EPA-600/2-77-044
February 1977
Environmental Protection Technology Series
ENVIRONMENTAL ASSESSMENT OF STEELMAKING
FURNACE DUST DISPOSAL METHODS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park,. North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental 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-600/2-77-044
February 1977
ENVIRONMENTAL ASSESSMENT
OF STEELMAKING FURNACE DUST
DISPOSAL METHODS
by
George E. Weant, ni and M.R. Overcash
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
Contract No. 68-02-1325, Task 61
Program Element No. 1AB604
EPA Task Officer: Robert V. Hendriks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
TABLE OF CONTENTS
FIGURES vi
TABLES viii
ACKNOWLEDGMENTS x
1. Introduction , , . , 1
2. Conclusions ................... 2
3. Recommendations . ,,,.... 3
4. Characterization of the Problem .............. 4
4.1 Sources of Residue 4
4.2 Quantities of Residue 4
4.2.1 Estimation of 1974 Residue Generation 8
4.2.2 Estimation of Future Residue Quantities .... 8
4.3 Residue Characterization ... 13
4.3.1 Particle Identification 14
4.3.1.1 Visual Inspection and Light Microscopy 14
4.3.1.2 Electron Microscopy .... 16
4.3.1.2.1 Open Hearth Furnace Samples 16
4.3.1.2.2 Electric Furnace Samples 19
4.3.1.2.3 Basic Oxygen Furnace Samples 29
4.3.2 Specific Gravity ............... 32
4.3.3 Particle Size Analysis ..... 32
4.3.4 Chemical Analysis 38
4.3.5 Source of Dust Constituents 40
4.3.5.1 Ores 41
4.3.5.2 Scrap 42
4.3.5.3 Additives 43
5. Sol utilization Tests 45
5.1 Procedures and Results . 45
5.2 Discussion 60
iv
-------
TABLE OF CONTENTS (cont'd)
6. Handling and Disposal of Steelmaking Furnace Residues ... 77
6.1 Handling 77
6.2 Disposal 77
6.2.1 Waste Heaping 78
6.2.2 Residue Utilization 78
6.2.3 Recycling 78
6.3 Regional Recycling Facilities 79
7. Assessment of the Enviromental Impact of Disposal Practices 82
7.1 Pathways for Pollutant Movement 83
7.1.1 Air Transport 84
7.1.2 Surface Runoff 87
7.1.3 Subsurface Migration 91
7.2 Pollutant Interactions 94
7.2.1 Metal-Water Interactions 94
7.2.2 Metal-Soil Interactions 96
7.2.3 Metal-Plant Interactions 99
7.3 Specific Disposal Cases 100
7.3.1 Heaps 100
7.3.2 Gullies or Natural Depressions 105
7.3.2.1 Mill Waste Placed in Gullies or Natural
Depressions with No Retention Structure 105
7.3.2.2 Mill Waste Placed in Gullies or Natural
Depressions with a Retention Structure 106
7.3.2.3 Determination of Effects 108
7.3.3 Lagoons 108
7.3.4 Landfill 109
8. References Ill
APPENDICES
A. Specific Gravity Determinations 114
B. Particle Size Analysis 116
C. Spark Source Mass Spectrometry Determinations 119
D. Sol utilization Test Procedure 121
-------
FIGURES
Figure Page
1 Geographical distribution of open hearth furnaces 5
. 2 Geographical distribution of electric furnaces ....... 6
3 Geographical distribution of basic oxygen furnaces 7
4 Total steel production projected through 1985 . 10
5 Steel production by furnace types projected through 1985 . . 11
6 Examination of the No. 2 open hearth furnace sample ..... 17
7 Examination of the carbon flakes in the No. 2 open hearth
furnace sample . . . .. ........... 18
8 Electron microscopy investigation of the No. 14 open hearth
furnace sample 20
9 Further characterization of the No. 14 open hearth furnace
samplie „ 21
10 Electron microscopy of the No. 18 electric furnace sample . . 23
11 Electron micrographs and X-Ray of the No. 5 electric furnace
sample 24
12 Results of the examination of the spherical particles in the
No. 5 electric furnace sample 25
13 Examination of carbon flake at 450x . 26
14 Titanium cube at 6000x ................... 27
15 Examination of carbon flake at 9000x ..... 28
16 Electron microscopy examination of the No. 11 BOF furnace
sample 30
17 Examination of the No. 10 BOF furnace sample 31
18 Photo and iron map of carbon flake and adhering iron agglo-
merates in the No. 10 sample 33
19 The relationship between the quality of cobalt extracted and
the quantity of cobalt in the residues 62
20 The relationship between the quantity of manganese extracted
and the quantity of manganese in the residue 63
21 The relationship between the quantity of antimony extracted
and the quantity of antimony in the residue 64
22 Solubility of zinc compounds ..... 64
vi
-------
FIGURES (cont'd)
Figure Page
23 Relationship between soluble zinc and pH 65
24 The relationship between soluble zinc and pH versus zinc
solubility constraints 66
25 Relationships of-soluble zinc as a function of pH after two
contact periods versus solubility constraints 67
26 Relationships between soluble manganese and pH 69
27 The effects of pH on the solubilization of manganese .... 70
28 The relationship of soluble lead with pH 71
29 The relationship of soluble cobalt with pH 72
30 The relationship between soluble chromium and pH 74
31 The relationship between soluble cadmium and pH 75
32 The relationship between the quantities of calcium in the
residues and their effects on pH 76
33 Potential evapotranspiration vs. mean annual precipitation
(inches) 93
34 Migration of pollutants to groundwater 101
35 Lateral movement of pollutants from waste heaps to surface
water 102
36 Subsurface movement of pollutants from darned waste heaps . . 107
VI1
-------
TABLES
able
^^^^^•^••••B
1
2
3
4
5'
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Estimation of 1974 Steelmaking Furnace Residues
Estimation of Steel Production, Residue Generation, and
Disposed Residues for 1974 Through 1985
Furnace Type and Location and Pollution Control Equipment
for 18 Samples Collected
Results of Visual Inspection and Light Microscopy of Dust
Samples ......... ......
Summary of Open Hearth Furnace Particle Characteristics . . .
Characteristics of Electric Furnace Samples ....
Major Characteristics of Basic Oxygen Furnace Samples ....
Specific Gravity of Residue Samples
Particle Size Analyses (Percent) . .
Results of SSMS Analysis of Steelmaking Residues (yg/g) . . .
Elemental Composition of Steelmaking Dusts Based on Reported
Values
Probable Sources of Substances Comprising Steelmaking Dusts .
1974 Consumption of Scrap and Pig Iron by Iron and Steel
Furnaces
Additives and Their Uses
1974 Consumption of Fluxes .................
Solubilization Data for Antimony ....
Solubilization Data for Cadmium
Solubilization Data for Chromium
Solubilization Data for Cobalt . .
Solubilization Data for Lead ....
Solubilization Data for Manganese
Solubilization Data for Inorganic Mercury
Solubilization Data for Total Mercury .... . .
Solubilization Data for Nickel ...
Solubilization Data for Selenium . .
Page
8
12
13
14
16
19
29
32
35
39
40
41
43
43
44
46
47
48
49
50
51
52
53
54
55
viii
-------
TABLES (cont'd)
Tables Page
26 Solubilization Data for Vanadium 56
27 Solubilization Data for Zinc 57
28 Effects on pH of Leaching Tests 58
29 Results of Analyses Performed on Second Set of Leachates ... 59
30 Percentage of Total Metals Extracted from Steelmaking Furnace
Residues 61
31 Estimates of Residue Volumes 82
32 Estimates of Potential for Environmental Impacts 83
33 Relationships between Detachment and Transport of Soil
Particles by Rain 88
34 Water Solubility of Metals from the Airborne Dust of 18 Steel
Mills 95
35 Elemental Composition of Steel Mill Airborne Dust and of a
Mineral Soil 97
A-I Specific Gravity Estimates 115
ix
-------
ACKNOWLEDGMENTS
The authors wish to express their gratitude to a number of individuals
for their help on various aspects of this report. Particular emphasis
should be placed on the assistance received from Mr. William Benzer of
AISI and environmental control officials from many steel companies in the
U.S. for their help in acquiring residue samples for this study.
Appreciation is also expressed to the various individuals who helped
perform the residue analyses. These include Dr. R. M. Statnick and
Mr. Hunter Dougherty of EPA, IERL/RTP and Dr. V. V. Cavaroc of NCSU's
Department of Geosciences.
Special appreciation is extended to Dr. Philip Singer of UNC's
Department of Environmental Sciences and Engineering for his help in the
analysis of analytical data and Mr. Richard Jablin for his overall input
of information.
-------
SECTION 1
INTRODUCTION
Residues collected by air pollution control equipment on steelmaking
furnaces is considered a potential multimedia environmental problem. This
dust is frequently high in tramp metals and is not suitable for recycle.
Thus, the residues are discarded, creating potential environmental problems
as well as resulting in a waste of natural resources. Although recycling
methods have been developed, they are presently in limited use for economic
reasons.
The potential environmental problems associated with the residues result
g
from its quantity (estimated to be approximately 1.9 x 10 kg in 1976), its
fine particle size allowing wind erosion and entrainment, and the potential
for water extraction of metals and organic materials leading to ground and
surface water contamination. This report focuses on the potential of environ-
mental contamination by characterizing the problem, examining handling and
disposal problems, and by examining the residue .and disposal characteristics
that provide the potential for affecting the environment.
-------
SECTION 2
CONCLUSIONS
A number of conclusions can be drawn from this study.
1. The potential exists for environmental degradation from the
disposal of steelmaking furnace residues. There are many
factors that influence this potential. These include particle
size and shape, handling and disposal practices, wetness of
disposed residues, type of disposal configuration, and clima-
tological and topographical factors of each disposal site.
The variation in these factors dictate a site-by-site investi-
gation of environmental problems.
2. Generalizations of residue characteristics from a furnace type
cannot be made. Even within a furnace type, residue character-
istics have great variability. The residues show variations in
particle size and shape, consistency, and chemical composition.
3. The steelmaking furnace sources of these residues are located
in well defined areas of the U.S. Therefore, problems associated
with residue disposal are generally localized.
4. A possibility exists for the profitable operation of regional
recycling plants in several locations around the country,
including Chicago and Pittsburgh and possibly Philadelphia
and Dallas.
5. The disposal method plays an important part in producing the
potential for adverse environmental effects. A trade-off can
exist depending on the type of pollution to be avoided (i.e.,
air, surface runoff, or subsurface migration). The greatest
potential for air emissions comes from piles and the least
potential from lagoons and possibly landfills if they are
covered by a soil layer. For runoff, the greatest potential
comes from piles with impermeable bases and from filling gullies
or natural depressions without dams; the least comes from land-
fills. For subsurface migration, the most potential comes from
lagoons and gullies or natural depressions with dams with no
overflows or drains, while the least comes from piles with
impermeable bases and lined landfills. However, other factors
such as topography and climate play an appreciable role.
-------
SECTION 3
RECOMMENDATIONS
Adverse environmental impacts can result from the disposal of furnace
dusts. A detailed study of residue disposal practices should be initiated
based on the following steps:
1. A detailed assessment of the disposal steps with emphasis
on gathering information in the field on the nature of the
residues, wet and dry differences, configuration and nature
of specific residue accumulationss drainage patterns, surface
erosion, seepage patterns, etc,
2. Additional studies of the furnace residues including better
definition of particle sizes; causes for pH changes in
solutions (notably soluble iron and calcium); solubility
tests using pH buffer systems; evaluation of secondary
reactions in the water systems; evaluation of significant
pathways into the environment by examining chemical
compositions of solutes, changes in valence states,
formation of colloids, and bonding to organic species;
evaluation of particle emissions at each stage of the
handling/disposal process; and leaching tests through
soil columns.
3. Field monitoring of existing disposal sites to evaluate
disposal site configurations as an influence on effluents.
4. Evaluation of control practices and alternatives.
5. Investigations of the fate of antimony, gallium, and ger-
manium in soil-water-piant systems, since no information
exists on their fate.
-------
SECTION 4
CHARACTERIZATION OF THE PROBLEM
The disposal of residues collected by air pollution control equipment
on steelmaking furnaces presents multimedia environmental problems because
of both the quantity and the physical and chemical properties of the disposed
waste. Approximately 62 percent of the residue is collected in dry form
(from electrostatic precipitators and baghouses), while the remaining 38
percent is collected as a sludge from various types of wet scrubbers. The
basic oxygen furnace accounts for about 80 percent of the sludge produced and
the open hearth furnace for the remaining 20 percent. Sludge from electric
furnaces is essentially negligible.
The problems involved with the disposal of steelmaking residues are
characterized by examining the sources, quantities, and characteristics of
these residues.
4.1 SOURCES OF RESIDUE
Basically, three types of furnaces are used in the steelmaking process:
open hearth, electric, and basic oxygen. The geographical distribution of
these furnaces is shown in Figures 1 through 3.* Each mark designates the
location of a steel facility with at least one furnace of the type represented
by the figure.
4.2 QUANTITIES OF RESIDUE
Estimations of the total quantities of residues generated by steelmaking
furnaces are hampered by variations in the reported data on both the quantities
produced and quantities recycled and by the lag time associated with the pub-
lishing of steel production data—the latest data are from 1974.3 The reported
quantities (in kilograms of residue per 1000 kg of steel produced) vary from 11
to 20, 5.5 to 14, and 11 to 21 for open hearth, electric, and basic oxygen
*
Produced from information extracted from the Directory of Iron and Step!
of the United States and Canada. 1974.2 *~ e'
-------
en
Figure 1. Geographical distribution of open hearth furnaces,
-------
Figure 2. Geographical distribution of electric furnaces.
-------
Figure 3. Geographical distribution of basic oxygen furnaces.
-------
furnaces, respectively. The quantities recycled vary from 3 to 6 percent for
electric furnaces and 15 to 25 percent for oxygen furnaces. The quantity
recycled from open hearth furnaces is considered to be negligible. ' "
4.2.1 Estimation of 1974 Residue Generation
Residue generation for 1974 from each of the three types of steel-
making furnaces was estimated using an average of the residue generation
factors (kg/1000 kg of steel) from the reported literature and the 1974
steel production statistics. Quantities subjected to disposal were then
calculated using disposal percentages of 100, 95, and 80 for open hearth,
electric, and basic oxygen furnaces, respectively. The results are presented
in Table 1.
TABLE 1. ESTIMATION OF 1974 STEELMAKING FURNACE RESIDUES
Furnace
Type
Open Hearth
Electric
Basic Oxygen
TOTAL
Steel
Production
(109 kg)
32.2
26.0
74.0
132.2
Residue
Generation
(kg/1000 kg)
14.5
10.5
18.0
Residue
(106 kg)
466.9
273.0
1332.0
2071.9
Quantity
Recycl ed
(106 kg)
0.0
13.7
266.4
280.1
Disposed
Residue
(106 kg)
466.9
259.3
1065.6
1791.8
4.2.2 Estimation of Future Residue Quantities
The estimation of future residue quantities of disposed steelmaking
furnace residues depends on an accurate projection of future steel production,
the assumptions that residue generation rates will be constant, that new steel-
making technology will not significantly impact on current steelmaking practices,
and that recycling methods will not be widely utilized. The accurate projection
of future steel production is difficult because steel production is affected
by a multitude of economic factors related to raw materials, production, and
end uses. However, there seem to be few major developments that could
8
-------
significantly alter the current growth of the industry, and a simple linear
projection of the current growth pattern should project future trends over
the next decade with reasonable accuracy.
To accomplish this projection, a regression analysis was performed on
total steel production from 1960 through 1974 (Figure 4). The scatter lines
g
represent plus and minus 8.2 x 10 kg. The linear projection of this line
through 1985 shows an annual growth rate of approximately 2 percent, a value
consistent with demand estimates produced in one report.
Q
To meet the projected production in 1985 (165.7 x 10 kg), the steel
industry must increase capacity. Since 1960, the capacity has grown at an
average rate of 0.5 percent, well below the 2.0 percent growth in production.
The projection of steel production by furnace type involved the use of
several assumptions. Basic oxygen furnaces are rapidly replacing the open
hearths, causing a shift in the character of the industry in relation to scrap.
The industry has become an overall scrap producer rather than a scrap consumer.
This trend has been a major factor in the increased usage of the scrap
consuming electric furnace.
Many steel companies have decided to keep their open hearth furnaces in
order to maintain their scrap bal-ance. This factor has resulted in a continual
linear decline of the projection of steel production by open hearths for
several years followed by a leveling of the trend line at about 2 percent of
the total steel production.
The projection of open hearths at 2 percent of total production is,
justified by the projected steadily increasing production from electric fur-
naces, following many years of steady increases amounting to approximately 1
percent of total steel production per year. It is reasonable to assume that
this pattern will continue.
Basic oxygen steelmaking was assumed to account for the remainder of the
projected increased steel production. As the open hearth production levels
out and the electric production remains on a steady increase, the basic oxygen
production will slip slightly. These growth projections are shown in Figure 5.
Using these projections, residue generation rates, and the disposal per-
centages previously established, the estimates of disposed residue from 1974
to 1985 have been calculated and are shown in Table 2.
-------
Jf
at
O
O
£
175
170
160
ISO
140
130
o
I "20
110
100
90
1870 I <
Year
Figure 4. Total steel production projected through 1985.
10
-------
•a
o
a.
u
k.
-------
TABLE 2., ESTIMATION OF STEEL PRODUCTION, RESIDUE GENERATION, AND DISPOSED RESIDUES FOR 1974 THROUGH 1985
OPEN HEARTH
Year
1974
5
6
7
8
9
1980
1
2
3
4
5
Steel
Production
(109kg)
32.2
27.3
23.8
20.0
16.0
11.9
7.5
3.1
3.1
3.1
3.1
3.1
Residues
006kg)
467.0
396.0
344.6
289.4
231.5
172.4
109.2
44.7
44.7
44.7
44.7
44.7
Disposed
Residues
(106kg)
467.0
396.0
344.6
289.4
231.5
172.4
109.2
44.7
44.7
44.7
44.7
44.7
PROJECTED
VALUES BY FURNACE TYPES
ELECTRIC
Steel
Production
(109kq)
26.0
28.8
30.8
32.7
34.9
37.1
39.3
41.6
43.9
46.7
48.8
51.3
Residues
(106kg)
273.4
302.0
323.0
343.9
366.8
389.6
412.5
437.3
461.0
490.6
512.5
539.1
Di sposed
Residues
(106kg)
259.7
286.9
306.8
326.7
348.4
370.1
391.8
415.4
438.0
466.0
486.9
512.2
BASIC OXYGEN
Steel
Production
(109kg)
74.0 '
80.7
85.9
89.8
94.5
99.4
104.3
109.3
110.2
110.4
110.9
111.2
Residues
006kg)
1332.5
1451.7
1533.3
1616.6
1701.5
1789.7
1877.9
1917.7
1979.1
1987.3
1995.4
2002.1
Disposed
Residues
(106kg)
1066.0
1161.4
1226.7
1293.3
1361.3
1431.7
1502.3
1574.2
1583.3
1589.9
1596.4
1601.6
TOTALS
Steel
Production
(109kg)
132.2
136.7
139.6
142.5
145.4
148.3
151.2
154.0
156.9
159.8
162.8
165.7
Residues
(106kg)
2073.0
2149.7
2201.0
2250.0
2299.8
2351.7
2399.6
2449.7
2484.9
2522.7
2552.7
2586.0
Disposed
Residues
006kg)
1792.7
1844.2
1878.2
1909.4
1941.1
1974.2
2003.4
2034.3
2066.1
2100.6 *
2128.0
2158.5
-------
4.3 RESIDUE CHARACTERIZATION
To characterize the residues from steelmaking furnace air pollution
control equipment, residue samples were collected from 18 steel facilities
around the Country. The furnace type, location, and type of pollution
control equipment are presented in Table 3.
