WATER POLLUTION CONTROL RESEARCH SERIES • 14010 FOA 02/71
Carbonate Bonding
of Coal Refuse
ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFICE
-------
WATER POLLUTION co: TRoL RESEARCil SERIES
The Water Pollution Control T esearch Repc rts describe
the results and progress in the control and abatenent
of pollution in our Nation’s waters. They provide a
central source of information on the research, develop-
nEnt, and dexmonstration activities in the Environmental
Protection Agency, 1iater Quality Office, through inhouse
research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.
A tripilicate abstract card sheet is included in the
report to facilitate information retrieval. Space is
provided on the card for the user’s accession nuz*er
and for additional uniterms.
Inquiries pertaining to Water Pollution Control Research
P ports should be directed to the Head, Project Keports
Systc , Office of Research and Oevciopnent, £nvironr ental
Protection Age y, Water Quality Office, Washington, D. C.
-------
Carbonate Bonding of Coal Refuse
by
Black, Sivalls & Bryson, Inc.
Applied Technology Division
135 Delta Drive
Pittsburgh, Pennsylvania 15238
for the
ENVIRONMENTAL PROTECTION AGENCY
WATER QUALITY OFFICE
Grant 14010 FOA
February 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price 60 cents
-------
FWQA REVIEW NOTICE
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
-------
ABSTRACT
A laboratory study of the operating variables which affect the properties
of carbonate bonded coal refuse has been made. The carbonate bonding
process utilizing coal refuse as a fill material consists of mixing
coal refuse with water and lime hydrate, compacting the mixture, and
reacting it with a carbon-dioxide-rich gas to form a coherent structure
bonded by a matrix of calcite crystals. The resulting carbonate bonded
coal refuse can be used in road building or as a coal refuse pile sealant
to minimize acid mine water pollution.
Four types of coal refuse were investigated--a relatively unoxidized and
highly oxidized bituminous coal refuse and a relatively unoxidized and
highly oxidized anthracite coal refuse. It was found that compressive
strengths of 2200 to 4400 psi were obtained for the four types of coal
refuse investigated using up to 12 percent lime hydrate and 9 to 15 per-
cent water. In general, the compressive strength of the carbonate bonded
coal refuse increases with increasing lime hydrate content, reaction
time and carbon dioxide concentration in the carbonate bonding reaction
gas
The air and water permeability of carbonate bonded coal refuse was found
to be comparable to concrete. An approximate cost comparison between
carbonate bonded coal refuse and other construction materials and techniques
indicated that the carbonate bonding process utilizing coal refuse is the
most economical means available for coal refuse pile sealing and road
building
This report was submitted in fulfillment of Research Grant No. 14O1OFOA
between the Environmental Protection Agency and the Applied Technology
Division of Black, Sivalls & Bryson, Inc.
Key Words: Carbonate bonding, coal refuse, coal refuse binder, refuse
pile sealant, pollution abatement, road building
1
-------
CONTENTS
Page
Conclusions vi
Recommendations vii
Introduction 1
Raw Materials Used 5
Carbonate Bonding Technique and Apparatus 9
Experimental Laboratory Study 11
Effect of Carbonate Bonding Parameters on Compressive
Strength 11
Effect of Lime Hydrate Content on Compressive Strength ii
Effect of Curing Time on Compressive Strength 12
Effect of Water Content on Compressive Strength 17
Effect of Particle Size on Compressive Strength 17
Effect of Carbon Dioxide Concentration and Carbonate
Bonding Reaction Time on Compressive Strength 20
Additional Properties of Carbonate Bonded Coal Refuse 24
Air permeability of Carbonate Bonded Coal Refuse 24
Water Permeability of Carbonate Bonded Coal Refuse 28
Freeze-Thaw Testing on Carbonate Bonded Coal Refuse 30
Suggested Operating Parameters for Carbonate Bonding Process 32
Applications of Carbonate Bonded Coal Refuse 33
Acknowledgments 39
Appendices 41
Determination of Air Permeability 41
Freeze-Thaw Testing of Coal Refuse 43
Ut
-------
FIGURES
Figures Page
1 Carbonate Bonding Reactor System 10
2 Effect of Lime Hydrate Content on Compressive
Strength for Unburnt Bituminous Coal Refuse 13
3 Effect of Lime Hydrate Content on Compressive
Strength for Burnt Bituminous Coal Refuse 13
4 Effect of Lime Hydrate Content on Compressive
Strength for Unburnt Anthracite Coal Refuse 14
5 Effect of Lime Hydrate Content on Compressive
Strength for Burnt Anthracite Coal Refuse 14
6 Effect of Curing Time on Compressive Strength for
Unburnt Bituminous Coal Refuse 15
7 Reduced Compressive Strength vs. Curing Time 16
8 Effect of Particle Size and Water Content on
Compressive Strength for Unburnt Bituminous 18
9 Effect of Particle Size and Water Content on
Compressive Strength for Burnt Bituminous 18
10 Effect of Particle Size and Water Content on
Compressive Strength for Unburnt Anthracite 19
11 Effect of Particle Size and Water Content on
Compressive Strength for Burnt Anthracite 19
12 Effect of Specific Surface Area on Compressive
Strength 21
13 Effect of Carbon Dioxide Concentration and Reaction
Time on Compressive Strength for Unburnt Bituminous 22
14 Effect of Carbon Dioxide Concentration and Reaction
Time on Compressive Strength for Burnt Bituminous 22
15 Effect of Carbon Dioxide Concentration and Reaction
Time on Compressive Strength for Unburnt Anthracite 23
16 Effect of Carbon Dioxide Concentration and Reaction
Time on Compressive Strength for Burnt Anthracite 23
17 Air Permeability Apparatus 25
V
-------
FIGURES - CON’T
Figures Page
18 Air Permeability vs. Average Particle Diameter 27
19 Water Permeability Apparatus 29
20 Effect of Freeze-Thaw Cycles on Compressive
Strength for -10, Mesh Unburnt Bituminous
Coal Refuse 31
21 Effect of Freeze-Thaw Cycles on Compressive
Strength for -20 Mesh Unburnt Bituminous
Coal Refuse 31
22 Effect of Freeze-Thaw Cycles on Compressive
Strength for -60 Mesh Unburnt Bituminous
Coal Refuse 31
V I
-------
TABLES
No. Page
I Screen Analysis of Coal Refuse 5
II Specific Surface Area of Coal Refuse 6
III Chemical and Screen Analysis of Dolomitic Lime Hydrate 6
IV Air Permeability of -10, -20, and -60 Mesh Unburnt
Bituminous Coal Refuse 26
V Water Permeability of -10, -20, and -60 Mesh Unburnt
Bituminous Coal Refuse 28
VI Paving Costs of Various Materials 35
VII Paving Costs of Carbonate Bonding Process 35
VIII Table of Nomenclature for Air Permeability 41
IX Schedule of Freeze-Thaw Test 43
I
-------
CONCLUSIONS
In general, the excellent properties of carbonate bonded coal refuse
obtained in this study indicate great potential for carbonate bonding as
a means of eliminating acid mine drainage by coal refuse pile sealing
and utilization of coal refuse in road building. In particular, the
following conclusions may be drawn from the results of this study:
1. Compressive strengths of 2200 to 4400 psi were obtained for the coal
refuse types under investigation using up to 12 percent lime hydrate
content and water contents in the range of 9 to 15 percent. The
compressive strength of the carbonate bonded coal refuse increases
with increasing lime content, carbonate bonding reaction time and
carbon dioxide concentration in the carbonate bonding reaction gas.
The compressive strength also increases with decreasing coal refuse
particle size due to a wider particle size distribution.
2. Air and water permeability of carbonate bonded coal refuse was
found to be comparable to concrete.
3. An approximate cost comparison indicated that carbonate bonded coal
refuse is significantly less expensive than other construction
materials for sealing coal refuse piles or for load bearing applications.
The cost of carbonate bonding 2-inch to 15-inch thick layers of
coal refuse was estimated to range from $0.83 to $2.48 per square yard.
For sealing coal refuse piles,a 6-inch carbormte bonded layer
will be comparable to a four foot layer of dirt. Sealing with a
four foot layer of dirt is estimated to cost $1.33 whereas a 6-inch
thick carbonate bonded seal was conservatively estimated at $1.28
per square yard. For secondary road construction, a 6-inch layer of
carbonate bonded coal refuse would cost approximately $1.28 as compared
to $4.68 per square yard for a 2-inch asphalt, 4-inch subbase road.
Carbonate bonded coal refuse used as a subbase in road construction
is significantly less expensive than conventional materials. A 4-inch
subbase of carbonate bonded coal refuse would cost $1.05 as compared
to $1.56 per square yard for conventional materials. If a 12-inch thick
layer of carbonate bonded coal refuse is covered with a 2-inch layer
of asphalt to produce a first class road, the cost would be $5.17 as
compared to $7.62 per square yard for a first class asphalt road.
4. Laboratory results indicate that commercial application of the
carbonate bonding process should present no unusual problems.
Although the particle size of the coal refuse to be carbonate bonded
affects the resulting compressive strength, there will be little or no
preparation required for the coal refuse prior to carbonate bonding.
