
Hockin kiln
Kawasaki
NSC
Midrex
SL-RN
Midrex
Midrex
Midrex
HyL
HyL
HyL
Midrex
HyL
Midrex
Purofer
Mi drex
Midrex
Midrex
HyL
Midrex
Total
-------
TABLE 2 (Continued).*
Annual Capacity of Direct Reduction Plants 1977, 1980, 1985 (4).
Projects Planned for Operation Between 1981 and 1985
Plant
No. Country
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
USA
USA
Canada
Argentina
Bolivia
Brazil
Brazil
Brazil
Brazil
Brazil
Mexico
Venezuela
Italy
Spain
Spain
Algeria
Egypt
Tunisia
Turkey
Iran
Iran
South Africa
New Zealand
USA
USA
El Salvador
Argentina
Argentina
Brazil
Brazil
Brazil
Brazil
Colombia
Ecuador
Greece
Abu Dhabi
Libya
Morocco
India
Australia
Total
*<& American Soci
Company and location
Texas Ferro reduction
Gulf Coast Consortium
Interprovincial Iron Co.
Gurmendi
SIDERSA (Santa Cruz)
COSIGUA II
COFAVI
USIBA II
IMBITUBA(Sta.Catarina)
Piritim II
TAMSA II
FIOR de Venezuela
Adriatic Consortium
Sid. Gibraltar
PREPELSA (Huelva)
SNS (Jijel)
Government (He! wan)
Government (Gabes)
EDAS
NISIC (Bandar Abbas)
NISIC (Esfahan)
SCAW & Anglo-American
New Zealand II
Republic Steel
(Massillon)
Republic Steel
(Gadsden)
Government (Acujutla)
HIPASAM (Punto Colorado)
Lucini
DEDINI
IKOSA-Pains
Mendez Junior
Mannesmann
ACENOR (Barranquilla)
EOJASIDOR
SIDERHELLAS
Government
Government (Misurata)
Annual rated
Probable capacity of
Reductant DRI, kt
Natural gas
Natural gas
Coal
Natural gas
Natural gas
Gasified oil
Gasified oil
Natural gas
Gasified coal
Coal
Natural gas
Natural gas
Natural gas
Liquified natural
gas
Liquified natural
gas
Natural gas
Natural gas
Natural gas
Natural gas
Natural gas
Coal
Coal
Coal
Coal
Coal
Natural gas
Natural gas
Coal
By-product gas
By-product gas
By-product gas
Natural gas
Natural gas
Natural gas
Natural gas
Government Natural gas
Government (AndraPradesh)Coal
Hamersley Iron Co.
Natural gas
900
1500
400
400
200
350
350
300
400
200
300
2000
800
450
500
1200
400
800
400
2800
1200
200
200
360
360
160
400
400
400
160
350
400
200
400
300
400
500
300
60
1200
22600
ety for Metals and the Metals Society (London) 1977.
-------
8. There will be a deficit in scrap availability by 1980 (10). Direct
reduction sponge iron is viewed as a good substitute feed material.
9. The time required for commissioning a DR/EAF steel plant is about
2 to 3 years compared with 5 to 7 years for a conventional inte-
grated steel plant (9).
10. Air pollution problems are minimized in a gas direct reduction
plant because all gases are cleaned and recycled. One of the main
air pollution producing stages in the conventional blast furnace
is cokemaking which is not required in a gas D.R. facility (9).
11. Transferability of technology has been demonstrated by fast start-
ups of new plants. This represents an important economic advantage
12. Energy requirements are less for the DR-EAF process than for the
conventional coke-blast furnace-BOF process (9).
13. Production facilities have been constructed to produce over a
million metric tons per year of D.R. iron; i.e., two D.R. furnaces
are presently used to accomplish this. It is anticipated that
single furnaces with capacities of a million tons per year will be
possible in the near future (9).
Terms and Nomenclature
Many articles have been published on direct reduction processes. The
nomenclature used by the various authors is sometimes confusing and contra-
dictory. Therefore, some of the terms and their various meanings will be
presented in this section.
The charge fed to the reduction reactor is an iron oxide containing
material. It may be ore lumps, ore fines, pelletized agglomerates, formed
briquettes, or a combination of several of these materials. In the litera-
ture, the feed material may be simply characterized as ore, lump ore,
nature ore, mineral ore, lump oxide, pellets, pelletized ore, pelletized
oxide, or pelletized fines.
The product of the direct reduction reaction is a reduced material that
contains metallic iron, some form of iron oxide, some iron carbide, and
gangue that was contained in the original feed. This product is called
sponge iron, luppen, metallized iron, metallized product, reduced iron,
direct reduced iron, or a trade name, such as Midrex iron or HyL iron.
The reducing reactions remove oxygen from the iron oxides. A measure
of the success of accomplishing this is described in several ways.
Percent metallization or degree of metallization or metallization
is defined as
8
-------
Met = % Fe (as metal) inn
Met< % Fe (total) x 10°
It is the percent of iron present in the feed material that has been
converted to metallic iron.
Equivalent metallization is defined as
% Eq. Met. = % Met. + 6 (%C in sponge).
This term is used in presentations on the HyL process but has not been
widely accepted.
Degree or Percent of Reduction is defined as:
Oxygen removal from feed
Original oxygen combined with iron
The terms percent reduction and percent metallization are not the same,
and percent metallization or simply metallization is the normally reported
value; i.e., metallization is more descriptive of how effective the reduction
was in producing metallic iron.
It is also important to know how much metallic iron is present in the
final product; i.e., the weight of metallic iron divided by the weight of
product.
Wt(Fe)
% Fe = Mire; x 100
Wl( product)
The metallization and the metallic iron percentages are normally not
the same numerical value; e.g., for a feedstock that contains 65 percent
iron that is reduced to give a 90 percent metallization, the metallic iron
content is about 86 percent. Bertram (11) points out that the two terms,
metallization and metallic iron, have sometimes been used interchangeably
and, therefore, care should be taken to understand how the author has
performed his calculations.
-------
SECTION 2
DIRECT REDUCTION PROCESSES
Introduction
More than 1200 patents have been issued covering various aspects of
proposed direct reduction processes. Nearly 100 direct reduction schemes
were examined by iron and steel producers between 1950 and 1975 (7). Of the
many proposed and piloted processes only 12 major designs have progressed
to commercial application. At the end of 1978 there were 55 direct reduction
plants in the world (1). Thirty-seven of these plants use a gas reductant.
Twenty-five of the 37 plants are based on either the HyL or Midrex process.
Eighteen of the 55 plants use a solid reductant. Fourteen of the solid
reductant systems are rotary kiln operations and use coal or coke breeze
as the reductant.
Kalla and Steffen (1) present an interesting display of the distribution
of process types used in sponge iron production (July 1977). See Figure 2.
Note that most of the world's sponge iron is produced by gas reduction
processes.
As already noted, there are two broad classes of direct reduction
processes; i.e., gas reductant systems and solid reduction systems. A
further sub-division is normally made in the literature according to the
type of furnace; i.e., shaft, rotary kiln, or fluidized bed. Individual
direct reduction systems will be described in the following section.
Systems
A summary of sponge iron production capacity by the type of process
is presented in Table 3.
The gas reductant systems that use a shaft vessel as the reactor are
the Midrex (Korf Industries), Purofer (Thyssen), Armco (Armco Steel), and
Wiberg processes. These are all moving bed reactors; i.e., the solid feed-
stock moves in a direction countercurrent to the flow of reducing gas.
The processes differ mainly in the way the reducing gas is produced and
circulated and the shape of the shaft. Also a major difference in the
Purofer process is that the product is discharged hot into a briquetting
system whereas all the other processes cool the product before discharge.
The HyL (Hojalata y Lamina) process uses retort reactors. The process
is a batch operation and the charge is prereduced, reduced, cooled, and
10
-------
Figure 2. Types of Direct Reduction processes (1)
(Reproduced with permission of Iron and Steel International.)
11
-------
TABLE 3. SPONGE IRON PRODUCTION CAPACITY BY PROCESS TYPES (1978)
Gas Reductant
(80.2%)
Shaft Furnaces (38.7%)
Midrex (27.4%)
Purofer (7.6%)
Armco (3.1%)
Wiberg (0.6%)
Retort Furnaces (26.7%)
HyL (26.7%)
Fluidized Bed Furnaces (10.7%)
HIB (7.0%)
FIOR (3.7%)
Others (4.1%)
Solid Reductant
(19.8%)
Rotary Kiln (16.0%)
SL/RN (7.9%)
Japanese (6.7%)
Krupp (1.4%)
Retort (2.5%)
Others (1.3%)
12
-------
discharged as separate operations.