TABLE 3. FURNACE TYPE AND LOCATION AND POLLUTION CONTROL EQUIPMENT
FOR 18 SAMPLES COLLECTED
Furnace
Type
Open Hearth
Electric
•
Basic Oxygen
Sampl e
Number
1
2
3
8
T4
17
4
5
6
7
12
13
18
9
10
11
15
16
Furnace
Location
Indiana
California
Ohio
Ohio
Ohio
Pennsylvania
Pennsylvania
Texas
Pennsylvania
Pennsylvania
Pennsylvania
Illinois
Ohio
California
Indiana
Indiana
Ohio
Pennsylvania
Type of Pollution
Control Equipment*
ESP
ESP
ESP
ESP
WS
ESP
BH
BH
BH
BH
BH
BH
ESP
ESP
ESP
BH
WS
ESP
BH = Baghouse, ESP = Electrostatic Precipitator, WS = Wet Scrubber
These samples were analyzed to determine physical and chemical pro-
perties. Particle identification, size determinations, specific gravity
determinations, and elemental analyses were performed. In addition, possible
sources of elements in the occurring residue were investigated.
13
-------
4.3.1 Particle Identification
Visual inspection, light micrsocopy, and electron microscopy were
used both to characterize and to identify distinct particles in the dust
samples. Each of the 18 samples was subjected to visual inspection and
light microscopy while six arbitrarily chosen samples were examined
by electron microscopy.
4.3.1.1 Visual Inspection and Light Microscopy--
Light microscopy observations were performed at SOX and 200X.
These results are presented in Table 4. Terms describing particle size
are relative to other samples only.
TABLE 4. RESULTS OF VISUAL INSPECTION AND LIGHT MICROSCOPY OF DUST SAMPLES
Sample
Number
Color
Consistency and Description of
Distinguishable Particles
1 Rust Brown
2 Rust Brown
3 Rust Brown
4 Chocolate Brown
5 Grayish Brown
6 Chocolate Brown
7 Rust Brown
Agglomerated rust brown particles and "faceted"
spheres of pyritic-like particles
Very fine agglomerated rust brown and pyritic-
type particles—some separate irregular pyritic
particles
Agglomerated rust brown and separate irregular
pyritic particles
Extremely fine agglomerated brown particles--
white calcium particles—small pyritic colored
particles
Unagglomerated pyritic-like spheres with rust
brown particles adherring, irregular pyritic
masses—carbon flakes
Agglomerated brown particles—pyritic-1 ike parti'
cles occur as (1) irregular, angular to sub-
angular, (2) shards, and (3) spheres
Agglomerated rust brown particles—irregular,
crystalline, white calcium particles with ad-
herring red hematite particles—irregular and
spherical pyritic particles—separate red
hematite
continued
14
-------
TABLE 4. (Cont'd)
Samp]e
Number Color
Consistency and Description of
Distinguishable Particles*
8
9
10
11
12
13
14
15
16
17
18
Rust Brown
Chocolate Brown
Rust Brown
Steel Gray
Chocolate Brown
Grayish Brown
Chocolate to
Rust Brown
Steel Gray
Chocolate Brown
Rust Brown
Chocolate Brown
Agglomerated brown particles—some cubic pyrite
Agglomerated brown particles—approximately 30
percent white calcium particles with adhering
red hematite—irregular pyritic particles
Agglomerated brown particles--white, crystalline
calcium particles with adhering red hematite-
irregular pyrite
Agglomerated spherical particles
Aggomerated brown particles—small amount of
irregular, white calcium with adhering red hema-
tite—large and small, irregular, spherical, and
angular pyrite—small black chunks, possibly
magnetite
Agglomerated, extremely fine particles—small,
irregular and elongated pyritic particles-- .
black, irregular, and spherical particles, pro-
bably magnetite
Very fine agglomerated red brown and pyritic
particles—gray particles, possibly carbon
Extremely fine-grained agglomerated masses of gray
pyritic-type particles—distinct particles not
distinguishable at 200x
Agglomerated red-brown particles—extremely large
chunks of flaky material with color of pyrite but
form of carbon—crystalline, white calcium with
adhering red hematite—small irregular pyrite—
black, spherical particles, possibly magnetite
Agglomerated rust brown particles, extremely fine-
grained—small pyritic particles embedded in agglo-
merated masses—small, embedded black particles,
possibly magnetite
Very fine agglomerated rust colored particles--
white calcium and gold pyritic particles in
agglomerated masses
Terms describing particle size are relative to other samples only.
15
-------
4.3.1.2 Electron Microscopy-
Two samples from each of the three types of furnaces were
examined by electron microscopy and associated X-ray diffraction
analysis.
4.3.1.2.1 Open Hearth Furnace Samples — Both open hearth furnace samples
were chiefly composed of agglomerated, spherical iron particles. In addition,
Sample No. 2 contained irregular carbon flakes, while Sample No. 14 contained
irregular calcium particles. X-ray analysis showed that Sample No. 2 was
composed of several elements not found in Sample No. 14. These included Mg,
Si, P, Cl, and Cu. Table 5 summarizes the characteristics of the two
samples.
TABLE 5. SUMMARY OF OPEN HEARTH FURNACE PARTICLE CHARACTERISTICS
Characteristic Sample No. 2 Sample No. 14
Consistency Agglomerated masses of iron Agglomerated masses of
spheres, irregular carbon Fe/S spheres, large,
flakes irregular Ca/S particles
X-ray Analysis Mg (trace), Si, P (trace), S, K (trace), Ca, Cr
S, Cl, K, Ca, Cr (trace), (trace), Mn, Fe, and Zn
Mn, Fe, Cu (trace), and
Zn (trace)
Particle Size Range Iron—up to 2.3 ym Iron--0.4 - 0.75 ym
Carbon—up to 183 ym Calcium—0.56 - 5.6 ym
The No. 2 Sample—Figure 6a is a 153X photograph of the agglomerated
sample with a carbon flake in the center, 6b is an iron map of the area that
clearly outlines the carbon flake, and 6c is an X-ray spectrum of the sample.
Further examination of the particle in the center of the photo showed
a flaky carbon particle with agglomerated iron particles adherring to its
surface as shown in Figure 7a. An X-ray spectrum of this particle is shown
in 7b, revealing the presence of chlorine and iron. The iron signal comes
from both the surrounding and the adhering iron particles. Figure 7c is
chlorine map, showing the dispersed nature of the chlorine and suggesting that
the chlorine has been adsorbed by carbon.
16
-------
(a) Photo of sample at 153x.
(b) Iron map of (a)
(c) X-Ray spectrum of the sample showing Si,
P (trace), S, K, Ca, Cr (trace), Mn, Fe,
Cu (trace), and Zn (trace).
Figure 6. Examination of the No. 2 open hearth furnace sample.
-------
(a) Photo of carbon flake at 2200x.
(b) X-Ray spectrum of carbon flake.
oo
(c) Chlorine map of carbon particle.
Figure 7. Examination of the carbon flakes in the No. 2 open hearth sample.
-------
Jhe No. 14 Sample — This sample is composed of agglomerated masses of
spherical iron and irregular iron and calcium particles as shown in Figure
8a. The results of iron, calcium, and sulfur mappings are shown in Figures
8b, 8c, and 8d respectively. The sulfur mapping was made possible because
of the high amount of sulfur in the sample (Figure 9b). The maps distinguish
between iron and calcium particles and also indicate that the iron and
calcium occur as sulfur compounds.
Figure 9 shows the sample at 13.500X and an X-ray spectrum of the sample.
Particle sizes in this sample range from 0.04 to 0.75 ym.
4.3.1.2.2 Electric Furnace Samples— The two electric furnace samples
differed markedly, with the No. 5 sample showing distinct, unagglomerated
particles and the No. 18 showing agglomerated masses. The No. 5 sample con-
tained cubes of titanium, irregular carbon particles, and irregular, but
well-rounded iron particles in addition to the spherical iron particles and
irregular calcium particles shown by both of the samples.
Other differences include the particle size ranges and the occurrence
of phosphorous, titanium, and chromium in the X-ray analysis of the No. 5
sample. The X-ray analysis also showed a much stronger calcium peak. A sum-
mary of the characteristics of these two samples is shown in Table 6.
TABLE 6. CHARACTERISTICS OF ELECTRIC FURNACE SAMPLES
Characteristic
Sample No. 18
Sample No. 5
Consistency
Particle Size
X-Ray Analysis
Agglomerated masses of
spherical iron and irregu-
lar calcium particles
Iron—0.06 - 2.08 urn
Calcium--0.5 - 1.0 y
Mg (trace), Si, S, Cl, K,
Ca, Mn, Fe, Zn
Unagglomerated spherical iron
rounded, irregular iron; ir-
regular, sub-angular to sub-
rounded calcium; irregular
carbon; and cubic titanium
particles
Iron—0.04 - 62.09 ym
Calcium—0.28 - 37.93 um
(longest length)
Carbon—up to 900 um
(longest length)
Titanium--3.5 - 13.9 um
Mg (trace), Si (trace), P
(trace), S, Cl, K, Ca, Ti,
Cr, Mn, Fe, Zn
19
-------
(a) Photo of sample at 4500x.
(b) Iron map of sample.
i 3
(c) Calcium map of sample.
(d) Sulfur map of sample.
Figure 8. Electron microscopy investigation of the No. 14 open hearth furnace sample,
-------
rsj
*
(a) Photo of sample at 13,5QOx,
(b) X-Ray spectrum of sample showing S, K
(trace), Ca, Cr (trace), Mn, Fe, and Zn,
Figure 9. Further characterization of the No. 14 open hearth furnace sample.
-------
The No. 18 Sample -- This sample is composed of agglomerated masses of
spherical iron and irregular calcium particles in approximate percentages of
92 and 2. Figure 10 shows two magnifications and an X-ray of the sample.
The No. 5 Sample -- This sample is composed of unagglomerated particles.
However, adherence of smaller particles to larger ones is commonplace in this
sample. The sample (Figure 11) is composed of both spherical and irregular
but rounded particles composed of iron with much smaller calcium and iron
particles adhering, carbon flakes with titanium, and separate irregular
calcium particles.
Further examination of the spherical particles shown in Figure lla led
to the assumption that these were iron spheres with irregular calcium parti-
cles adhering to their surface. The result of this examination is shown in
Figure 12.
Figure 12a shows four spherical particles located in the bottom left
corner of Figure lla. Each sphere shows adhering1 particles with the center
sphere showing larger adhering particles. Figure 12b is an iron map of the
particles while Figure 12c is a calcium map. In 12c, the calcium shows up
much stronger for the center particles. This is probably a result of the
larger adhering particles that are most likely calcium. The conclusion
drawn from this examination is that the spheres are most likely iron with
adhering calcium particles.
Further examination of a carbon flake in Sample No. 5 revealed the
adhesion of spherical iron, irregular calcium, and cubic titanium particles.
Figure 13 shows the first step in the examination process taken at 450X.
Figure 13a shows the surface of the carbon flake and the adhering particles,
13b is an iron map of this surface (the locations of iron are scattered and not
concentrated),, and 13c is a calcium map showing the same scattering as b.
Figure 13d is a titanium map showing the concentration of titanium in areas
occupied by cubic particles as shown in 13a. The cubic titanium particles
range in size from 3.5 to 13.9 urn.
Figure 14 is a titanium cube at 6000X. This particle is a 7.4 pm cube.
Figure 15 gives the results from a further examination of the carbon
flake at 9000X. In 15a, spherical and irregular particles adhering to the
carbon flake are shown, 15b is an iron map showing a distinct iron concentra-
tion where the adhered particles lie, and 15c, the calcium map, also shows a
22
-------
(a) Photo of the sample at 450x showing
agglomerated masses.
UJ
(b) Photo of the sample at 3,800x taken at the center
of (a) and showing the morphology of the agglomerated
particles. Distinct particles are in the size range
of 0.06 to 2.08 u-
(c) X-Ray of the sample showing Mg (trace), Si, S,
Cl, K, Ca, Mn, Fe, and Zn.
Figure 10. Electron microscopy of the No. 18 electric furnace sample,
-------
(a) Photo of spherical and irregular, well rounded
iron particles at 225x. Shows small particle
adherence to larger ones. Largest spheres are
approximately 62 y in diameter.
(b) Photo of irregular carbon particle at 90x.
Particle is approximately 900 y at largest
lenath.
ro
(c) Photo of irregular calcium particle at 2250x. (d)
Particle is approximately 38 y at longest length
X-Ray spectrum of sample showing Mg (trace),
Si (trace), P (trace), S, Cl, K, Ca, Ti, Cr (trace),
Mn, Fe, and Zn.
Figure 11. Electron micrographs and X-Ray of the No. 5 electric furnace sample.
-------
(a) Photo of spheres at 900x.
Shows irregular adhering particles,
(b) Iron map of particles in (a)
r j
CJl
(c) Calcium map of particles in (a).
Figure 12. Results of the examination of the spherical particles in the Mo. 5 electric furnace sample.
-------
(a) Photo of carbon flake surface at 450x. Notice
cubic titanium, and scattered spherical and
irregular particles adhering to the surface.
(b) Iron map of (a)
IV-
a i
(c) Calcium map of (a).
(d) Titanium map of (a)
Figure 13. Examination of carbon flake at 450x.
-------
Figure 14. Titanium cube at 6QOOx,
27
-------
(a) Photo of carbon flake and ahering particles at
9000x. Distinguishable particles range in size
from 0.4 to 1.06 p.
Iron map.
ro
oo
(c) Calcium map.
Figure 15. Examination of carbon flake in sample No. 5 at 9000x,
-------
distinct concentration of calcium in the areas of the adhered particles. With
the aid of these maps, iron and calcium particles can be distinguished in
some cases.
4.3.1.2.3 Basic Oxygen Furnace Samples -- The two basic oxygen furnace samples
were similar. The No. 10 sample showed agglomerated masses and carbon flakes
while the No. 11 sample showed large spherical particles and agglomerated masses,
In the X-ray analysis, the No. 10 sample showed Si, Pb, Cl, and K in addition
to S, Ca, Mn, Fe, and Zn. The No. 11 sample showed these latter elements
and Mg and Ni. A summary of the particle characteristics is given in Table
7.
TABLE 7. MAJOR CHARACTERISTICS OF BASIC OXYGEN FURNACE SAMPLES
Characteristics
Sample No. 11
Sample No. 10
Consistency
X-ray Analysis
Particle Size
Range
Large spherical particles and Agglomerated masses and
agglomerated masses of
smaller particles
S, Ca, Mn, Fe, Ni (trace),
and Zn (trace)
Up to 35 ym
large irregular carbon
flakes
Si, S, Pb, Cl, K, Ca,
Mn, Fe, Zn
Carbon -- up to 740 urn
Spherical Iron — 0.02
0.24 m
The No. 11 Sample — Figure 16 shows the No. 11 BOF sample and the results
of the X-ray analysis. The sample is characterized by large spherical and
irregular-shaped particles and by agglomerated masses of smaller particles.
The X-ray spectrum shows more calcium than iron with small amounts of S,
Mn, Ni, and Zn.
The No. 10 Sample -- Examination of the No. 10 basic oxygen furnace
sample revealed basically two types of particles—agglomerated, spherical
iron and irregular, flaky carbon. Figure 17 shows photographs of the agglo-
merated masses at various magnifications from 460 to 23,OOOX and an X-ray
spectrum of the sample. Examination of the agglomerated sample revealed mostly
spherical iron particles of the size range of 0.02 to 0.24 ym (Figure 17c).
X-ray analysis revealed very little calcium plus Si, S, Pb, Cl, K, Mn, Fe, and Zn.
29
-------
-.-•
o
(a) Photo of sample at 576x.
(b) X-Ray of sample showing S, Ca, and Fe.
Figure 16. Electron microscopy examination of the No. 11 BOF sample.
-------
(a) Photo of sample at 460x.
(b) Photo of sample at 4600x.
(c) Photo of sample at 23,000x. Shows
distinct particles in the size range
of 0.02 to 0.24 u-
(d) X-Ray of sample showing Si, S, Pb,
Cl, K, Ca, Mn, Fe, and Zn.
Figure 17. Examination of the No. 10 BOF sample.
-------
Figure 18a shows a photograph of a carbon particle at 120X. An iron map
(18b) confirms the composition of the agglomerates adhering to and lying
around the carbon flake.
4.3.2 Specific Gravity
The specific gravity of the eighteen samples was determined and is
presented in Table 8. These results represent a bulk specific gravity. The
procedure employed is discussed in Appendix A.
TABLE 8. SPECIFIC GRAVITY OF RESIDUE SAMPLES
.,,
Sample Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Specific Gravity
4.07
3.59
5.20
3.86
4.40
4.06
4.22
3.90
4.06
4.29
4.75
4.12
4.57
3.99
4.24
3.77
4.12
4.26
4.3.3 Particle Size Analysis
Particle size analysis was attempted on each of the eighteen samples
by a combination sieve/pipette technique. A description of the method
employed and the raw data is given in Appendix B. After completing the
size analysis of the first group of samples, electron microscopy results indi-
cated that the pipette analysis was providing grain size values much coarser
than those of the actual samples. Analysis was discontinued at this point.
Data bias such as exhibited by the pipette analysis usually can be traced
to incomplete initial dispersion of the original sample and/or flocculation
32
-------
.
-
(a) Photo of carbon flake at 120x.
(b) Iron map of carbon flake.
Figure 18. Photo and iron map of carbon flake and adhering iron agglomerates in the
No. 10 BOF sample.
-------
(or clumping) of the dispersed grains related to surface charges. Incomplete
dispersion probably was not a major factor because of the method employed. Two
identical samples were dispersed in 1000 ml cylinders. The first had been
soaked overnight, then stirred gently with a glass rod before introduction into
the cylinder. The second, after soaking overnight, was treated to 10 minutes
in a laboratory blender before introduction into the cylinder. No visual
difference could be detected in their rate of settling.
Clumping of individual particles to create larger grains while within
the dispersed mixture appears to be the source of the apparent coarse bias
of the pipette data. The cause of this flocculation of the sample material
is not known, and could not be controlled by addition of standard peptizers to
the mixture. In the initial pipette analysis, some sodium oxalate was added
as a deflocculant agent. When it later became apparent that flocculation
nonetheless had occurred, some brief comparative experiments were run using
two different samples. Varying concentrations of hydrochloric acid, sodium
oxilate, and Calgon had no visually detectable effect on rate of sedimentation
vis-a-vis water alone. Addition of NfyOH, however, caused almost instantaneous
flocculation and settling. Non-water suspensions, including Varsol and grain
alcohol, were no improvement and would not disperse.
The immediate flocculation in NJfyOH, but lack of response to other pep-
tizers might suggest clumping is a result of processes other than true floccu-
lation. The volume of material suspended was sufficiently small to
minimize dragging of smaller grains in the wake of coarser grains. Therefore,
the rate of clumping would seem to preclude aggregation by simple mutual mass
attraction. The physical character of the settling clumps under low magnifi-
cation suggests a loose chain-like or lattice-like arrangement of particles.
Perhaps, with the high iron content of the grains, they may be acting as small
natural magnets.
The results of the completed size analysis are shown in Table 9.
The grain size data based on wet sieving reflects true grain size per-
centages. Size data of the finer material, based on pipette, at best must be
considered as a relative "effective sedimentation grain size" in water. Here,
the flocculation characteristics of a sample seem to be consistent enough to
indicate that the material will settle out in water as if it had the tabulated
34
-------
TABLE 9. PARTICLE SIZE ANALYSES (PERCENT)
CO
en
Sample
Number
<4*
>62.5
1 2.2
2 2.7
3 0.6
4 0.4
5 93.8
6 1.5
7 2.1
8 2.1
9 24.8
10 14.6
11 70.6
12 7.2
13 3.8
14 1.7
15 4.9
16 24.1
17 4.2
18 1.0
Replicates
7
7R
2.1
2.2
4-5
62.5-31.2
1.7
3.4
j , _,„
4.9
6.9
12.4
5.5
7.8
0.9
4.9
4.9
5-6
31.2-15.6
96.4
1.0
,._ m v
88.0
17.4
14.9
70.7
12,4
V
11.3
88.0
88.8
SIZES
6-7
15.6-7.8
0.5
1.3
4.8
60.8
1.8
19,3
r v
42.1
86.1
4.8
0.6
7-8 8-9 9-10
7.8-3.9 3.9-1.95 1. 95-. 975
0.9
0.2 0.3
0.2
0.3
0.3
0,3 0,4
31,2 1.6
0.5 0.1
0,2
3.4 0.2 ' —
i Sample
. 19:ll—2 Weight
.975-. 487 (gm)
23.955
26.889
30.588
25.250
95.065
25.867
19.925
29.190
26.681
26.722
29.875
26.611
24.728
25.353
17.923
32.183
26.810
26.313
19.925
18.455
continued
-------
TABLE 9 (Cont'd)
Samole
Number
15
15R
0
&
*
hw cia«
<4*
>62.5
4.9
4.6
o 7
£.. /
9 7
f- • /
(Q- <
4-5
62.5-31.2
7.8
3.7
,v „,,..,.