Compressive strength did not exhibit severe changes with water
content in the 9 to 15 percent range. Consequently, there will be no
need for critical control of the water content in the commercial
application of this process. In addition, it was found that compressive
strength was not substantially increased by utilizing pure carbon
dioxide; therefore, an inexpensive burner will suffice as a source
of carbon dioxide in the application of the process.
i x
-------
RECOMMENDATIONS
The proposed applications for carbonate bonded coal refuse and the
suggested commercial process techniques are postulated on the encouraging
but limited results of this study. Additional work will be required to
demonstrate the carbonate bonding process on a large scale. Depending on
the availability of funds, it is recommended that at least one of the
following three process demonstration studies be conducted.
1. It is recommended that a small coal refuse plot be paved to
demonstrate the commercial utility of the process as a coal refuse
pile sealant and road building material. For demonstration purposes,
the test plot should be a minimum of 20 feet by 20 feet. Based on
the results from the demonstration plot, an engineering evaluation
of the •process will be made with pertinent recommendations as to
further application.
2. It is recommended that the process be demonstrated on a full-size
actual coal refuse pile. By sealing the pile, not only will acid water
formation be prevented but the necessary techniques for large scale
construction could be developed. Based on results of this demonstration,
an engineering evaluation of the process will be made, construction
standards and procedures will be developed and process costs
established, for future application of the process.
3. It is recommended that the process be demonstrated by construction
of a secondary road in the vicinity of a coal refuse pile. This
demonstration will reduce the size of the selected coal refuse pile,
provide additional roads for the community, establish road building
costs, and develop road building techniques for application
of the carbonate bonding process.
4. As a part of any of the above projects, a study should be undertaken
to determine feasibility of producing lime hydrate suitable for
carbonate bonding from limestone available near coal refuse sources.
The purpose of this work will be to minimize cost of transporting
lime hydrate to point of use.
5. An additional part of any of the above programs should include a
study of material stability in an actual acid mine drainage environ-
ment.
Once the above demonstration studies are completed, carbonace bonding
costs will have been established
xi
-------
INTRODUCTION
The project involves the investigation of a process that has several
applications in combating water pollution and other problems created by
coal refuse piles. The need for such a process arises from the fact that
as rain water percolates down through the spoil bank, the reaction
between pyrites, air, and water generates acid-forming constituents
which are carried in the water table or streams surrounding the locale.
Thus, to minimize this serious pollution problem, an effective process
for sealing off spoil banks or utilizing coal refuse is required.
One application of the process is the sealing of coal refuse piles with
a layer of carbonate-bonded refuse to exclude air and moisture from the
pile, and thus eliminate the formation of acid drainage. A second
application of this process is the elimination of refuse piles by uti-
lizing the refuse material as filler in the construction of carbonate
bonded road pavement and in the manufacture of low-cost bricks and
similar products.
In this process, lime hydrate is throughly blended with coal refuse (or
spoil bank materials) and then treated with carbon dioxide to form strong
calcium carbonate (calcite) crystals. The calcite crystals act as a
cementing matrix that turn the coal refuse into a coherent and imper-
vious structure. The carbonate bonding process relies on the simple
reaction that occurs between carbon dioxide and lime hydrate. In the
reaction:
Ca (OH) 2 + CO 2 = Cac O 3 + H 2 O
carbon dioxide reacts with lime hydrate to produce calcium carbonate
(calcite) and water. The reaction is well known and occurs naturally
upon the exposure of lime hydrate to the atmosphere. However, the rate
of conversion of lime hydrate to calcium carbonate is relatively slow
because of the small percentage of carbon dioxide contained in the air.
It is of interest to note that although the reaction in air is slow, the
reaction proceeds rapidly in concentrated carbon dioxide. Because of
its high strength, the calcite crystal affords an opportunity for ex-
ploiting lime hydrate as a cementing material to bind coal refuse into a
coherent structure.
Most of the experimentation prior to this work has been conducted with
various types of soils and not with coal refuse materials. Therefore,
this bench-scale experimental study was undertaken to determine the
operating parameters for generating a strong carbonate bonded coal refuse
and to determine the feasibility of the carbonate bonding process for
the elimination of acid mine water formation from coal refuse piles. In
particular, the effects of lime hydrate content, moisture content,
partial pressure of carbon dioxide in carbonating gas, carbonation time,
and the external surface area of coal refuse on the compressive strength
— 1—
-------
of the coal refuse were determined. In addition, the air and water
permeability and the freeze-thaw resistance were determined for carbonate
bonded coal refuse.
Before proceeding with a discussion of the experimental work, it would
be well to discuss some of the theoretical aspects of the carbonate bond-
ing reaction and the results from prior experimentation with other fill
materials. The relationship between the compressive strength of car-
bonate bonded materials and process operating variables is as yet not
fully understood. Prior experimentation on soils, clays, and sands
indicates that compressive strength increases as:
(1) The particle size of the fill material and lime hydrate decreases
(2) An optimum water content is selected for each specific fill material
(3) The external surface area of the fill material increases
(4) The carbon dioxide partial pressure in the carbonating gas is
balanced so that the rate of heat generation within the composite
equals the rate of heat loss to the surrounding gas phase
(5) The time of carbonation increases, and
(6) The lime hydrate concentration increases
Decreasing the particle size (along with a wider particle size distri-
bution) of the fill material decreases the porosity of the composite
which is to be carbonate bonded. Decreasing the porosity yields a more
dense composite and enhances the formation of a continuous calcite
matrix. As the compressive strength of the composite is mainly a
function of the continuity of the calcite crystal structure, a minimum
porosity condition increases the strength of the carbonate bonded
material.
As the surface area of the lime hydrate increases, its rate of solution
in water is increased. This causes the carbon dioxide to react with a
saturated lime hydrate solution and results in a rapid precipitation of
a fine calcite crystal matrix. The rapid formation of a multitude of
crystals enhances additional crystaline growth so that bridging and
growth can occur between crystals with a resulting increase in strength.
In general, the greater the external specific surface area of the fill
material, the greater will be the rugosity or roughness of the surface of
the particles. This condition provides numerous bridging channels be-
tween particles for the growth of calcite crystals and increases the
strength of the carbonate bonded material.
The effect of the partial pressure of carbon dioxide in the carbonating
gas on the carbonation reaction is directly related to the gas phase mass
transfer of carbon dioxide to the liquid film. Prior experimentation has
- 2
-------
shown that the rate of the carbonate bonding reaction is directly pro-
portional to the partial pressure of carbon dioxide. Since the carbon-
ate bonding reaction is highly exothermic, the heat released during
this reaction can be controlled by varying the partial pressure of
carbon dioxide in the gas phase. Depending upon the water content, and
the thermal characteristics and packing of the fill material, it is
possible to crack the resulting carbonate bonded material by causing
the reaction to occur too rapidly. Due to a high carbonation rate suf-
ficient heat is generated to produce steam within the interior of the
composite. If the packing is too dense, rapid steam evolution from the
carbonate bonded material is inhibited and can result in sufficient
internal pressure to crack or rupture the composite. Consequently, the
partial pressure of carbon dioxide must be optimized so that internal
pressure cracks are not generated and the strength characteristics of
the carbonate bonded material can be maintained.
-3-
-------
RAW MATERIALS USED
Four types of coal refuse were investigated, a relatively fresh (unburnt)
bituminous, a highly oxidized (burnt) bituminous, a relatively fresh
unoxidized (unburnt) anthracite, and a highly oxidized (burnt) anthracite.
The bituminous samples were obtained from the Kittaning coal seam of the
Roaring Creek Coal Company* in Norton, West Virginia. The anthracite
samples were obtained from the Reading Coal Company* located in Potts-
yule, Penna. Each of the samples was initially screened to remove
lumps and yield a -10 mesh sample. A portion of the -10 mesh sample was
subjected to grinding to obtain a -20 mesh and -60 mesh size fraction
for use in subsequent experimentation. The screen analysis for the -10,
-20, and -60 mesh samples for each type of coal refuses is presented in
Table I. TABLE I
Screen Analysis of Coal Refuse
UNBURNT BITUMINOUS BURNT BITUMINOUS
In Table II is shown the external surface
fraction and type of coal refuse.
area (cm 2 lgram) for each size
* Mention of commercial products or organizations does not imply
endorsement by EPA
-10
Mesh
-20
Mesh
-60
Mesh
-10
Mesh
-20
Mesh
-60
Mesh
Screen Sample
Sample
Sample Screen Sample
Sample
Sample
20
69.77.
20
38.17.
40
15 67.
48 97.
40
23 47
30 9%
60
4 7%
18 67
60
8 370
16 77
100
3 47.
11 37.
27 8%
100
8 37
12 47.
52 87
200
3 5%
8 47.
56 6%
200
9 270
12 07
39 77
325
4.57.
6.37.
325
8.570
4.270
PAN
3.17
8.3%
9.3%
PAN
12.77.
19.5%
3.3%
UNBURNT ANTHRACITE
BURNT ANTHRACITE
-10
Mesh
-20
Mesh
-60
Mesh
-10
Mesh
-20
Mesh
-60
Mesh
Screen Sample
Sample
Sample Screen Sample
Sample
Sample
20
57.87.
20
43.97.
40
20.87.
43.1%
40
22.9%
33.57.
60
6.6%
19.6%
60
9.67
19.4%
100
5 07
10.7%
45 7%
100
8 5%
13 8%
34 47
200
4.4%
10.0%
50.6%
200
8.2%
13.0%
61.3%
325
6.57.
3.57.