There are two commercial fluid bed processes, the HIB (high iron
briquette - U. S. Steel) process and the FIOR (fluidized iron ore reduction -
Esso Research) process.
The main process using solid reductants is the rotary kiln. The rotary
kiln systems used on an industrial scale are the SL/RN (Stelco, Lurgi,
Republic Steel, National Lead) and Krupp processes. The iron containing feed
material is either lump ore or pellets. This differentiates the two pro-
cesses from the Japanese rotary kiln reactors that use pelletized steel pi ant
dusts from the feedstock; e.g., Kawasaki process, Nippon Steel's Koho process,
and Sumitomo Metals SPM and SDR processes.
In March 1978 a rotary kiln facility was started up in Rockwood, TN.
The process (Azcon Corporation) utilized pellets as the iron source and coal
as both the energy and reductant sources.
Another process that utilizes a rotary kiln is the ACCAR (Allis Chalmers
Controlled Atmosphere Reduction) process. Oil and natural gas are the heat
and reducing sources.
A few other low production solid reductant systems are in operation;
e.g., the Kinglor-Metor and the Hoganas processes. The Kinglor-Metor
process is a shaft furnace that uses coke, coke breeze, or lignite as the
reductant and external heating as the energy source. Muffle furnaces are
used to reduce batch charges of ore and coke breeze in the Hoganas process.
Gas Reductant Systems
More than 80 percent of the world's sponge iron is provided by gas
reductant systems. About half of this amount is produced in shaft-type
furnaces; about a third is produced in retort furnaces. The remainder is
produced in fluid-bed reactors.
The feed material to gas reductant systems can be lump ore, pellets, or
ore fines. The shaft furnaces require lump ore, pellets, or briquetted
fines. The fluid bed processes require ore fines. The ore feed size
requirements are depicted graphically in Figure 3.
The feed material to a direct reduction plant is carefully controlled;
not only the feed material particle size distribution but also the feed
chemistry. Each plant operation has its own chemical specifications for the
material it buys. The specification is influenced by not only the type of
direct reduction reactor system but also by the subsequent end use. An
illustrative specification is an iron content of at least 65 percent, a
gangue content (Si02+Al203)/Fe ratio < 5 percent; and a phosphorous and
sulfur content < 0.05 percent and < 0.01 percent, respectively.
The Midrex Process
From published literature it appears that the Midrex process is being
13
-------
too
001
Groin sin (mm)
Figures. Feed material size, distribution (1)
(Reproduced with permission of Iron and Steel International.)
14
-------
used at 10 locations. The companies using the process, their location,
and production capacities are given in Table 4. About 7 million tonnes
per year of sponge iron is produced by this process. Midrex (12) predicts,
based on present contracts and agreements, that the production capacity
by 1981 will be 17 million tonnes/year.
The largest single gas reductant furnace in the world is now on
stream at Contrecoeur, Canada; it is a single reactor with a rated
capacity of 600,000 tonnes/year (9). This increases the production
capacity of Sidbec-Dosco's two reactors to over 1 million tonnes per year
of direct reductant iron.
Detailed process information is available on the two Midrex facilities
in the United States (13-16), the Sidbec-Dosco facility in Canada (17-
19) and the Dalmine Siderca, Sidor, and Acindar plants in Latin America
(19). All of the plants are based on modules that have a 400,000 tonnes/year
capacity except the Sidbec-Dosco II facility (17) which has a 400,000
and a 600,000 tonne/year module. Only a brief general description of
the process will be presented here. See the noted references for further
details.
The Midrex standard flowsheet is presented in Figure 4. The process
consists of a reduction shaft furnace and a gas reformer plant. The
shaft furnace is 16 feet (4.88 meters) in diameter and is rated at
400,000 tonnes/year. The reactor is divided into a reduction zone and
a cooling zone. In the reducing zone iron oxide is reduced as it moves
countercurrent to the injected reducing gas. The reactions of interest
are the reduction reactions and the formation of iron carbide reaction:
3 + 3(CO+H2) -»- 4Fe + 3(C02+H20)
(reduction) (800-1 000°C)
3Fe + CO + H2 -* Fe3C + H20
(carburizing)
The product gas exits the furnace near the top. It is cooled,
cleaned by a scrubber to remove the particulates, mixed with new natural
gas, and passed to a reformer to form more CO and H2-
The reformer furnace contains long alloy tubes heated externally.
The reactions of interest are:
CH4 + C02 -> 2CO + 2H2
(11 00-11 50°C)
CH4 + H20 -> CO + 3H2
The top gas from the reducing furnace is the source of the required oxi-
dants; additional air or steam is not required for the reforming reactions
Some of the top gas which contains about 75 percent CO + H2 is used as
15
-------
TABLE 4. MIDREX PROCESS (1,4,12,20)
Operator
Oregon Steel
Georgetown Ferreduction
Hamburger Stahlwerke
Sidbec-Dosco I
Sidbec-Dosco II
Dalmine Siderca
Si dor I
Sidor II
Acindar
NISIC
Qatar Steel
British Steel
Location
Portland, OR
Georgetown, SC
Hamburg, West Germany
Contrecoeur, Canada
Contrecoeur, Canada
Campana, Argentina
Mantazas, Venezuela
Matanzas, Venezuela
Villa, Argentina
Ahwaz, Iran
Doha, Qatar
Hunterston, Gr. Britain
Total (1978)
Capacity, tonnes/yr
400,000 (1969)*
400,000 (1971)
40Q,000 (1972)
400,000 (1973)
600,000 (1977)
330,000 (1976)
400,000 (1977)
1,200,000 (1978)
(3-400,000 plants)
420,000 (1978)
1,200,000 (1978)
(3-400,000 plants)
400,000 (1978)
800,000 (1978)
(2-400,000 plants)
6,950,000**
* Start date
** The total capacity is based on published literature figures. However,
from private communications with J. Bradley, Midrex Corporation this
figure should be 6,065,000 tonnes per year (August 29, 1979).
16
-------
Process
Gas Mist I *
Eliminator
Oxide
Feed
Cooling Gas
Mist Compressor
Eliminator
Process
Gas
Compressor
Figure 4. Midrex standard flowsheet (18).
(Reproduced with permission of Midrex Corp.)
-------
fuel for firing the reformer chambers.
Hot metallized sponge iron descends into the cooling zone of the fur-
nace. It is cooled by a closed loop of circulating inert or natural gas.
The cooled iron product is discharged continuously at about 30-50 C.
The iron product may be further treated to prevent it from reoxidizing
during storage. Midrex has developed processes that are used for passivating
the product; e.g., a controlled slow reoxidation of the iron surface and a
cold briquetting process that reduces the total surface area so that reoxi-
dation is of minor importance. However, in both cases the passivated
product must be stored in areas protected from rainwater and/or seawater
during transportation.
It is of interest to compare the energy requirements of gas reductant
systems. Kalla and Steffen (1) report the results shown in Table 5. The
Midrex, Purofer, and Armco processes require about the same energy, with
the Midrex requiring slightly less than the other two. The Midrex 600,000
tonne/yr (Sidbec II) plant has reported even lower energy use, 2.5 Gcal/
tonne Fe for gas consumption usage (21).
The HyL Process
The Hojalata y Lamina S.A. (HyL) process is currently (end of 1978)
being used at nine locations. The companies using the process, their loca-
tion, and production capacities are given in Table 6. About 7 million
tonnes per year of sponge iron is produced by this process. Projections (4)
are that capacity will be increased to about 9 million tonnes/yr by 1981.
Detailed process information is available on the HyL facilities in
Mexico (22-24). Only a brief general description will be presented here.
See the references for further details.
The HyL process operates on a fixed bed principle. The ore remains
stationary in a closed retort. The gas phase is changed according to the
operation to be performed in the reactor; i.e., drying, preheating, reduct-
ion, and cooling. A flow sheet for the process is shown in Figure 5 (7). A
process module consists of four reactors. Each reactor is about 17 feet
(5.18 meters) in diameter and 49 feet (14.9 meters) in height. Three of the
four reactors are always connected in series (with respect to gas flow) while
the fourth is loaded or unloaded. The steps in the reduction cycle consist
of first reactor - loading and unloading, second reactor - initial reduction,
third reactor - final reduction, and fourth reactor - cooling and carburiz-
ing.