/ •• 1
—A. nn nino-H*
5-6
31.2-15.6
12.4
0.7
ii.M i in V
. \
a analucic
SIZEJ
6-7
15.6-7.8
42.1
50.0
avv>m«i inrlli
^
7-8
7.8-3.9
32.1
39.7
^atoc annw
8-9 9-10
3.9-1.95 1. 95-. 975
1.6
0.8 0.5
wimato of-for'Hue ninotl
in n1 Sample
...IQlll— 2 Weight
.975-. 487 (gm)
17.923
19.667
?6 889
CU . OO.7
30 037
•o ci 70 vanna
CO
en
phi scale
micrometer scale
-------
grain size distribution. One caution in such a usage is that flocculation
rate may be concentration sensitive. In the event it is not, the data show
that if introduced into water systems, the residue will settle much faster
than its grain size (true) would suggest.
The problems in obtaining particle size data led to an examination of
the literature for supplementary information. Results of this examination
follow.
OPEN HEARTH FURNACE DUST , REFERENCE
DATA
50 percent less than 5 micrometers 1
Size (urn) Weight Percent 8
0-5 64.70
5-10 6.79
10-20 11.90
20-44 8.96
>44 7.65
1-3 7.3 9
0,5-1 28,4
0.15-0.5 49.5
-0.15 , 14.8
+5 69
1-5 34
-1 60
ELECTRIC FURNACE DUST
93 percent less than 0.5 micrometers 1
Size (ym) Weight Percent* 8
0-5 67.9-71.9
5-10 6.8-8.3
10-20 6.0-9.8
20-44 7.5-9.0
>44 6.3-6.5
*Based on two furnaces
0-5 43.3-72.0 8
5-10 10.5-37.8
10-20 2.7-8.0
20-40 1.6-14.6
>40 0 -18.0
*Based on five furnaces
37
-------
REFERENCE
90 to 95 percent less than 0.5 micrometer 10
90 percent less than 1.0 micrometer 10
100 percent less than 3.0 micrometer 10
70 to 71.9 percent less than 0.5 micrometer 10
BASIC OXYGEN FURNACE DUST
85 percent less than 1.0 micrometer 1
Size (urn) Height Percent 10
<0.5 20
0.5-1.0 65
1.0-15.0 15
<5 8.9
5-10 9.1
10-20 39.9
20-30 28.8
>30 13.3
4.3.4 Chemical Analysis
Each of the 18 residue samples was subjected to elemental analysis
by spark source mass spectrometry (SSMS). The results are shown in Table 10.
The preparation and analysis procedures for the samples are described in
Appendix C.
The determination of some elements is precluded or restricted
by interferences or other factors. Those elements thus influenced are as
follows:
Indium, Erbium, and Tantalum—used as source parts and
internal standards
Fluorine (19F+)—overlap by 57pe+3
Sulfur--zinc and oxygen interferences
Aluminum—interference from 54pe+2; therefore, estimated by
27A1+2 line. Estimates good to order of magnitude
only.
Selenium—suffers interference from FeC2+ species and only
estimated when Se 77, 78, and 80 isotope ratios are
within experimental uncertainty.
38
-------
TABLE 10. RESULTS OF SSMS ANALYSIS OF STEELMAKING RESIDUES (yig/g)
(Intent 1
Alunlmim • l.?00
Antimony 44
Arsenic 210
Barium 89
Bismuth
Cadmium
Calcium 12,300
Cerium
Cesium 2.4
Chlorine 750
Citron! un 1,980
Cobalt ITO
Copper ?.850
flallum 155
Germanium
Iron Major
Lanthanum
Lead 1 ,510
Magnesium 8,300
Manganese 8 .620
Molybdenum 71
Nickel 328
Niobium
Phosphorous 3.500
Potassium 11.500
Rubidium
Scandium I.I
Selenium
Silicon 8,650
Silver 35
Sad linn 5,500
Strontium 21)
Tellurium
Tin 113
Titanium 294
Vanadium 410
Vttrlun
Zinc 5,500
Zirconium
WEN.
2
- 1.500
106
550
—
--
--
7,700
--
--
„
1.800
11
2,550
121
284
Major
--
7,150
3,600
9.620
112
260
--
1,240
13.300
32
6.1
—
8,500
26
3.5SO
31
650
233
234
172
—
81 .700
JKARTH
3
1.800 -
40
78
24
--
«
7,330
--
—
530
2,820
127
1,260
61
128
Major
--
4,120
4,660
7.940
133
250
--
830
4,070
15
7.5
.-
8,360
20
8.000
IB
..
195
200
104
--
182.000
8 14
730 - 1,400
89
7? 210
17 30
40
..
3.390 18.600
15
--
540 1 ,000
826 1.260
29 183
215 2.470
71 240
105 250
Major Major
.-
101 10.300
3,050 17.000
2,800 7.600
74 200
100 300
..
800 2.420
3.920 1 .450
16 7
2.9 17
—
4.640 4.350
2fl
2,990 2.340
14 195
._
500
104 380
84 IJ5
--
813 76,900
00
17 4
1,400
54
75
74
..
350
23,700
20
1.3
3. SOD
1.810
135
1,720
SS
74
445,000
--
17,600
38.900
132.000
54
116
--
2,320
14,200
41
15
--
4.350
30
11. TOO
55
..
200
320
170
— -.
97.900
-•
- 4,100
109
.69
336
103
..
103,000
46.5
4.4
2,130
181.000
3,100
7.550
250
68
392.000
8.9
4.820
• 32,000
125,000
6,850
46,400
160
—
54,000
no
5
2,440
- 11,800
116
31,000
94
19
100
200
S50
40
12.400
67
5 6
- 2,700 - 400
78
8.6 164
42 364
44
..
39,000 192,000
20
2.8
274 15.300
2,770 27,900
37 1.600
5?0 4.070
17 93
58
Hnjor 479,000
513 17,200
4.520 86.700
10.200 92.400
108 8.500
230 420
21
3IMI 1.120
232 17.900
78
28
156
7.560 66,500
60
320 16,300
34 174
__
19 269
250 - ?00
105 112
14
t.B.40 140.000
21
ELKJRIC
7
500
185
200
478
205
1.044
92.800
6.1
--
2,380
51)0
5.730
47
24
310,000
.,
108.600
11,800
•6, TOO
n
no
13
"55
9.350
62
12
300
55,800
130
0,950
124
__
1.920
660
100
8
294.000
90
12 13
• 1,400 - 1.370
140 2CO
145 250
272 600
84
710 980
89,400 68.200
6
3.6 4
7.630 16.800
8.500 3.120
266 2.200
5,560 2,280
76 130
52 78
402,000 400,000
__
25,700 39,500
44,800 20.1(10
91.400 45,700
446 ICO
2.070 310
36 29
890 1.550
25,600 13,000
74 52
39 39
630
43,900 16.000
80 120
26.400 12.200
147 140
104
865 730
1.100 866
183 89
7 II
375,008 Major
108 ' 130
18 . 9
• 4,500 ' 66
74
250 57
38 132
..
.-
11,400 .61.500
I
750 984
1,090 100
56 590
2,370 550
134 SO
126
Major Major
..
S.700 105
9,700 2,420
10.200 5.BBO
550
300 39
.-
2,230 1.040
14.800 4.270
45 24
8 19
.-
91,600 14,300
34
10,100 4,740
20 34
__
305
265 307
70 104
--
62.900 1,700
—
ii AS ic nxrwii
10 II 15 1C
- 1.100 - 730
20
30 136
78
..
35,000 71,000
.-
1,790 400
420 1,220
37 1,450
650 > 459
24 200
79
Major Major
„_
4.760 2,100
8,000 33.200
25,500 11.500
54
69 100
.-
840 1,660
5.270 300
22
6.3 66
—
17,700 30.000
30
5,190 400
43 290
__
65
450 1.420
154 2111
90.000 4.2RO
- 450 1,4110
21
81 25
39 3S
__
..
35.900 VI ,600
--
1.580 430
563 330
mo wi
600 140
?68 1',
87 If.
Major Hajor
__
6, UK) 711)
8,530 12,?(1H
39,500 13. Ml
' 64 611
126 6?
--
2.630 470
1.620 4,960
6 II
7.8 t?
--
29,000 18.4110
32
440 1 ,6.W
51 2110
--
81
40H 4IKI
576 1 | 3
51 .TOO 3. Mi
u>
vo
-------
Cadmium—uncertainty associated with results because of
known volatility losses in sparking electrodes.
Sample No. 4—high Cr, Ni, Fe, and Mo precluded normal
determination of Mg, Si, and Ti due to overlap of +2
masses, and results good only to an order of magnitude.
A search of available literature for comparative analysis was conducted.
The results of this search are presented in Table 11.
TABLE 11. ELEMENTAL COMPOSITION OF-STEELMAKING,DUSTS
BASED ON REPORTED VALUES6'11'1^'13'14
Element
Open Hearth
CONCENTRATION (yg/g)
Electric
Basic Oxygen
Aluminum
Antimony
Arsenic
Calcium
Carbon
Chromi urn
Copper
Fl uori de
Iron
Lead
Magnesium
Manganese
Nickel
Phosphorous
Potassium
Selenium
Silicon
Sodi urn
Sulfur
Tin
Zinc
400 - 6700
<100
<200 - <300
4800 - 26,000
2300 - 12,000
920 - 1000
15,200
270,000 - 680,000
500 - 25,000
2000 - 186,000
200 - 4900
820
5200 - 25,000
<100
2500 - 110,000
1300 - 9000
1000 - 27,000
<100
300 - 151,000
1500 - 42,000
<100
200
600 - 302,000
2800 - 74,000
1400 - 97,000
1000 - 2700
7500 - 12,700
217,000 - 558,000
trace - 51,000
4000 - 126,000
5000 - 64,000
300 - 29,000
740 - 2900
1000 - 32,000
<100
1000 - 46,000
500 - 41,000
200 - 10,000
<100
trace - 248,000
trace - 36,000
<100 - <200
<200
3000 - 237,000
trace - 6100
80 - 500
10,000 - 13,000
324,000 - 866,000
trace - 18,000
800 - 86,000
800 - 20,000
1000
100 - 12,000
<100
1900 - 130,000
<100 - 4700
<100 - 20,000
trace - 4600
trace - 151,000
4.3.5 Source of Dust Constituents
The substances that compose the dusts collected by air pollution
control equipment can originate in the ore charged to the blast furnace, in
the scrap charged to the steelmaking furnace, or in additives used to control
the process or impart special properties to iron or steel. A listing of the
40
-------
probable sources of the various substances is given in Table 12. Source
identification is described below.
TABLE 12. PROBABLE SOURCES Of SUBSTANCES COMPRISING STEELMAKING DUSTS
Substance
Aluminum
Antimony
Arsenic
Calcium
Chromium
Cobalt
Col umbi urn
Copper
Iron
Gold
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Phosphorous
Potassium
Selenium
Silica
Silver
Sodi urn
Sulfur
Tin
Titanium
Tungsten
Vanadi urn
Zinc
Ore
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
SOURCE
Scrap Additive
X
X*
X X
X
X
X X
X
X X
X**
X X
X
X X
X
X**
X
X
X
X
X
Reference
15,16
17
14, 15, -17
15
14,15,16
15,16
16
14,15,16
14,15
16
14,15,16
15,16
14,15,16
14
16
15,16
14,15,16,17
14
14,17
15,16
16
14
14,15
15,16
15,16
16
15,16.17
14,15
Flux
**
Additive and flux
4.3.5.1 Ores —
The iron ores used in steelmaking come from a variety of U.S. and foreign
9 6
sources. In 1974, total consumption of iron ore was 131.1 x 10 kg (144.5 x 10
net tons), of which approximately 65 percent came from U.S. sources and 35
41
-------
3
percent'from foreign sources. Approximately 85 percent of the U.S. pro-
duction came from the Lake Superior District. Here banded, sedimentary ore
deposits are predominately composed of iron and silica with small amounts of
18
alumina, sodium, potassium, and minor elements. The major ranges in this
district are the Mesabi, Cuyuna, Gogebic, Marquette, and Menominee-.
Both the occurrence and the quantities of the various metals vary from one
iron deposit to another. Factors involved in these variations include type
of deposit, parent material of the deposit* and post depositional geological
influences on the deposit, including metamorphism. These variations also
occur within a single deposit. Examples of these quantity variations include:
iron 32-55.5 %
silica 3-45 %
aluminum 0.8-^2.6 %
calcium 0*02-2.3 %
magnesium 0.02-3.0 %
phosphorous 0.04-0.54 %
manganese 0.17-4.5 %
4.3.5.2 Scrap--
Steep scrap can originate from a variety of sources, although two
grades of scrap exist. Number one scrap is considered essentially "pure",
while number two scrap is considered "contaminated" and may contain copper
wire, rubber, wood, etc. Table 13 shows the 1974 consumption of scrap and
pig iron by type of furnace.^
Depending on the type of scrap used, ft is possible for most elements
to originate from the scrap. For example, lead and zinc probably originate
from galvanized scrap, while copper probably comes from the copper wiring in
auto-serap.
42
-------
TABLE 13. 1974 CONSUMPTION OF SCRAP AND PIG IRON BY IRON AND
STEEL FURNACES'3
CONSUMPTION (109 Kq)
Furnace Type
Open Hearth
Basic Oxygen
Electric
Blast
Subtotal
Industry Total*
Scrap
17.1
24.0
26.2
4.0
71.3
72.7
Pig Iron
21.2
59.3
1.0
—
81.5
83.9
Total
38.4
83.4
27.2
4.0
153.0
156.7
Includes consumption by cupolas, direct casting, and by other
furnaces. Totals may not equal due to rounding off.
4.3.5.3 Additives --
Certain substances are added to the furnaces to control the process,
to impart special properties to the iron or steel, and to control iron or
steel properties. Table 14 lists these additives and their uses.
TABLE 14. ADDITIVES AND THEIR USES16
Substance Use
Aluminum Control grain size and remove impurities and gases
Calcium Basic flux in form of limestone (calcium carbonate)
and neutral flux in form of fluorspar (calcium
fluoride)
Chromium Resist corrosion, harden, and increase high-tem-
perature strength
Cobalt Increase high-temperature strength and make per-
manent magnet steels
Columbium Resist heat,strength, and improve mechanical pro-
perties
continued
43
-------
TABLE 14 (Cont'd).
Substance Use
Copper Improve machinability of iron and resist atmospheric
corrosion of steel
Lead Improve machinability
Magnesi-um Basic flux in form of dolomite (magnesiun carbonate)
and improve strength and ductibility of cast iron
Manganese Remove gases from steel (small amounts), increase
strength and toughness (1-2 percent Mn), and
greatly increase toughness and abrasion resistance
(12 percent Mn)
Molybdenum Decrease temper brittleness and increase strength,
ductibility, and shock resistance
Nickel Increase toughness without brittleness
Silica Acid flux, deoxidizer, and increases resistance to
corrosion, heat, and wear
Tin Resist wear in iron
Titanium Control grain size and remove gases and impurities
Tungsten Increase high temperature strength
Vanadium Increase strength, ductibility, and resiliency
3
Table 15 shows the 1974 consumption of fluxes by the steel industry.
Consumption information for the other additives could not be located.
TABLE 15. 1974 CONSUMPTION OF FLUXES3
CONSUMPTION (106 Kg)
Furnace Type
Blast
Open Hearth
Basic Oxygen
Electric
Fluospar
88.0
412.8
83.4
Limestone
13,316.8
1,469.7
594.2
244.9
Lime
—
456.3
5,963.0
796.5
Other
—
216.8
706.7
73.5
Total
13,316.8
2,230.8
7,685.8
1,198.4
TOTAL 584.2 15,625.6 7,215.8 997.0 24,431.8
44
-------
SECTION 5
SOLUBILIZATION TESTS
A major portion of this study was devoted to an examination of the
possibility for water extraction of certain metals and organic compounds
from the steelmaking furnace residues, since these residues are exposed to
water at their disposal sites.
5.1 PROCEDURES AND RESULTS
Each of the 18 samples was subjected to 10 successive water washings
by the procedures outlined in Appendix D. After each washing, a portion of
the solute was removed and analyzed by atomic absorption for antimony,
arsenic, cadmium, chromium, cobalt, lead, manganese, mercury (both total and
inorganic), nickel, selenium, vanadium, and zinc. Cumulative solubilization
data for these metals are presented in Tables 16 through 27. Arsenic
was not detected in any of .the samples. The pH values for each sample and
solubilization run are presented in Table 28. The initial pH of the deionized
water was 6.7.
A new one-wash solubilization test was performed on each of the 18
samples for detection of organics in the solutes. To analyze for the pre-
sence of organic compounds, a carbon analysis was performed. This was
followed by a gas chromatography-mass spectrometry analysis of selected
samples. Additional parameters such as conductivity and acidity/basicity
were also determined. The results of these analyses are presented in Table
29 and are followed by a discussion of specific organic compounds identified
by gc-ms in each sample.
45
-------
TABLE 16. SOLUBILIZATION DATA FOR ANTIMONY
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (Mg)
1
4,914.0
5,310.0
4,100.0
2,363,4
966.0
4,367.6
5,338.0
1,529.5
6,842.5
6,670.0
7,840.0
2,352.0.