325
7.270
2.370
PAN
5.470
10. 1%
.2%
PAN
6.970
13. 1%
2.0%
-5-
-------
TABLE II
Specific Surface Area of Coal Refuse
Surface Area, cm 2 /gram
Sample - 10 Mesh - 20 Mesh - 60 Mesh
Unburnt Bituminous 163 763 2249
Burnt Bituminous 998 1231 1959
Unburnt Anthracite 186 705 1775
Burnt Anthracite 294 535 1533
The external surface area was determined using an air permeability
technique in which the Carman-Kozeny equation was used to relate the
external surface area to the pressure drop of air passing through a
packed bed of coal refuse.
The lime hydrate used in these experiments is a dolomitic hydrate and
was obtained from the Ohio Lime Company,* Woodville, Ohio. A typical
analysis for this superfine lime hydrate is presented in Table III.
TABLE III
Chemical and Screen Analysis of Dolomitic Lime Hydrates
TYPICAL CHEMICAL ANALYSIS
Guaranteed
Minimum Ave r ge
Calcium Oxide 46.877. 49.207.
Magnesium Oxide 33.987. 34.007.
Calcium Hydroxide 62.047. 65.007.
Silicon Dioxide .237. .207.
Iron Oxide .077. .107. max.
Aluminum Oxide .167. .407. max.
TYPICAL SCREEN ANALYSIS
Fineness
Passing 100 U.S. Mesh 100.007.
Passing 200 U.S. Mesh 99.787.
Passing 300 U.S. Mesh 99.647.
Passing 325 U.S. Mesh 99.507.
607. finer than 20 microns
207. finer than 14 microns
87. finer than 10 microns
* Mention of commercial products or organizations does not imply
endorsement by EPA
-6-
-------
This type of lime hydrate was selected because prior experimentation has
shown it to be suitable for carbonate bonding. Other types of limestones
might have been used with more or less success; however, an investigation
of the effect of lime hydrate type on the quality of carbonate bonded
coal refuse was beyond the scope of this study.
—7—
-------
CARBONATE BONDING TECHNIQUE AND APPARATUS
Tile preparation of coal refuse for carbonate bonding consisted of mixing
in sufficient water to bring the water content to a specified value.
This mixture was allowed to stand at least 24 hours to permit an equil-
ibrium of the water throughout the mixture. Lime hydrate was then
thoroughly blended into the coal refuse. To permit a rapid scanning of
parameters and to minimize the quantities of materials and size of the
carbonate bonding apparatus, the lime hydrate coal refuse mixture was
placed in a small die and compacted to yield a 1 1/8 inch diameter by
1 inch high briquette. A compaction load of 6000 psi was used to form
the briquettes of these experiments. In the commercial application of
the carbonate bonding process lower compaction loads would be used in the
order of 1000 psi. It was decided to use higher compaction load for
experimental purposes so that the briquettes would not have loose edges
and would be of exact dimensions. The briquetting pressure was then re-
moved and the soft, easily crushed briquette was placed in a reactor and
contacted with a carbon dioxide rich gas for a specified time. The
briquette was removed and the compressive strength determined. In this
manner, the compressive strength of a carbonate bonded coal refuse was
determined for a particular coal refuse and particle size, lime hydrate
and water content, carbon dioxide concentration and carbonate bonding
time.
The carbonate bonding reactor is shown schematically in Figure 1. Tanks
of carbon dioxide and nitrogen were used as the source of these gas
constituents. Experiments were run with carbonate bonding gas containing
20, 60, and 100 percent carbon dioxide concentrations. The carbon dioxide
and nitrogen were individually metered, combined in a mixing chamber,
and admitted to the carbonate bonding reactor. The reactor consisted of
a plexi-glass tube, 2 inches I.D. by 24 inches long. The formed
briquettes were arranged on a metal grate which slid into the reactor
tube. Six reactor tubes were used to permit segregation of the various
coal refuse briquettes. For all experimentation, a gas flow of 0.55
standard cubic feet per minute was used which resulted in a superficial
gas velocity of 0.43 feet per second in each individual reactor tube.
The •procedure used to carbonate bond the coal refuse briquettes consisted
of placing the briquettes in the reactor tubes. The flow rates of
nitrogen and carbon dioxide were then adjusted to the desired flows.
During this initial adjustment the flow of the resulting gas mixture was
made to bypass the reactor tubes. Once the desired gas mixture was
obtained, a stop watch was started and the bypass closed causing the gas
mixture to enter the reactor tubes. When the specified carbonate bonding
reaction time was reached the carbon dioxide was cut off and the nitrogen
allowed to flow through the reactor tubes for several additional minutes
to flush out all remaining reaction gas. The carbonate bonded coal
refuse briquettes thus formed were removed and either tested for their
compressive strength or subjected to other tests.
-9-
-------
Rubber
Stopper
Gas Outlet
Glass Beads
Portable
Rotameter
Line To
Check
Individual
Gas Flows
Reactor System
FIGURE 1-CARBONATE BONDING REACTOR SYSTEM
Plexiglass Tube
Grate Briquette
Carbon Dioxide
Rich Gas
Inlet
Rotame ter
/
Tank
I ; -
Valve
L
Nitrogen
Tank
-10-
-------
EXPERIMENTAL LABORATORY STUDY
For convenience in presentation, the experimental laboratory study was
divided into two categories--those experiments which pertain to the effect
of various carbonate bonding parameters on the compressive strength of
carbonated coal refuse and those experiments which were used to determine
the physico-chemical properties of the resulting carbonate bonded coal
refuse. In the first category, the effect of coal refuse type and
particle size, lime hydrate and water content, carbonate bonding reaction
time, and carbon dioxide concentration, on the compressive strength of
the carbonate bonded coal refuse were ascertained. In addition, the
effect of curing time on compressive strength will be discussed. The
curing time is defined as the total time elapsed after carbonate bonding
is officially completed (removal from reactor). During this study it was
observed that the compressive strength increased significantly for some
time after carbonation. Included in the second category of experiments
were air and water permeability and freeze-thaw testing of carbonate
bonded coal refuse.
Effect of Carbonate Bonding Parameters on Compressive Strength
The structural integrity of carbonate bonded coal refuse is a prime
consideration of any of its possible applications. Since the process
is being considered for the sealing of coal refuse piles, the sealing
layer must have sufficient strength to withstand vehicular and pedestrian
traffic, Consequently, in this short feasibility study it was expedient
to select one strength property as representative of the structural
integrity of the resulting carbonate bonding material. For this study
compressive strength was selected because of its applicability toward
the intended use of the carbonate bonding process and because of its
ease of determination. The determination of the compressive strength
of a solid material is normally accomplished under conditions in which
the rate of application of load is precisely controlled. In view of
the limited objectives of this study and the numerous uncontrollable
parameters associated with carbonate bonding, this degree of precision
was deemed unnecessary. Consequently, a simple technique was devised
to determine compressive strength. The briquette was placed in a press
and the press plates closed until a small load resulted on the briquette.
The press operator then applied a slow but approximately reproducible
rate of loading on the briquette until failure occurred. The load at
failure was recorded and used to calculate the compressive strength of
the briquette.
Effect of Lime Hydrate Content on Compressive Strength
To determine the effect of lime hydrate content on the compressive
strength of the resulting carbonate bonded briquettes, a series of
carbonate bonded coal refuse briquettes were prepared in which the lime
hydrate content was varied between 3 and 12 percent by weight and the
water content varied in the rangeof 3 to 12 percent by weight. From
—1 l_
-------
prior experimentation with other fill materials (soils, ores etc.) it
has been found that optimum carbonate bonding strength normally occurs
in these ranges of lime hydrate and water. The percent of water was
based on the original dry weight of the coal refuse (and thus corresponds
to the moisture that will be measured in exposed refuse piles) whereas the
percent of lime hydrate was based on the weight of wet coal refuse. The
compressive strength as a function of the percent of lime hydrate at
the optimum percent of water is shown in Figures 2 to 5 for each coal
refuse (-10 mesh). For these experiments a 20 percent carbon dioxide
gas mixture and a 4 hour carbonate bonding reaction time was used. Also,
the compressive strength was determined immediately after carbonate bond-
ing was completed (zero curing time). Figures 2 to 5 clearly indicate
that the compressive strength increased as the percent of lime hydrate
increased for each of the refuse materials studied.
Effect of Curing Time on Compressive Strength
In Figures 2 to 5 it is seen that the maximum compressive strength ob-
tained for each coal refuse was about 1500 psi at 12 percent lime hydrate,
and maximum strengths occurred at 12, 12, 9, and 12 percent water for
unburnt and burnt bituminous and unburnt and burnt anthracite coal refuse
respectively. The resulting compressive strengths were rather low
judging from prior experimental results using other fill materials for
carbonate bonding which yielded compressive strengths in excess of 2000
psi. Consequently, additional experiments were conducted to determine
if there is a curing time effect on compressive strength.
To show the effect of curing time on compressive strength, carbonate
bonded briquettes were prepared in the usual manner except that the
compressive strength was determined as a function of the time elapsed
after carbonate bonding was completed. This was done for each coal refuse
at 12 percent lime hydrate and the optimum percent of water. For these
experiments the -10 mesh particle size fraction, a 100 percent carbon
dioxide gas mixture and a 4 hour carbonate bonding time were used.
Compressive strengths were determined at various times after carbonate
bonding was completed for a total time of 600 hours. In Figure 6 the
compressive strength as a function of curing time for unburnt bituminous
which is shown is typical for all the coal refuses studied. The com-
pressive strength increases significantly during the first 150 hours of
curing time and then levels off to a relatively constant value.