Each retort is discharged every 9-12 hours. The sequence of treatment
occurring in each reactor is: fresh reformed gas is used to cool the reduced
product; it then passes on to another retort where final reduction takes
place; and gas from this reactor is passed through a third reactor where
preheating of the charge and preliminary reduction occurs. Gas from this
stage of the operation still contains enough CO + H« to be used as fuel
for firing the reformer combusters. Note that the product gas from the last
18
-------
TABLE 5. ENERGY REQUIREMENTS FOR THE GAS REDUCTANT SYSTEMS (1)
Reducing Agent
Gross Energy Consumption Total
Gas Electricity Equivalent
Process
Midrex
HyL
Pruofer
Armco
FIOR
Blast Furnace
H2
53
75
47
68
75
(25)
CO
35
14
45
20
14
C02 H20 (CH4+N2)
2 5
8
2 3
2 9
8
5
3
3
1
3
(Gcal /Tonne Fe)
3
4
3
3
4
.1
.6
.3
.4
.0
(kWh/tonne Fe)
155
15
130
35
150
(Gcal /tonne Fe)
3
4
3
3
4
2
.2
.6
.4
.4
.1
.8
-------
TABLE 6. HyL PROCESS (1,2,4,22)
Operator
HYL I
HYL II
HYL III
TAMSA
HYL I
HYL II
USIBA
SIOOR I
SIDOR III
SIDOR IV
NISIC III
Krakatau-Ferrosteel
Tika
Iraq
Location
Monterrey, Mexico
Monterrey, Mexico
Monterrey, Mexico
Vera Cruz, Mexico
Puebla, Mexico
Puebla, Mexico
Bahia, Brazil
Matanzas, Venezuela
Matanzas, Venezuela
Matanzas, Venezuela
Ahwaz, Iran
Kota Baja, Indonesia
Solwezi , Zambia
Khor El Zubeir, Iraq
Capacity, tonne/yi
95,000 (1957)*
260,000 (1960)
475,000 (1974)
280,000 (1967)
315,000 (1969)
625,000 (1977)
250,000 (1974)
400,000 (1977)
350,000 (1977)
2,000,000 (1978)
400,000 (1978)
500,000 (1978)
300,000 (1978)
600,000 (1978)
* Start date
Total (1978)
6,850,000
20
-------
Fuel Gas
o
O
Reactor A
Loading
and
Unloading
o
O
Reactor B
Initial
Reduction
Legend
Dehumidifier
Gas Heater
k Gas Flow
o
O
Reactor C
Final
Reduction
\
Reformer
Dehumidifier
o
O
Reactor D
Cooling
and
Carburizing
m *
Figure 5. The HyL process (27).
(Reproduced with permission of Iron and Steel Engineer.)
21
-------
reactor is not recycled for reforming as is the product gas from the Midrex
furnace.
The HyL product is normally controlled at about 85 percent metallization
and contains up to 2.5 percent C. It is claimed that the product is nonpyro-
phoric (22) The energy requirements were reported by Kail a (Table 5) to be
4.6 Gcal/tonne Fe. This is somewhat higher than the value reported by Labee
(22) for the HyL Puebla II plant; i.e., 3.4 Gcal/tonne product (which is
approximately equivalent to 4 Gcal/tonne Fe).
Other Processes
The Midrex and HyL processes accounted for over 88 percent of the
metallized iron produced by gas reduction processes in 1978. The other
gas reduction processes, their production capacities, and locations are
presented in Table 7.
The Purofer (26) and Armco (27) processes are both shaft furnace pro-
cesses that are similar in principle and design to the Midrex process. Both
processes involve the flow of solids countercurrent to the flow of reducing
gas. The feeds are similar to the Midrex and the results of metallization
essentially the same, 92-94 percent. The differences are that the Armco
process does not recycle the reducing gas and the Purofer process discharges
its product hot. The hot product is briquetted before release and is there-
fore nonpyrophoric. Another difference is that the Purofer process can use
either oil (gasified) or natural gas as the reductant.
Two other gas reduction processes use fluid-bed reactors; i.e., the HIB
(high iron briquette) process (28) and the FIOR (fluidized iron ore reduc-
tion) process (29). Both are U.S. developed processes but commercialization
has taken place in foreign countries; see Table 7. The FIOR process produces
a product that is 88-93 percent metallized whereas the HIB process is design-
ed for the product to be used in ironmaking rather than steelmaking, hence
its metallization is much lower; i.e., 70-75 percent Fe. The products from
these processes are a fine powder that has been produced at low temperatures
and is, therefore, pyrophoric. The product is briquetted to prevent
reoxidation during storage.
Solid Reductant Systems
Approximately 20 percent of the world's sponge iron capacity is provided
by solid reductant systems (1978). Most of the capacity is in rotary kiln
reactors (16%) and a small amount is retorts (2.5%) and vertical shaft
furnaces (1.3%). The solid reductant systems, their production capabilities,
and their location are presented in Table 8. (There are several discrepan-
cies in the production figures given by references 1, 2, and 4.) In addition
to those processes listed in Table 8, a number of rotary kiln furnaces
treat steel plant dust, see Table 9.
Approximately 90 percent of the production from solid reductant systems
is by rotary kiln processes, 2,170,000 tonnes/yr from virgin ores and
1,725,000 tonnes/yr from mill wastes. According to Miller (4), "There has
22
-------
Purofer
Purofer
ARMCO
FIOR
HIB
TABLE 7. GAS REDUCTANT PROCESSES
EXCLUDING MIDREX AND HYL (1978) (1,2,4,22)
Process Operator
COSIGUA
NISIC
ARMCO
Fior de Venez
MONORCA
Location
Purofer Thyssen-Purofer West Germany
Brazil
Iran
USA
Venezuela
Venezuela
Total (1978)
Capacity, tonnes/yr
150,000 (1970)*
350,000 (1977)
330,000 (1970)
330,000 (1972)
400,000 (1976)
650,000 (1973)
2,210,000
*Start date
23
-------
TABLE 8. SOLID REDUCTANT SYSTEMS (1978) (1,2,4)
Process
Location
Reactor Type
Capacity, tonnes/yr
Hoganas
Hoganas
SL/RN
SL/RN
SL/RN
SL/RN
Krupp
High veld Kiln
Highveld Kiln
Hockin Kiln
Kinglor-Metor
Sweden
USA
New Zealand
Brazil
Canada
Peru
S. Africa
S. Africa
S. Africa
USA
Italy
Muffle
Muffle
Rotary Kiln
Rotary Kiln
Rotary Kiln
Rotary Kiln
Rotary Kiln
Rotary Kiln
Rotary Kiln
Rotary Kiln
Shaft
170,000
70,000
120,000
60,000
360,000
100,000
150,000
1,000,000
300,000
100,000
40,000
Total (1978)
2,470,000
-------
TABLE 9. ROTARY KILN PROCESSES FOR TREATING STEEL PLANT WASTES (1,2,4)
Process
SL/RN
Sumi tomo
Kawasaki
NSC
Location
Fukayama, Japan
Wakayama, Japan
Kashima, Japan
Mizushima, Japan
Chiba, Japan
Hirohata, Japan
Capacity, tonnes/yr
490,000
340,000
290,000
310,000
85,000
210,000
Total (1978) 1,725,000
25
-------
been no general acceptance of the rotary kiln processes. Solutions for
frequent electrical and mechanical breakdowns and for complex operating
difficulties have been neither easy nor consistently successful, and start-
up periods have been longer than planned."
It should be noted that the SL/RN plant in Canada (30) has been shut
down since May 1976 (4). An additional facility, not listed in Table 8,
uses a mixture of coal and oil; i.e., the Allis Chalmers Controlled Atmos-
phere Reduction (ACCAR) rotary kiln (50,000 tonnes/yr). A second ACCAR kiln
is located at Sudbury, Canada (capacity 340,000 tonnes/yr) but the reductant
and fuel are a mixture of oil and natural gas. It has been out of operation
since October 1976 (4).
According to Miller (4) there appears to be reason for some optimism for
rotary kiln processes: "There has been an increasing number of encouraging
reports since 1975 from the modified SL/RN unit in New Zealand, the Krupp
plant in Benoni, and the Highveld operation in South Africa, and especially
from several kiln operations with steelplant waste-material charges in Japan.