2,387.t)
1,577.6
3,900.4
3,900.0
633.6
1,626.0
2
4,914.0
6.087.0
4.100.0
2,673.9
966.0
4,367.6
5,338.0
1,529.5
6,842.5
6,670.0
8,248.0
2.352.0
2,387.0
2,729.6
3,900.4
3,900.0
633.6
1,857.0
Run Number <
3
5,803.0
6,748.5
4,500.4
2,673.9
1,896.0
5,544*6
5,395.4
2,833.0
10,044.1
8,779.7
8,378.8
4,586.0
2,387.0
2,837.5
6,146.8
6,555.0
1,004.4
1,861.7
4
6,476.2
7,681.5
4,727.2
2,673.9
1,896.0
5,694.8
5,395.4
3,366.0
10,120.3
8,813.2
8,378.8
5,087.4
2,407.2
2,837.5
6,195.8
12,203.4
1,004.4
1,861.7
5
7,483.5
8,522.3
5,213.2
3,142.0
1,937.4
5,761.2
5,612.4
4,035.9
10,120.3
9,289.0
9,098.8
5,600.1
2,407.2
3,611.0
6,238.5
12,203.4
1,395.0
1,873.7
6
8,184.6
8,734.8
5,505.3
3,208.0
2,508.9
6,527.9
5,836.4
4,166.8
10,377.7
9,663.4
9,438.8
6,240.6
2,554.8
4,213.0
6,689.7
12,487.1
1,395.0
2,148.1
7
8,601.5
8,734.8
5,505.3
3,851.5
2,556.6
7,461.6
5,836.4
4,934.4
11,190.2
10,457.5
10,182.8
7,998.0
2,668.4
5,413.6
8,087.1
14,035.9
1,632.6
2,361.1
8
9,819.8
9,638.9
5,988.9
3,851.5
2,556.6
7,461.6
6,599.6
5,444.0
11,582.2
11,776.7
10,182.8
8,282.2
3,193.8
5,413.6
8,087.1
14,035.9
1,639.4
2,626.3
9
9,819.8
9,638.9
5,988.9
3,851.5
4,806.2
7,782.5
6,599.6
5,759.8
11,740.6
11,776.7
11,153.0
9,585.3
3,206.5
6,326.2
8,688.8
14,290.9
1,909.7
2,914.5
10
10,693.4
9,638.9
6,841.5
3,851.5
*> 4,806. 2
7,782.5
7,625.4
5,759.8
11,939.1
13,448.3
11,675.1
9,665.1
3,490.7
6,742.2
8,688.8
14,290.9
1,909.7
3,862.5
Total Concentration
(ug metal /g residue)
85.55
77.11
57.73
30.81
38.45
62.26
61.00
46.08
95.51
107.59
93.40
77.32
27.93
53.94
69.51
114.33
15.28
30.90
-------
TABLE 17. SOLUBILIZATION DATA FOR CADMIUM
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (|ig)
Run Number
1
1,207.5
89.3
1,220.0
15.3
246.0
514.6
0.0
2
1,207.5
89.3
1,220.0
15,3
.-
246.0
514.6
0.0
3
1,264.7
97.0
1,244.6
15.3
246.0
514.6
•-•
0.0
4
1,324.1
115.8
1,244.6
15.3
"
246.0
514.6
15.7
5
1,387.8
126.9
1,244.6
15.3
246.0
514.6
15.7
6
1,425.9
133.2
1,252.1
15.3
246.0
514.6
15.7
7
1,447.5
144.4
1,252.1
15.3
246.0
514.6
15.7
8
1.480.2
150.6
1,252.1
15.3
246.0
514.6
15.7
9
1,488.0
150.6
1,252.1
15.3
246.0
514.6
15.7
10
1,488.0
150.6
1,252.1
15.3
246.0
514.6
15.7
Total Concentration
dig metal/g residue)
11.90
1.20
10.02
0.12
1.97
4.12
0.13
-------
TABLE 18. SOLUBILIZATION DATA FOR CHROMIUM
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE COHCENTRATION (iig)
Run Number
1
10.5
22.5
18.0
13.600.0
0.0
1,147.5
12.8
5,985.0
8.4
0.0
0.0
109.2
2
10.5
42.5
18.0
16.507.0
0.0
1,169.3
12.8
5.985.0
25.5
46.7
0.0
109.2
3
10.5
42.5
29.9
16.507.0
55.3
1,613.5
12.8
5,985.0
64.8
46.7
0.0
109.2
4
10.5
42.5
29.9
17,553.4
145.5
1,695.4
18.8
6.231.0
99.7
46.7
0.0
109.2
5
10.5
42.5
35.3
17,553.4
285.6
1,695.4
25.3
6,231.0
107.0
46.7
0.0
109.2
6
10.5
57.2
37.0
17,553.4
440.8
1.732.1
25.3
6,231.0
127.2
46.7
0.0
112.0
7
10.5
71.7
40.6
17,553.4
606.4
1,788.3
29.3
6,318.6
137.3
46.7
0.0
112.0
8
10.5
71.7
40.6
17.553.4
779.3
1,835.9
29.3
6,396.2
159.6
46.7
0.0
113.6
9
10.5
71.7
40.6
17,903.2
945,6
1,922.3
29.3
6.445.5
159.6
46.7
0.0
121.5
10
10.5
71.7
40.6
18.081.6
1,060.8
2,093.0
53.2
7.201.4
188.9
46.7
29.7 <
125.1
Total Concentration
(pg metal/g residue)
0.08
0.57
0.32
' 144.65
8.49
16.74
0.43
57.61
1.51
0.37
0.24
1.00
-------
TABLE 19. SOLUBILIZATION DATA FOR COBALT
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (ug)
Run Number
1
525.0
1,020.0
557.6
397.8
0.0
805.2
958.8
226.1
969.0
527.8
1,288.0
180.0
465.0
232.0
656.6
555.0
88.0
138.0
2
693.0
1,251.0
557.6
397.8
230.4
1,179.3
958.8
226.1
1,125.6
806.8
1,288.0
453.6
465.0
443.2
656.6
555.0
181.0
138.0
3
1,124.8
1,251.0
803.3
397.8
230.4
1.414.7
958.8
226.1
1,125.6
806.8
1,309.8
650.3
465.0
443.2
1,031.0
1,074.2
304.6
302.5
4
1,124.8
1,359.9
851.9
397.8
230.4
1,414.7
1,007.9
226.1
1,125.6
820.2
1,657.8
683.0
510.5
443.2
1,129.0
1,900.4
304.6
302.5
5
1,379.8
1,787.6
851.9
397.8
230.4
1,423.0
1,035.9
226.1
1,346.7
1,296.0
1 ,657.8
683.0
510.5
740.7
1,129.0
1,900.4
304.6
394.5
6
1,552.0
1,787.6
851.9
397.8
230.4
1,525.9
1,035.9
226.1
1,346.7
1,296.0
1,657.8
683.0
510.5
894.7
1,241.8
2,101.7
304.6
394.5
7
1,632.7
1,787.6
1,170.5
397.8
230.4
1,525.9
1,369.4
226.1
1,346.7
1,356.3
1,657.8
683.0
510.5
894.7
1,282.6
2,172.1
304.6
643.0
8
1,973.3
1,787.6
1,170.5
397.8
448.8
1,782.1
1,433.0
499.1
1,640.7
1,463.0
1,823.4
683.0
794.5
1,155.1
1,302.4
2,183.9
461.0
650.8
9
1,973.3
1,787.6
1,170.5
593.6
600.8
1,864.9
1,433.0
604.4
1,640.7
1,601.6
1,854.2
683.0
845.3
1,265.0
1,340.6
2,281.7
461.0
673.4
10
2,376.5
1,787.6
1,170.5
641.0
600.8
1,864.9
1,433.0
922.9
1,839.2
1,800.6
1,854.2
683.0
845.3
1,348.2
1,419.8
2,281.7
742.6
,709.4
Total Concentration
(ug inetal/g residue)
19.01
14.30
9.36
5.13
4.81
14.92
11.46
7.38
14.71
14.40
14.83
5.46
6.76
10.79
11.36
18.25
5.94
5.68
•p.
V£>
-------
TABLE 20. SOLUBILIZATION DATA FOR LEAD
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (ug)
Run Number
1
210.0
150.0
82.0
514.5
305.0
68.0
598.5
195.5
25,350.0
0.0
144.0
241.8
0.0
5,880.0
4,380.0
0.0
0.0
2
476.0
150.0
114.1
672.9
395.3
68.0
598.5
195.5
25,350.0
284.0
144.0
241.8
0.0
6,480.0
6,823.6
0.0
270.6
3
780.8
150.0
186.9
1,029.4
727.0
92.6
776.3
207.1
59,056.3
536.6
1,324.2
241.8
282.2
7,809.6
7,449.0
0.0
270.6
4
1,071.2
150.0
186.9
1,437.6
727.0
92.6
919.8
207.1
94,879.6
680.6
1,324.2
241.8
282.2
7,809.6
7,449.0
. 62.0
270.6
5
1,275.2
150.0
186.9
1,796.4
727.0
•
92.6
1,610.0
207.1
96,928.9
776.6
1,544.0
282.1
282.2
8,596.6
8,687.4
62.0
270.6
6
1.422.8
150.0
186.9
2,139.3
801.8
92.6
1,610.0
207.1
128.344.9
776.6
1,781.9
282.1
317.2
9,207.5
9,062.6
62.0
270.6
7
1,611.1
150.0
186.9
2,505.0
801.8
92.6
1.610.0
207.1
128,344.9
2,396.6
1,781.9
367.3
317.2
9,207.5
9,837.0
62.0
270.6
8
1,611.1
150.0
186.9
3,269.4
904.3
92.6
1.610.0
207.1
160.934.7
2,614.0
2,066.1
367.3
317.2
10,019.3
10,031.7
62.0
333.0
9
1,727.7
150.0
186.9
3,862.2
904.3
92.6
1,610.0
207.1
160.934.7
2,614.0
2,619.4
367.3
317.2
10,019.3
10.078.4
62.0
333.0
10
1,772.5
150.0
186.9
4,473.2
904.3
92.6
1,610.0
207.1
183,449.7
3,363.1
3,710.0
753.0
317.2
10,019.3
10,078.4
182.60
333.0
Total Concentration
lug metal/g residue)
14.18
1.20
1.50
35.79
7.23
0.74
12.88
1.66
1467.60
26.90
29.68
6.02
2.54
80.15
80.63
1.46
' 2.66
01
o
-------
TABLE 2]. SOLUBILIZATION DATA FOR MANGANESE
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
6
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (,,g)
Run Number
1
23,730.0
4,875.0
0.0
0.0
813.8
24.400.0
74.1
0.0
399.5
0.0
15.4
312.0
979.6
0.0
0.0
21.0
113.5
0.0
2
23,730.0
4,875.0
0.0
3.5
813.8
24,400.0
74.1
0.0
399.5
0.0
15.4
312.0
979.6
21.1
0.0
74.6
113.5
38.3
3
23,730.0
4,875.0
13.7
8.7
815.3
25,069.8
w
81.5
0.0
399.5
6.6
92.8
733.5
979.6
21.1
8.6
88.8
113.5
38.3
4
23,978.2
4,948.1
13.7
8.7
878.1
25.069.8
81.5
106.6
399.5
13.9
92.8
733.5
979.6
21.1
14.2
126.6
251.2
38.3
5
23,978.2
4,948.1
13.7
8.7
894.7
25.069.8
84.3
135.0
399.5
121.6
97.6
767.2
988.2
21.1
°14.2
126.6
327.4
38.3
6
23,978.2
4,974.2
13.7
8.7
915.0
25,069.8
89.1
176.2
399.5
121.6
97.6
897.1
995.6
21.1
14.2
126.6
327.4
38.3
7
23,978.2
5,006.5
13.7
8.7
946.8
25,069.8
107.5
176.2
.399.5
151.7
220.0
897.1
1,052.4
21.1
14.2
152.1
327.4
38.3
8
23,984.7
5,020.9
13.7
8.7
967.1
25,080.1
107.5
203.5
428.9
151.7
495.3
897.1
1,063.8
21.1
41.0
152.1
343.1
38.3
9
23,984.7
5,020.9
13.7
8.7
992.9
25,080.1
110.7
203.5
428.9
151.7
495.3
920.3
1,099.3
21.1
41.0
245.6
413.8
47.9
10
24,031.8
5.038.3
34.0
32.4
1.030.5
25.080.1
159.7
419.1
273.9
251.2
617.9
1,298.1
1,275.9
100.2
41.0
245.6
834.9
51.5 >
Total Concentration
dig metal/g residue)
192.25
40.31
0.27
0.26
8.24
200.64
1.28
3.35
2.19
1 2.01
4.94
10.38
10.21
0.80
0.33
1.96
6.68
0.41
-------
TABLE 22. SOLUBILIZATION DATA FOR INORGANIC MERCURY
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (ug)
Run Number
1
290.0
840.0
744.0
2
290.0
7,560.0
744.0
3
290.0
7,560.0
992.4
4
290.0
8,484.0
1,694,4
5
290.0
9,300.0
1 ,993.4
6
290.0
9,300.0
2,108.2
7
290,0
9,300.0
2,108.2
8
290.0
9,300.0
2,108.2
9
290.0
9,300.0
2,108.2
10
290.0
9,300.0
2,108.2
Total Concentration
(tig metal/g residue)
2.32
74.40
16.87
ro
-------
TABLE 23. SOLUBILIZATION DATA FOR TOTAL MERCURY
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (Mg)
Run Number
1
1,508.0
6,440.0
420
1,364
2
1,508.0
10,724.0
636.6
1.684
3
1.909.!
10,724. t
917.6
1,684
4
1,969.8
11,528.0
917.6
1,987
5
2,188.4
12,776
917.6
2,849.5
.
6
2,581.8
13,755.2
917.6
2,849.5
7 '
*
2.581.8
15,075.2
917.6
3,701.5
8
3,590.6
15,779
917.6
3,701.5
9
3,590.6
15,856
917.6
3,701.5
10
\
3,590.6
18,534.6
917.6
3,701.5
Total Concentration
(lig metal/g residue)
28.72
148.28
7.34
29.61
01
CO
-------
TABLE 24. SOLUBILIZATION DATA FOR NICKEL
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (wg)
Run Number
1
73.5
285.0
63.0
97.6
47.6
40.6
0.0
31.0
0.0
Z
73.5
285.0
63.0
97.6
47.6
40.6
0.0
31.0
0.0
3
73.5
285.0
63.0
97.6
47.6
40.6
0.0
31.0
0.0
4
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
5
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
6
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
7
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
8
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
9
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
10
73.5
285.0
63.0
97.6
47.6
40.6
54.5
31.0
54.0
Total Concentration
(pg metal/g residue)
0.59
2.28
0.50
0.7B
0.38
0.32
0.44
0.25
0.43
cn
-------
TABLE 25. SOLUBILIZATION DATA FOR SELENIUM
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (ug)
Run Number
1
420.0
150.0
82.0
93.6
997.5
122.0
10,676.0
133.0
170.0
203.0
980.0
162.0
142.5
58.0
98.0
300.0
0.0
0.0
t
s
420.0
286.5
210.4
03.6
997.5
122.0
11,493.0
226.0
257.0
203.0
980.0
162.0
174.6
67.6
98.0
300.0
55.8
59.4
3
534.3
286.5
210.4
250.8
997.5
196.9
11,493.0
364.3
291.8
363.6
980.0
162.0
174.6
92.5
242.0
300.0
55.8
59.4
4
573.9
550.9
364.3
276.6
997.5
218.4
11,825.5
364.3
291.8
363.6
1,040.0
314.6
210.0
150.6
* 368.0
440.4
93.0
59.4
5
637.7
550.9
364.3
327.2
1,121.7
218.4
11,825.5
547.0
298.5
363.6
1,064.0
446.5
210.0
150.6
368.0
440.4
93.0
59.4
6
637.7
550.9
364.3
380.0
1,121.7
274.5
11,825.5
696.6
298.5
609.6
1,064.0
446.5
218.2
150.6
368.0
440.4
93.0
59.4
7
637.7
550.9
417.4
380.0
1,121.7
274.5
11,837.0
696.6
373.5
609.6
1,064.0
446.5
218.2
150.6
368.0
440.4
93.0
59.4
8
637.7
550.9
417.4
380.0
1,121.7
397.5
11,837.0
696.6
373.5
609.6
1,064.0
446.5
218.2
150.6
368.0
440.4
93.0
59.4
9
637.7
550.9
417.4
380.0
1,121.7
397.5
12,038.4
696.6
373.5
609.6
1,064.0
446.5
218.2
150.6
368.0
440.4
93.0
59.4
10
I
t
637.7
550.9
417.4
380.0
1,121.7
397.5
12,038.4
696.6
373.5
609.6
1,064.0
446.5
218.2
150.6
368.0
440.4
93.0
59.4
Total Concentration
(pg raetal/g residue)
5.10
5.40
3.34
3.04
8.97
3.18
96.31
5.57
2.99
4.88
8.51
3.57
1.75
1.20
2.94
3.52
0.74
0.48
OH
on
-------
TABLE 26. SOLUBILIZATION DATA FOR VANADIUM
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (pg)
Run Number
1
428.4
0.0
182.0
74.4
107.8
0.0
2
:|,
F(
1
428.4
0.0
182.0
80.8
107.8
0.0
3
428.4
0.0
182.0
80.8
107.8
0.0
4
428.4
0.0
182.0
80.8
107.8
91.8
5
428.4
0.0
182.0
80.8
107.8
91.8
6
428.4
0.0
182.0
80.8
107.8
91.8
7
451.4 .
100.5
182.0
80.8
107.8
91.8
8
451.4
100.5
182.0
80.8
107.8
91.8
9
451.4
100.5
182.0
80.8
107.8
91.8
10
451.4
100.5
182.0
80.8
107.8
91.8
Total Concentration
(iig metal/g residue)
3.61
0.80
1.46
0.65
0.86
0.73
01
cr>
-------
TABLE 27. SOLUBILIZATION DATA FOR ZINC
Furnace
Type
Open Hearth
Electric
Basic Oxygen
Sample
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
CUMULATIVE CONCENTRATION (|ig)
Run Number
1
96,600
114.0
41.0
7.8
65.625.0
54.168.0
61.2
266.0
1.411.0
783.0
882.0
936.0
1.736.0
7.5
2S4.8
133.5
334.4
0.0
2
96.600
114.0
630.6
10.1
65,625.0
54,168.0
61.2
266.0
1,411.0
783.0
967.2
936.0
1,736.0
23.9
420.4
281.0
471.1
291.1
3
102.924.6
114.0
714.3
44.2
65.625.0
57,752.5
94.8
390.4
1,411.0
1,164.1
2.048.5
8.531.4
1.737.4
44.6
544.2
364.8
471.1
291.1
4
102.924.6
114. C
964.6
44.2
80,084.7
57,752.5
94.8
960.3
1,413.5
1.237.1
2,048.5
8.531.4
1,737.4
58.6
637.3
1,157.0
722.2
291.1
5
110,549.1
114.0
964.6
44.2
83,134.5
57.752.5
94.8
1,199.9
1.422.3
1.856.4
2.251.3
10.542.9
1,737.4
62.9
637.3
1.157.0
862.6
297.5
6
111.557.7
141.8
1,713.3
44.2
85,166.5
57,752.5
101.2
1,285.9
1,443.0
1,856.4
2,251.3
12,861.5
1,737.4
62.9
637.3
1,255.8
862.6
300.8
7
123,904.8
141.8
1,730.1
44.2
87,344.8
57,752.5
108.1
1,346.5
1.458.0
2,099.6
4.622.5
12,861.5
2,102.3
76.8
725.1
1,320.9
862.6
300.8
8
123,904.8
141.8
1,730.1
44.2
89,435.2
57,752.5
120.8
1,402.9
1,529.4
2.099.6
7.198.6
13,728.3
2,214.5
76.8
851.8
1,368.7
, 903.4
303.9
9
124,601.4
141.8
1.730.1
44.2
91.198.4
57.752.5
125.1
1,457.6
1.555.4
2,099.6
7.198.6
14.099.6
2,311.0
99.6
1,092.4
1,424.0
1.052.4
303.9
10
127.132.6
141.8
2.211.2
101.0
92,722.9
57,785.3
131.8
1,844.7
2.266.0
2.656.8
8.099.8
19.940.9
2.816.5
126.7
1 .092.4
1.424.0
1.138.1
369.9
Total Concentration
(119 metal/g residue)
1.017.06
1.13
17.69 '
0.81
742.18
462.28
1.05
14.76
18.13
21.25
64.80
159.53
22.53
1.01
8.74
11.39
9.11
2.96
01
-------
TABLE 28. EFFECTS ON pH OF LEACHING TESTS
Ul
00
pH OF SAMPLES
Furnace Type
Open Hearth
Electric
Basic Oxygen
Sampl e
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
1
7.2
8.6
9.5
10.0
5.7
7.6
12.0
12.4
10.6
12.8
12.7
8.9
7.9
12.7
12.5
12.6
10.5
12.7
2
7.2
8.0
7.4
8.3
7.1
7.3
11.2
11.7
8.0
12.5
11.9
8.4
8.1
12.3
11.9
12.1
8.2
12.2
3
7.3
8.4
8.7
8.5
7.0
8.0
11.3
11.6
10.0
1.21
11.8
9.3
9.2
12.3
11.6
11.6
9.6
11.6
4
7.4
8.5
9.5
10.0
7.2
7.9
12.1
11.9
10.2
12.6
12.1
10.8
7.1
12.4
12.4
12.4
9.7
12.3
BY LEACHING RUNS*
5
7.6
8.9
9.1
9.5
7.2
8.0
12.1
11.8
10.2
1.25
11.9
10.8
8.6
12.5
11.8
12.4
9.1
12.3
6
7.1
9.9
9.3
9.2
6.8
7.9
12.1
11.1
9.9
11.8
11.4
10.5
8.4
11.5
11.8
11.5
9.0
11.3
7
7.4
9.3
9.0
9.5
6.8
7.6
11.9
10.7
10.5
11.9
11.0
10.5
8.5
11.9
11.5
11.6
9.0
11.9
8
6.0
8.2
9.2
9.5
6.8
8.8
11.8
10.9
10.3
11.8
11.4
10.4
7.5
11.6
11.5
11.6
8.2
11.4
9
6.4
7.7
7.9
10.2
7.3
9.1
11.9
11.4
10.8
12.2
11.9
11.0
8.6
12.4
12.2
12.4
9.0
12.3
10
6.4
7.9
8.2
9.9
6.2
9.8
11.9
11.4
10.9
12.1
11.8
11.0
8.9
12.3
12.0
12.2
8.7
12.4
Initial pH of deionized water was 6.7.
-------
.TABLE 29. RESULTS OF ANALYSES PERFORMED ON SECOND SET OF LEACHATES
Carbon Analysis
Furnace Sample (.mgA)
Type Number Inorganic Organic
pH
Acidity/Basicity Conductivity
(equivalents KOH/x,) (m mhos/cm)
Ooen
Hearth
El ectri c
Basic
Oxygen
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
3
9
7
5
0
0
26
2
10
14
21
0
0
4
2
0
4
2
19
18
21
27
16
11
310
23
166
41
279
59
181
58
21
35
13
12
7.0
7.2
8.6
9.2
6.0
6.5
11.9
12.4
9.2
12.8
12.6
7.3
6.5
12.6
12.5
12.5
8.1
12.5
0.00
0.64
0.64
0.03
0.00
0.00
1.71
1.69
0.06
7.01
3.10
0.01
0.00
2.74
2.58
2.47
0.01
3.36
11.5
8.6
10.8
5.7
2.3
11.1
24.5
5.1
20.5
44.3
29.8
20.5
17.0
13.8
14.6
7.2
0.8
13.1
Blank
6.5
3.0*
Expressed as u mhos.
