EPA-670/2 73 093
NOVEMBER 1973
Environmental Protection Technology Series
An Appraisal of
Neutralization Processes to
Treat Coal Mine Drainage
Office of Research and DevBlopment
U.S. Environmental Protection Agency
Washington. D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate furtber
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
.. 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 instrumentati.on, 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 Office of Research and
Development, EPA, 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 counnercial products constitute
endorsement or reconmtendation for use.
-------
EPA-670/2-73-093
November, 1973
AN APPRAISAL OF NEUTRALIZATION PROCESSES
TO TREAT COAL MINE DRAINAGE
By
Harold L. Lovell
Pennsylvania State University
University Park, Pennsylvania 16802
Project 14010 EFN
Program Element 1BB040
Project Officer
Ronald D. Hill
Mine Drainage Pollution Control Activities
National Environmental Research Center
Cincinnati, Ohio 45268
Prepared fpr
Office of Research and Demonstration
U.S. Environmental Protection Agency
Washington. D.C. 20460
For Mia by the Superintendent of Document*, U.S. QOYwnment Printing Office, Washington, D.C. 20402- Price $3.60
-------
ABSTRACT
The treatment of four different coal mine drainage waters was studied
utilizing neutralization processes. Each of the unit operations, from
water collection to sludge disposal, was considered by drainage
reaction with eight different reagents. The process variations
required by each reagent was possible with the speciallydesigned
500,000 gpd facility.
Limestone has the lowest delivered cost per neutralization equivalent
and may be utilized with waters containing up to 500 mg/i iron II,
which includes most coal mine drainages. A rapid settling, dewater
able sludge is formed. Lime can treat any drainage with efficiency,
but requires control to avoid excess consumpticn by reaction with
aluminum and magnesium. Lime forms a voluminous sludge which is
difficult to handle. Iron II oxidation may be accomplished with air
in alkaline systems or via autotrophic bacteria in acidic systems.
Unitized sludge recycle process forms dense, dewaterable sludge from
highly mineralized waters.
Thickeners are preferred to settling lagoons for sludge separations
in larger plants. Sludge devatered by drying basin, filtration or
centrifugation requires the least landfill disposal area. Alter-
natively, nearby abandoned deep or surface mines may be advantage-
ously used for settled sludge disposal.
This report was submitted in fulfillment of Grant Number 114010 EFN,
under the partial sponsorship of the U.S. Environmental Protection
Agency, and The Department of Environmental Pesources, Commonwealth
of Pennsylvania. Work was completed as of October 1971. Work was
performed by Pennsylvania State University.
11
-------
CONTENTS
Page
Ahstr t
List of Figures V
List of Tables ix
cknowledgemeflt 5 xiii
Sect ions
I ConclUSiOflS 1
11 RecommendatiOns 6
iii Introduction 7
IV Description of Pilot Plant 23
V Objectives 39
\rI Treatment of Coal Mine Drainage with 41
Hydrated Lime
rjj Treatment of Coal Mine Drainage with 45
Calcined Lime
VIII Treatment of Coal Mine Drainage to Produce 48
High Density Sludge by Recycle
ix Treatment of Coal Mine Drainage with Caustic 60
Soda
x Treatment of Coal Mine Drainage with Soda Ash 63
xi Treatment of Coal Mine Drainage with 67
DolomitiC Reagents
xii Treatment of Coal Mine Drainage with 72
Limestone
xiii Biochemical Oxidation of Iron II 99
XIV Sludge Control Ifl Coal Mine Drainage 116
Treatment
iii
-------
CONTENTS (Continued)
Sections Page
XV References 223
XVI Glossary 232
XVII Appendices 235
iv
-------
FIGURES
Page
1 Relative Reaction Rates for Various Alkali 17
Reagents
2 pH Range Resulting from Various Alkali Reagents 17
3 Simplified Schematic Drawing of Treatment Plant 25
4 Aerial Photograph of the Experimental Mine 31
Drainage Treatment Facility
5 Hydraulic Profile of the Experimental Mine 32
Drainage Treatment Facility
6 External Unit Operations of the Experimental Mine 33
Drainage Treatment Facility
7 External Piping Arrangement of the Experimental 34
Mine Drainage Treatment Facility
8 Retention Time Studies - Oxidation Tanks 36
9 Introduction of Lime Slurry into Flash Mixer 37
10 Flow Sheet for Producing High Density Sludge 50
11 Operational Data from Densator Recycle Studies 52
12 Mine Drainage Recycle Sludge 55
13 Steady State Densator Flow Rates for Various 57