The sponsors of coal-based processes believe, therefore, that the break-
through by their designs is very close."
Another recent development also lends optimism to the use of coal based
rotary kiln processes; i.e., the Western Titanium, Ltd. coal fired Hockin
process (32). This process was developed in Australia and is now being
demonstrated by the Azcon Corporation at its Rockwood, TN plant. It is a
rotary kiln process that uses coal for both the reductant and the fuel. No
other fuel is required.
A comparison of the energy requirements for solid reductant systems is
presented in Table 10. To compare these values with gas reductant systems
refer to Table 5. The average energy requirement per tonne of iron produced
by direct reduction (all processes included) falls within the range
3.8 + 1.0 Gcal/tonne Fe. It has been reported that the production of pig
iron by the blast furnace uses approximately 2.8 Gcal/tonne Fe (1973 figure,
Kono (25)). This energy value is for Japanese practice and is, likely,
considerably less than United States energy consumption. According to
Dailey (33), the best North American blast furnace practice (1972) required
5 Gcal/tonne Fe. Depending on the basis of comparison, one can state that
direct reduced iron can be (a) produced at a lower energy input than pig iron
(e.g., any direct reduction practice compared to "best" North American blast
furnace practice) or (b) produced at essentially the same energy input
required for "best" blast furnace produced pig iron (e.g., Japanese blast
furnace practice compared to any direct reduction practice). At one time
or another all of these claims have been reported in the literature. Often
the assumptions used in making the comparisons are not clearly stated and
care must be exercised in using the results.
The SL/RN Process
Rotary Kiln processes are similar in principle. Kilns are long cylin-
drical refractory lined vessels that can be rotated about their longitudinal
axis. The vessel is tilted at about a 3 percent slope. Feedstocks
26
-------
TABLE 10. ENERGY REQUIREMENTS FOR SOLID REDUCTANT SYSTEMS
Total Equivalent Energy Consumption
Process Gcal/tonne Fe Reference
SL/RN
Krupp
Hockin
Kawasaki
Accar
Coal /Oil
Oil
Gas
3.7
3.8
3
3.5
2.8
3.0
3.5
- 4.8
- 4.8
.2
- 4.2
- 3.3
- 3.3
- 3.8
1, 34
1, 34
32
34
35
35
35
Kinglor-Metor 4.2 5, 36
27
-------
(usually lump or pelletized ore), fluxes, and coal are fed in the high end of
the vessel. The ore is preheated and reduced as it is moved by the vessel's
rotation and gravity. The reduced material is discharged continuously from
the lower end of the kiln into a cooling or briquetting system.
The rotary kiln processes differ mainly in design of the heating system
and in the selection of operating conditions such as type of ore, type of
coal, feed rates, gas and solid temperatures, and oxygen potential.
The SL/RN process will be briefly described here as an illustration of
rotary kiln processes. Further detailed descriptions are readily available
in current literature for all the rotary kiln processes (11,20,28,30,37,38).
A flow diagram (7) is presented in Figure 6. The selection of feed
materials is very important. The reactivity of the ores is carefully tested
before they are accepted for use in the reactor. The gangue content,
base/acid ratio, sulfur, phosphorus, alkaline metals, and heavy metal conten-
are all specified and tested before an ore is purchased. The physical
properties are also important; e.g., strength, size. "Contrary to a
misconception that is more widespread than it should be, direct reduction
will not work with 'any' ore" (7).
The type of coal is also specified and tested before acceptance for use
in a solid reductant kiln. It must have a proper reactivity, a specified
low sulfur, phosphorus and volatile content, and a minimum fusion temperature.
A coal that does not have a proper fusion temperature can agglomerate and
stick to the reactor walls and rather quickly shut down the reactor.
The kiln temperature is controlled by a series of burners along the
length of the kiln. Ore and coal are charged into the kiln (up to 100 meters
long). Limestone and dolomite may be included if the sulfur content is to
be maintained at low levels.
The solid mixture is heated to about 1000°C (this varies depending on
ore and fusion temperature of the coal) as it moves along the kiln. The
residence time is 3-4 hours and the metallization is from 92-95 percent. The
discharged product is cooled and screened, and the iron is magnetically
separated from the coal char.
The waste gases are further burned in an afterburner or are scrubbed
and the waste gas flared.
Other Sol id Reductant Systems
There are a few nonrotary kiln processes; e.g., the Hoganas and Kinglor-
Metor (39) processes. The Hoganas process is based on the use of muffle
furnaces and the Kinglor-Metor process is an externally heated shaft furnace
system. Neither process is considered suitable for large scale development
and, therefore, will not be discussed here.
28
-------
TO
"IRON-ORE CONCENTRATES
e=g _^ PELLET1ZATION
rr
EXHAUST GAS
HARDENING AND
OXIDATION OF PELLETS
DUST
EMERGENCY ,
STACK
AIR
....... ..,........ .....-
\iiJ2J .>-.:__...-.. ROTARY KILN. .- V '. "I L
.^....-.:x:-.:.:..:v..- .. ...
COAL
WATER
rrOi IP
IT
SCREEN
MAGNETIC SEPARATOR
RECOVERED COAL CHAR
COAL SEPARATOR
WATER
MAGNETIC SEPARATOR
SPONGE IRON
ASH AND RESIDUES
Figure 6. The SL/RN process (7).
-------
Direct Reduction in the United States
The near future prospects for U. S. adoption of direct reduced iron-
making is at this time questionable. The present direct reduced iron
capacity in the United States (4) is 1.1 million tonnes per year. This
production capacity is projected (4) to increase to 4.32 million tonnes per
year by 1985 (Table 2). About 2.4 million tonnes per year of the projected
increase is based on natural gas reduction units. The proven reserves in
the lower 48 states amount to 200 trillion cubic feet (Tcf). Our current
rate of consumption is 19 Tcf per year. Therefore, unless the quantity of
proven reserves increases dramatically, a supply of only 10 years exists.
Even if an optimistic view is taken that the reserves can be doubled by
acquiring gas from Alaska and Mexico and/or by production of synthetic gas
from coal, only a 20 year supply at current consumption rates could be
possible. Therefore, coal or coal derived products appear to be the only
fuel type that can be realistically considered for direct reduction use (40,
41).
Hayes (42) estimates that by 1985 the United States production of coal
will be 818+18 million tonnes per year. The reserves of coal are estimated
by the Bureau of Mines to be 397 billion tonnes (42). Of the available coal,
Hayes estimates that 227 billion tonnes is recoverable. It, therefore,
appears from the projected supply and demand estimates that coal will be an
energy form available for many years.
Coal-based direct reduction systems have been discussed previously in
the section, "Solid Reductant Systems." There are commercial plants present-
ly producing iron. Recent progress in the technology of solid reductant
processes, particularly rotary kiln reactors, was reviewed at a recent Office
of Technology Assessment (O.T.A.) Seminar on New Techniques in Steelmaking
(41). The presentations on the SL/RN (43) and ACCAR (44) rotary kiln
processes suggested that the major technical problems experienced in opera-
ting rotary kilns have been solved.
The successful demonstration of a new technology is an important step
toward the adoption of that technology by industry. But the successful
demonstration of new technology does not ensure that it will be adopted even
if it is shown to be superior to the old technology (45). Consideration of
the factors that impact on the near future adoption of direct reduction
technology in the United States is beyond the scope of the present project;
e.g., impact of foreign steel imports, availability and price of scrap,
expected replacement of old facilities, and need for new capital formation.
However, these factors are being considered in a comprehensive Office of
Technology Assessment study, "Impact of Technology on the International
Competitiveness of the United States Steel Industry" (46). The results of
this study will be available in the Fall, 1979. Conclusions of the study
and resultant policy decisions may have an important impact on the near
future development (or nondevelopment) of direct reduction ironmaking in
the United States.
-------
SECTION 3
POTENTIAL POLLUTION PROBLEMS
One of the attractive features of the direct reduction route to iron-
making is that the processes can be effectively controlled to prevent major
emissions to the environment. The following discussion outlines several
potential pollution problems but each of the potential problems can be
eliminated by good process design and control.
Gas Reductant Systems
The shaft furnace gas reductant systems produce some particulate mater-
ial. The origin of the particulate material is feed deterioration brought
about by abrasion and reduction reactions.