The samples showing the highest TOC (numbers 4, 6, 12, and 18) were
analyzed by gc-ms. The results showed the types of products that would be
expected from the distillation of mineral-or petroleum-based lubricating
oils. For example, in the number 4 sample, the majority of the organics were
C,g to C-Q alkanes, a C22 substituted aromatic acid, and C2g and C32 branched
chain aliphatic acids. For sample number 12, the majority were CIQ to C26
alkanes, C32 and C2g branched chain aliphatic acids, and a C22 branched chain
substituted aromatic chain. No polynuclears or organomercurys were found.
59
-------
5.2 DISCUSSION
The explanation of what is occurring in the solid/liquid system over a
period of ten washings is elusive. A straightforward correlation between the
concentration of a metal in the solute versus either the quantity of metal in
the dust or the pH changes of the solute after each washing is complicated by
the possibility of numerous secondary reactions occurring once the meprals are
extracted from the solids. Examples of secondary reactions include reprecipi-
tation of metals as oxides, hydroxides, or even carbonates; oxidation of metals
to a higher oxidation state, as in the case of Mn(II) and Fe(II); and trace
element adsorption on and co-precipitation with a newly precipitated solid
phase—e.g., MnOg and Fe(OH)3 have relatively high ion-exchange capacities.
Other secondary reactions are discussed in Section 7.2.1, Metal-Water Interaction.
The amount of the elements extracted from the solids is a very small per-
centage of the total amount available in the suspended residues except for anti-
mony and in some cases cobalt (Table 30). The high percentages related to
antimony can only be explained be differences in analysis methods and analysis
error. There is no correlation between the quantity of cobalt extracted and the
quantity in the solid phase (Figure 19); very little correlation for chromium,
lead, manganese, nickel, selenium, or zinc (Figure 20); and some correlation
for antimony and vanadium (Figure 21). This lack of correlation implies that
dissolution was governed by inherent solubilities of the compounds.
The solubility of metals is known to be pH dependent, as illustrated in
Figure 22 for zinc. The concentration of zinc found in solution after one
contact period and the pH are shown in Figure 23. If the solubility constraints
for zinc oxide (or zinc hydroxide) from Figure 22 are plotted on the data in
Figure 23, Figure 24 results. This correlates fairly well considering that
equilibrium constants used in defining the solubility domain are for 25°C and
zero ionic strength, solution conditions which are somewhat different than the
actual experimental solutions.
Figure 25 shows the data points for soluble zinc as a function of pH
after two contact periods. It is apparent that the extraction of zinc :
from the suspended dust particulates is determined by the solubility in
zinc hydroxide which in turn is controlled by pH. Ztnc :solubilization is
60
-------
TABLE 3Q. PERCENTAGE OF TOTAL METALS EXTRACTED FROM STEELMAKING FURNACERESIDUES
ov
PERCENTAGE OF METAL EXTRACTED
Residue
Number
1
2
3
8
14
17
4
5
6
7
12
13
18
9
10
11
15
16
Antimony
194.4
72.7
144.3
N
14.8
86.7
56.0
N
122.4
58,2
66.7
29.7
37.7
N
347.6
N
72.8
N
Cadmium
K
—
—
N
2.9
N
—
—
—
—
0.2
N
...
—
N
___
...
Chromium
0.004
0.03
0.01
...
__-
—
0.08
0.3
0.06
0.02
0.7
0.05
0.03
.«.
0.06
—
—
0.3
Cobalt
10.6
46.1
7.4
19.0
2.6
11.1
0.4
19.9
0.9
2.5
5.6
0.2
12.1
1.8
30.7
1.3
0.7
1.2
Lead
0.9
0.02
0.04
.__
0.3
0.04
0.02
2.5
0.01
1.4
0.1
0.08
0.09
2.4
1.7
3.8
0.02
0.3
Manganese
2.2
0.4
0.003
0.009
0.1
0.2
0.001
0.03
0.002
0.004
0.005
0.002
0.1
0.01
0.001
0.02
0.02
0.003
Inorganic
Mercury
—
—
...
— .
...
N
N
N
...
—
—
—
Total
Mercury
—
...
—
...
—
...
...
N
N
N
N
...
—
—
—
Nickel
0.2
0.9
...
...
0.2
0.7
0.0008
—
...
0.2
...
0.1
0.08
...
—
>0.4
—
Selenium
N
N
N
N
N
N
3.9
N
1.9
1.6
N
0.6
N
N
N
N
N
N
Vanadium
...
...
—
0.7
..>.
—
0.7
0.8
—
0.9
...
0.6
0.3
—
Zinc
19.5
0.001
0.01
0.1
1.0
0.5
0.008
0.3
0.01
0.007
0.02
<0.03
0.04
0.06
0.01
0.3
0.02
0.09
not reported In solid sample
not found In extracted liquid
-------
ioor
en
ro
u
o
w
X
UJ
o
o
10
jo
0 o18
o1
o
17
12
O
o7
o9
J6
o
15
O1
O4
I02
Cobalt in Residue
Figure 19. The relationship between the quantity of cobalt extracted and the quantity of
cobalt in the residues.
-------
io3r-
CFl
a.
•o
o
2
10'
.18
0°
©
o1
OS
O4
I0a
Manganese in Residue (pg/g)
10°
Figure 20. The relationship between the quantity of manganese
extracted and the quantity of manganese in the residue.
63
-------
J,o'
X
Ul
§
10'
Antimony in Residua (\iq/q)
10*
Figure 21. The relationship between the quantity of
antimony extracted and the quantity of
antimony in the residue.
o
o
PH
Figure 22, Solubility of zinc compounds.
19
64
-------
IOJ
d4
I02
10
ISI
I ,0°
"o
to
Iff1
'T
0
o
9 10
P H
II
,0?
10
t 16
12 * 13
^
Figure 23. Relationship between soluble zinc and pH.
65
-------
10'
o'4
I02
^ .0'
o>
E
o
c
N
» 10°
.a
icr'
0
is
I I I \ I
©V 0"
G9
I 16 I
10 II 12 V 13
Figure 24. The relationship between soluble zinc and pH versus
zinc solubility constraints.
66
-------
\0<
I02
C 10
^
o>
£
10'
JD
o
CO
©'
13
_L
9 10
PH
II
12
13
Figure 25. Relationships of soluble zinc as a function of pH after
two contact periods versus solubility constraints.
67
-------
2+
appreciable in neutral solution, zinc being present as the free Zn ion,
and in strongly alkaline solutions where zinc is present in the anionic
Zn(OH)3" and Zn(OH)42" forms.
Figure 26 shows the concentration of manganese in solution as a function
of pH after one contact period. In contrast to the case with zinc, the
concentration of manganese continues to decrease with increasing pH even in
the strongly alkaline solutions. It is worth noting that Mn(II) is relatively
soluble, while Mn(IV) is relatively insoluble and readily precipitates as
solid Mn02. It is reasonable to assume that manganese is extracted out of
the dusts in the Mn(II) oxidation state. Since the solutions were shaken and
exposed to oxygen in the atmosphere, it is plausible to surmise that some
oxidation of the Mn(II) to Mn(IV) occurred with the resultant precipitation
20
of Mn02. The kinetics of Mn(II) oxidation by oxygen are strongly pH dependent,
and it is unlikely that there was much oxidation and subsequent precipitation
below pH 9. In alkaline solutions Mn(II) oxidation is relatively rapid, and
it would seem that the low manganese concentrations observed for all samples
at pH greater than 11 can be attributed to Mn(II) oxidation to unsoluble
Mn02. In oxygen-containing solutions of high pH, the concentration of soluble
Mn will be small, and Mn02(s) is the most stable (predominant) form of Mn.
To illustrate the effect of pH further, Figure 27 shows three slices across
Figure 20 for three different concentrations of manganese in the dust. The
total amount of manganese extracted out of the solid phase by the ten washings
is seen to be very much a function of pH with more metal leached out, in
general, at lower pH values.
Figure 28 shows the concentration of soluble lead as a function of pH
after one leaching period. At high pH, the dissolved lead concentration
increases, presumably due to'the species Pb(OH)o in a fashion similar to
the situation with zinc and Zn(OH)j. (For the case of manganese, the species
Mn(OH)g does not appear since the manganese(II) probably gets oxidized and
precipitates at the high pH values.) In the pH range 6 to 11, there is
apparently little variation in lead concentration with pH. This is in contrast
to the observations with zinc and manganese where the dissolved metal
concentrations were appreciable below pH 9. The reason for this occurrence
is not known, nor is it known why the concentration of dissolved cobalt is
independent of pH after one contact period, as shown in Figure 29.
68
-------
io-
I02
c:
E 10 '
"- O14
o>
in
e:
a
o>
a ,
2
10°
0)
•g
"o
CO
icr1
o'e,
O2
o16
_
o13 o6
-
o16
m
o"
•MO
® ? uu!
I
i i i 11 i i
7 8 9 10 II 12 1
P H
Figure 26. Relationships between soluble manganese and pH.
69
-------
io-
o>
0>
a.
•o
* I02
0
o
I*
X
LU
o>
(O
0>
c
a
a>
a
10 '
i^^
••
1
I
C
•B
_
— .
O
-
1
pH 7.2
£? pM 8.6
18
O pH 7.9
pH 5.7
©
O%H 12.4
O"pH 12.6
| 'SpH.2.6
G pH 8.9
SPH 10.5
o1'
O^H 12.8
1
17
O P« T.6
-
pH 12.7
pH 10.6
©4 PH 12.0
1
IO4 10s IO6
Manganese in Residues (pg/g)
Figure 27. The effects of pH on the sol utilization of manganese.
70
-------
53^-
IU
I02
o>
s
^ 10'
_J
e14
-------
I02
10
S
o
o
—
jQ
,0°
10"
0-
80
7
-»2 ^.3 *JO
O O O
9
©" **
o16
o's os
J I I I I I t
7 8 9 10 II 12 13
PH
Figure 29. The relationship of soluble cobalt with pH.
72
-------
Variable relationships between concentration and pH occur for many of
the other elements. For inorganic mercury, total mercury, and vanadium, a
lack of sufficient data points does not allow a trend analysis. For antimony,
selenium, and nickel, concentrations after one contact period appear to be
independent of pH as in the case of cobalt (Figure 29).
In the case of chromium, concentrations after one contact period
appear to increase with increasing pH while they appear to decrease with
increasing pH for cadmium (Figures 30 and 31). These results are consistent
with solubility constraints.
It is apparent that the concentration of most trace metals leached
from the dusts is dependent upon the pH of the solutions. Most of the dusts
were alkaline in nature and caused the pH of the solutions in which they
were suspended to increase to varying degrees. Since solution pH is such
an important parameter defining trace metal behavior and solubility, it would
be interesting to determine the basis for this alkaline behavior. If the
metals are present in the dusts in the sulfate form, for example, their
+2 2
dissolution would have no effect on pH (e.g., MnS04 •*• Mn + $04 ) but their
subsequent precipitation would cause a decrease in pH (e.g., Mn+2 + 20H" •>
Mn(OH)2(s)). If the metals were present in the dusts as oxides and they
were highly soluble metals, as for example calcium, their dissolution would
cause the pH to increase (e.g., CaO + H20 •* Ca + 20H~). A plot of the cal-
cium concentration in the dusts versus the pH resulting after one contact
period showed only relatively low correlation between increasing pH and
increasing calcium content in the dusts (Figure 32). The amount of Ca+
actually extracted from the dusts was not measured. A better correlation of
+2
pH with Teachable calcium may have resulted if Ca was quantified.
73
-------
ioV
io2
0»
I
1
o
w
6 10'
V)
10°
O6
o4
-12
0
07
Ol I I I I I I
7 8 9 10 II 12 13
PH
Figure 30. The relationship between.soluble chromium an.d pH.
74
-------
IU
10 '
~
"N.
o>
E
1
•a
0
O
IS
3
OT 10°
*<
©
™» ^
©'*
o18
•^iv
14
Q6'
1 1 1 I I I i
7 8 9 10 II 12 1
pH
Figure 31. The relationship between soluble cadmium and pH,
75
-------
solo-
's
o»
a.
I05
«
•a
•o
-------
SECTION 6
HANDLING AND DISPOSAL OF STEELMAKING FURNACE RESIDUES
The handling and disposal of steelmaking residues are complicated by
their physical and chemical properties. Generally, handling and disposal
practices are those normally associated with solids and sludges from other
types of operations.
6.1 HANDLING
The handling of dry steelmaking dusts presents environmental problems
similar to those of other materials, i.e., wind blowing during transport
'and transfer. This dusting is compounded in the steel industry because of
the fine nature of the residues. In some cases, the dry dusts are either
mixed with water in a pug mill to form a semi-wet product or mixed with
sludges from other iron and steel processes to facilitate handling.
Wet residues present a particular handling problem since prior to
disposal, these residues are dewatered. Settling ponds or tanks are not
effective because of the fine size of the suspended particles. Therefore,
classifiers and thickeners followed by filters are used prior to settling.
Additional operations may include the use of flocculants for settling and
the adjustment of pH for acidity.
Along with fines, a coarse residue known as "sand" is recovered from
the basic oxygen furnace control equipment. These sludges do not present
any unusual problems and are usually separated in settling ponds or tanks,
sometimes with the aid of rakes.
6.2 DISPOSAL
Essentially, three methods exist for the disposal of residues. The
most commonly used method is heaping, either for direct disposal or for
future use; the least used method is the utilization of the residues in
other products; and the final method is recycling back into the process.
77
-------
6.2.1 Waste Heaping
Waste heaping has been estimated to account for as much as 93 percent
of these residues. The heaping can occur either on-site or off-site. In
most cases the residues are applied directly to the land although in some
cases clay or tar linings are used to protect groundwater from waste
Teachings. In one instance, a Hypalon liner was used in a sludge lagoon.
6.2.2 Residue Utilization
As mentioned previously, utilization of the residues to make other
materials is not widespread. Some uses of these residues include paint
pigments, paving, bricks, iron molds, base stone for road building, building
fillers, wall and floor tiles, and "blueing" hydrangeas although markets are
limited.5'21
6.2.3 Recycling
Recycling of steelmaking furnace residues is restricted because of
physical and chemical properties of the residues. At the present time,
approximately 7 to 12 percent of the total furnace residues are recycled.
The potential exists for recycling the residues to the blast furnace, the
sintering operation, or to the steelmaking furnaces.
The chief restrictions on residue recycling to the blast furnaces are
the physical fineness of the residues that causes them to be carried out of
the furnace by the top gases unless the residues are agglomerated. The zinc
and lead content of the residues cause problems with the refractory linings
of the furnace and with the quality of the steel produced. In general, 0.2
percent zinc is considered to be an allowable maximum.
Recycling to the sinter plant could become a viable alternative if certain
difficulties are overcome. These include special handling and storage facilities
to minimize the environmental problems associated with these operations,
the careful and uniform proportionment of the residues into the sinter so as
not to introduce an excessive amount that would weaken the sinter or destroy
its quality, and the control of the amount of zinc-containing sinter entering
the blast furnace.
78
-------
The nature of the sintering process does not allow the removal of zinc
from the residue prior to introduction into the blast furnace. Generally, the
blast furnace can accept a maximum of 0.17 kg zinc per 1000 kg (0.33 pounds
per ton) of molten iron without substantially affecting the equipment. This
restriction necessitates carefully designed proportionment and feed controls.
This is especially true if the burden materials also contain zinc.
Residues can also be recycled to the steelmaking furnaces without
adversely affecting equipment, but with certain restrictions. This recycling
can lead to a buildup of zinc that must be controlled by bleeding. The build-
up of other substances such as lead, copper, and sulfur may prove deleterious
to some grades of steel.
Other factors must be considered. The use of recycling introduces
additional oxides into the furnace that absorb heat from the refining
operation. In a BOF, for example, more molten iron would be required in the
charge, and molten iron is in short supply in many plants. In addition,
recycling equipment and the bleed systems must be designed to provide adequate
environmental controls.
6.3 REGIONAL RECYCLING FACILITIES
The construction and operation of regional recycling facilities could
be feasible in some cases depending on the type of process, the costs, and
the geographical location. The most promising technology involves the mixing
of the residues with a carbonaceous material, such as coking sludge, and
pelletizing, drying, and heating at about 1093°C (2000°F) in a reducing
atmosphere. This procedure results in the removal of approximately 90 per-
cent of the lead and 90 to 95 percent of the zinc. The resulting pellet can
then be charged to either the blast or the electric arc furnace.
Several processes provide the technology necessary to produce suitable
furnace feed pellets. Of these, the one that is in the most advanced stage
of development is the Inland Steel-Heckett-Krupp Process. Others include:
—SL/RN Process, Steel Co. of Canada, Lurgi Gesellschaft fur
Chemic and Huttenswessen, Republic Steel Corp. and National
Lead Co.
79
-------
—Kawaski Process, Kawaski Steel Co.
—Allis-Chalmers Process, Allis-Chalmers Mfg. Co.
--"Heat Fast" production process, Surface Combustion Co.,
Surfac Comb., Div. of Midland-Ross Corporation in collaboration
with M. A. Hanna Co.
—Lummus Direct-Reduction Process, Lummus Co.
~ORCARB-Electric Furnace Process, Swindell-Dressier Corporation
—Arthur G. McKee & Co. Patent
--Wellman Engineering Co. Patent
—Bethlehem Ferro-Tech Process
—Bureau of Mines Process
—Inazaki Process (Japanese)
—Yawata Process
—Alan English Patent (No. 3,386,816)
—Sponge Iron Process of U.S. Steel Co., Geneva Works
A recycling system can be economically feasible when large amounts of
residues are available for treatment, such as with regional recycling plants.
The economics of a 363 x 10 kilograms per year (400,000 tons per year) plant
are as follows:
Capital Costs — $18,000,000
This does not include the environmental costs or the costs for
a 25-acre industrial site, which can be major items. For example,
land costs in four possible regional sites are:
Site Cost Range ($/acre)
Chicago 50,000 - 98,000
Pittsburgh 28,000 - 87,000
Philadelphia 50,000 - 131,000
Dallas 22,000 - 131,000
Charges in capital costs should follow a 0.60 to 0.65 size factor.
80
-------
Operating Costs -- $34.00 per ton residue
Includes amortization, fuel, labor, transportation, and main-
tenance. Operating costs should follow a straight line
function with 40 percent of costs constant and 60 percent
increasing in direct proportion with the size decrease of the
plant.
Revenues — $40.00 per ton residue
Assuming equal revenue values from iron units (0.625 units,
85 to 90 percent iron per residue unit) and zinc units
(0.05 units, 60 percent zinc per residue unit)
Net Return — $6.00 per ton residue
This return provides for a write-off of 8 years. A conserva-
tive 10 year write-off is recommended due to uncertainties
in the cost calculations.
Using the above assumptions and conditions, the economic break-even
plant size would be approximately 281 x 10 kilograms per year (310,000
tons per year). Four steelmaking areas of the United States generate enough
residues to support regional recycling plants. These include:
Chicago 544.3 x 106 kg/yr (600,000 ton/yr)
Pittsburgh 435.5 x 106 kg/yr (480,000 ton/yr)
Philadelphia 283.0 x 106 kg/yr (312,000 ton/yr)
Southwest (Dallas) 239.5 x 106 kg/yr (264,000 ton/yr)
The Philadelphia area may be doubtful because of a borderline net return
potential and because of the widespread locations of the plants and the
associated higher transportation costs. In the southwest area, residue
quantities are low but both general economic conditions and the higher return
on zinc units may provide favorable conditions.
81
-------
SECTION 7
ASSESSMENT OF THE ENVIRONMENTAL IMPACT OF DISPOSAL PRACTICES
A major effort has been concentrated on controlling air pollutants
from steelmaking furnaces to meet emission standards. While reducing
the net emissions to the atmosphere, this control and the subsequent
disposal of the collected residues has created environmental problems in
itself due to the exposure of the residues to environmental elements. These
problems are compounded by the large quantities of disposed residues and the
multiple disposal sites (Table 31).