Specifically, for the unburnt bituminous coal refuse, the compressive
strength increased from about 1500 psi to 4400 psi after a curing time
of approximately 150 hours. In order to make these findings applicable
to a wide variety of coal refuses, Figure 7 was prepared and is a plot
of the reduced compressive strength, defined by (S - S 0 )I(S 00 -S 0 ),
as a function of the curing time in hours: where,
S = compressive strength at any time
S 0 compressive strength at time zero (immediately after
carbonate bonding)
-12—
-------
I I I
0 3 6 9 12
7 Lime Hydrate
FIGURE 2-EFFECT OF LIME HYDRATE CONTENT ON COMPRESSIVE
STRENGTH FOR UNBURNT BITUMINOUS COAL REFUSE
1000
500 —
0
6
12
7 Lime Hydrate
FIGURE 3-EFFECT OF LIME HYDRATE CONTENT ON COMPRESSIVE
STRENGTH FOR BURNT BITUMINOUS COAL REFUSE
2000
1500
1000
500
U)
w
‘-1
4J
cI
U)
a)
w
0
0
U)
a)
U)
U)
a)
E
0
0
2000
1500
I
3
9
-13-
-------
3
6 9
.1
12
7 Lime Hydrate
FIGURE 4-EFFECT OF LIME HYDRATE CONTENT ON COMPRESSIVE
STRENGTH FOR UNBURNT ANTHRACITE COAL REFUSE
7 Lime Hydrate
FIGURE S-EFFECT OF LIME HYDRATE CONTENT ON COMPRESSIVE
STRENGTH FOR BURNT ANTHRACITE COAL REFUSE
2000
1500
1000
500
0
a)
aD
a)
I-i
U )
a)
a)
a)
a)
I-i
0
0
a)
aD
a)
U)
a)
a)
a)
a)
I
2000
1500
1000
500
0
6
9 12
-14-
-------
Curing Time, Hours Since Carbonate Bonding
FIGURE 6-EFFECT OF CURING TIME ON COMPRESSIVE STRENGTH
FOR UNBURNT BITUMINOUS COAL REFUSE
5500
4500
*
U,
-1
U)
w
‘ -4
CID
w
U,
U)
Q)
‘-1
0
0
2’
•1
0 100 200 300 400 500
600
-------
0
.4
•1
NO
) -1
ww
EN
1 l .,-4
8
.
Cl ) Q)
w I - I
CI )
Cl ) w
I)
U)
0”
EQ
-l I
;‘c •4
-------
S 00 = final compressive strength at times beyond 150 hours
This figure indicates that a curing time of approximately 150 hours is
required to develop the ultimate compressive strength. An investigation
of this curing time phenomenon was beyond the scope of this study; how-
ever it is believed that a recrystallization of the calcium carbonate
crystal matrix is occurring to yield a stronger structure. Based on
this finding, compressive strength determination for all subsequent
experimental investigations was performed on briquettes which were cured
at least 150 hours.
Effect of Water Content on Compressive Strength
The effect of coal refuse particle size and water content on carbonate
bonded compressive strength is shown in Figures 8 to 11. In these ex-
periments the coal refuse contained 12 percent lime hydrate which was
reacted with pure CO 9 for eight hours. Considering only the results for
the -10 mesh material, it is seen that the maximum compressive strength
results at the higher water content. Specifically, the optimum water
content is 12 percent for all coal refuses except the burnt bituminous
which increased to the maximum percent of water investigated (15
percent). Based on these limited results, it appears that coal refuse
will be amenable to carbonate bonding at 12 to 15 percent water. The
results of Figures 8 to 11 show a gradual variation in compressive
strength with changes in water content. These results are encouraging
since it appears that quite useable compressive strengths are obtained
without resorting to unrealistic water content controls which would be
a decided disadvantage to the large scale application of this process.
It is also advantageous that the observed optimum water contents occur
at relatively high values, because in the commercial application of the
process it is more practical to add water rather than dry the coal refuse.
Of course, lime rather than lime hydrate could be used if lower water
content levels were required. Lime would reduce the free moisture by
reacting with water to form lime hydrate.
The unburnt bituminous coal refuse exhibited a compressive strength-water
content relationship which was unexpected. The compressive strength did
not exhibit it5 usual functional dependence (i. e., increase as particle
size decreases), but the optimum moisture varied in a somewhat random
manner. Specifically, the optimum water content for maximum compressive
strength was 12, 9, and 12 percent for the -10, -20, and -60 mesh
particle size fractions respectively. These results are unusual and as
yet cannot be explained.
Effect of Particle Size on Compressive Strength
Figures 8 to 11 also indicate the effect of particle size on compressive
strength. Again excluding the unexplainable results for unburnt
bituminous, it is seen that the compressive strength increases as the par-
ticle size of the coal refuse decreases. This is expected since a de-
crease in particle size (along with a wider size distribution) normally
—17—
-------
4000
3000
2000
4000
3000
2000
FIGURE 9-EFFECT OF PARTICLE SIZE AND WATER CONTENT ON
COMPRESSIVE STRENGTH FOR BURNT BITUMINOUS
FIGURE 8-EFFECT OF PARTICLE SIZE AND WATER CONTENT ON
COMPRESSIVE STRENGTh FOR UNBURNT BITUMINOUS
8 10 12 14 16
7 Water
“ -I
. 1 - I
I -i
E
0
U
‘ -I
4J
w
I -I
cl
a)
‘-4
ci )
a)
‘-4
0
U
6 8 10 12 14
7o Water
16
6
-18-
-------
7, Water
FIGURE 10-EFFECT OF PARTICLE SIZE AND WATER CONTENT ON
COMPRESSIVE STRENGTH FOR UNBURNT ANTHRACITE
7 Water
FIGURE 11-EFFECT OF PARTICLE SIZE AND WATER CONTENT ON
COMPRESSIVE STRENGTH FOR BURNT ANTHRACITE
+
0
4000
3000
2000
1000
4000
3000
2000
1000
8 10 12 14
4)
riD
4)
-1
C l ,
U )
‘a)
0
U)
0..
a)
4J
C,)
4)
‘-4
‘I )
C ,)
4)
c i .
E
0
0
16
-60 esh
0
8 10 12 14
16
-19-
-------
results in a more dense composite thus enhancing the formation of a strong
continuous calcite crystal matrix. Additionally, the higher surface
area permits more calcite-to-refuse contacts for cementing and holding
the materials together. The effect of surface area on compressive
strength is shown in Figure 12 where the compressive strength is plotted
as a function of the specific surface area (cm 2 /gram) for the -60 mesh
particle size fractions of all the coal refuses except the unburnt
bituminous. The carbonate bonding conditions were exactly the same as
for Figures 6 and 7 and a 12 percent water content was used. Based on
these findings, the smallest economically feasible particle size coal
refuse should be used in the commercial application of the process to
obtain maximum strength (provided the composite material is not so dense
as to retard the transport of carbon dioxide into the material, and to
prevent the venting of steam formed by the highly exothermic nature of
the carbonate bonding reaction). If steam is prevented from rapidly
exiting the carbonate bonded material, the internal pressure will build
up and cause cracking with a resulting loss in strength. As explained
previously, the rate of the carbonate bonding reaction can be controlled
by varying the partial pressure of the carbon dioxide in the gas. The
effect of carbon dioxide concentration and carbonate bonding reaction
time on compressive strength will be discussed in the next section.
Effect of Carbon Dioxide Concentration and Carbonate Bonding Reaction
Time on Compressive Strength
To determine the effect of carbon dioxide concentration and carbonate
bonding reaction time on compressive strength, three separate experiments
were conducted using a 20, 60, and 100 percent carbon dioxide gaseous
mixture. BriquetteS (12 percent lime hydrate, 12 percent 1120) were re-
moved from the reaction tubes after 2, 5, and 8 hours for each of the
above gaseous mixtures. In Figures 13 to 16 are shown the results of
these experiments for the -10 mesh mesh coal refuse. As is generally the
rule, the compressive strength increases as the carbon dioxide concen-
tration increases for a given carbonate bonding reaction time. This
occurs because the extent of carbonation increases with carbon dioxide
concentration. As seen in Figures 13 to 16, both the 60 percent and
100 percent carbon dioxide gaseous mixtures yield higher compressive
strengths than the 20 percent gaseous mixture. However, for long carbon-
ation times (less than or equal to S hours) compressive strengths tend
to converge to a comnxn value. This is to be expected since the ultimate
strength is a strong function of the degree of carbonation. Figures 14
and 15 indicate that for burnt bituminous and unburnt anthracite the 60
percent 02 gaseous mixture yields a higher compressive strength than the
100 percent CO 2 gaseous mixture. This is probably caused by the car-
bonate bonding reaction occurring too rapidly at the higher concentration.
Steam generation was too rapid causing a multitude of hair line fractures
which have reduced the compressive strength.