Recycle Operating Conditions
14 A Limestone-Iron Neutralization System 77
15 Limestone Reactor Retention Time 86
16 Distribution of Power Requirements for Limestone 87
Reactor
17 Alternate Unit Operations for Limestone 91
Neutralization
V
-------
FIGURES (Continued)
No. Page
18 Biochemical Iron Oxidation - Limestone 93
Neutralization System
19 Deposit on Surface Biochemical Reactor 101
20 Microphotograph of Culture of Ferrobacillu.s 104
ferrooxidans Employed to Oxidize Iron II
21 Electron Microphotograph of Ferrobacillus Cell 105
22 Iron II Oxidation Rates with Batch Inoculation 111
of Mine Drainage
23 Surface Biochemical Reactor Performance 113
24 Sludge Settling Rate - Solid-Fluid Separation 119
Influent Proctor No. I Water
25 Sludge Settling Rate - Solid-Fluid Separation 120
Influent Proctor No. 2 Water
26 Sizing and Thickener Cost Estimation Data 141
27 Settling Lagoon Weir Overflow Showing Sludge 144
Settling Pattern
28 Settling Lagoon Calculated Retention Time 146
29 Sludge Deposition in Settling Lagoon 148
30 Mine Water Sludge Coating on Precoat Druni Filter ioi
31 Cutting Sludge Cake from Vacuurt Filter 161
32 Test Leaf Filtration Equipment 163
33 Test Leaf Filter for Top Loading Process 163
34 Sleeve for Top Loading Filtration 163
35 Solids Filtration Rate as a Function of Drum Speed 167
36 Variation of Filtrate Flow Rate vs. % Solids in 168
Filter Feed at Constant Solids Filtration Rate
vi
-------
FIGURES (Continued)
No. Page
37 Relationship of Corresponding Solids and 169
Liquid Filtration Rates at Various Levels of
Solids Content in the Filter Feed Slurry
38 Sludge Production Rates and Filter Area 185
Requirements
39 Approximate Installed Costs of Vacuum Drum 186
Precoat Filters (1971)
40 Costs of Vacuum Drum Precoat Filtration as a 188
Function of Solids Filtration Rate Capacity
41 Sewage Sludge Drainage Rates 191
42 Pond Liquid Level Dropping Rate vs. Pond Bottom 193
Drainage Rate Capacity
43 Precipitation and Reservoir Evaporation Data 194
for Western Pennsylvania and Eastern Ohio
44 Drainometer Test Unit 196
45 Summary of Sludge-Drainometer Dewatering Rates 199
46 Photograph of Sludge Drying Basin 202
47 Newly Constructed Sludge Drying Basin Showing 202
Sludge Influent
48 Introduction of Settled Sludge into Sludge Drying 202
Basin
49 Sludge in Dewatering Process 203
50 Dewatered - Drying Sludge 203
51 Dried Sludge in Covered Drying Basin 203
52 Removing Dewatered Sludge from Drying Basin 20.8
53 Bone-Dry Sludge Showing Minimal Penetration of 208
S and
vii
-------
FIGURES (Continued)
No. Page
54 L)ewatered Sludge Drying in Land Disposal Area 208
55 Cubic Foot of Dewatered Sludge 209
56 Frozen Dewatered Sludge 209
57 Dried Sludge After Freezing 209
58 Sludge Volume and Drainage Rates Sludge 212
Drying Bed
59 Settling Rate of Limestone-Produced Sludge 213
Removed from Bottom of Settling Lagoon
60 Dewatering Rates from Frozen Sludge 214
61 Calculator for Determining Drying Basin 217
Loading Rates and Sludge Volumes
62 Drainage Basin Area Calculator 218
63 Sludge Drying Basin Area Calculator 219
64 Suggested Layout for Dual Sludge Drainage 221
Basin System
65 Volume Reduction in Dewatered Sludge When Dried 222
viii
-------
TABLES
Table Page
1 Mine Drainage Treatment Reagent Data 12
2 Reagent Cost Data 13
3 Analyses of Reagents Used During Operating 14
Program
4 Regression Coefficients of the Linear Response 20
Surface for the Related Response of Iron II
Oxidation Rates
5 Basis of Design of the Hollywood Facility 24
6 Sources of Mine Drainage, Estimated Volumes, 26
Constituents and Characteristics
7 Treatment Plant Design Conditions 27
8 Major Treatment Plans 28
9 Construction Costs for the Experimental Mine 30
Drainage Facility
10 Summary of pH Control Chart Variation - 46
Calcined Lime Treatment - Process 10
11 Densator Recycle Test - Operating Parameters 56
12 Densator Recycle Sludge Analyses 58
13 Solubility of Calcium Carbonate in Water in 75
Contact with Air (16°C)
14 Solubility of Calcium Sulfate Di-Hydrate 76
15 Solubilities in the Quaternary System 76
CaCO 3 + MgSOi - CaSU + MgCO 3
16 Potentialities for Limestone Treatment of 81
Mine Drainage Waters
17 Characteristics of Limestone Sludge from 84
Settling Lagoon
ix
-------
TABLES (Continued)
Table Page
18 Neutralization of CMD by Limestone in the 95
Rotary Reactor
19 Estimation of Costs - Limestone Treatment 97
Process
20 Chemical Analysis of Sludge from Biochemical 103
Oxidation of Coal Mine Drainage