TABLE 31. ESTIMATES OF RESIDUE VOLUMES
OPEN HEARTH
Year
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
Disposed
Residues1
(106 kg)
466.9
396.0
344.6
289.4
231.5
172.4
109.2
44.7
44.7
44.7
44.7
44.7
Vol ume2
(103 m3)
112.5
95.4
83.0
69.7
55.8
41.5
26.3
10.8
10.8
10.8
10.8
10.8
ELECTRIC
Disposed
Residues1
(106 kg)
259.3
286.9
306.8
326.7
348.4
370.1
391.8
415.4
438.0
466.0
486.9
512.2
Volume3
(103 m3)
61.7
68.1
72.9
77.6
82.8
87.9
93.1
98.7
104.0
110.7
115.7
121.7
BASIC OXYGEN
Disposed
Residues1
(106 kg)
1065.6
1161.4
1226.7
1293.3
1361.3
1431.7
1502.3
1574.2
1583.3
1589.9
1596.4
1601.6
Volume4
(103 m3)
252.6
275.2
290.7
306.5
322.6
339.3
356.0
373.0
375.2
376.8
378.3
379.5
Total
Volume
(103 m3)
426.8
438.7
446.6
453.8
461.2
468.7
475.4
482.5
490.0
498.3
504.8
512.0
Disposed residues taken from Table 2.
2Based on an average bulk density of open hearth furnace residues of 4.15 x
103 kg/m3 (see Table 8).
3Based on an average bulk density of electric furnace residues of 4.21 x 103 kq/m3
(see Table 8).
''Based on an average bulk density of basic oxygen furnace residues of 4.22 x
103 kg/m3 (see Table 8).
82
-------
7.1 PATHWAYS FOR POLLUTANT MOVEMENT '
The residues from steelmaking furnaces are normally discarded in above-
ground waste piles, natural depressions or gullies, lagoons, or landfills.
Pollutants from these residue piles can be dispersed into the enviroment
by air transport, surface runoff, and subsurface migration. Estimates of
the relative probabilities of potential environmental impacts by pollutants
from the various disposal configurations are shown in Table 32.
TABLE 32. ESTIMATES OF POTENTIAL FOR ENVIRONMENTAL IMPACTS
Disposal Technique for
Steel Furnace Dusts
Heaps - permeable base
- impermeable base
Gullies and Natural Depressions
- no dam
- dam with no overflow drain
- dam with overflow
- dam with drain
Lagoons
Landfill - lined
- un lined
PATHWAY
Air
Transport
1
1
3
3
3
3
5
4-5
4-5.5
FOR POLLUTANT
Surface
Runoff
2
1
1
3
2
2
3
5
5
MOVEMENT
Subsurface
Migration
3
5
3
1-2
2
, 4
1
4-5
2
Relative probability of transport: 1 - very acute; 2 - —; 3 ; 4 - —;
5 - little movement potential —initial sampling sufficient.
83
-------
7.1.1 Air Transport
Air transport of solid pollutants starts with entrainment of the
particles. Entrainment can occur at each stage of the handling/disposal
process. The potential for entrainment must be a function of the handling
procedures of the plant, the characteristics of the residues (i.e., wet
or dry, size, shape, etc.), transportation from the plant to the site,
characteristics of the disposal site, and local climatic and topographic
conditions.
Entrainment can begin at the collector transfer site and continue in the
holding area, during the transfer to the disposal area, and in the disposal
area. The emissions from each process are a function of existing conditions,
the transfer mechanism, and the^exposure to atmospheric processes. A
quantification of these emissions is not readily available.
Entrainment of the dust particles arises from wind erosion. This
commonly occurs during the .transport of the dust to the disposal area
by truck or by the action of the wind or traffic in the landfill area. The
dust particles are picked up when the wind energy, expressed as the shear
stress, at the ground exceeds a threshold value which is dependent upon the
22
particles' size, shape, and density. The shear stress, T, is given by:
T =
where u* is the velocity of the air at a characteristic distance above the
ground dependent on the surface roughness, Z . The wind speed, u, at ane-
mometer height Z, is related to u* by the turbulent and thermal properties of
the lower 10 meters of the atmosphere. The simplest relationship is:
u/u* = 1/0.4 - ln(Z/ZQ).
Changes in roughness length from one location to another or at different loca-
tions at a site can make a substantial difference in the wind energy and,
therefore, the emissions. The threshold velocity required to get a dust par-
ticle airborne should be linearly related to the density of the material and
84
-------
its aerodynamic cross-section presented to the wind. Some general relation-
ships about threshold wind speed and particle pickup have been presented.23
In this study, it was found that winds^of 22.4 m/sec (50 mph) or more were
required to set in motion particles as large as 2000 ym; particles between
1000 and 2000 urn could be transported under ideal conditions by wind speeds
of 15.6 to 20.1 m/sec (35 to 45 mph); winds between 4.9 and 13.4 m/sec (11
and 30 mph) transport particles having sizes of 80 to 1000 ym; and as the
particle diameter decreased below 80 ym, the threshold speed increased,
so that particles of 2 ym diameter or less are not moved by winds below 22.4
m/sec (50 mph). The reversal of threshold speed is caused by the smaller
grains being sheltered by the larger ones and increased adhesion of the
small particles.
In the case of steelmaking residues where the majority of particle
sizes are consistently below 80 ym, the effects of sheltering by larger
particles would be negligable. However, the effect of the lack of sheltering
is not quantifiable but is Instrumental in the lowering of the threshold
wind speed.
Surveys of particulate emissions from open coal piles showed a threshold
wind speed of about 5.4 m/sec (12 mph) was needed to initiate emissions.
The emission rate was dependent upon an exponential power of the wind speed
between 2.5 and 3.0, when the wind speed exceeded the threshold value.
Various characteristics of disposal sites can affect emissions from
residue waste heaps. The. most important characteristics are configuration,
topographic location, and the moisture content of the residue. Also, the
effect of time on heaps and the presence of plant life can reduce emissions.
The configuration of the residue heaps affeets eramission rates. The
above-ground dumped wastes in cone-shaped, tall piles or flat, fill-type
mounds are more susceptible to wind erosion than waste landfills, lagoons,
or gullies. The use of multiple small heaps rather than one large heap
increases dusting because of the exposure of a larger surface area to the
wind. In the case of a landfill residues can be transported by air'ff
daily covering of the filled dust with ,a :soil layer is not practiced.
Lagoons or settling ponds should not present a dusting problem unless
they are filled and evaporation causes the exposure of dried material to
85
-------
the wind. The disposal of residues in gullies or natural depressions
would reduce their exposure to winds; however, wind movement can occur in
small gullies, and a certain amount of wind channeling may occur if the
gullies are large enough, thus producing wind erosion. This is a complex
problem that is mostly site dependent.
Other factors that must be considered are the presence of tree buffers
and siting on hillsides or in valleys.
The moisture content of the residues has an important influence on
dusting. Laboratory data indicate that there is a decrease in the number
22
of particles that become airborne as solid moisture increases. The
practice of disposing of steel residues in a wet or semi-wet form thus reduces
the chance of dusting.
The effect of time on the surface of waste heaps could have an important
influence on dusting. Older piles, exposed to weather effects over a longer
period of time, may produce a crusty surface by agglomeration or binding due
to water.
The presence of plant life on waste heaps can reduce dusting. Even
with the high metals content of steelmaking residues, plants should eventually
grow on older undisturbed heaps.
Once airborne, particle characteristics such as size, shape, and density
play an important part in the transport of these pollutants from the piles.
The size distributions for residues from steelmaking furnaces vary substan-
tially for a given process and in many cases for residues from the same type
of process. ' ' The examination of dusts by electron microscopy revealed
individual spherical iron particles in the size ranges of 0.04 to 62.09 ym,
0.02 to 35.00 ym, and 0.04 to 2.30 ym for electric, basic oxygen, and open
hearth furnaces, respectively. These individual particles were frequently
agglomerated into larger particles. In addition to the iron spheres, irregular
carbon particles up to 900 ym long were found in all types of residues;
irregular calcium particles up to 183 ym long in the electric and open hearth
residues; and 3.50 to 13.90 ym cubes of titanium in one electric furnace residue.
Once these particles become airborne, some may remain for long periods.
Studies of atmospheric dust have shown that the settling velocity of particles
of 150 ym or greater is high enough to cause nearly instantaneous removal from
86
-------
the air, particles in the 1 to 150 ym size range fall out at a time dependent
upon the inverse square of the radius, and particles smaller than 1 ym are
nearly permanently airborne, having residence times of from 9 to 99 years.22
The consideration of particle size ranges similar to those encountered by
the electron microscopy examination (i.e.., 0.01 to 10.00 ym) provides large
terminal velocity variations. For a density of 10 g/cm3, at atmospheric
temperature and pressure, a 10 ym particle has a terminal velocity of 2.95
cm/sec allowing it to fall one meter in approximately 34 seconds. The 0.01 ym
particle has a much smaller terminal velocity of 2.95 x 10"6 cm/sec. At this
rate, the smaller particle will take more than one year to fall one meter.
Terminal velocities are also affected by both particle density and shape.
Density has a nearly linear effect on terminal velocities while the effect
of particle shape is essentially unknown due to both a lack of research and
the variations in the measurement of equivalent particle size. Generally,
the velocity relationships for spheres approximate (within ± 20 percent)
those of irregularly shaped particles for Re < 50 when particle shapes are
not extreme and particle size is measured as the diameter of equivalent
25
spheres.
7.1.2 Surface Runoff
Surface water runoff resulting from precipitation may transport sus-
pended particles or dissolved material from the furnace dust waste deposition
site. Heavy metal transport represents an immediate hazard to stream and
reservoir water quality. Also, heavy metals may be transported from the
original disposal site to surrounding land and vegetation.
The magnitude of transport by surface runoff and the relative importance
of surface runoff compared to subsurface water movement depends largely upon
the methods of disposal. Wastes buried in soil pits do not present a surface
runoff problem, whereas waste piles exposed to precipitation are;subject to
particle dislodgement and transport as surface runoff. Also, any water that
infiltrates the waste pile and confronts an impermeable base will be more
likely to produce runoff containing dissolved waste material. Waste retained
87
-------
in gullies or depressions are..Impacted by runoff from upland portions of the
watershed and this can greatly magnify the liquid contacting the waste in
regions of moderate to high rainfall. Thus, the principles-of runoff must
be considered for direct runoff or water falling on the dust and for runoff
aggravated by the dust storage location. In addition, overflow or drainage
from lagoons, darned gullies or depressions may result in overland flow or
channel flow of metal-laden water below the dam.
Factors which have been shown to affect erosivity of soil will most
likely affect the erosivity of waste material. Therefore, one can hypo-
thesize that the physical characteristics of the waste piles that will affect
the erosivity and the water waste contact time include particle size and
specific gravity, cohesive forces between particles, and shape of the pile
(mainly slope length and percent).
The waste particles are smaller and heavier than most soil particles
and relationships developed for predicting soil detachment by raindrops and
soil transport by surface runoff would have to be utilized with extreme
caution. The relationships between detachment and movement of soil
particles by raindrops to*".the particle sizes are given by Table 33.
X
TABLE 33. RELATIONSHIPS BETWEEN DETACHMENT AND TRANSPORT OF SOIL
PARTICLES BY RAIN
Particle Size Relative Amount of Detachment
Particle (ym) and Transport (%)
Coarse Sand
Medium Sand
Fine Sand
Very Fine Sand
Silt
590 -
250 -
175 -
50 -
2 -
840
420
250
100
50
30.0
77.2
100.0
61.0
21.0
A higher percentage of fine sand was detached than coarse sand, and
more coarse sand than silt. Most of the steel works waste particles are clay
size (> 2 ym) but have specific gravities of 3 to 5 while soil mineral particles
-------
vary in specific gravity from about 2.5 to 2.75. It is difficult to say
whether the metal particles would behave more like silt or fine sand. It
was noted in one study that the smaller particles were compacted, the surface
sealed, and a film of water developed at the surface which helped dissipate
the energy of the falling raindrops and reduced the amount of particle
detachment.
The relative amounts of infiltration and surface runoff are dependent
upon the compacting, crusting, aggregating, or agglomerating occurring at the
surface. A crusted surface may produce sheet runoff which can reduce dis-
lodgement of particles by reducing the amount of raindrop impact energy
transferred to dust particles, but any sheet runoff also provides a means of
transport for dislodged particles. A crusted surface may crack and induce
gully or rill erosion. Thus, the erosion effect of crusting or compacting
can best be determined by first carefully observing differences in pile sur-
faces and second, noting the relative movement of surface materials which has
already occurred.
Other factors which will affect the surface nature of the pile are the
initial texture and moisture content of the wastes when first piled and the
surface changes due to wetting or drying. Dry particles may not wet easily
without a mixing action to overcome fine particle cohesive forces; therefore,
if the material is deposited dry, the raindrops may tend to bead and runoff.
The shape and distribution of piles will determine entrapment and sur-
face storage of water which can result in less direct runoff and more infil-
tration. Greater side slopes increase water velocity and, therefore, erosion
potential. Erosion on the pile will also depend on length of slope. The
effect of degree and the length of slope have been combined into a generalized
equation of predicting soil erosion under field conditions:
X= (C)(S)1-4(L)1'6
where X is the soil loss, S is the land slope in percent, L is the length
of the slope and C is a constant that depends upon the infiltration rate,
physical properties of the soil, intensity and duration of the rainfall, and
other factors. Thus, for identical pile characteristics, erosion will be
a function of the characteristics of the precipitation. The dislodging of
89
-------
soil particles by raindrops to several rainfall characteristics^ is related
by
Dislodging = f[precipitation intensity x time x dropdc°£Ss-sLtion
2
x {drop velocity) ].
This indicates that climatic conditions of the disposal site will influence
surface runoff and erosion because of different rainfall amounts and patterns
of intensity and drop kinetic energy. In addition, the relative amounts of
precipitation and evaporation vary with regions causing some regions to have
moisture excess. This normally means more surface runoff and overflow of
reservoirs (including waste lagoons). These climatic factors can be assessed
from available weather data.
Once the particles have been detached and transported to the base of
the pile, any further transport will depend largely upon the runoff and
erosion potential of the surrounding area. In order to transport the steel-
works waste particles, certain minimum or threshold velocities will have to
be obtained. An equation for threshold velocity of unigranular materials
ranging in diameter from 0.35 to 5.7 mm and in specific-gravity from 1.83 to
29
2.64 has been developed. This equation is
Vt=0.5d4/9 (G-l)1/2
where Vt is the threshold velocity (ft/sec), d is the particle diameter (mm),
and 6 is the specific gravity. Using G = 5 and d = .002 mm, the predicted
Vt is 0.02 m/sec (0.06 ft/sec). Overland flow velocity is typically 0.30
to 0.61 m/sec (1 to 2 ft/sec); therefore, this equation would predict that
these waste particles would be transported. However, caution must be used
since particles of this size and density were not used in developing the
equation.
90
-------
Using previously developed equations for predicting soil detachment and
transport, it can be proposed that most of the same factors such as rainfall
intensity, raindrop kinetic energy, slope, and slope length of pile will
influence transport of these waste particles. Factors such as surface
crusting, infiltration, and solubility of wastes can potentially have a large
effect. While snow melt is not covered here, it-certainly would be a con-
tributory factor in snow regions. Relatively little is known of this runoff
mode and specific experimental consideration would be necessary to determine
these effects. Therefore, the initial steps in determining the environmental
impact from the steelworks waste piles should probably be to observe the waste
piles to detect runoff and erosion, and to sample water and sediment that may
indicate transport of the waste from the deposition site.
7.1.3 Subsurface Mi grat ion
A certain amount of water that comes in contact with steel mill
furnace dusts will leave the area as subsurface water. This water may move
vertically into the groundwater or laterally to finally enter surface waters
such as a stream or lake. Waters may also move laterally to a recharge area,
then vertically to a groundwater aquifer. It is assumed in this section that
these waters may carry materials that are toxic or at least potentially harm-
ful. Several examples of groundwater contamination due to the movement of
30
chromium from disposal sites for electroplating wastes have been documented.
While there are probably many other references to groundwater contamination by
metals that are contained in mill waste, no attempt to perform an exhaustive
search of the literature has been made. Instead, the general conditions
and factors affecting the subsurface movement of any materials from a disposal
site and methods that could be used to detect such movement are discussed. The
existence of toxic materials in the mill wastes, their solubility in water,
and the extent to which they will be tied up by the soil particles are covered
in the following sections of the report.
It seems clear that any subsurface movement of pollutant materials in soil
will take place in soluble form. The subsurface movement of particulate matter
in soil beyond a centimeter or-two is almost certain to be negligible. Experiments
with soils have shown negligible translocation of even clay sized particles
91
-------
31
(< 2 urn) for short-term experiments. Many of the materials present in mill
wastes have been shown to be soluble in water and can thus move with the water
through the soil profile and beyond.
The directions and rate of movement of pollutant-laden waters from mill
waste disposal sites will depend on several factors. These include factors
that are dependent on the pollutant being transferred, site factors,- soil
factors, and climatological factors. One important factor in the first
category is sorption of the pollutants on the soil particles. Some materials
will be tightly "sorbed" and will not move through the soil while others
such as the Cl" ion move freely with the soil water. Obviously those materials
that are held tightly will pose less of an immediate problem than those which
move more readily. Possible chemical or biochemical reactions which change
the form of "sorbed" ions and increase their mobility should be considered in
this regard.
Site factors include slope, distance to groundwater, site elevation, and
surrounding streams, lakes, and wells (both present and future). The
characteristics of the disposal site (piles, landfills, lagoon, etc.) are
also included here and will be discussed in detail in a later section of this
report (Section 7.3). The site factors determine the direction and magni-
tude of prevailing hydraulic gradients and thus the movement of pollutant-
laden waters.
Soil factors include those soil properties that govern the rate of
subsurface water movement and the dispersion of solutes as the water moves.
These include the hydraulic conductivity (both saturated and unsaturated),
the dispersion coefficient which is dependent on both the soil and the solute,
lenses, layering, and the depth and thickness of restricting layers.
Climatological factors are of obvious importance since the pollutants
are carried by water. In a high rainfall area deep seepage to groundwater
aquifers is more likely than in an area where potential evaporation exceeds
rainfall (Figure 33). Thus subsurface movement of pollutant materials will
be slower in drier regions, and pollutants will be less likely to be flushed
out of the soil-water system; by infiltrating waters.
92
-------
to
420
430
450
+ Potential evapotranspiration more than
mean annual precipitation
- Potential evapotranspiration less than
mean annual precipitation
Figure 33. Potential evapotranspiration vs. mean annual precipitation (inches).
-------
There are essentially two methods of determining the extent of subsurface
movement of pollutant materials from mill waste disposal areas. Either distri-
bution of wastes in the soil-water system can be measured on a site-by-site
basis, or the problem can be approached from the theoretical standpoint with
the development of methods or models to describe pollutant movement in terms
of the factors discussed above. The latter approach would be the most
general and would have the advantage of being applicable to alternative sites
and land disposal methods as well as existing sites. However, it would be
extremely difficult to develop and verify a prediction model because of the
wide range of conditions that now exist. Some general models for solute
movement in soils and groundwater bave already been developed. Others are
in the development processes. However, as a first approach it would appear
best to actually measure distribution of wastes in the underlying soils and
aquifers near the sites to assess the likely environmental impact of such
disposal methods.- Methods for making these measurements are outlined in the
discussion of various types of disposal sites (Section 7.3).
7.2 POLLUTANT INTERACTIONS
Metal pollutants removed from the disposal site by water systems can
interact with the intercepted water, soils, or plants. Soils and plants can also
act as sinks for the metals and can serve as the introducers of the metals into
the food chain.
7.2.1 Metal-Mater Interaction
Metal ions in water undergo extensive hydrolysis and complex formations.
Several of the metals exist in different oxidation states in the redox range
of most aqueous systems. The pH-E, plots of various metals like iron and zinc
"V? ^?
in solution phase have been elucidated by several studies. There are many
chemical species of each metal which vary with the reaction (pH) and redox (E.)
condition of the acquatic systems. At pH's near 7.0 with a redox of nearly
200 mV, zinc and iron in pure water exist as ionic species, Zn++ and Fe++.