The effect of carbonate bonding reaction time is as expected- - the
compressive strength increased in each case as the time increased. More
-20-
-------
4000
3000
a)
4J
2000
a)
a)
E
0
o 1000
• Burnt Bituminous
+ Unburnt Bituminous
Burnt Anthracite
0 I I
500 1000 1500 2000
Specific Surface, cm 2 /Gram
FIGURE 12-EFFECT OF SPECIFIC SURFACE AREA ON COMPRESSIVE STRENGTH
+
I
+
x
-21—
-------
0
1
2
I
I
4 6 8
Time, Hours
FIGURE 13-EFFECT OF CARBON DIOXIDE CONCENTRATION AND REACTION
TIME ON COMPRESSIVE STRENGTH FOR UNBURNT BITUMINOUS
3000
2500 -
200(
0
2
6
8
Time, Hours
FIGURE 14-EFFECT OF CARBON DIOXIDE CONCENTRATION AND REACTION
TIME ON COMPRESSIVE STRENGTH FOR BURNT BITUMINOUS
4500•
4000
350( -
nnt —
x
bO
w
5-I
4J
5- ’
0
0
“-4
C’)
. 1 )
bO
s-I
C/:3
a)
- .4
0 ,
0,
0)
s-I
0
0
4
I
-22--
-------
2500
2000
1500
1000
O 2 4
Time, Hours
FIGURE 15-EFFECT OF CARBON DIOXIDE CONCENTRATION AND REACTION
TIME ON COMPRESSIVE STRENGTH FOR UNBURNT ANTHRACITE
3000 -
2500
2000
1500
0 8
Time, Hours
FIGURE 16-EFFECT OF CARBON DIOXIDE CONCENTRATION AND REACTION
TIME ON COMPRESSIVE STRENGTH FOR BURNT ANTHRACITE
I I I
2 4 6
U)
0
4J
U)
a)
—I
U)
U )
a)
0
0
I I ____
6 8
. ‘.-
U)
4 J
a)
C / )
.,-4
U)
U)
a)
0
0
x
-23-
-------
important is the observation that for an 8-hour carbonate bonding time
the compressive strength is not significantly different at the various
carbon dioxide concentrations. For example, the compressive strength
of unburnt bituminous after 8 hours of carbonate bonding varies between
4200 and 4350 psi for carbon dioxide concentrations of 20, 60, and 100
percent. Thus, compressive strength (obtained under optimum conditions)
is a strong function of the degree, or extent, of the completion of the
carbonate bonding reaction. This result is encouraging because in the
commercial application of the process it is economically more feasible
to utilize a 20 percent carbon dioxide gaseous mixture derived from the
combustion of a carbonaceous fuel rather than expensive pure CO 2 .
Additional Properties of Carbonate Bonded Coal Refuse
In addition to strength, the resulting carbonate bonded coal refuse must
possess other requisite properties to produce suitable construction
materials, The carbonate bonded product must be able to prevent or
minimize the permeation of water and air (oxygen) to inhibit acid mine
water formation and must also possess a resistance to weathering. If
carbonate bonded coal refuse cannot maintain its structural integrity
during repeated freezing and thawing, then it cannot be an effective
construction material. The properties that were investigated and presented
in this section are: the air and water permeability of coal refuse and
the compressive strength after repeated freeze-thaw cycles. A search of
the ASTM standards did not yield an applicable standard determination for
air and water permeability, therefore, special equipment and procedures
were established for these determinations which were based on the commonly
accepted theory of permeability. Where applicable, cement briquettes were
used as a basis for comparison. These cement briquettes were prepared
using one part (by weight) Portland Type One cement, 2.75 parts Ottawa
Sand and ) part water. The cement and sand were dry mixed, the water
added slowly, and the wet mixture blended thoroughly. The cement briquettes
were allowed to cure 28 days.
Air Permeability of Carbonate Bonded Coal Refuse
The air permeability of carbonate bonded coal refuse briquettes was
determined using the apparatus schematically represented in Figure 17.
The apparatus consisted of a 1¾ inch I.D. by 24 inch high plexiglass
chamber on the bottom of which was cemented a coal refuse briquette. The
top of the chamber was sealed with a plexiglass plate containing a
pressure gauge. The air permeability was defined as the volumetric flow
rate of air (ft 3 /hr) per cros sectional flow area (ft 2 ) per unit
pressure driving force (lb/ft /ft).
The chamber was pressurized with gas and the change in chamber pressure
with time was measured to determine the permeability of the coal refuse
briquette. Initial chamber pressures of 30, 15 and 5 psig were used.
For an experimental run the chamber pressure was allowed to decrease to
a pressure of 1 psig before termination. An average air permeability was
then determined using the results of all the experiments.
As indicated above, the experimental procedure to determine air permeability
-24-
-------
Pressure
Gage
Plexiglass
Tube
t te
FIGURE 17-AIR PERMEABILITY APPARATUS
Sealant
-25-
-------
consisted of using a coal refuse briquette as a plug in a sealed chamber
and the chamber pressure was determined as a function of time. This is
an unsteady state operation in which the pressure driving force is
continually changing with time. The mathematical treatment required to
obtain the air permeability is somewhat involved and is presented in the
Append ix.
In this experiment the air permeability for -10, -20, and -60 mesh
unburnt bituminous coal refuse briquettes was determined. For com-
parative purposes, experiments were also run on a cement briquette of
the same dimensions as the coal refuse briquettes. The results are
presented in Table IV where the air permeability, the ratio of the
permeability of the coal refuse briquettes to the cement briquette,
and the average particle diameter of the coal refuse (determined from
the screen analysis presented in Table I) are presented.
TABLE IV
Air Permeability of -10, -20, & -60 Mesh
Unburnt Bituminous Coal Refuse
Air* Permeability of Coal Refuse! Average Particle**
Sample Permeability Permeability of Cement Diameter
Unburnt
Bituminous
-10 Mesh 2.7 x l0 9 .046
Unburnt
Bituminous
-20 Mesh 6.0 x l0 2 .013
Unburnt
Bituminous
-60 Mesh 2.4 x 10 0.8 .008
Cement 3.0 x l0
* ft 3 /(hr - ft 2 - (lb - ft 2 /ft ))
** Diameter in centimeters
It is seen for the -10 mesh briquettes that the permeability ratio is 9
times that of the cement briquette whereas the ratio is 2 and 0.8 for
the -20 and -60 mesh coal refuse. This infers that the smaller particle
size fractions (—20 mesh or lower) carbonate bonded coal refuse compare
quite well with cement and would provide an adequate seal against air
permeation. In order to make the air permeability results generally
applicable to all types of coal refuses, Figure 18 was prepared. Figure
18 is a plot of the air permeability for unburnt bituminous coal refuse
vs. the average particle diameter and can be used to estimate the
permeability of a coal refuse provided the particle size distribution is
known.
-26-
-------
10
4_ i —
c. 1
4 _ i
4 -1
2
4-1
cn
1
4 _ i
0 .03 .05
Average Particle Diameter, cm
FIGURE 18-AIR PERMEABILITY VS. AVERAGE PARTICLE DIAMETER
I
.01 .02
04
-27-
-------
Water Permeability of Carbonate Bonded Coal Refuse
The water permeability of coal refuse was determined in much the same
manner as air permeability. The apparatus for the water permeability
determination is shown in Figure 19,and consists of a l inch I.D.
by 24 inch high plexi-glass primary chamber sealed on top with a plexi-
glass plate containing a valve and on the bottom with a coal refuse
briquette. For this test, an additional chamber is attached to the
primary chamber and located beneath the briquette to collect the water
permeating through the briquette and to prevent unrealistic permeation
rates due to convective air flows about the exposed surface of the
briquette. In addition, water permeating through carbonate bonded coal
refuse will probably enter into surroundings which will normally be
saturated with water vapor. The lower chamber, therefore, provides a
water vapor saturated atmosphere to better simulate actual conditions
and provide more meaningful results. The lower chamber contains a small
hole on its side to vent evaporated water and prevent a build up of
pressure.
The water permeability test was conducted by initially filling the
primary chamber with water to a level of 20 inches and, on a daily basis
measuring the amount of water lost due to permeation. After the quantity
of water loss was determined, this amount was added back into the
primary chamber to maintain a constant pressure driving force.
This test was run for 14 days and an average rate of water permeation
determined. For all practical purposes this test was run on a steady
state basis in which the water pressure driving force was constant;
consequently, the water permeability of coal refuse can be directly cal-
culated since the flow area of the briquette, the thickness of the
briquette, the constant water pressure driving force, and the average
rate of water permeation are known.
The water permeability of -10, -20, and -60 mesh unburnt bituminous coal
refuse and a cement briquette is shown in Table V.
TABLE V
Water Permeability of Coal Refuse
Permeability
Sample Permeability* Ratio**
Unburnt Bituminous, -10 Mesh 5.5 x 10 4.0
Unburnt Bituminous, -20 Mesh 2.5 x l0 1.8
Unburnt Bituminous, -60 Mesh 2.5 x 1O 1.8
Cement 1.3 x io
* Water Permeability in lbs/(hr. — ft. 2 - (lbs.Ift. 2 /ft.))
** Permeability of Coal Refuse/Permeability of Cement
-28-
-------
Va1
/ ‘ Water
— ‘ Leve1
Plexiglass
Tube
Primary Chamber
Briquette
Secondary
Chamber
Sealant /
Vent
— _ JJ,J/ 7 1
FIGURE 19-WATER PER1 4EABILITY APPARATUS
-29-
-------
Water permeability is the mass flow rate of water per crossectional
flow area per unit pressure driving force. Also shown in Table V is the
water permeability ratio defined as water permeability of coal refuse
divided by the water permeability of cement. It is interesting to note
that carbonate bonded coal refuse compares quite well with cement in
prevention of water permeation. This is particularly so at the lower
particle size fractions of -20 and -60 mesh where the permeability ratio
is 1.8. Results indicate that carbonate bonded coal refuse would
provide an adequate seal against water permeation and minimize acid mine
water formation.