21 Culture Media for Iron Oxidizing Bacteria 108
22 Growth Stimulation of Iron Oxidizing Bacteria 110
in Natural Mine Drainage
23 Detailed Conditions of Batch Inoculation 112
24 Summary of Sludge Settling Data from 122
Hollywood Plant Tests
25 Analyses of Mine Drainage Waters Evaluated 125
with SUPERFLOC Reagents
26 Sludge Settling Rates with SUPERFLOC 126
Polyelectrolytes. Water Origin - Bennf tts
Branch
27 Sludge Settling Rates with SUPERFLOC 127
Polyelectrolytes. Water Origin - Proctor
No. 1
28 Sludge Settling Rates with SUPERFLOC 128
Polyelectrolytes. Water Origin -
Proctor No. 2
29 Sludge Settling Rates with DECOLYTE 129
Polyelectrolytes
30 Screening Tests with ATLASEP Flocculants 131
31 Sludge Settling Rates with ATLASEP Flocculants 131
32 Thickener Flow Rate Parameters 133
x
-------
TABLES (Continued)
Table Page
33 Sludge Rate Parameters for 25-Foot Thickener 134
34 Typical Thickener Performance Data 136
35 Observed Thickener Performance - Sizing Data 137
36 Suggested Optimum Design Parameters for 140
Thickener Operation
37 Approximate Thickener Equipment Costs 140
38 Daily Thickener Operation Costs 142
39 Typical Settling Lagoon Performance - 152
Bennetts Branch Water
40 Typical Settling Lagoon Performance - 152
Tyler Run Water
41 Typical Settling Lagoon Performance - 153
Proctor No. 2 Water
42 Typical Settling Lagoon Performance - 153
Proctor No. 1 Water
43 Settling Lagoon Area and Volume Requirements 155
44 Settling Lagoon Excavation Cost Estimates 156
45 Comparison of Thickener and Settling Lagoon 157
Sizes and Costs for Various CMD Capacities
46 Filtration Rates for Several Varieties of 159
Sludges
47 Operational Parameters for Filtration Unit 162
Operation - 1 -Lollywood Pilot Plant
48 Filtration Test Data 165
49 Hollywood Plant Vacuum Filter Cycle 166
Cal ibrat ions
xi
-------
TABLES (Continued)
Table Page
SO Chemical Analyses of Sludges Used in 171
Filtration and Drainometer Studies
51 Mine Water Neutralized Sludge Solids 172
Filtration Rates
52 Vacuum Filter Leaf Tests of Sludge from 173
Limestone Neutralized Coal Mine Drainage
53 Filtration Rate Test Data for Recycle Sludge 175
54 Summary of Tests on Pilot Plant Rotary Vacuum 177
Filter
55 Chemical Analysis of Filter Cake from CMD Sludge 178
56 Summary of Filtration Leaf Tests Using Flocculants 180
57 Pilot Plant Precoat Rotary Vacuum Filter Tests 181
with Strongly Flocculated Feed Slurries
58 Recommended Vacuum Filter Design Parameters 182
59 Estimated Filtration Costs for Various Levels 184
of Solids Filtration Rates
60 Operating Parameters for the Merco Bowl 189
Centrifuge
61 Summary of Drainometer Tests 198
62 Effect of Drainage Rate and Influent Solids 200
Content on Drying Bed Size
63 Operating Parameter Limits of Drying Beds 204
64 Sludge Drying Basin Evaluation 210
65 Sludge Drying Bed Design Parameters 216
66 Typical Sludge Drying Bed Operational Parameters 220
xii
-------
AC KNOWLEDGEMENTS
The activities surrounding the Hollywood Facility program were exten-
sive and required the cooperation of many groups and individuals.
The role of each was most important and is gratefully acknowledged:
Sponsors : The former Pennsylvania Coal Research Board - Department
of Mines and Mineral Industries: Honorable H. Beecher
Charmbury, Secretary; Dr. D.R. Maneval, Project Officer,
Cooperating Agency - Pennsylvania Sanitary Water Board.
Now: Pennsylvania Department of Environmental Resources,
Honorable Maurice K. Goddard, Secretary: Donald E. Fowler,
Project Officer.
United States Environmental Protection Agency. Water
Quality Office - Ronald D. Hill, Project Officer.
Dr. Raymond Thacker.
Site : Provided by New Shawmut Mining Company, Mr. Anthony Palumbo,
President.
The University:
Trustees : C. Albert Shoemaker, Walter W. Patchell.
Adminigtrat&on : Robert A. Patterson, Ralph E. Zilly, Dr. E.F,
Osborn, Dr. Harry Zook, Mr. Ralph Montgomery.
Physical Plant : Walter Wiegand, John Miller, John Jellison.
Institute for Research Land and Water : Dr. John C. Frey, Dr. Bruce E.
Jones.
College of Earth and Mineral Sciences:
Administration : Dr. CharlesL. Hosler, Dean; and Dr. M.E,
Bell, Dr. H.B. Charmbury, Assistant Deans; Dr. William
Spackman, Director Coal Research Section.
Project Staff : Dr. Harold L. Lovell, Project Director.
Research Staff : Fred Leitzel, David Kaelin, W. Lee Duguay,
Donald Gummo, Robert I. Lachman, Gerald McHugh, Terry E.
Stauffer, Richard C. Weller.
Analytical : Elaine M. Heilman, Norman H. Suhr, Joseph B.