The gas exiting the shaft furnace contains unreacted CO, H2, CH4,
reacted products, C02, H20, and dust particles. This gas is normally treated
in wet scrubbers and the cleaned gas is recycled to the gas reforming unit
and subsequently back to the reactor furnace. The particulate solids in the
scrubber waters are recovered and then briquetted, pelletized, or disposed
of in solid waste storage areas.
It is important to note that gas reductant systems operate under condi-
tions that are unfavorable for formation of polynuclear aromatic hydrocarbons
(PAH); e.g.,
1. Low order hydrocarbons are used in the reforming process (Cfty) and
they are not likely to combine to form the high order PAH's.
2. Reforming is conducted by surface catalyst techniques which do not
promote formation of high order molecules.
3. The reduction reactor is operated at temperatures (=950°C) below
which PAH's form.
Even if PAH's are formed in the reduction furnace they probably are
effectively removed from the gas streams along with the scrubber sludge. The
fate of PAH's in the scrubber water is not known. It is suspected that some
of the PAH material will dissolve in the aqueous phase (the low molecular
weight compounds) and some will remain with the solids either as condensed
particulate organic solids (the high molecular weight compounds) or as ad-
sorbed species on the solids. That fraction of the PAH material that is not
removed by scrubbing should not be a problem because the gas stream is
either recycled to the reduction furnace or is combusted in a separate
chamber.
31
-------
Even if PAH's are formed and removed by scrubbing, they would most
likely not be a problem at those facilities that recycle their solids to the
direct reduction furnace and their water to the scrubber system. Some faci-
lities do not recycle the sludge solids but dispose of them in solid waste
storage areas. The final fate of any PAH material (if it even exists) in the
solid waste is not known.
As is true at all facilities that have ore storage, wind blown dust is
a potential pollution problem. The sponge iron product is not a pollution
problem because it is normally placed in protective enclosures or covered
with canvas to prevent reoxidation by exposure to moisture.
The literature suggests that gas reductant systems can utilize the pro-
duct from a coal or oil gasification plant (47). If this occurs, the poten-
tial pollution problem associated with gasification plants must also be
considered (48).
Another associated source of potential pollution needs to be considered;
i.e., pelletization. It is estimated (1) that 90-150 million tonnes per year
of ore, that has acceptable chemical composition for use in direct reduction,
is available. However, about 70 percent of this material is fine-grained.
The fine-grained material is useful for fluid-bed applications but the size
range necessary for use in the fluid-bed reactors is fairly narrow. There-
fore, a significant fraction of the ore will most likely have to be agglomer-
ated before it can be used. Emissions from iron ore mining, beneficiation,
and pelletizing have been studied by Midwest Research Institute for the EPA.
Potential emission sources are discussed in reference (49).
Note that pelletization of ore fines is not unique to direct reduction
feed materials: it is also used extensively for preparation of blast furnace
feed material. In fact, only about 3 percent of the pellets produced at
present are used in direct reduction processes (2).
Solid Reductant Systems
The gas from a rotary kiln coal reductant system could contain CO, C02»
H2, H20, SOX, NOX, hydrocarbons, trace metals, and particulate matter. Most
systems combust natural gas or oil above a coal/ore mixture to maintain the
desired temperature. Two systems use coal both as the reductant and as the
combustion fuel; i.e., Azcon Corporation's Hoskin process reactors in
Australia and Rockwood, TN. The effluent gas phase composition from rotary
kiln reactors is very dependent on the coal chemistry and mineralogy.
Stringent requirements must be imposed on the coal to limit both the sulfur
and trace metal content. The particulate matter from these reactors is
usually high in flyash and char. Some systems combust the exit gas in an
afterburner chamber to reduce the concentration of CO, H2, char, and hydro-
carbons in the exit streams. The afterburner is normally followed by a
wet scrubber device. The collected scrubber product may be dewatered and
the solids pelletized or briquetted and recycled to the kiln, but usually
they are simply disposed of in solid waste areas.
The product from the kiln is screened and the iron is magnetically
32
-------
separated from the char and other waste solids. The other waste solids are
calcium sulfate (if the feed contains some sulfur, it can be controlled by
adding lime to the charge) and a slag-like material from the coal The
larger char material is recovered and recycled to the kiln. The fine char
material is disposed of along with the slag and gypsum in solid waste storage
areas.
The potential pollution problems are:
1. The survival of PAH's formed in the coal-ore bed. It has been well
documented that PAH materiel is created and/or evolved during pyrolysis and
combustion of coal (50-62). Coal fired heating and power generating plants
are estimated to contribute over 30 percent of the BaP (Benzo-alpha-pyrene,
a member of the Polynuclear Aromatic Hydrocarbon group of chemicals that is
known to be carcinogenic) emitted to the United States air environment each
year (62), and over 45 percent to the world's air. It is, therefore, reason-
able to suspect the rotary kiln processes of being potential Polynuclear
Aromatic Hydrocarbon emitters.
The gas phase above the rotary kiln bed is normally oxidizing and will,
therefore, be expected to combust the organic material if the retention
times are long enough. The entrance region of the kiln appears to be an
ideal place for the formation of PAH compounds; i.e., the material is being
preheated and volatile constituents are being evolved; soot is also being
evolved (in effect the material is smoldering). The PAH compounds evolved
into the gas phase will be oxidized if the retention times in the oxidizing
zone above the bed are sufficiently long. However, near the feed end of the
kiln the retention times will be very short. Also, it appears likely that
PAH compounds readily adsorb on fine particulate matter, in particular soot,
and that the PAH compounds are protected and survive if the retention times
are short.
2. The survival of PAH's from the combustion of oil and coal above the
coal/ore mixture. The survival of organic compounds evolved during the com-
bustion of the fuel is expected to be very low. The kiln gas temperature
(32) varies from approximately 1200°C at the burner end to 500°C at the gas
exit end. An exposure at greater than 900°C for at least 0.3 sec is neces-
sary for destruction of the PAH material. The gas flow rates generally allow
an average residence time in the reactor of from 5 to 9 sec. Therefore, the
destruction of most combustion formed PAH material is expected to occur
during steady state operation. However, emission of PAH compounds during
start-up and shut-down periods is expected to be considerable.
As stated earlier, some rotary kiln processes treat the kiln gases in
an afterburner and then scrub the gas in wet scrubbers. Therefore, the PAH
material that survives the combustion is most likely collected in the scrub-
ber sludge. Its fate in subsequent processing or in solid disposal sites is
uncertain at this time.
The Minimum Acute Toxicity Effluent (63,64) value for BaP disposal to
land is 0.006 yg/g. This is not a regulation-restricted emission value but
indicates minimum concentrations suspected of potential health effects. One
33
-------
should be aware that this value is considered controversial because the natu-
ral background of BaP in soils and sediments is often quoted to be much
greater (65-67).
3. Trace element and sulfur oxide emissions from the coal fired react-
ors. Reactors (only one in U.S. at present) that combust coal above the
surface of a coal/ore mixture emit some of the sulfur and most of the trace
elements from the rotary kiln. A portion of these elements are removed by
the attached control devices.
The sulfur content in the feed coal is specified to be low; i.e.,
usually less than 0.5 percent. Example calculations using the operating data
presented by Cassidy and MacKay (32) for a coal fired rotary kiln (assuming
0.5 percent sulfur in the feed coal and a heating value of 10,000 Btu/lb
coal) yield a sulfur emission factor of 0.5 Ib sulfur/MM Btu. In terms of
sulfur concentration, emission is 0.6 grams per cubic meter. New Source
Performance Standards (NSPS) have not been considered for direct reduction
reactors. However, if we assume that regulations similar to the proposed
NSPS for electric utility steam generating units (68) apply to a coal fired
direct reduction reactor, then emission levels above 0.2 Ib sulfur/MM Btu
would have to be controlled to meet an 85 percent reduction in input sulfur
content. About 20 percent of the inlet sulfur will be associated with the
slag phase (69) exiting the reactor and should be effectively removed from
the gas stream by most particulate control devices. A portion of the sulfur
will be "gettered" from the gas stream while still in the reactor by lime in
the ore bed. The fraction extracted by this mechanism cannot be estimated
because the reaction is dependent on such kinetic factors as efficient gas/
solid contact and residence time. It is anticipated, however, that sulfur
oxide emissions can be controlled without special add-on sulfur removal
equipment.