Freeze-Thaw Test on Carbonate Bonded Coal Refuse
Freeze-thaw tests were performed on a series of -10, -20, and -60 mesh
unburnt bituminous coal refuse briquettes. The freeze-thaw test was based
on the ASTM designation; C290-67 entitled, “Standard Method of Test for
Resistance of Concrete Specimens to Rapid Freezing and Thawing in Water”.
The freeze-thaw test, as modified for this study, is presented in the
Appendix. The objective of this test was to determine compressive
strength of coal refuse briquettes as a function of the number of freeze-
thaw cycles to which they had been subjected. Figures 20 to 22 present
the briquette compressive strength as a function of the number of freeze-
thaw cycles for -10, -20, and -60 mesh unburnt bituminous coal refuse.
The compressive strength in the 20 to 35 freeze-thaw cycle range tends
to stabilize at a relatively constant value of 700 to 900 psi for each of
the particle size fractions.
It must be noted that subjecting a composite material of small dimensions
(1-1/8 inch diameter by 1 inch high briquettes) to 35 freeze-thaw cycles
is an extreme test. The ASTM standard test method contemplates the use
of specimens not less than 3 inches nor more than 5 inches in width and
depth and not less than 14 inches nor more than 16 inches in length.
The normal erosion of surface edges which occurs during freeze-thaw
testing does not seriously effect large specimens, whereas for small
briquettes the effect on strength is quite dramatic as evidenced by the
results of this study. An examination of the coal refuse briquettes after
35 cycles revealed an erosion of about ¾ inch of material almost completely
around all edges. This reduced the area on which the compressive
strength test load was applied causing the briquette to fail at lower
loads and indicate a lower compressive strength based on the original
one square inch area. For comparison, cement briquettes were also
subjected to freeze-thaw testing. The cement briquettes had an initial
compressive strength of 2000 psi which reduced to 1200 psi after 35
freeze-thaw cycles. An examination of the cement briquettes after 35
cycles revealed an erosion of about 1/8 inch of material almost completely
around all edges.
Even under extremely unfavorable test conditions, the carbonate bonded
coal refuse yielded compressive strengths in the order of 700-900 psi.
Consequently, it should adequately withstand the effects of severe
-30-
-------
4000
3000
2000
1000
U)
a)
‘-1
C l )
a)
U)
U)
a)
0
0
3000
2000
1000
U)
a)
4- i
C r4
C l
“-4
U)
U)
U
0
0
3000
2000
1000
0 10 20 30 40 0 10 20 30 40 0 10 20 30 40
Freeze-Thaw Cycles
(--10 Mesh)
FIGURE 20
Freeze-Thaw Cycles
(--20 Mesh)
FIGURE 21
Freeze-Thaw Cycles
(--60 Mesh)
FIGURE 22
4000
“ - 4
U)
4-i
a)
4-i
C t )
a)
- ,- -1
U)
C l)
a)
I -i
0
0
I -
EFFECT OF FREEZE-THAW CYCLES ON COMPRESSIVE STRENGTH FOR UNBURNT BITUMINOUS COAL REFUSE
-------
changes in temperature.
Suggested Operating Parameters for Carbonate Bonding Process
The economics of coal refuse carbonate bonding depend primarily on four
controllable factors: the particle size of the coal refuse, quantity of
lime hydrate required, carbon dioxide concentration and time of carbon-
ation.
In the present study, particle size effects were evaluated over a rather
narrow range limited by the size of the briquette die employed. However,
based on the results obtained, it is reasonable to expect that the
carbonate bonding process for the conversion of coal refuse to paving
materials will accommodate fill material containing a top size of one
inch. To maintain air and water permeability conditions as developed in
this report, it will only be necessary to specify a particle size dis-
tribution so that minimum porosity (i.e., maximum density) will be
achieved. Data are available in literature that will permit an estimate
of the proper size consist to use in commercial applications.
Depending upon strength criteria established for commercial applications,
this study has shown that a maximum of 12 percent lime hydrate will yield
compressive strengths approaching 4400 psi for coal refuse. Should
lower compressive strength be acceptable, the amount of lime hydrate can
be decreased accordingly.
It has been found that adequate carbonate bonded coal refuse strengths
have been obtained when a 20 percent carbon dioxide gaseous mixture is
used for reaction times of 8 hours. Assuming that faster set times are
not required, a simple inexpensive carbon dioxide gas generator can be
used to yield carbon dioxide concentrations of about 15 to 20 percents
Faster set times will require the use of carbon dioxide generators
producing a higher purity gas.
When using coal refuse material, the data indicate that a moisture con-
tent of about 9 to 15 percent will be required to obtain optimum
strength. This moisture requirement can readily be realized at the site.
-32—
-------
APPLICATIONS OF CARBONATE BONDED COAL REFUSE
The primary objective of this study was to determine the feasibility of
utilizing coal refuse as a fill material in the carbonate bonding process.
The results of this study show that carbonate bonded coal refuse has
the necessary strength properties for construction applications and the
necessary sealant properties to effectively prevent the permeation of
air and water which cause the generation of acid constituents in water.
The applications of carbonate bonded coal refuse are numerous. The
carbonate bonded process utilizing coal refuse could be employed to (1)
construct secondary roads utilizing the formed carbonate coal refuse as
is, (2) provide a strong inexpensive base material for major road
construction, (3) provide a first class road when used in conjunction with
a thin layer of tar or similar material, (4) provide a strong inexpensive
sealant layer on existing coal refuse piles, and (5) make bricks. These
applications, except (4) above, have the advantage that coal refuse can
be worked off thus eliminating unsightly coal refuse piles. Also the
current production of coal refuse could be utilized in the process to
prevent the formation of new or the enlargement of old coal refuse piles.
Although no work has been completed on the field testing of the car-
bonate bonding process employing coal refuse or spoil bank material, the
bench scale study and earlier experimentation (on a 4 foot square patch
of carbonate bonded soil) show that the following proposed commercial
concept is feasible for both sealing coal refuse piles and constructing
roads. The coal refuse or spoil bank material is tilled to a predeter-
mined depth, depending on the particular application. During the tilling
operating, lime hydrate and water (if necessary) in the proper pro-
portions, are introduced into the tilled material. The tilling action
not only loosens the fill, but is also effective for blending and making
a homogeneous mixture of lime and fill material. Once the tilling and
blending step has been completed, perforated plastic pipe is placed on
four foot centers underneath the surface of the blended material. The
perforated plastic pipe is expendable and remains in place under the
carbonate bonded material. The plastic pipe is connected to a portable
carbon dioxide generator which can be a burner for coal refuse, coal,
oil, or any other carbonaceous fuel which generates products of combustion
containing carbon dioxide. Once the plastic pipe is in place and connected
to the generator, the blended material is then tamped down by means of a
roller. The flue gases are pumped under pressure (approximately five
pounds per square inch) into the perforated plastic pipe. At this time
the portable carbon dioxide generator is started and gaseous carbon
dioxide along with the inert components of the flue gas are pumped into
the perforated pipe. As the carbon dioxide permeates upward through the
lime hydrate-fill material mixture, the carbonate bonding reaction starts
and hardens the blend. The conversion of the lime hydrate to the car-
bonate form occurs with the visible evolution of steam. When the steam
has stopped emanating from the pavement, the reaction is completed and
the spoil bank or coal refuse material will be surfaced hardened with a
-33-
-------
concrete-like covering. At present the effect of the combustion products
other than carbon dioxide on the carbonate bonding reaction is not known.
Since these gases will consist mainly of nitrogen, unburnt oxygen and
water vapor, it is anticipated that there will be little or no effect.
If future work indicates that the water vapor decreases the carbonation
reaction rate, the initial water content of the coal refuse may be
adjusted to compensate for the water contained in the flue gas.
As is evident from the foregoing description, the process is relatively
simple using existing road grading equipment, requires no unusual equip-
ment or highly skilled labor, and uses locally available materials (with
the exception of the lime hydrate) thus minimizing material handling.
To carry out the process, the following items of equipment will be
required
1. A portable carbon dioxide generator
2. Expendable perforated plastic pipe
3. Earth moving and compacting equipment
4. A roto-tiller attachment for tractors
The earth moving and compacting equipment are not unique to the carbonate
bonding process and are considered part of the normal inventory of
maintenance construction equipment. If the work is carried out as out-
lined above, a rotary tiller attachment to the tractor will be needed.
Such a tilling attachment is readily available as a stock item.
The portable carbon dioxide generator is also readily fabricated. The
equipment will essentially consist of a burner and a blower. The burner
which can operate on a convenient fuel (coal, coal refuse, bottled gas,
fuel oil, or other carbonaceous fuels) would generate products of com-
bustion containing 15 to 20 percent carbon dioxide. These flue gases
will be compressed to about five pounds per square inch for introduction
into the perforated plastic pipe. If rapid setting of the carbonate
bonded material is required, concentrated carbon dioxide can be gen-
erated by the addition of an amine absorber to the above equipment.
Economics will dictate the type of carbon dioxide generator.
Cost for using carbonate bonding as a method for sealing coal refuse
piles is difficult to establish, because full scale paving tests have
not been completed. However, by assuming that labor costs for carbonate
bonding are comparable to paving techniques which require similar
operations, an approximate cost estimate can be made.