Bodkin, Jessie W. Tieman, Margaret Miskovsky, Robert K.
Brown, David R. Maneval, Jr., Robert E. Raver.
Secretarial : Loretta Marley, Lynn Kapeghian, Eileen Grimes,
Elaine Horgas, Janice Johnston, Sharon Luce, Paula May,
Judy Schellenberger, Karen Copenhaver.
xiii
-------
Records : Jane Dolsen, Vesta Immel, Patty Podwysocki .,
Margaret Whiting.
Drafting and Photography : Donald 1. Krebs, Ruth M. Krebs,
Linda M. Senefeld.
Shop : Torsten V. Bjalme, George E. Reisinger, Robert Rodgers,
John C. Rudy.
Plant Staff : Richard Thomas, Chief of Operations; James
Netterbiade, Mechanic and Plant Operator.
Plant Operators : Edward Schollenberger, Wesley Smeal,
Barry Walker.
Technicians and Maintenance : Fred Graham, Dennis Guido,
James Gummo, Rea Gununo, Bruce Mattern, Lee Mattern,
William Scheffley, Charles 0. Wilson, Gardner Wissinger.
Technical Advisors : Dr. Robert Stone - Microbiology, Prof.
Rupert Kountz - Civil Engineering, Thomas Rucinsky -
Electron Microscopy, and Dr. L.E. Casida - Optical Micros-
copy.
Technical Consultants : Universal Technical, Tnc. - R.J. Reese,
P.E. - Design and Construction Liason; D.C. McLean, P. 13. -
Data Evaluation; Gannett, Fleming, Corddry, and Carpender,
Inc. - Bruce Gerber, Anthony Miorin, Kenneth Meyers and
Russel Klingensmith - Detail Design and Contracts.
xiv
-------
SECTION I
CONCLUSIONS
Any of the coal mine drainages (CMD) considered in these studies may
be treated by neutralization processes to meet State specifications
in Pennsylvania: e.g., maintain a p11 between 6.0 and 9.0, have less
than 7 mg/i iron, and have no titratable acidity. Most of the alumi-
num content of the waters is also removed and under some conditions,
there is a reduction in sulfate, magnesium, and manganese content.
The product waters usually have no appreciable decrease in total
dissolved solids and contain increased cation concentrations of the
neutralizing reagent. The treated waters may be utilized where a
hard water is suitable. They support aquatic life and are much
less pernicious than the original mine drainages.
The treatment may be accomplished using any of the reagents studied -
provided adequate process conditions are established. However, the
efficiency of reagent utilization, the unit treatment cost, the
required processing steps and the quality of the water product will
vary widely among the several reagents.
Each treatment plant must be individually designed. Neutralization
reagent selection, based on feed water quality, flow levels, and
site conditions, is logically the initial decision. Reagent selection
should also be justified by its process compatibility, reliable availa-
bility, transportation considerations, and cost at the proposed plant
location. Continuous treatment capability must be insured.
The neutralization reagent determines the required unit operations
and many of their parameters, thus the basic plant design.
The reagent limestone has the lowest delivered cost of those con-
sidered. This value approximates 1.92 x lO_6 cents per gallon per
mg per liter acidity which is less than 40 percent of the next
higher cost reagent, calcined lime. Excepting dolomite, limestone
has the slowest reactivity with mine drainage and is most subject
in its response to physical parameters as particle size and hardness.
The physical parameters establish the mode of limestone utilization
and ultimately its efficiency of use. The CMD reaction with lime-
stone forms bicarbonates, carbonates and carbonic acid whose concen-
trations must be controlled to allow the p1-1 of the water to rise.
This latter response may limit selection of limestone in respect
to the iron II concentration of the mine water. Waters containing
up to 100 mg/l Fe II may be effectively treated using limestone
without separate iron oxidation in the simplest system. Waters
with Fe II contents between 100 and 500 mg/i also may be effectively
1
-------
treated with limestone but require prior iron oxidation. At still
higher iron II levels, oxidation rates may limit pragmatic treatment.
Preneutraljzatjon oxidation iron II extends the limits of limestone
application, as does postlimestone neutralization additions of hy-
drated lime slurry. Most known mine waters may be effectively treated
with limestone. Overtreatment using limestone is impossible. A dense,
rapid-settling, dewaterable sludge results. Reagent usage efficiency
can exceed 80 percent.
Hydrated and calcined lime are comparable reagents and are the least
expensive reagents capable of treating any coal mine drainage in a
feasible system. They react rapidly. When these reagents are
properly reacted and controlled, nearly perfect reagent utilization
efficiency is attainable. However, difficulties in control of reagent
addition increase with the acidity of the water. The pH of the lime-
treated water defines lime consumption and iron II oxidation rates.
The pH-lime consumption relationship in these waters is complex since
further reactions with aluminum and magnesium may occur. Overtreatment
easily results using calcium hydroxide and may lead to unacceptable
effluents and very high reagent costs. These reagents usually form
a voluminous, gelatinous slurry of the impurities which settles to a
low density sludge. These sludges gel on aging. They have poor
handling and dewatering characteristics.