Thus, the potential for ion movement can be increased with Eh and pH shifts
likely under furnace dust disposal.
94
-------
Metal ions react and sorb with the organic molecules and may result in
formation of soluble chelates such as ironrhumate and insoluble complexes such
as calcium-humate. Soluble metal ions and chelates are mobile and constitute
the most common dispersion route of metals with potential impact on surface
as well as groundwater.
Under alkaline (pH 10-12) and oxidizing conditions, such as the steel
mill dust, certain metals transform to their oxidized state such as Fe(OH3),
Fe203, Mn02, etc. Kinetically, oxidation of Mn2+ is relatively slow. There-
7*4* Oi
fore, Mn tends to persist longer in aerated waters than does Fe . These
oxidized states of many metals may eventually lead to formation of colloids
and precipitates under high ionic strength and osmotic concentrations. Under
reducing conditions, metal ions tend to transform to lower oxidation states
which are more soluble and stable under acid conditions.
Significant amounts of antimony, mercury, cobalt, lead, zinc, chromium,
selenium, and manganese solubilize in water washings of airborne dust from
steel mills (Table 34).
TABLE 34. WATER SOLUBILITY OF METALS FROM THE AIRBORNE DUST OF 18
STEEL MILLS
CUMULATIVE CONCENTRATON (ppm)
IN 10 WASHINGS
El ement
Pb
Zn
Co
Cr
Mn
Sb
Se
Hg
Cd
V
Ni
Maximum
1437.9
1145.3
410.0
253.9
228.1
184.9
173.7
136.1
13.7
6.5
1.9
Minimum
2.0
0.9
4.0
0.1
0.2
22.2
0.9
10.0
1.3
1.1
0.5
Remark
Only Electric
Furnaces
95
-------
Considerably greater amounts of metals may be solubilized in acidic waters
or waters with organic and inorganic impurities as might be encountered at
a variety of disposal sites especially when mixed with other wastes. The-
presence of mercury in water extract of metallic dust from electric process
steel mills indicates its soluble form as ions Hg , Hg++. Mercury, lead,
and zinc appear to be the most hazardous of all elements in metallic dust
from environmental considerations of surface water pollution.
7.2.2 Metal-Soil Interactions
The soil matrix consists of mineral particles in clay (< 2 y), silt
(2-50 y) and sand (> 50 u) size fractions; organic matter at varying stages
of decomposition; living microorganisms; water with soluble ions and com-
plexes; and air. All these soil components interact with metals reaching the
solid system in various forms and amounts.
Chemically the airborne steel mill.dust is abundant in iron, zinc,
chromium, calcium, manganese, and/or lead in contrast to soils, the mineral
fraction of which is made up of alumino-silicates (Table 35). Metallic
dusts from 18 different steel mills possessed a wide range of elements in
varying proportions in as yet undetermined chemical combination or form.
Some of the salient interactions that metals undergo in soils are out-
lined below.
Ion exchange reactions and adsorption—Soils with high cation exchange
capacity such as clay and organic soils would absorb and retain considerable
amounts of metallic ions. The order of adsorpt-ion iof metal ions on soil
colloids is approximately Cu>Pb>Ni>Co>Zn>Ba>Rb>Sr>Ca>Mg>Na. Thus constituents
of furnace dust can be screened using the least absorbed species,
Metal fixation reactions—Chemical reactions in which metal ions bind to
the soil inorganic fraction irreversibly or substitute isomorphically for
other ions of similar radii from mineral structure, do occur and are an essential
feature of chemical weathering phenomena and clay formation:
M2+ * M - X -> (XMX)
(soluble) (exchangeable) (fixed) ,
96
-------
TABLE 35. ELEMENTAL COMPOSITION OF STEEL MILL AIRBORNE DUST AND
QF A MINERAL SOIL
El ement
Iron, Fe
Zinc, Zn
Chromium, Cr
Calcium, Ca
Manganese, Mn
Lead, Pb
Silicon, Si
Potassium, K
Nickel, Ni
Magnesium, Mg
Sodium, Na
Chloride, Cl
Molybdenum, Mo
Copper, Cu
Aluminum, Al
Phosphorus, P
Cobalt, Co
Selenium, Se
Tin, Sn
Scandium, Sc
Cadmium, Cd
Arsenic, As
Vanadium, V
Carbon, C
Antimony, Sb
Silver, Ag
i .._
METALLIC
Maximum
> 50
> 32.5
18.1
14.2
13.2
10.86
6'.7
5.4
4.64
4.5
3.1
1.6
0.685
0.57
0.45
0.35
0.31
0.244
0.192
0.11
0.104
0.055
0.053
0.03
0.026
0.013
•**ll'*'lll''**l*i^—f*^^*'l'll*iiii^'ll^^a*a^^**^^^^^****mmmii*mm*tlm
DUST %
Minimum
31
0.08
0.018
0.34
0.288
0.011
0.44
0.02
0.004
0.36
0.03
0.03
0.005
0.014
0.007
0.04
0.003
0.016
0.002
tr*
0.035
0.003
0.008
tr*
0.002
0.002
^^•'^^•••^^•••^•^•^HMIH^HHWHH'I^HM^AVVI^^VBB^HH
Soil, %
4.0
0.02
0.01
3.6
0.1
0.01
27.7
2.6
0.004
2.1
2.8
0.01
tr*
0.01
8.1
0.07
0.001
tr*
tr*
tr*
tr*
0.002
0.01
2.0
0.001
tr*
tr refers to trace, < 10 ppm in total sample.
97
-------
Manganese and aluminum, for example, may substitute for iron and silicon,
respectively, in silicate clay structures, creating a permanent change.
*
Thus, metal ions are fixed permanently and irreversibly. This pathway
represents a desirable route since, in this form, metals become unavailable
for movement or uptake.
Precipitation—Reaction of metal cations with anions like phosphate
carbonate, and sulfide leads to formation of insoluble precipitates with
extremely low solubility products. Lead in solution may react with phos-
phates and molybdates, forming very insoluble precipitates. The metallic
colloid particles may aggregate and the precipitate in high-salt waters by
forces of coagulation and flocculation. The order of insolubility of various
metallic sulfides is Hg>Cu>Sn>Pb>Ni>Co>Cd>Zn>Fe>Mn>Mg>Ca.
Chel_ation_--Coordinate bonding between a metal ion and chelating ligand
in soil may form complexes of varying solubility constants. The soluble
metal complexes may be anionic, cationic, or without any charge dependeng on
the pH of the media. Formation of low molecular weight soluble metal-
organic complexes increases the mobility of metals, and the chances of sur-
face and groundwater contamination are greatly increased.
The insoluble chelates, on the other.hand, irreversibly bind and prevent
the metal pollution of water. The order of stable chelate formation of
various metals in Hg>Cu>Ni>Pb>Co>Zn>Cd>Fe>Mn>Zn>Mg>Ca.
Oxidati'on-reducti'on--Most metallic ions undergo change in their oxidation
state depending on the redox status of soil. Soil redox is strongly influenced
by moisture aeration relationships, wetting and drying, and nature and extent
of microbial activity. Certain metals under aerobic conditions oxidize
quickly and may form metallic oxides like Fe203, Mn02, etc. These trans-
formations may be directly mediated by soil microorganizms. The oxidized state
of metal ions is virtually immobile, thereby reducing the probability of
water pollution. Reduced forms of metal ions are soluble and more subject to
move with leaching water than are oxidized forms.
Methylation-demethylation—Certain metals like mercury, arsenic, selenium,
and cadmium are biochemically methylated and demethylated in soil and sediment
"systems. Methylated metal species are more mobile and toxic as contaminants
than elemental or ionic forms. Methyl-mercury can move with water flow to
plant roots or to streams where it can enter the food chain or may be enzy-
matically demethylated into elemental forms.
98
-------
Mineralization-immobilization--Certain metal ions such as iron,
manganese, copper, zinc, calcium, magnesium, potassium and molybdenum are
essential in trace amounts to various microbiological functions and synthesis
of biotnass. Effects of different nonessential metals on soil microbes have
not been well studied. Silver and mercury are most toxic followed by
cadmium, nickel, lead, chromium, and barium.
7.2.3 Metal-PIant Interactions
Plants require 16 elements for growth and essential metabolic functions.
Of these, iron, copper, zinc, manganese, calcium, magnesium, potassium,
molybdenum, cobalt, phosphorous, and chlorine are present in the metallic
dust. Non-essential elements present in the dust are arsenic, silver,
cerium, cadmium, chromium, silicon, lead, aluminum, barium, sodium, nickel,
vanadium, tin, scandium, yttrium, zinc, niobium, and antimony. Very little
is known about the interaction of non-esential metal elements with plants.
Metal ions may reach the foliage of plants in the form of dust or dirt blown
by wind. In solution phase, dissolved metal ions or complexes may reach the
plant roots by root interception or contact (e.g., Hg, Cu, Ni, etc.), mass
flow of solute metal with water (e.gr, Fe, Ca, Mg, Al, etc.), and diffusion
of metal ions in accord with anvion activity gradient (e.g., Zn, Mn, etc.).
Once the metal ions and complexes reach the root surface, they may
compete with other ions for carrier site uptake. Plant species differ
markedly in accumulation of and requirement for different metals. Entry of
metals into plant roots and further trans!ocation to tops may be categorized
into the following modes:
1. Metals like iron form strong, slowly exchangeable chelates at the
root epidermis and move intact from epidermis to leaves.
2. Metals like copper and chromium form similar chelates as iron
with little or no ability of chelate to enter the xylem.
3. Metals like zinc and manganese remain soluble in the chelator
pool and the chelators hold very little control on their
movement.
4. Metals like lead and aluminum may be precipitated by phosphate
or may form a mixed chelate precipitate with the ion exchange
surfaces in the root.
99
-------
5. Metals like mercury and nickel may bind to very stable, slowly
exchanging macromolecules or ion exchange surfaces and not be
soluble enough to reach the xylen parechyma cells.
Increased mobility of metal ions by solubilization or formation of soluble
metal-chelates also increases their availability to plants.
Above a certain critical concentration or accumulation threshold level,
the metal may exert a toxic influence on plants and animals. An attempt to
classify the metals based on their 50 percent lethal dose has been made and
34
their modes of toxic effects have been discussed. The order of problem
metal toxicity to oats is Ni>Cu>Co>Zn>Mn, whereas, to sugar beet it is Co>Cu>
Zn>Ni>Mn. Some of the metals like Cd and Se may accumulate in plants at
levels that are lethal for animal and human consumption. Plants may, exert
influence on solubility and transformation of metals in the rhizosphere.
Initially, accumulation of large amounts of metals from steel mill wastes may
prohibit vegetation growth at the dumping site. Sooner or later, metal-resis-
tant flora and fauna would evolve with ecotypes adapting to metal rich environ-
ments. Plant species vary in evolving ecotypes tolerant to one metal and not
to the other.
7.3 SPECIFIC DISPOSAL CASES
This section examines specific disposal cases for steelmaking residues.
The potential for environmental effects and methods for sampling and evaluation
are presented.
7.3.T Heaps
Heaps are the most commonly used method of residue disposal. Most heaps
have permeable bases although some heaps with impermeable base "liners" do exist.
Infiltrating water can carry waste material below the soil surface and from
there to either groundwater (Figure 34) or laterally to surface waters (Figure
35). Lateral movement is the critical element when restricting layers are at
a shallow depth, if the heap is located on an incline or in a position where
water is moving laterally from other surfaces, and in the case of heaps with
impermeable bases where no downward movement exists.
The magnitude of- this problem will be affected somewhat by infiltration and
seepage through the waste material. If the material is a very fine dust, it
100
-------
wiu,-
deraectlc, Induttrial
at el»y crater tvpply.
Contaminated ._
= , - —^
Zone
Figure 34. Migration of pollutants to groundwater.
101
-------
o
ro
UNSATURATED
ZONE
Contaminated
Zone
Figure 35. Lateral movement of pollutants from waste heaps to surface
water.
-------
may repel water and thus reduce the problem of seepage through the pile and
into the soil underneath. This would increase the problem of erosion and
movement of the material via surface runoff, however. A permeable base will
tend to reduce the amount of surface runoff, especially if water infiltrates
through the pile easily.
Determination of waste transport from heaps by surface runoff can be
made by observations and by sampling of water and sediment. Heaps can be
observed for development of gullies or movement of material toward the base.
Surface soil samples should be taken upslope and downslope of the heap and
analyzed for variation in heavy metal concentration. If there is a distinct
gully or channel near the heap which has periodic flow, water samples
should be taken. Sediment samples from the channel's bed should be taken
with distance from the heaps. Water samples should be taken from any ponded
area that may Include drainage from the waste site. Nearby streams should
be sampled upstream and downstream of the disposal site during base flow
and during stormwater runoff events. Sampling frequency will depend
upon whether automated samplers are used, but some samples should be taken
during the first hour of a rainfall runoff event when erosion potential
is probably highest.
Some laboratory experiments may help if modeling is desired. Small
heaps of wet or dry wastes with various slopes and slope lengths could be
exposed to simulated rainfall to determine erosion potential and changes in
the waste pile surface characteristics. However, the extent of the environ-
mental pollution problem due to surface runoff needs to be quantified before
developing elaborate experiments in the laboratory.
Under well-drained conditions of a heap with a permeable base, an
aerobic environment would prevail and many soil fungi, actinomycetes, and
bacteria, may actively invade the dust metals at the heap-soil surface.
Soil microorganisms may assimilate and utilize some of the metals. In acid
soils at pH below 6.5, metals like sodium, calcium, magnesium, arsenic,
rubidium, antimony, and strontium may become soluble and may possibly contri-
bute to pollution of surface waters. At pH above 6.5, little groundwater
pollution is expected. Wetting and drying of the pile may lead to oxidation
of iron, aluminum, and manganese in the steel mill dust to form ferric
103
-------
oxide, aluminum oxide, and manganese dioxide. These sesquioxides can adsorb
other heavy metals but there is very little likelihood of their movement
unless severe erosion of heap and soil takes place. Loss of organic matter
by intense decomposition may create circumstances where copper is lost in
drain waters and may ultimately contaminate the streams with copper.
Electric process steel mill wastes possess, high molybdenum and mercury
which require special management practices to prevent the pollution of
surface and groundwater. Molybdenum presents a pollution problem under
alkaline conditions when it is most soluble and mobile, whereas, mercury
becomes mobilized by the formation of methyl mercury.
Various factors affecting aerobic transformations and transport of
metal ions are: temperature; rainfall intensity; soil-metal heap contact
area; composition and characteristics of metal heap; soil characteristics
and pH; redox conditions, fluctuations, and associated transformations
of metal ions in metal-heap-soil systems; and biological and biochemical
forces of weathering that produce organic substances that dissolve,
chelate, or stabilize metals in residue heaps.
The present distribution of pollutant materials can be obtained by
taking soil samples at various depths both close to and laterally removed
from the heap. Care should be taken'so as not to contaminate samples with
waste materials near the surface. The device described by Terry, et al. for
taking soil samples from specific zones would appear suitable for this pur-
35
pose. It may require modification for deep sampling. Sampling should
be concentrated in the direction of lateral groundwater movement which could
possibly be determined from surrounding wells, streams, etc.
If possible, both water and soil solids should be analyzed at each
sampling depth. This would indicate the proportion of material that is
tied up on the soil solids as well as that which is mobile.
Groundwater should be sampled on a continuing basis from observation
wells placed in the vicinity of the heap. Existing wells in the area should
be sampled and analyzed for the various pollutants.
The most critical case would be a shallow water table (10-30 ft) with
permeable soils such as sands or sandy loams. These sites might be found in
the Great Lakes area. On the other hand, pollutant concentration would tend
104
-------
to be higher in tighter soils, and the zone of contamination would move
more slowly. The hazard here would be movement of the contaminated zone
over several decades and its subsequent effect on water supplies.
The long-term fate of the heaped waste materials may also have impli-
cations so far as groundwater is concerned. The following case demonstrating
this point is cited.30
"A unique case of groundwater contamination by chromium
occurred in the city of Srandville (Michigan) west of Grand
Rapids. In this incident, the city drilled a public supply well
in the glacial-drift deposits along the Grand River. To protect
the well from flooding during periods of high water, the casing was
extended several feet into the air and the land surface was raised by
filling with sand and gravel. In time, chromium was detected in the
water. This resulted in considerable consternation, since there
was no apparent source of chromium contamination in the vicinity.
Investigation by the Grandville Superintendent of Water revealed
• that the sand and gravel fill used to raise the land surface at
the well was taken from a former dumping grounds for electroplating
wastes I The river was in flood stage shortly before the contamina-
tion appeared. Water from the river obviously had leached the
chromium from the fill and carried it into the aquifer."
7.3.2 Gullies or Natural Depressions
This case may be most hazardous of all so far as water pollution is
concerned because water from a larger watershed area passes over or through
the waste with the potential to both erode the actual material and to trans-
port soluble species downstream and through the soil.
7.3.2.1 Mill Waste Placed in Gullies or Natural Depressions with No
Retention Structure—
This case has maximum potential for surface runoff. Contamination
can occur from both erosion of solid material and runoff water contaminated
from having gone through filled waste material. Again, there is subsurface
seepage both laterally and vertically as described above for above-ground
heaps. Both permeable and impermeable bases apply. A similar sampling
procedure as described above for groundwater and for soil can be used for
this case.
105
-------
In this case, there is a large amount of surface runoff and erosion
because the material has been located in a natural drainage channel. Rainfall
events would be the critical time for sampling water upstream and downstream.
Between rainfall events, sediment should be sampled downslope of the waste
pile.
7.3.2.2 Mill Waste Placed in Gullies or Natural Depressions with a Retention
Structure--
Runoff from wastes in gullies is an obvious problem so a dam may
have been put in the gully to prevent direct runoff (Figure 36). Also the
mill waste itself may have redistributed to form a dam which results in
the same situation. In this case the downard seepage would be maximized.
Water would be ponded behind the' dam or retention facility and slowly
move either laterally or vertically from the heap. The increased infiltra-
tion under the wastes could cause a localized rise in the water table and •
lateral movement of the contaminated water. Where this situation already
exists, drilling and sampling vertically under or close to the waste would
be the best way of determining potential hazard to groundwater as discussed
above for heaps. If the disposal site is located on a hilly area or if there
are wells or natural drains such as streams or canals in the area, there may
be considerable lateral water movement in the saturated zone.
For this case, investigations should be concentrated on groundwater and
lateral seepage rather than runoff. Runoff would occur only from overflow
from large rainfall events, or from seepage through the dam. The overflow
would be more critical.
A trickle tube placed near the top of the dam would increase the potential
for runoff, but there would still be the same problem with subsurface move-
ment. In this case, one must determine the effects of both subsurface movement
and surface runoff. Samples should be taken at the source of the overflow
and downstream whenever overflow occurs.
A drain placed in the lower part of the retention structure would reduce
surface water movement because water would no longer stand behind the dam.
However, the concentration of solutes in the runoff water would be increased
since all of this water would have filtered through the waste heap. Erosion
106
-------
Drain
' i I i /';
Aquiclude
Figure 36. Subsurface movement of pollutants from darned waste heaps
107
-------
of the particles would be reduced since water velocities are reduced, but
there still would be more subsurface movement than with heaps because water
could pond behind the structure for short periods of time.
7.3.2.3 Determination of Effects —
Sampling of soil and water should be as discussed for heaps above.
More emphasis is placed on subsurface sampling than with heap disposal.
Drilling holes for deep sampling should be done carefully and should not
penetrate restrictive (very slowly permeable) layers which are called aqua-
eludes. Drilling through an aquaclude for sampling purposes might provide
a path for the movement of pollutant-laden waters below the restrictive layer.
Drilling should be done by a groundwater hydrologist or geologist who can
identify the aquacludes. If.it is necessary to sample beneath aquacludes,
the drill holes should be filled securely after sampling.
For all the cases involving disposal in gullies or natural depressions,
airborne transport would be reduced because of the less exposed nature of
the dust piles. However, wind movement occurs in gullies and small channels,
hence wind pickup is possible. The soil, soil-water, and plant interaction
would be similar to the previous case, once material has left the dust pile.
7.3.3 Lagoons
The main problems with lagoons are with seepage and/or overflow. The
relative magnitude of each component will be case dependent. If the lagoon
is not lined, seepage to groundwater aquifers and laterally to wells or
natural drains would be higher than any other case. This is due to the fact
that water is always in the lagoon and available for seepage.