Sulfur content in the coal used as a reductant in the coal/ore bed has
been shown to be effectively controlled by the addition of lime or dolomite
to the charge mixture (32). Gypsum is formed and can be separated from the
sponge iron and disposed of in a solid landfill site.
The partitioning behavior of trace elements during coal combustion has
been studied (59). The fate of trace elements falls into three general
classes: those normally partitioned to the slag phase are Al, Ba, Ca, Ce,
Co, Eu, Fe, Hf, K, La, Mg, Mn, Rb, Sc, Si, Sm, Sr, Ta, Th, and Ti; those
normally partitioned to the fly ash are As, Cd, Cu, Ga, Pb, Sb, Se, and Zn;
and those normally remaining in the gas phase are Hg, Cl, and Br. As and Cd
are usually more concentrated in the fly ash particle sizes less than 10
micrometers, so concentration occurs in the size fractions that are least
effectively removed by control devices. The concentration of trace elements
varies in coals and the release of trace elements during combustion varies
greatly (60,61). It is certain, however, that great care should be exercised
in designing a control system to include good removal of fly ash particles
which, most likely, will contain toxic material. National Emission Stand-
ards for Hazardous Air Pollutants (NESHAP) are being considered for As, Cd,
and Pb (59). Proposed standards have been generated for Be (63).
34
-------
For example, if one assumes that the concentrations of trace metals
given by Thompson and Harrison (59) are representative of coal fuels and
that values for combustion coal feed rates given by Cassidy and MacKay (32)
are appropriate, then uncontrolled emissions of As, Cd, Pb, and Be can be
ca I cu I ated .
The concentrations of trace metals in low sulfur coals are:
As 9.3 - 13 ppm
Cd 0.1 - 2.4 ppm
Pb 8.3-12 ppm
Be 1.2 - 2.8 ppm
The process feed rates are:
Combustion coal 0.70 tonnes/hr
Air (At Standard Conditions 4980 m3/hr
0°C, 1 Atm.)
The calculated trace metal uncontrolled emission values are:
As 1085 - 1517
Cd 12 - 280
Pb 968 - 1400
Be 140 - 326
Control devices will decrease these concentration levels, and dispersion
dilution will further decrease the concentration of trace elements emitted to
the environment. The NESHAP proposed standard for Be is 0.01 yg/m3 as the
outplant concentration maximum; therefore, the effect of the control device
and dispersion dilution must be to decrease the Be concentration at least
14,000 times.
Recommendations
The gas reductant systems appear to be environmentally clean processes,
particularly those that recycle the reductant gas. It is recommended that
samples of scrubber effluents be collected and characterized at least through
the E.P.A. Source Assessment Sampling System (S.A.S.S.) Level 1 analytical
scheme.
The rotary kiln solid reductant systems are of particular interest
because of the large U.S. coal reserves. All systems that involve coal and
oil combustion have been shown to be potential PAH, sulfur oxide, and trace
35
-------
metal emitters. It is recommended that an initial sampling be made to ascer-
tain the uncontrolled and controlled emission rates for PAH, SOX, and trace
metals. As an initial effort, a preliminary sampling program should be
performed to provide mass balances on these effluents from the rotary kiln
and scrubber system.
The only coal fired rotary kiln process in the United States is the
Azcon reactor at Rockwood, TN. At present it has neither an afterburner
nor scrubber. Sampling of this system for kiln emissions would be desirable.
36
-------
SECTION 4
STEELMAKING USING METALLIZED PRODUCT
Direct Reduction - Electric Arc Furnace (DR-EAF) Steelmakino
The major use of sponge iron produced by direct reduction processes is
/5A!?"6 ?f 1>0n in electr1c arc steelmaking furnaces. Two mini mills
114,16) ln the U.S., Georgetown Steel and Oregon Steel, use a substantial
fraction of sponge iron in their charge. Both use a charge of sponge iron
and scrap, about 50 percent of each.
If the optimistic projections hold that 10 percent (100 million ton-
nes) of the world's iron will be sponge iron (7) by 1985, and that 30 percent
(300 million tonnes) of the world's steel will be produced by the electric
arc steelmaking process (70) significant further growth in the DR-EAF
process systems can be anticipated.
The literature quotes a number of advantages for the DR-EAF combination
process. Several quoted advantages are presented below:
1. A variety of excellent processes have been developed and proved in
commercial operation. These are based upon the moving-bed shaft
furnace, the fixed shaft furnace, fluidized beds and the rotary
kiln. All represent good technology for tonnages of 900 to 1800
tonne/day. Additional plants can be added as needed, and they can
be used with no initial overcapacity as would occur for a modern
blast furnace-coke plant-BOF complex (5).
2. For socio-economic reasons, developing nations desire domestic
steel industries; limited initial tonnages are the guideline and
direct reduction provides moderate scale production at moderate
capital investment. Moreover, many of these nations having
reserves of high-grade iron ore are deficient in metallurgical
coals, but have abundant supplies of natural gas and petroleum.
These latter forms of reductant and energy can be used in -only
limited quantities in blast furnaces but are ideal for many
direct reduction processes (5).
3. In developed nations, opportunities occur for mini- and midi-steel
mills with direct reduction product supplementing local scrap.
These applications will be hampered in the future by high costs and
shortages of natural gas or petroleum from which reducing gases can
be produced, but processes for economically gasifying coals will
37
-------
doubtlessly be a favorable future factor (5).
4. The gross consumption of energy per ton of liquid steel is compara-
ble for the method of combining direct reduction and the electric
furnace with the blast furnace and the basic oxygen furnace (7).
5. The time required for commissioning a DR-EAF steel plant is about
two to three years compared to five to seven years for an integrated
steelworks (9).
6. Miller summarized a report by the World Bank comparing DR-EAF to
the blast furnace-basic oxygen furnace (BF-BOF):
a. The quality of common steels produced by DR-EAF and BF-BOF
practices was approximately equal;
b. The estimated capital costs of facilities for pelletizing, iron-
making, steelmaking, and continuous slab casting of 3-4 million
tonnes/year at Matanzas favor a DR-EAF plant by nearly 40 per-
cent over a comparable BF-BOF installation;
c. The production cost, excluding fixed charges and income taxes,
for a tonne of carbon steel slabs is approximately 20% lower
when produced by a DR-EAF operation than when made by the BF-BOF
route;
d. The average rate of return on investment is from 2.5 to 3 times
as great for the DR-EAF as for the BF-BOF (4).
7. Korf extols other attributes of DR-EAF systems with conventional
BF-BOF steelmaking. Steel production at an EAF plant becomes
independent of scrap, a raw material that is volatile in pricing,
quality and supply (21).
8. A number of qualities are quoted for the use of sponge iron (versus
scrap) as a feed material to electric arc furnaces (71):
a. chemical composition is known exactly
b. chemical composition is uniform
c. contains no undesirable impurities
d. permits dilution with low cost (more available) scrap
e. easy to transport and handle
f. permits automatic continuous charging
g. increases furnace productivity
h. less noise during melt down
38
-------
i. more predictable price structure.
9. Air pollution problems are minimized in a closed cycle gas direct
reduction plant because all gases are cleaned and recycled. Effect-
ive control of electric arc furnace steelmaking is possible but care
must be exercised to assure a nonpolluting operation.
The major disadvantage normally quoted for the direct reduction pro-
cesses is that they are limited to small production units, whereas blast
furnaces are high production units. This argument has been disputed by
Dancy (72) and others: "To put the size of this (Midrex) direct reduction
plant into perspective in relation to modern blast furnaces, it will produce
about 8.0 tonnes/day/m3 of working volume (bustle to stockline) as compared
with only 2.4 tonnes/day/m3 of a modern 5000 m3 blast furnace. Furthermore,
its production of over 1800 tonnes/day is only a little less than that of
the 'average1 blast furnace in Japan as recently as 1967 and about the same
as that of the 'average1 blast furnace in the USA as recently as 1975."
A second disadvantage for direct reduction processes is that they cannot
treat a wide variety of ore types, as can the blast furnace. However, direct
reduction processes should not be thought of as a means of eliminating con-
ventional iron and steelmaking processes, but a supplement for special
applications.