Hydrated lime may be reacted as a solid or made into an aqueous
suspension, whose concentration is a process variable. The latter
form is preferable.
Pebble-sized calcined lime should be slaked prior to treatment with
the water. Although less expensive per unit neutralization equivalent,
calcined lime is recommended only when consumption levels exceed 25
tons/week. Lower levels make slaking control difficult due to
available equipment size.
Dolomite and its manufactured products perform similarly to high cal-
cium reagents, except they are usually more costly and less available.
They exhibit no apparent advantage in sludge volume or characteristics,
except with excessively polluted waters. In that case, calcium sul-
fate deposition within the process system may be more controllable,
but at the sacrifice of a less acceptable process effluent. Dolomite
is less reactive than limestone, primarily due to its greater hard-
ness. This low reactivity limits its application to waters which
are very lightly mineralized. These reagents may lead to undesirable
increases in magnesium content of the treatment effluents.
Sodium hydroxide is utilized as a solution and offers mechanical
advantages. It is the most reactive, expensive and dangerous reagent
to utilize. Its cost can vary widely with reagent physical form,
2
-------
with treatment plant location, and with transportation considerations.
Sodium hydroxide is even more sensitive to the control difficulties
cited for calcium hydroxide slurries and is also capable of near-
perfect use efficiency. The resulting sludge characteristics are
usually equivalent, but may be less desirable than those formed with
calcium hydroxide. The sludge may contain less calcium sulfate and
the plant effluent may be considerably different than with lime
reagents. Sodium hydroxide is especially desirable as a short time,
emergency control reagent to prevent unacceptable or accidental dis-
charges.
Sodium carbonate, in the same neutralization cost range as solid
sodium hydroxide, is utilized as a solution prepared prior to reaction.
Although of high reactivity, the ensuing equilibria involving carbon-
ates, bicarbonates and carbonic acid require decarbonation to allow
the pH of the waters to rise and reactions to be completed. This
reaction complication, which is similar to that exhibited by naturally-
alkaline coal mine drainage, when added to those difficulties cited
for calcium hydroxide slurries, makes reaction control in the use of
this reagent difficult. At least 100 percent excess reagent con-
sumption can be anticipated. In properly reacted systems, the sludge
characteristic may be somewhat superior to hydroxide reagents.
The mode of alkali addition and its reaction with the feed mine
drainage waters, as established by the reaction control procedure
(as a pH probe and its location in the system), are prime factors
in determining reagent utilization efficiency.
A process was studied which recycles the impurity sludge. It pro-
vides an effective procedure to treat coal mine drainage by developing
a crystalline, dense sludge. This process reduces sludge volume.
The procedure is most applicable to highly-mineralized waters having
a high iron II content. Sludge disposal problems are usually ex-
tensive for such waters. The process holds major advantage in situ-
ations where minimum plant space is available. The final sludge
product, resulting with lime reagent, has densities in excess of
16 percent solids by weight. The sludge is desirably responsive
to dewatering by vacuum filtration or sludge drying basin.
Separate air oxidation of iron II (at levels above 100 mg/i) is
nearly essential with most of these reagent systems. It is prefera-
bly performed in an alkaline system. Procedures to estimate the
rate of this oxidation are given and show that the oxidation rate
dependency involves pH, temperature, and concentrations of iron II
and aluminum. Inadequate retention times for this oxidation are
more critical than oxygen transfer rates which are readily attained.
The characteristics of this reaction, especially the changing rate
with iron II concentration, requires its careful integration within
a plant design.
3
-------
Biochemical oxidation of iron II in an acid system, prior to neutral-
ization, is feasible and extends the range of waters to which limestone
as a reagent may be applied. In addition to the reaction parameters
cited for air oxidation, the biochemical system requires consideration
of cell concentrations and growth rates as controlled by cell metabo-
lism and interferences. The surface biochemical reactor permits oxi-
dation rates (in thousands of mg/i/br) which are many fold faster
than deep vat systems. The surface reactor system is easy to control
and has little or no operational costs while the deep vat system has
undesirable reaction rate limitations and is more subject to inter-
ferences. Further study of the surface biochemical reaction regarding
temperature and sludge deposition is desirable.
The formation of large volumes of low density impurity sludges may
be minimized to realistic levels during the initial water treatment.
Limestone can be used to treat waters with iron II concentrations
of less than 500 mg/l while more severely polluted waters may be
effectively treated using lime in a sludge recycle system to yield
more desirable sludge levels.
Sludge settling rates are about 2 ft/hr but vary widely with feed
water quality, reagent choice and mode of utilization. These settling
rates may be increased by addition of high molecular weight, polymeric,
weakly anionic flocculants.
Either a settling lagoon or thickener may be satisfactorily utilized
for solid-fluid separation of impurity sludges. As plant capacities
exceed 300,000 gpd, thickeners become increasingly favored from both
operational and economic bases.
Settling lagoons become filled with sludge. It is difficult and
costly to remove the settled sludge. Procedures to implement sludge
removal from lagoons or obviate it, increase costs favoring the use
of thickeners.
Observations indicate that bottom-fed sludge solid-fluid separation
techniques utilizing limestone-produced sludges can drastically
reduce size requirements for settling lagoons and thickeners.