The critical periods would occur during overflow from excess rainfall.
Seepage may also produce surface runoff below the dam. Overflow should be
sampled at the overflow channel and downstream to determine transport of
material. Generally because of the pit-like construction of lagoons,
surface runoff into the lagoon should be minimal.
Air transport effects' would be minimal for lagoon disposal because of
the constant water cover on the furnace dust.
108
-------
In the inundated environment around the metallic waste in the bottom
of the lagoon, anaerobic conditions will be associated with the reduction of
metal ions and compounds. The anaerobic bacterial population would pre=
dominate in the total biomass.of the system. The reductive reactions and
formation of organic acids in the submerged solids would solubilize metals
which would then be more subject to leaching as reduced metal ion species or
soluble chelates. The soluble constituents of metal dust would diffuse in
the direction of the activity gradient of the metal species and move as
solute with water to lower horizons and possible to groundwater.
Under flooded conditions, fermentation of organic materials results in
the initial production of organic acids which is followed by formation of
bicarbonate, carbonate, sulfides, and other reduced states of metal ions.
As a consequence,, at first there is appreciable mobilization of zinc, man-
ganese, iron, lead, nickel, copper, cobalt, vanadium, and magnesium but
later on some of the metal ions precipitate as carbonates (lead) and as sul-
fides (iron, mercury, copper, tin, lead, nickel, cadmium, cobalt, manganese)
in the advanced stage of anaerobiosis. Some metal ions, e.g., copper, nickel,
etc., do irreversibly complex with organics. Copper can chelate with both
-COOH and -SH groups. Under flooded lagoon situations, no mobilization of
chromium, titanium, and molybdenum is expected.
The extent of pollutant movement from the lagoon can be measured by
sampling soil and water phases as discussed above. Drill holes for determining
vertical pollutant movement should be made close to but not in the lagoon.
Again, care should be exercised in deep sampling because pumping of wells near
a lagoon may speed up the lateral movement of contaminated waters. In humid
regions, lagoons may overflow especially if they are in tight soils because
rainfall often exceeds evaporation. For these cases, runoff must be
considered.
7.3.4 Landfill
For a lined landfill with surface water excluded by diversions of
because of location on high ground, there should be no soil-water movement
problems. Again, one should check soil and water samples near such landfills
as a possible best technology.
109
-------
Surface water should be diverted from inclined landfills located on
high ground. No interception of subsurface water would be anticipated
since the only water movement is due to rainfall striking the surface of
the heap. There could thus be both deep and lateral seepage which is
essentially the same as the lagoon discussed above but with less water
to move out. Such landfills present no problem with surface runoff unless
seepage can produce surface runoff somewhere downs!ope from the landfill
site.
Air transport could exist if daily covering of filled dust with a soil
layer was not practiced. Operation with soil covering is similar to standard
sanitary landfill practices presently used for municipal refuse.
Where the unlined landfill is located in the path of natural subsurface
flow, water could move through the waste and carry pollutants to receiving
streams or to groundwater recharge areas. Here there is a need for monitoring
wells downstream from the landfill. This is essentially a special case of
the pile insofar as subsurface water is concerned but potentially more of a
problem because the landfill can intercept subsurface flow thus presenting
an opportunity for increased dispersion.
110
-------
SECTION 8
REFERENCES
1. Jablin, R., Personal communication, July 10, 1976.
2. American Iron and Steel Institute, Directory of Iron and Steel Works
of the United States and Canada, 1974, 33rd Ed., AISI, Washington, 1974.
3. , Annual Statistical Report, 1974, Washington, 1975.
4. Pasztor, L. and S. B. Floyd, Jr., "Pollution Abatement Residues Arising
in the Iron and Steel Industry," paper presented at the Third National
Conference on Complete Water Reuse, Cincinnati, June 27-30, 1976.
5. Barnard, P. G., A. G. Starliper, W. M. Dressel, and M. M. Fine, "Recycling
of Steelmaking Dusts," in Proceedings of the Third Mineral Waste
Utilization Symposium. Chicago, March 14-16, 1972.
6. Pasztor, L. and S. B. Floyd, Jr,, "Managing and Disposing of Residues
from Environmental Control Facilities in the Steel Industry," EPA Final
Report, Contract No. R-803619, June 1976.
7. U.S. Bureau of Mines, Commodity Data Summaries. 1976. USBM Publication
No. 1975-603-755/58, Washington, 1976.
8. Danielson, J. A. (ed), Air Pollution Engineering Manual. 2nd Ed., EPA
AP-10, May 1973.
9. O'Mara, R. F., "Dust and Fume Problems in the Steel Industry," Iron and
Steel Eng.. October 1953, pp. 100-106.
10. Oglesby, S., Jr. and 6. B. Nichols, "A Manual of Electrostatic Precipitator
Technology," Part II—Application Areas, NAPCA, PB 196-381, Aug. 25, 1970.
11. Goldberg, -M., letter communication and enclosed materials, Bureau of Air
Pollution Control, Allegheny County Health Department, Pittsburgh, Pa.
May 11, 1976
12. Hogan, J. C., "Physical and Chemical Characterization of Refining Furnace
Flue Dusts," paper presented at Symposium on Iron and Steelmaking Number
2, McMaster University, Hamilton, Ontario, Canada, May 16, 1974.
13. Kaerchen, L. T. and J. D. Sensenbaugh, "Air Pollution Control for an
Electric Furnace Melt Shop," Iron and Steel Engineer, May 1974, pp. 47-51.
14. Jablin, R., Personal communication, July 13, 1976.
15. Katari, V., G. Isaacs, and T. W. Devitt, "Trace Pollutant Emissions from
the Processing of Metallic Ores," EPA-650/2-74-115, October 1974.
16. Reno, H. T. and F. E. Brantley, "Iron", in Mineral Facts and Problems,
1970 Edition, U.S. Bureau of Mines Bull. 650, 1970, pp. 291-314.
Ill
-------
17. Mason, B., Principles of Geochemistry. 3rd Ed., John Wiley & Sons, Inc.,
New York, 1966, p. 183.
18. Klemic, H., H. L. James, and E. D. Eberlein, "Iron", in United States
Mineral Resources, Geological Survey Professional Paper 820, U. S. Dept.
Interior, USGS, 1973, pp. 291-306.
19. Stumm, J. J. and J. J. Morgan, Aquatic Chemistry, Wiley-Interscience,
New York, 1970.
20. Morgan, J. J., "Chemical Equilibrium and Kinetic Properties of Manganese
in Natural Waters," in Principles and Applications of Water Chemistry,
John Wiley & Sons, New York, 1967, pp. 561-624.
21. Brinn, D. G., "A Survey of the Published Literature Dealing with Steel
Industry In-Plant Fines and Their Recycling," NTIS, PB-236-359, August
1974.
22. "Sand and Dust", Engineering Design Handbook, Environmental Series. Part
Three Induced Environmental Factors, Chapter 3, Army Material Command,
AMCP 706-711, Alexandria, Virginia, January 1976.
23. Clements, T., "A Study of Windblown Sand and Dust in Desert Areas,"
Tech. Report ES-8, U.S. Army, AD-47 036, Natic, Mass., 1963.
24. "Coal Storage," Source Assessment Document No. 2, Monsanto Research
Corp., November 1974.
25. Lapple, C. E., Fluid and Particle Mechanics, 1st Edition, University of
Delaware, Newark, Delaware, March 1954.
26. Ekern, P. C., "Raindrop Impact as the Force Initiating Soil Erosion,"
Soil Science Society Proceedings, 1950, pp. 7-10.
27. Zingg, A. W., "Degree and Length of Land Slopes as it Affects Soil"Loss
in Runoff," Agricultural Engineering, vol. 21, 1940, pp. 59-64.
28. Ekern, P. C., "Problem of Raindrop Impact Erosion," Agricultrual Engineer-
ing, vol. 28, January 1953, pp. 23-25.
29. Mavis, F. T., et al., "The Transportation of Detritus by Flowing Water,"
Studies in Eng., Bull. 5, State University, Iowa, 1935.
30. Deutsch, M., "Incidents of Chromium Contamination of Groundwater in
Michigan," Proceedings - 1961 Symposium Ground Water Contamination,
Tech. Report W61-5, USDHEW PHS, Robert A. Taft Center, Cincinnati,
Ohio, 1961, pp. 98-104.
31. Swartzendruber, D., R. W. Skaggs, and D. Wiersma, "Characterization of
the Rate of Water Infiltration into Soil," Tech. Report 5, Purdue
University Water Resources Research Center, Lafayette, Indiana, 1968.
32. Spague, J. B., "Metals in Water," Metals in the Biosphere, Proceedings
of a symposium at the University of Guelph, Guelph, Ontario, Canada, 1975.
33. Manahan, S. E., Environmental Chemistry. Willard Grant Press, Boston,
Mass., 1972.
112
-------
34. Bowen, M. J. H., Trace Elements in Biochemistry, Academic Press, London
and New York, 1966Ti
35. Terry, D. U, C. B. McCant, and F. G. Averette, "A Sampler for Taking
Soil Samples from Specific Soil Zones," Soil Science. 118(6), 1975,
pp. 402-404.
113
-------
APPENDIX A
SPECIFIC GRAVITY DETERMINATIONS
RESULTS
Specific gravities were estimated for the 18 samples provided. The
results are tabulated in Table A-l.
PROCEDURE
Each sample was split in a Sepor sample splitter, oven dried, and then
allowed to return to room temperature in a dessicator. Splits were then
weighed and placed in 50 ml cylinders. Water was added to each cylinder to
the 40 ml line using a graduated 50 ml burette. Water displaced by the sample
was 40 ml minus water introduced from the burette. Specific gravity was then
estimated as weight of sample split divided by weight of water displaced.
CHECKS AND CAUTIONS
Entrapment of air by the fine material in the cylinder was minimized by
introducing 10 ml of water initially, then slowly adding the sample material.
The slurry was then briefly vibrated before bringing the water level to the
40 ml mark.
Water used was demineralized, and a check was run on its specific gravity
(Table A-l). The difference between expected and observed values are within
the error limits expected by the technique (1 burette drop = 0.06 gm).
Tests of the technique (Table A-l) were made using the mineral quartz
(s.g. = 2.65), both as a single crystal (sample A) and as a very fine sand/
silt (sample B). In both cases, error for the technique is about 1 percent.
Two uncontrolled sources of variation could effect the results. Composi-
tion could vary with grain size and small bubble inclusions within grains
could change proportion with grain size. The specific gravity estimates,
therefore, are averages for the bulk sample.
114
-------
TABLE A-I. SPECIFIC GRAVITY ESTIMATES
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
(Qtz. sand)
A.
B.
Wt. of Sample
Split (gm)
12.421
14.121
7.025
14.498
12.892
6.376
11.394
12.100
13.649
17.826
8.499
13.178
13.217
14.682
11.752
11.541
11.122
10.915
23.825
29.791
Water Displaced
(ml)
3,05
3.93 .
1.35
3.76
2.93
1.57
2.70
3.10
3.36
4.16
1.79
3.20
2.89
3.68
2.77
3.06
2.70
2.56
8.89
11.32
Wt. Water
(gm/ml)
T.OOO
ii
11
"
"
1.000
II
II
II
II
1.000
II
II
II
II
1 .000
II
II
1.000
II
Specific Gravity
Estimate
4.07
3.59
5.20
3.86
4.40
4.06
4.22
3.90
4.06
4.29
4.75
4.12
4.57
3.99
4.24
3.77
4.12
4.26
2.68*
2.63
**
1
(Water Used): 40.146 gms; 40.33 ml vol; S.G.= 0.996
*Quartz = 2.65, expr. error
**Within margin of burette error, assumed 1.0 subsequently
115
-------
APPENDIX B
PARTICLE SIZE ANALYSIS
RESULTS
Particle size distributions for the 18 samples provided were presented
in Table 9. Table values are in weight percentage by size classes. Size
classes are divided on the phi particle scale (phi = -Iog2 mm), and equivalent
micron size ranges are given. Particles coarser than 4 phi were determined
by wet sieving; those finer by pipette analysis where the table is completed.
For those samples not analyzed when the study was terminated, the finer
than 4 phi row contains a double-headed arrow. The arrow shows the sedi-
mentation size range based upon visual sedimentation.
PROCEDURES
Each sample was split on a Sepor sample splitter, its dry weight
determined, and placed to soak 'overnight in a covered beaker.
The splits were then washed through a No. 230 U. S. Standard Sieve (4
phi size boundary). Gentle rubbing of the sieve fraction often was
necessary to break apart aggregates. The fraction passing the sieve was
transferred to a 1000 ml cylinder for subsiquent pipette anlaysis The
fraction collected on the screen was oven dried, cooled, and weighed. This
computed percentage is in column 1 of Table 9.
The finer than 4 phi fraction of each split was transferred to a
1000 ml cylinder. Sodium oxilate was added to each cylinder as a peptizer.
The cylinders then were allowed to;sit overnight and come to equilibrium
with controlled room temperature (22°C). The following morning, the fractions
were analyzed by standard pipette analysis, based upon Stokes1 law of
settling velocities of spheres through a water column. The resultant summary
percentage values are in Table 9. Details of the pipette procedure follows.
116
-------
Pipette analysis was discontinued after half of the samples had been
run because of flocculation problems. The effective grain size range of
the finer fraction for the remaining samples was estimated roughly by
stirring the 1000 ml cylinder mixtures, then timing how long before
a marked card could be seen through the water column (about 10 cm depth).
These size ranges are depicted in Table 9 by double headed arrows.
REPLICATIONS
Three samples were randomly selected for replicate analysis, subject
to the limitation that one be from each of the three groupings provided
(OH, Electric, EOF). Sieve replications were good in all samples; pipette
is good in No. 7. Pipette discrepancies in No. 15 may reflect incomplete
disaggregation, as the sample contained hard clumps when sieving which
were difficult to break down.
PIPETTE ANALYSIS
Pipette analysis is a standard sedimentation technique for analysis
of grains finer than sand size. Its major area of utilization is in soil
spheres through a water column:
«
V = 2/9 • (d1 - d2) gr2/n
where: V = settling velocity (cm/sec)
d, = particle density (gm/cc)
d2 - fluid density (assumed 1.0 gm/cc)
g = gravity (assumed 980 cm/sec2)
r = particle radius (cm)
n » fluid viscosity (assumed 0.01 poise)
Since velocity also equals distance (or depth) divided by time, knowing the
particle density one can solve for the time necessary for a particle of a
given diameter to have fallen through a given distance of water column.
117
-------
Utilization of this law in size analysis typically involves the following
steps:
a. the material is thoroughly mixed through a column of water
of known volume (1000 ml in this case). The instant external
mixing ceases is time "zero", and it is assumed the sediment
at this instant is homogeneously dispersed through the water
column;
b. at a predetermined time a 20 ml pipette is inserted to a pre-
determined depth and a sample extracted. This sample can
contain only grains of a diameter less than the calculated size;
c. the 20 ml extraction sample is placed in a beaker, and the
water evaporated off;
d. the weight of the sediment contained in the 20 ml extraction
is determined, and corrected for peptizer present. Since
the sample is only a fraction of the suspended volume (20 ml
of 1000 ml), it must be proportioned up (times 50 in this
case). The result is the weight of material finer than a
given size in the suspension;
e. weight between selected size limits is determined by subtracting
the "finer than" weight of the lower boundary from that of the
upper boundary. Weight percentage in each size class may then
be calculated.
References include:
Krumbein, PettiJohn, Manual of Sedimentary Petrography, 1938.
Schweyer, Sedimentation Procedures, Fla. Univ. Ind. Expr. Sta. Bull.
54, 1952.
118
-------
APPENDIX C
SPARK SOURCE MASS SPECTROMETRY DETERMINATIONS
SAMPLE PREPARATION
The samples, as received, varied widely in their gross appearance and
visual homogeneity. Representative portions of each sample were taken and
dried to constant weight, then ground in a mixing mill to improve the
homogeneity of the material and to afford better mixing with the graphite
support material. One sample, number 15S showed a significant weight loss
on drying.
Five milligrams of the dried sample were mixed with fifty milligrams
of graphite, the erbium internal standard added; and the resultant mixture
evaporated to dryness under an infrared lamp. The dried mixture was shaken
further in a mixing mill to achieve homogeneity. The electrodes were then
compacted.
•
Notice was taken of some segregation of sample and graphite on shaking
in the mixing mill if the sample had not been ground before mixing. Evalua-
tion of several mixing methods was performed, and it was found that all that
incorporated a pre-grinding of the sample material were satisfactory (judging
from the reproducibility of the results).
ANALYSIS
The usual graded series of exposure were taken (10" - 300 nC, in steps
of a factor of 3) by photoplate. One photoplate was exposed per sample, due
to the time limitations specified in the request for analysis and also due
to a scarcity in the supply of photoptates. The exception to this is sample
no. 1, on which the mixing technique/homogeneity study was performed on
foar sample electrode sets.
119
-------
CONCLUSIONS
The results reported here of the analysis of 18 samples by SSMS
yielded the normal precision that can be expected from SMSS photoplate
analysis. Noting the usual caveat concerning the sample homogeneity require-
ments of SSMS, the results should be as good as the sampling.
120
-------
APPENDIX D
SOLUBILIZATION TEST PROCEDURE
PROCEDURE
Eighteen steelmaking furnace residues were subjected to sol utilization
by contacting 125 grams of sample with 250 milliliters of deionized water.
The mixture was shaken for 48 hours and allowed to stand for 16 hours to
facilitate solid/liquid separation. A portion of the solute was drawn-off
and analyzed. The pH of the solutes were also measured.
The amount of solute drawn-off was replaced with fresh deionized water.,
and the process was repeated. A total of ten contact periods was used.
CONCENTRATION CALCULATIONS
With only a portion of the solute drawn off at each contact, a certain
amount of "residual" concentration would be left in the liquid. This residual
would affect the concentration of the successive contacts. Thus, the residual'
must be accounted for in the calculation of subsequent concentrations. This
is done by the following:
(250 - Quantity Removed )(Concentration) =(Residual.
250
A plus and minus 15 percent of the residuals was also calculated to provide
an error range. The concentration reported was then a difference between the
analysis result and the ± 15 percent range.
121
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-044
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Environmental Assessment of Steelmaking Furnace
Dust Disposal Methods
5. REPORT DATE
February 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOH(S)
G.E. Weant m and M.R. Overcash
8. PERFORMING ORGANIZATION REPORT l>
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P.O, Box 12194
Research Triangle Park, NC 27709
10. PROGRAM ELEMENT NO.
1AB604
11. CONTRACT/GRANT NO.
68-02-1325, Task 61
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERS
Task Final; 3-12/76
14. SPONSORING AGENCY CODE
EPA/6QP/13
is. SUPPLEMENTARY NOTES DSRL-RTP project officer for this report is R.V. Hendriks, Mai3
Drop 62, 919/549-8411 Ext 2557.
16. ABSTRACT
The report gives results of a study to examine the nature of steelmaking
furnace residues and disposal techniques, and to assess potential problems asso-
ciated with residue disposal, a potential multimedia environmental problem. The
study established a preliminary basis for impact evaluation. Solubilization tests of
18 furnace residue samples showed that the amount of metals extracted from the
solids is a small percentage (less than 10 percent) of the total amount available in
the residues, except for antimony and (in some cases) cobalt and zinc. Organic
analysis of the solute showed low total organic carbon, except in four electric
furnace samples. Further analysis of these four samples showed the types of
products normally associated with the distillation of mineral- or petroleum-based
lubricating oils. The study concludes that potential exists for environmental degra-
dation from the disposal of these furnace residues. Due to variations in residue
characteristics, climatic and topographic conditions, and disposal site configuration
a site by site investigation is necessary to further evaluate these environmental
problems. The study also concludes that, although residue recycling is not economi
cally feasible on a plant by plant basis, regional recycle plants can be operated
profitably in some areas.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Pollution
Iron and Steel
Steel Making
Furnaces
Dust
Dust Control
18. DISTRIBUTION STAT
Unlimited
Residues
Industry
Disposal
Waste Disposal
Circulation
Lubricating Oils
EMENT
b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Environmental Assess-
ment
Unclassified
20. SECURITY CLASS (This page)
Unclassified
C. COSATt
13B
11F
13H
13A
11G
Field/Group
.
11H
21. NO. OF PAGES
132
22. PRICE
EPA Form 2220-J (9-73)
122
------- |