Other Uses of Metallized Product
The major use of metallized product is as a feed material to the elect-
ric arc steelmaking furnace, as noted in the previous section. Other uses
have been proposed and tested on, at least, a pilot plant scale. These
suggested uses are:
a. To increase production and decrease coke requirements in a blast
furnace. Miller (4) notes that the literature (73) shows that
blast furnace productivity is increased and coke rate decreased by
5-6 percent for each 10 percent increase in furnace burden metal-
lization.
b. To serve as a coolant source in the BOF process for temperature
control.
c To replace the use of scrap in EAF steelmaking processes, in cast
iron foundry electric furnaces (73) and cupolas (74,75), and in the
BOF process.
Miller (4) proposes that every integrated steelworks should have a
direct reduction capability. Its product could be used in one of the ways
listed above. Its utility would be to supply prereduced iron wherever it
was required. Its source of feedstock would be steelplant fines, dusts from
pollution control devices, and supplemental ore. Its reductant would be
Ske oven gas, blast furnace gas, gasified coal gas or, if a rotary kiln
process, coke, coal, or coke breeze.
39
-------
REFERENCES
1. Kail a, U. and R. Steffen, "Direct Reduction: Progress and Plans", Iron
and Steel Int.. V. 50, No. 5, pp 307-315, October 1977.
2. Greenwalt, R. and J. Stephensen, "The Role of Agglomeration in Direct
Reduction Processes", Agglomeration 1977, AIME, V. 2, pp 765-784, 1977.
3. Miller, J. R., "The Inevitable Magnitudes of Metallized Iron Ore", Iron
and Steel Engr.. V. 49, No. 12, December 1972.
4. Miller, J. R., "Use of Direct-Reduced Iron Ore and Balanced Integrated
Iron and Steel Operations", Ironmaking and Steel making, V. 4, No. 5,
pp 257-264, May 1977.
5. Davis, W. L., Jr., "Hicap Direct Reduction Process", Iron and Steel
Engr., V. 55, No. 3, pp 42-50, March 1978.
6. Personal Communications with W. Griffins, Heel a Mining Company,
September 1978.
7. Miller, J. R., "The Direct Reduction of Iron Ore", Scientific Am..
V. 235, No. 1, pp 68-138, July 1976.
8. Miller, J. R., "Global Status of Direct Reduction", Iron and Steel Engr..
V. 54, No. 9, pp 45-50, September 1977.
9. Staff, "Sidbec-Dosco Scales Up and Captures Top-Tonnage Title in Direct
Reduction", 33 Metal Producing, pp 40-43, December 1978.
10. Thorton, D. S., "Requirements of Scrap for Steelmaking and the Increased
Use of Scrap Processing", Iron and Steel Int., V. 50, pp 221-225,
October 1977.
11. Bertram, J. M., "What-How-Who-Where - Direct Reduction", Iron and Steel
Engr., V. 49, No. 7, pp 31-40, July 1972.
12. Midrex Brochure, 1979, Midrex Corporation, Charlotte, NC.
13. Wiliars, H. M. and R. C. Madden, "The Development of a New Steelmaking
Process - Utilising Highly Metallised Sponge Iron", Iron and Steel Int.,
V. 48, No. 4, pp 313-321, August 1975.
40
-------
14. Schroer, C A , "Operating a Midrex Direct Reduction Plant - Current
State of the Art", Iron and Steel Enq.. V. 53, No. 8, pp 21-25, August
I -/ / 0
15. Pietsch, W., "The Midrex Cold Briquetting System: An Economic Answer
to Direct Reduction Iron Fines Recovery", Iron and Steel Int , V 51
No. 2, pp 119-124, April 1978. ~ '
16. Jinks, W. 0., "Oregon Steel Mills - An Innovator in the Steel Industry"
Iron and Steel Enq., V. 54, No. 7, pp 29-33, July 1978.
17. Coyne, T. J., Jr., R. L. Hunter, and D. J. Werner, "An Update of Sidbec
II1 , SEAISI Symposium. Manila, Philippines, October 1978.
18. Caradine, T. R. and H. J. Klingelhofer, "Status of Midrex Direct
Reduction Plants in Latin America", Latin American Iron and Steel
Congress on Direct Reduction, Macuto, Venezuela, July 1977.
19. Coyne, T. J., R. L. Hunter, and D. J. Werner, "An Update of Sidbec II:
The First Midrex Series 600 Module", SEAISI Symposium. Manila,
Philippines, October 9-13, 1978.
20. Goodman, R. J., "Direct Reduction Processing in Canada - Its Status and
Future", UN Seminar, Bucharest, Romania, May 24-28, 1976.
21. Dayton, S., "Direct Reduction", EMJ. pp 80-84, January 1979.
22. Labee, C. J., "HYLSA's Puebla 2 - A Fantastic Success", Iron and Steel
Eng.. V. 54, No. 10, pp 26-29, October 1977.
23. Quintero, R. G., "Direct Reduction - The HyL Boom", Iron and Steel Int..
V. 47, pp 437-440, December 1975.
24. Rodriguez, F. A., "Concepts Relevant to Steelmaking with HyL Metallized
Pellets", Iron and Steel Eng.. V. 54, No. 1, pp 57-60, January 1977.
25. Kono, T., "Consumption and Conservation of Energy In The Japanese Steel
Industry", Iron and Steel Int., V. 50, pp 87-92, April 1977.
26. Staff, "First Industrial Application of Thyssen Purofer Direct Reduction
Process", Iron and Steel Eng., V. 54, No. 8, pp 87-88, August 1977.
27. Labee, C. J., "Armco's Direct Reduction Facility Operating at Full
Production", Iron and Steel Eng., V. 51, No. 11, pp 73-75, November 1974.
28. McGannon, H. E., Ed., "The Making, Shaping, and Treating of Steel",
Ninth Edition, United States Steel. 1971.
29 Oehlberg, R. J., "FIOR Process for Direct Reduction of Iron Ore", Iron
and Steel Eng.. V. 51, No. 4, pp 58-60, April 1974.
41
-------
30. Murdock, C. H., R. Littlewood, "The SL/RN Direct Reduction Plant at
Griffith Mine", Second Latin American Seminar on Direct Reduction",
Porto Alegre, Brazil, May 4-9, 1975.
31. Bold, D. A. and N. T. Evans, "Direct Reduction Down Under: The New
Zealand Story", Iron and Steel Int.. V. 50, No. 3, pp 145-156, July 1977.
32. Cassidy, P. W. and J. M. MacKay, "Development of the Hockin Process and
Its Application to the Direct Reduction of Iron Ore", Inst. Min. and
Met., Eleventh Commonwealth Mining and Metallurgical Congress, Paper 62,
May 1978.
33. Dailey, W. H., "Integrated Steel Plants of the Future", Iron and Steel
Eng:. V. 49, No. 4, pp 87-94, April 1972.
34. Staff, Direktreduktion von Eisenerz, Eine Bibliographisch Studie im
Auftrag der Kommission der Europaischen Gemeinschaften, Verlag
Stahleisen, M.B.H., Dusseldorf, Germany, 1976.
35. Rierson, D. W. and A. A. Albert, Jr., "Development of the Accar Process
at Allis-Chalmers, Company Report, All is Chalmers, Milwaukee, Wisconsin.
36. Ferrari, R. and F. Colautti, "The Kinglor Metor Process - Direct
Reduction Using a Solid Reducing Agent", Iron and Steel Eng., V. 52,
No. 5, pp 57-60, May 1975.
37. Meyer, G. and U. Bongers, "Reduction by Solid Fuels with the Krupp
Sponge Iron Process", 2nd Latin American Seminar on Direct Reduction,
Porto Alegre, Brazil, May 4-9, 1975.
38. Staff, "Krupp Waste Recovery Process", Metal Bulletin Monthly, p 31,
February 1977.
39. Ferrari, R. and F. Colantii, "The Kinglor Metor Process - Direct
Reduction Using a Solid Reducing Agent", Iron and Steel Eng., V. 51,
No. 5, pp 57-60, May 1975.
40. Field, L. I., "The Impact of Energy Conservation", Iron and Steelmaker.
V. 6, pp 8-16, January 1979.
41. Office Technology Assessment, O.T.A. Seminar on New Techniques in
Steelmaking, Washington, D. C., May 2-3, 1979.
42. Hayes, E. T., "Energy Resources Available to the United States, 1985 to
2000", Science, V. 203, January 1979.
43. Reuter, G., "Rotary Kiln Direct Reduction Plants Built by Lurgi", O.T.A.
Seminar on New Techniques in Steelmaking, Paper No. 3, Washington D.C.,
May 2, 1979.