Secondary sludge dewatering is usually indicated. In some cases,
economics favors discarding primary settled sludge in abandoned deep
mines or surface strip pits.
Secondary sludge dewatering may be accomplished by sludge drying
basin, filtration or centrifugation to form a product handleable
as a solid containing 15-50 weight percent solids. The sludge dry-
ing basin has negligible operating costs; centrifugation is versatile
in application while filtration has performance advantages.
4
-------
Effectiveness of these unit operations increases with feed solids
content. Vacuum precoat filtration is economically feasible only
when solids production rates exceed an attainable 100 lbs dry solids!
ft 2 /day. Filtration may be the only approach when the filtrate has
a specified use.
Top loading filtration techniques may be less expensive since thicker
cakes are achieved prior to blinding.
Freezing of wet, dewatered sludge modifies its physical structure
to accelerate further dewatering and volume reduction upon thawing.
Movement of dewatered sludge to a land-fill disposal area and con-
tinued outdoor drying can result in a further volume reduction (two-
thirds). This provides minimal disposal requirements. The dried
sludge may be further mechanically compacted, graded and seeded.
5
-------
SECTION II
RECOMMENDATI ONS
The biochemical-iron II oxidation-limestone neutralization treatment
system should be studied in continuous, year-round operation to
improve the reliability and extend the applicability of the process.
The areas to be emphasized:
A. Biochemical-iron II oxidation
1) Temperature control under winter conditions
2) Seek rates in deep vats equivalent to the surface-type reactor
3) Cleaning surface-type reactor media and/or minimize sludge
deposition.
B. Further quantify performance and seek increased capacity of rotary
limestone reactor.
C. Bottom-feed solid-fluid separations to confirm potential reduc-
tions in thickener or settling lagoon areas indicated in these
studies.
0. Top-feed filtration of impurity sludges.
E. Evaluate changes in sludge drying basin design to improve drainage
rates and to establish optimum usage cycles.
Study treatment of extremely mineralized mine waters with sulfate
contents over 3000 mg/i, seeking to minimize gypsum deposition by
considering the use of mixed reagents (magnesium and sodium).
6
-------
SECTION III
INTRODUCTION
This study concerns treatment of coal mine drainage (CMD) waters by
neutralization processes. Such drainage creates an unacceptable con-
dition in receiving streams to an ecologically-conscious society.
Major portions of the river systems of coalproducing states, especially
those with long histories of coal mining, as Pennsylvania and West
Virginia, are extensively polluted. There are currently no feasible
means to stop this water contamination, although a number of actions
are possible to reduce the volume and extent of this pollution loading.
Pennsylvania and other states have, in recent years, enacted regu-
lations which establish specifications for minimum water quality that
may be discharged from operating coal mines. The goal is to stop
the environmental degradation at the present levels and alleviate
the pollution through continuing programs. In most cases the existing
specifications can be met only by treatment of the water discharged
from coal mines. In some instances, state action programs have been
initiated to correct such discharges from abandoned coal mining
operations.
Although concern and research related to such treatment has been
evidenced for nearly fifty years, there has been inadequate practical
engineering experience and knowledge which can be efficiently,
effectively, and economically applied. The operators of existing
facilities believe their operational costs are excessive. Attempts
by industry and government to design and operate treatment facilities
to meet the regulations have frequently resulted in frustration and
have developed many questions.
The Pennsylvania Coal Research Board in December, 1965, requested The
Pennsylvania State University to acquire the necessary data to per-
mit effective, economical treatment. This information was to be
developed from a small, but full-scale treatment facility utilizing
the prevalent neutralization process concepts. This report presents
the observations from that operation.
The research program, which was subsequently cosponsored by the
Environmental Protection Agency, encompassed design and construction
of a mine water treatment plant capable of processing 500,000 gallons
of water per day. The plant was located at Hollywood, Pennsylvania
(Clearfield County). Supporting laboratory studies were made at the
University Park, Pennsylvania campus of The University. The experi-
mental plant would be suitable for obtaining cost data and amenable
7
-------
to experimentation with innovations. The results of plant operation
were to indicate performance and costs, including an evaluation of
process elements.
The plant design was based upon treatment of the four available
drainage waters, each with different characteristics. It provided
capability to utilize any feasible alkali. The unit operations in-
cluded ranged from water collection stations through the treatment
steps to sludge disposal. The design versatility permitted inter-
connection between these unit operations to consider several processes
including: (1) The Yellowboy Process (lime-air oxidation) with
provision for solid-fluid separation by either thickener or settling
lagoon, (2) Biochemical iron oxidation-limestone neutralization process
with the oxidation achieved in either deep tanks or in a surface-
type reactor, and (3) Recycle sludge process. Dewatering procedures
applicable to the settled sludge sought to evaluate potential methods
to control this troublesome and costly aspect.