Results of a survey of construction costs associated with other paving
methods are presented in Table VI, which shows paving costs as a function
-34-
-------
of construction material and depth of paving. If it is assumed that
carbonate bonding labor costs are comparable to those for asphalt,
the data on Table VI for 2-inch asphalt on 4-inch subbase indicate that
the labor costs for a 6-inch carbonate bonded pavement will be $0.85
per square yard. Assuming labor cost varies linearly with the
TABLE VI
Paving Costs of Various Materials*
Construction Material
Total Cost
Labor Cost
Material Cost
4-inch thick concrete
on 6-inch subbase
$7.87/sq. yard
$5.00/sq. yard
$2.87/sq. yard
2-inch asphalt on
existing base
3.12/sq. yard
0.56/sq. yard
2.56/sq. yard
2-inch asphalt on
4- inch subbase
4.68/sq. yard
0.85/sq. yard
3.83/sq. yard
3-inch asphalt on
12-inch subbase
7.62/sq. yard
1.37/sq. yard
6.25/sq. yard
4-foot layer
of dirt
1.33/sq. yard
1.33/sq. yard
No Cost
l- foot layer
of dirt 0.68/sq. yard
thickness of the seal over the range of 6 to 15 inches, labor costs for
the carbonate bonded process as a function of thickness were calculated
and are presented in Table VII.
TABLE VII
Paving Costs for the Carbonate Bonding Process
Thickness
Total Cost
Labor Cost
Material Cost
*Popper, H, “Modern Cost Estimating
York, p. 101 (1970), except cost of
communication.
Techniques”, McGraw-Hill Co., New
dirt which was obtained by private
0.68/sq. yard
No Cost
2
inches
$0.83/sq.
yd.
$0.63/sq.
yd.
$0.20/sq.
yd.
4
inches
1.05/sq.
yd.
.74/sq.
yd.
0.31/sq.
yd.
6
inches
1.28/sq.
yd.
.85/sq.
yd.
0.43/sq.
yd.
8
inches
1.60/sq.
yd.
.97/sq.
yd.
0.63/sq.
yd.
10
inches
1.82/sq.
yd.
1.08/sq.
yd.
0.74/sq.
yd.
12
inches
2.05/sq.
yd.
1.10/sq.
yd.
0.95/sq.
yd.
15
inches
2.48/sq.
yd.
1.37/sq.
yd.
1.11/sq.
yd.
-35-
-------
Because it is possible to substitute equipment costs for labor costs, and
vice versa, labor and equipment costs are combined in Table VII under
the heading “Labor Cost”. The “Labor Cost” also includes the cost
of earth tilling, labor and equipment required to crush coal refuse to
a particle size distribution suitable for carbonate bonding.
Material cost for carbonate bonding as a function of seal or paving
thickness is also presented in Table VII. In the material cost cal-
culations it was assumed that coal refuse is available on site at no
cost and is used as fill and as fuel to produce the carbon dioxide re-
quired in the process. Lime hydrate was assumed to cost $25 per ton
delivered and 12 percent lime hydrate was assumed to be used in the fill.
Perforated hose, used to distribute the carbon dioxide in the material
to be carbonate bonded was assumed to be located on four-foot centers
in the pavement. For carbonate bonded material thicknesses of more
than 6 inches, additional hose was assumed. Pending establishment
of specifications, the retail price of perforated hose, $0.04 per
linear foot was assumed. Using these cost data, the total cost (material
and labor) of carbonate bonding was calculated for various sealing or
paving thicknesses and is presented in Table VII.
For sealing of coal refuse piles, it is believed that a 6-inch thick
carbonate bonded layer will have air and water permeability and load
bearing characteristics comparable to a 4 foot layer of dirt. A
comparison of Tables VI and VII indicates that the total cost per square
yard of seal is slightly lower for the carbonate bonding process.
Sealing with the four foot layer of dirt costs $1.33 per square yard
whereas a 6-inch thick carbonate bonded seal costs $1.28 per square
yard. Since the costs were conservatively estimated, it would appear
that carbonate bonding offers considerable economic superiority for seal-
ing coal refuse piles.
At present, the State of Pennsylvania requires that a 1½ foot layer of
dirt be used to seal a coal refuse pile. The cost of a 1½ foot layer of
dirt is estimated to be $0.68 per square yard (Table VI) compared to a
cost of $0.83 per square yard for a 2-inch carbonate bonded layer
(Table VII). A 2-inch layer of carbonate bonded coal refuse is of a
much more permenant nature and will be far superior to a 1½ foot layer
of dirt as concerns air and water permeability and load bearing char-
acteristics. Consequently, it is believed that the $0.15 difference in
cost per square yard is a necessary additional cost to insure the
prevention of acid water formation.
If carbonate bonded coal refuse is used to construct a secondary road
then its cost should be compared with a 2-inch asphalt layer on a 4-inch
subbase (4.68 per square yard-Table VI). For a 6-inch carbonate
bonded layer, the cost is $1.28 per square yard (Table VII). These
results indicate that for secondary road building applications carbonate
bonding costs considerably less than asphalt paving.
-36-
-------
Cost of carbonate bonded coal refuse used as a strong base material
for road building can be compared to the cost of other materials from
data presented in Table VI. From Table VI, it is seen that a
difference of $1.56 ($4.68—$3.12) per square yard exists for a 2-inch
asphalt paving with and without a 4-inch subbase. The $1.56 difference
represents the cost of the subbase. A 4-inch subbase of carbonate
bonded coal refuse would cost $1.05 per square yard (Table VII). As
is evident, carbonate bonded coal refuse is significantly less expensive
than conventional subbase materials. If 12-inch thick carbonate bonded
coal refuse is covered with a 2-inch layer of asphalt to provide a first-
class road, its cost should be compared to alternatives of either 4-inch
concrete on 6-inch subbase or 3-inch asphalt on 12-inch subbase. For
a 2-inch asphalt layer on a 12-inch carbonated bonded subbase, the cost
of asphalt will be $3.12 per square yard (Table VI) and the cost of
the carbonate bonded subbase will be $2.05 per square yard (Table VII).
Thus the cost of such a first-class road is $5.17 per square yard, which
is significantly less than $7.87 and $7.62 per yard for the concrete
and asphalt alternatives, respectively (Table VI).
In summary, the above indicates that substantial economics can be realized
if the carbonate bonding process is utilized as a refuse pile sealant
or road construction technique. This should induce coal producers to
cover or utilize their coal refuse, thereby minimizing the formation of
acid water.
We suggest the process be demonstrated by actually sealing a coal refuse
pile to prevent acid water formation. This would also allow development
of large scale construction techniques necessary to apply the process.
Based on results of this demonstration, an engineering evaluation will
be made, construction standards and procedures developed, and costs
established for future application of the process.
-37-
-------
ACKNOWLEDGMENTS
The support of the project of the Environmental Protection Agency
and the help provided by Mr. Ernst Hall, Mr. Ronald Hill, Mr. Don O’Bryan,
and Mr. Robert Scott, Project Officer, is acknowledged with sincere
thanks.
-39-
-------
APPENDICES
Determination of Air Permeability
When a pressure difference exists across a porous solid, a flow of gas
through the solid will occur. This is not diffusional flow in the usual
sense but it can be described using the methods of diffusion. For the
permeation of gas through a coal refuse briquette, where the diameter of
the capillaries is small and the gas ‘velocity through the solid is small,
flow will be streamline and can be described by Poiseuille’s law for a
compressible fluid obeying the perfect gas law
2
N = d g a i - P 2 ) (1)
32ulRT
where a (P + P)/2. A table of nomenclature is presented in Table
Viii for the following discussion.
TABLE VIII
Table of Nomenclature for Air Permeability
A Area through which gas is flowing, ft 2
d Diameter of a capillary, ft 8 2
g Conversion factor, 4.17 x 10 lb. mass - ft/(lb. force - hr.)
1 Length of capillary, ft 2
N Rate of diffusion, lb. moles/hr. - ft.
n Moles, lb. moles
P Pressure, lb./ft 2 (abs.) 2
P 1 Gas pressure in pressure vessel, lb./ft (abs.)
P 2 Atmospheric pressure, lb./ft 2 (abs.)
P 0 Gas pressure in pressure vessel at time zero, lb./ft 2 (abs.)
a (P 1 + P 2 )/2, lb./ft 2 (abs.)
R Universal gas constant, 1543 ft. - lb./(lb mole - °R)
T Absolute Temperature, °R
t Time, hr.
u Viscosity, lb./(ft. - hr.)
V Volume of ressure vessel, ft
v Volume, ft
The perfect gas expression which relates the pressure, temperature and
volume of a perfect gas is expressed as
Pv = nRT (2)
The experimental procedure used to determine the permeability of coal
refuse briquettes consisted of using a briquette as a plug in a pressure
-4l -
-------
dt
vessel and determining the pressure within the vessel as a function of
time. This is an unsteady state operation in which the pressure driving
force is continually changing with time; therefore, a mass balance on the
gas contained in the vessel yields
(3)
where A is the area through which gas flow is occurring.