42
-------
44. Lepinski, J. A., "Raw Material and Product Flexibility of the ACCAR
System for Direct Reduction of Iron Ores", O.T.A. Seminar on New
Techniques in Steelmaking. Paper No. 1, Washington D.C., May 2, 1979.
45. Gold, B., "Tracing Gaps Between Expectations and Results of Technologi-
cal Innovations: The Case of Iron and Steel", J. Ind. Econ., V 25
No. 1, pp 1-28, September 1976.
46. Hirshhorn, J., et al., "Impact of Technology on the International
Competitiveness of the United States Steel Industry", O.T.A., Washington,
D.C., September 1979.
47. Johnston, H., "Oil Gasification is Teamed with Iron-Ore Reduction," Chem.
Engr., pp 124-125, March 26, 1979.
48. Ayer, F. A., Symposium Proceedings: Environmental Aspects of Fuel
Conversion Technology, II. EPA-600/2-76-149, NTIS No. PB 257-182,
USEPA, RTP, NC, June 1976.
49. Midwest Research Institute, Emissions from Iron Ore Mining. Beneficia-
tion. and Pelletizing. U.S.E.P.A., Industrial Environmental Research
Laboratory, RTP, NC, Contract 68-02-2113, Draft Report, March 1979.
50. Crittenden, B. D. and R. Long, "The Mechanism of Formation of Polynuclear
Aromatic Compounds in Combustion Systems", Carcinogenesis, V. 1, Edited
by R. I. Freudenthal and C. W. Jones, Raven Press, New York, N.Y.,
pp 209-223, 1976.
51. Herlan, A., "On the Formation of Polycyclic Aromatics: Investigation of
Fuel Oil and Emissions by High Resolution Mass Spec.", Combustion and
Flame. V. 31, No. 3, pp 297-308, 1978.
52. Blumer, M., "Polycyclic Aromatic Compounds in Nature", Sci. Am.. V. 234,
pp 34-45, 1976.
53. Bridbord, K., J. Finklea, J. Wagoner, J. Moran, and P. Caplan, "Human
Exposure to Polynuclear Aromatic Hydrocarbons", Carcinogenesis. V. 1,
pp 319-324.
54. PEDCO Environmental, "Assessment of Arsenic and Other Trace Contaminant
Emissions from Emerging Energy Technologies", Draft Report, U.S.E.P.A.,
Strategies and Air Standards Division, Research Triangle Park, NC,
Contract No. 68-02-2515.
55 Committee on Medical and Biologic Effects of Environmental Pollutants,
Vapor-Phase Organic Pollutants, Volatile Hydrocarbons and Oxidation
Products, National Academy of Sciences, Washington, D.C., 1976.
56. Salvesen, K. G., K. J. Wolfe, E. Chu, and M. A Herther, Emission
Characterization of Stationary NOX Sources: Volume I. Results, EPA-
600/7-78-120a, NTIS No. PB284-b20, USEPA, RTP, NC, June 19/8.
43
-------
57 Ibid., Volume 2. EPA-600/7-78-120b, NTIS No. PB285-429, June 1978.
58. PEDCo Environmental, Assessment of Arsenic and Other Trace Contaminant
Emissions from Emerging Energy Technologies, EPA Contract 68-02-2515,
Task Assignment No. 19, Draft Report, September 1978.
59. Thompson, W. E. and J. W. Harrison, Survey of Projects Concerning
Conventional Combustion Environmental Assessments, EPA-600/7-78-139,
NTIS No. PB285-188, USEPA, RTP, NC, July 1978.
60. Maloney, K. L., 6. L. Moilanen and P. L, Langsjoen, Low-Sulfur Western
Coal Use in Existing Small and Intermediate Size BoiTers. EPA-600/7-78-
153a, NTIS No. PB287-937, USEPA, RTP, NC, July 1978.
61. Colley, J. D., C. A. Muela, M. L. Owen, N. P. Meserole, J. B. Riggs, and
J. C. Terry, Assessment of Technology for Control of Toxic Effluents
From the Electric Utility Industry, EPA-600/7-78-090, NTIS No. PB283-716,
USEPA, RTP, NC, June 1978.
62. Suess, M. J., "Th6 Environmental Load and Cycle of Polycyclic Aromatic
Hydrocarbons", The Science of the Total Environment. V. 6, pp 239-250,
1976.
63. Cleland, J. 6. and G. L. Kingsbury, Multimedia Environmental Goals for
Environmental Assessment, Volume 1, EPA-600/7-77-136a, NTIS No. PB276-
919, USEPA, RTP, NC, November 1977.
64. Ibid.. Volume II, MEG Charts. EPA-600/7-77-136b, NTIS No. PB276-920.
USEPA, RTP, NC, November 1977.
65. Hites, R. A. and R. E. LaFlamme, "Sedimentary Polycyclic Aromatic Hydro-
carbons: The Historical Record", Science, V. 198, pp 829-831, November
1977.
66. Blumer, M., W. Blumer and T. Reich, "Polycyclic Aromatic Hydrocarbons
in Soils of a Mountain Valley: Correlation with Highway Traffic and
Cancer Incidence", Env. Sci. and Tech., V. 11, No. 12, pp 1082-1084,
November 1977.
67. Hites, R. A. and J. B. Howard, Combustion Research on Characterization
of Particulate Organic Matter from Flames, EPA-600/7-78-167, NTIS No.
PB291-314, USEPA, RTP, NC, August 1978.
68. Dunlap, R. W. and B. J. Goldsmith, "NSPS: Critique of Proposed Rule-
making", Env. Sci. and Tech., V. 13, No. 2, pp 172-178, February 1979.
69. Davis, J. A., Screening Study on Cupolas and Electric Furnaces in Gray
Iron Foundries, EPA Contract 68-01-0611, Task No. 8, Draft Report,
August 1975.
44
-------
70. Secretariat, "An Appraisal of Some of the Direct Reduction Processes
for the Production of Sponge Iron", Third Interregional Symposium on
the Iron and Steel Industry, Brasilia, Brazil, October 14-21, 1973.
71. Brown, J. W., "Direct Reduction - What Does It Mean to the Steelmaker",
Iron and Steel Eng., V. 54, No. 6, pp 37-46, June 1976.
72. Dancy, T., "The Evolution of Iron Making", Met. Trans., AIME, June 1977.
73. Maschlanka, W., G. Post and E. Eisner, "Utilization of Direct Reduced
Iron In Different Iron and Steel Production Processes", Int. Iron and
Steel Congress, Chicago, IL, April 1978.
74. Geek, H. G. and W. Maschlanka, "Use of Midrex Sponge Iron in the Induc-
tion and Cupolas Furnaces of Foundries", Sem. on the Utilization of
Pre-Reduced Materials in Iron and Steel Making, Bucharest, Romania,
May 24-28, 1976.
75. Pietsch, W. G. and R. P. Kreimendahl, "Use of Direct Reduced Iron in
Ironmaking", 2nd ILAFA Direct Reduction Conference, Macuto, Venezuela,
July 1977.
76. Editor, "First Industrial Application of Thyssen Purofer Direct Reduction
Process", Iron and Steel Eng., V. 54, No. 8, pp 87-88, August 1977.
45
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-80-036
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Direct Reduction: A Review of Commercial Processes
5. REPORT DATE
January 1980
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Larry G. Twidwell
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
See Block 12.
1AB604
11. CONTRACT/GRANT NO.
NA
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PEfllOD COVERED
Tnhouse Final; 1-7/79
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES Author Twidwell is no longer with the EPA. For report details,
contact Korman Plaks, Mail Drop 62, 919/541-2733.
. ABSTRACT
repOr^ gjves results of a review of direct reduction commercial pro-
cesses for ironmaking. The potential for environmental degradation appears to be
minimal. A detailed environmental assessment does not appear to be warranted. It is
recommended that samples of scrubber water and sludge material be collected and
characterized from several gas reductant reactor systems. It is also recommended
that rotary-kiln solid-reductant reactors be sampled to ensure that the final gas
effluent does not contain harmful concentrations of organic species, sulfur oxides,
and trace element contaminants; and that such systems be observed to ascertain that
fugitive emissions are not released from the kiln at harmful rates.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
Pollution
Iron and Steel Industry
Reduction
Scrubbers
Sludge
Kilns
Pollution Control
Stationary Sources
Ironmaking
Direct Reduction Pro-
cesses
13B
11F
07B,07C
07A,13I
13A
8. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
50
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-46-
------- |