Neutralization processes are designed to produce a water which will
have minimum detrimental effects on the bioecological system. The
treated waters must not detract from the appearance of receiving
streams and be acceptable for navigational, recreational and industrial
purposes. The development of potable-quality water for human con-
sumption has not been a primary objective. These processes adjust
the p 11 of the waters to levels above 6.0. This treatment creates
conditions which form insoluble compounds from the iron, aluminum
and sometimes calcium sulfate contained in the drainage. The in-
solubles must be removed from the treated water , The quality criteria
for such discharged waters established by the Sanitary Water Board
of the Commonwealth of Pennsylvania in 1964 have served as general
guides. These require that discharges to the streams of the Common-
wealth have: (1) a pH between 6.0 and 9.0, (2) iron content less
than 7 mg/liter, and (3) no titratable acidity. The Hollywood plant
was operated to meet these specifications.
It appears these criteria should result by reacting any inorganic
alkali providing it does not introduce other inimical components.
The volume of such waters which may require treatment reaches millions
of gallons daily, thus pragmatic considerations also necessitate the
utilization of processes having minimal requirements for capital
equipment, labor, process control, and operating costs. They must
also offer a high degree of reliability.
These processes can be conceived as cation exchange reactions in
which the hydrogen, iron, and aluminum ion species in the drainage
waters are replaced by calcium, magnesium, or sodium ions. The term
cation exchange reactions describes the replacement of the undesirable
8
-------
cation components of the drainage waters by more acceptable soluble
ions. The detrimental components are converted to insoluble sub-
stances. In no way does the term refer to processes labelled ion
exchange which involve a second phase, insoluble substrate. The
water products resulting from neutralization treatment are not
significantly improved in total dissolved solids content.
Other processes, described as demineralization procedures, do
result in higher quality water products (low dissolved solids).
They are more expensive and of limited application.
Although the number of neutralization processes that can be used to
achieve these minimal quality standards is limited, the factors in-
volved in choosing a process are more complex than commonly conceived.
Once the characteristics of the product waters are defined, the pro-
cess choice depends upon the loading of the drainage to be treated,
its volume, logistical problems of the water sources and the treatment
facility as relates to each potential reagent. The chemistry of the
reactions defines the engineering criteria. The cost and availability
of the various reagents establishes the pragmatic choice. The study
seeks an insight into these various determining factors.
Since waters dissolve various substances as they flow through coal
measures, the characteristics of the drainage differs widely with
strata mineralogy. The principal mineralization results from the
soluble oxidation products of the indigenous mineral pyrite. The
components of concern include the following cations: hydrogen,
ferrous, ferric, and aluminum. Sulfate is the predominating anion.
The species in which these ions exist in coal mine drainage waters
probably vary but they have not been fully defined. Furthermore,
under the in situ conditions, these components are undergoing
continuing internal reactions. These reactions may be enhanced
biochemically or by their contact with other strata, adding ions
such as calcium, magnesium, sodium, potassium, carbonate, bicarbonate,
and, to a lesser degree, manganese and chloride as well as a host
of other ions at trace level concentrations. The many interactions
and equilibria which exist in mineralized waters or may be developed
with the introduction of inorganic alkalies create systems that may
be described as buffered. The waters considered in this study,
whose analyses are detailed in Appendix A, are reasonably typical of
coal mine drainage Classes I and 11 ia They represent the preponderance
of coal mine drainages.
The generalized reactions which occur in the desired cation exchange
processes are of two types:
9
-------
1) Neutralization - The available hydrogen ions in the drainage
waters react with negatively-charged radicals of the reagents to
form weakly-ionized species, such as water. In this chemical sys-
tem, the solubility products of the iron and aluminum ionic species
are exceeded, thus reducing their solution concentrations.
2) Oxidation of iron II to iron III - The solubility product of
ferrous hydroxide is much larger than that of arric hydroxide. With
the predominance of the ferrous species in most drainage waters, the
provision of a greater hydroxyl concentration is necessary than required
for ferric systems to achieve acceptable levels of iron concentration.
The final p 1 - I of waters so treated may exceed the criterion for hydrogen
ion concentrations. During iron iT oxidation, additional acid species
form which represent an unacceptable pollution capability. Oxida-
tion of the iron to the higher valence during processing obviates
these problems. The iron II oxidation with oxygen from air tends
to proceed after neutralization. The oxidation rate increases with
pH at levels above 4.0.
NEUTRALIZATION
In these neutralization reactions, the hydroxyl ion is really the
only ion species feasible to form a weakly ionized, polar compound
with the hydrogen ion. The carbonate ion will react with the hydrogen
ion to form carbonic acid, which will dissociate to form water and
dissolved carbon dioxide. Because of the limited solubility at
ambient temperatures and pressures, the carbon dioxide can be removed
from the water. The carbonates also form insoluble compounds with
iron and aluminum species. Thus, the two anions, hydroxide and car-
bonate, are the fundamental reagent constituents in the treatment of
coal mine drainage waters. They are available as compounds of cal-
cium, magnesium, sodium, and ammonia at costs feasible for these
processes. Other cations form hydroxides and carbonates, but their
cost and availability eliminate them from consideration.
Ecological principles usually preclude the use of ammonium hydroxide
since the soluble ammonium ion serves as a nutrient for plant and
animal organisms whose growth in receiving streams is undesirable.