Expressions (1) to (3) form the basis of the mathematical treatment
required to determine permeability. Substituting equation (1) into (3)
yie ids
d 2 g A a (P 1 - P 2 )
dn — 4
dt 32 u1RT
Taking the differential of the ideal gas law, equation (2), with dv = 0
results in dn = Vdp 1 /RT and substituting this expression into (4) yields
dP 1 — p 1 p 2 (P -P)
— — -K 2 1 2 (5)
where K = A d 2 g/(32 Vul)
Rearranging equation (5) results in the following expression relating the
pressure within the vessel P 1 to time t:
t
dt
(6)
0
where P is the initial pressure in the vessel at time t = 0. Intergrat-
ing the°above expression between the indicated limits yields the following
relationship between the chamber pressure P 1 and t:
P 2 + p 0
P -P
In 2 °KP 2 t
+ P
1
P 2 - P 1
(7)
Expression (7) is used to determine K
as a function of time. The parameter
since experimentally P 1 is known
K is determined as the slope of
1
P 22 - P 1 2
P
0
-42-
-------
(P + P /P - P )I(P + P /P - P ) vs t
2 o2 o 2 12 1
By definition, the permeability of a porous solid is the rate of diffusion
per unit pressure gradient across the solid. The permeability P can be
defined using the following expression for which P has the units of
ft 3 /hr - ft 2 - (lb/ft 2 /ft) and z is the depth of the porous solid in
the direction of flow.
P = (N RT/P - P 2 /z) (8)
Substituting expression (1) into (8) yields a relationship between the
parameter K and permeability:
KVz (9)
Since K is known from (7) and V, z, and A are known, the permeability
can be established using expression (9).
Freeze-Thaw Testirgof Coal Refuse
The freeze-thaw experiments were conducted using a refrigerator in which
the refrigeration area was at 34°F and the freezing compartment at 3°F.
To initiate the freeze—thaw test, 5 briquettes each of-lO, —20, and —60
mesh unburnt bituminous coal refuse briquettes were soaked in 34°F water
about 3 hours and then placed in the freezing compartment overnight.
The freeze-thaw test required a rigid schedule in order to make the
results meaningful. This schedule is presented in Table IX.
TABLE IX
Schedule of Freeze-Thaw Test
Time Event
9:00 a.m. Remove from freezing compartment.
Place in thawing water.
Complete Cycle 9:45 a.m. Remove from thaw water. Place
in freezing compartment.
12:45 p.m. Remove from freezing compartment.
Place in thawing water.
Complete Cycle 1:30 p.m. Remove from thawing water.
Place in freezing compartment.
4:30 p.m. Remove from freezing compartment.
Place in thawing water.
-43-
-------
TABLE IX CON’T
Schedule of Freeze-Thaw Test
Time Event
Complete Cycle 5:15 p.m. Remove from thawing water.
Place in freezing compartment.
A freeze-thaw cycle consists of one thawing operation, and one freezing
operation. After 10 of these cycles, one of each of the 3 different
kinds of briquettes was removed from the system and allowed to air dry for
1 to 2 days. The compressive strength of these samples was then determined.
This procedure was repeated at 20, 25, 30 and 35 cycles. The data
obtained was the compressive strength as a function of the number of
freeze-thaw cycles for each kind of briquette.
-44-
-------
BIBLIOGRAPHIC: Applied Technology Div.
Carbonate Bonding of Coal Refuse, EPA Grant
No. 14O1OFOA
A laboratory study of the variables
which effect carbonate bonded coal re-
fuse has been made. The carbonate bond-
ing of coal refuse consists of mixing
coal refuse with water and lime hydrate,
compacting the mixture, and reacting it
with a carbon-dioxide-rich gas to form
a coherent structure bonded by a matrix
of calcite crystals. Carbonate bonded
coal refuse can be used in road building
or as a coal refuse pile sealant to
minimize acid mine water pollution.
Four types of coal refuse were in-
vestigated. Compressive strengths of
BIBLIOGRAPHIC: Applied Technology Div.
Carbonate Bonding of Coal Refuse, EPA Grant
No. l4OlOFOA
A laboratory study of the variables
which effect carbonate bonded coal re-
fuse has been made. The carbonate bond-
ing of coal refuse consists of mixing
coal refuse with water and lime hydrate,
compacting the mixture, and reacting it
with a carbon-dioxide-rich gas to form
a coherent structure bonded by a matrix
of calcite crystals. Carbonate bonded
coal refuse can be used in road building
or as a coal refuse pile sealant to
minimize acid mine water pollution.
Four types of coal refuse were in-
vestigated. Compressive strengths of
BIBLIOGRAPHIC: Applied Technology Div.
Carbonate Bonding of Coal Refuse, EPA Grant
No. 14O1OFOA
A laboratory study of the variables
which effect carbonate bonded coal re-
fuse has been made. The carbonate bond-
ing of coal refuse consists of mixing
coal refuse with water and lime hydrate,
compacting the mixture, and reacting it
with a carbon-dioxide-rich gas to form
a coherent structure bonded by a matrix
of calcite crystals. Carbonate bonded
coal refuse can be used in road building
or as a coal refuse pile sealant to
minimize acid mine water pollution.
Four types of coal refuse were in-
vestigated. Compressive strengths of
ACCESSION NO:
Key Words
Carbonate
Bonding,
Coal Refuse,
Coal Refuse
Binder,
Refuse Pile
Sealant,
Pollution
Abatement,
Road Building
ACCESSION NO:
Key Words
Carbonate
Bonding,
Coal Refuse,
Coal Refuse
Binder,
Refuse Pile
Sealant,
Pollution
Abatement,
Road Building
ACCESSION NO:
Key Words
Carbonate
Bonding,
Coal Refuse,
Coal Refuse
Binder,
Refuse Pile
Sealant,
Pollution
Abatement,
Road Building
-------
2200 to 4400 psi were obtained for the
coal refuse investigated using up to 12
percent lime hydrate and 9 to 15 percent
water, In general, the compressive
strength of the carbonate bonded coal
refuse increases with increasing lime
hydrate content, reaction time and car-
bon dioxide concentration in the carbonate
bonding reaction gas. The air and water
permeability of carbonate bonded coal
refuse was comparable to concrete.
An approximate cost comparison be-
tween carbonate bonded coal refuse and
other construction materials indicated
that the carbonate bonding process utiliz-
ing coal refuse is the least cost means
for coal refuse pile sealing and road
building.
2200 to 4400 psi were obtained for the
coal refuse investigated using up to 12
percent lime hydrate and 9 to 15 percent
water. In general, the compressive
strength of the carbonate bonded coal
refuse increases with increasing lime
hydrate content, reaction time and car-
bon dioxide concentration in the carbonate
bonding reaction gas. The air and water
permeability of carbonate bonded coal
refuse was comparable to concrete.
An approximate cost comparison be-
tween carbonate bonded coal refuse and
other construction materials indicated
that the carbonate bonding process utiliz-
ing coal refuse is the least cost means
f or coal refuse pile sealing and road
building.
2200 to 4400 psi were obtained for the
coal refuse investigated using up to 12
percent lime hydrate and 9 to 15 percent
water. In general, the compressive
strength of the carbonate bonded coal
refuse increases with increasing lime
hydrate content, reaction time and car-
bon dioxide concentration in the carbonate
bonding reaction gas. The air and water
permeability of carbonate bonded coal
refuse was comparable to concrete.
An approximate cost comparison be-
tween carbonate bonded coal refuse and
other construction materials indicated
that the carbonate bonding process utiliz-
ing coal refuse is the least cost means
for coal refuse pile sealing and road
building
-------
1 A Ce IOiI Numbe, Subjert F,vid & GrmIPT
SELECTED WATER RESOURCES ABSTRACTS
05D INPUT TRANSACTION FORM
—
Organization
j Environmental Protection Agency
Washington, D. C.
6_JTI(ie
Carbonate Bonding of Coal Refuse
lOJAIht1 .0 r(a)
Paul J. La osa
James A. Karnavas
Eugene A. Pelczarski
Project Dos ignation
14010 FOA
—
No t t
22 Citation
— - Environmental Protection Agency
14010 FOA 02/71 Washington, D. C.
23 Descriptors (Starred First)
• Acid mine water*, coal mine waste*, calcium carbonate*,
sealants*, carbon dioxide, pollution abatement, road
construction
25 Ideflt liar. (Starred First)
Carbonate bonding*, coal refuse binder*
27j ’ ’ A laboratory study of the variables which affect the properties of the car-
bonate bonded coal refuse has been made. The carbonate bonding process utilizing
coal refuse as a fill material consists of mixing coal refuse with water and lime
hydrate, compacting the mixture, and reacting it with a carbon-dioxide-rich gas to
form a coherent structure bonded by a matrix of calcite crystals. The resulting
carbonate bonded coal refuse can be used in road building or as a coal refuse pile
sealant to minimize acid mine water pollution.
Four types of coal refuse were investigated--a relatively unoxidized and
highly oxidized bituminous coal refuse and a relatively unoxidized and highly
oxidized anthracite coal refuse. It was found that compressive strengths of 2200 to
4400 psi were obtained for the four types of coal refuse investigated using up to 12
percent lime hydrate and 9 to 15 percent water. In general, the compressive strength
of the carbonate bonded coal refuse increases with’ increasing lime hydrate content,
reaction time and carbon dioxide concentration in the carbonate bonding reaction gas.
The air and water permeability of carbonate bonded coal refuse was found
to be comparable to concrete. An approximate cost comparison between carbonate
bonded coal refuse and other construction materials and techniques indicated that
the carbonate bonding process utilizing coal refuse is the least cost means
available for coal refuse pile sealing and road building.
Abstrac(Of
Paul J. LaRosa j lfl (ltU(jon Black, Sivalls & Bryson, Inc.
w IOZ CV. JUL.? ..•
w*$Ic
IgNO tOs WATER E$OIJRC S SCIENTIFIC INFORMATION CENrER
U. S. OEPARIMENT OF TNt INTERIOR
WASHINGToN. 0. . aoamo
• SPel 59 5 5 .35 5.3 )5
------- |