The physiological action of magnesium ion as a cathartic, especially
when associated with sulfate ion, makes its introduction into streams
undesirable ,
The reagents which may appropriately be considered for achieving the
cation exchange reactions necessary for neutralization of mine waters
are limited. These include: calcium carbonate (high calcium lime-
stone), calcium oxide (calcined, quick, or pebble lime), calcium
hydroxide (hydrated lime), calcium carbonate-magnesium carbonate
10
-------
(dolomite or dolomitic limestone), calcium oxide-magnesium oxide
(burnt or calcined dolomite), calcium hydroxide-magnesium oxide
(hydrated dolomite or partially hydrated dolomite), calcium hydroxide-
magnesium hydroxide (pressure hydrated dolomite), sodium carbonate
(soda ash), and sodium hydroxide (caustic soda). The oxides of cal-
cium and magnesium meet the requirements for reagent anions (e.g.
they function as hydroxides) since as anhydrides they react with
water to form hydroxides.
Pertinent data regarding these reagents are listed in Table 1; while
their costs, as experienced during the project studies, are cited
in Table 2. The chemical and particle size analyses of the reagents
used in these studies are given in Table 3.
Many other considerations become basic in choosing among these reagents.
In planning a treatment system, the selection of a reagent would be
the initial decision since this information forms the basis of the
processes, plant design detail, and economics. The selection is
established by the drainage water analyses and volume, plant loca-
tion and the reagent performance potential. These studies were
designed to provide, from operating experience, the criteria for the
reagent performance potential. Each of the following factors should
also be optimized for the reagent: source geographical distribution,
availability, transportation, handleability, cost, reactivity,
chemical and physical properties and their mechanistic action within
the treatment system as this affects the quality of the final water
product and the characteristics of the impurity sludges.
High calcium and dolomitic limestone are naturally-occurring mineral
products. The remaining reagents are all manufactured products
which partially accounts for their higher cost. Quality and re-
activity of high calcium and dolomitic limestone and their products
vary widely with origin and manufacturing process 81 . Calcined
materials contain five percent or more inert constituents which
effect costs and processes. The differences in chemistry between
dolomite and high calcium materials must also be considered. The
physical form of caustic soda 95 and soda ash (solid, flake, solution,
etc.) selected must be carefully evaluated. There are major cost
differences and they have an impact in determining reaction conditions,
ambient storage and reaction temperatures, transportation, handling,
storage, safety, etc. Since the demand for these reagents for mine
drainage treatment is in no way market controlling, consumption for
this purpose must be planned around prevailing commercial products;
e.g. particle sizes and qualities. The various reagents have definite
limits of interchangeability within a given plant design.
By-products of manufacturing processes, as carbide lime from
acetylene manufacture, may offer cost or process advantages but
11
-------
Table 1. MINE DRAINAGE TREATMENT REAGENT DATA
Theoretical
Neutralization Equivalent
lb/gallon/mg/ 1 acidity
Reagent Formula Formula Weight x l0
Limestone CaCO 3 100.09 8.3453
Calcined lime GaO 56.08 4.6758
Hydrated lime Ca(OH), 74.10 6.1783
Dolomite CaCO 3 .MgCO 3 184.42 7.6749
Calcined dolomite CaOMgO 96.40 4.6759
Hydrated dolomite Ca(OH) 2 Mg0 114.42 4.7701
Pressure hydrated Ca(OH) 2 Mg(0H) 2 132.44 5.5213
dolomite
Caustic soda NaOH 40.01 6.6719
Soda ash Na 2 C O 3 106.00 8.8377
-------
Table 2. REAGENT COST DATA
(19681971)
Reagent Cost ata
Reagent
S/ton
fob
source
Transpor
tation
c/ton mile
Delivered
Cost
Hollywood
$/ton c/lb
Hollywood
c/neutralization
equivalent
Delivered Cost
to Limestone on
Neutralization
valent Basis
Ratio
Equl
Iga1/mgILI 6
acidity x l0
Limestone
Stone sizes 1.50 3. 4.50 0.23 1.92 -
Waste 0.90 3. 3.90 0.20 1.67 0.87
Pulverized 3. 8.50 0. 143 3.59 1.87
Calcined lime
Bulk 16.0018.00 3. 21.00 1.05 4.91 2.56
Bagged 19.50 3. 23.50 1.18 5.52 2.88
Hydrated lime
Bulk 18.2520.75 3. 23.75 1.19 7.35 3.83
Bagged 21.2523.75 3. 28.25 1.141 8.71 4.54
Dolomite
Stone sizes 1.802.50 4. 10.00 0.50 3.84 2.00
Pulverized 7.00 14. 15.00 0.75 5.76 3.00
Calcined dolomite
Bulk 16.00 4. 24.00 1.20 4.82 2.51
Bagged 21.00 4. 29.00 1.45 5.83 3.04
Hydrated dolomite
Bagged 19.75 14. 29.50 1.48 7.06 3.68
Pressure hydrated
dolomite
Bagged 23.00 31.00 1.55 8.56 4.46
Caustic Soda
50% Solution
Tanker 66.00 3.30 22.02 11.47
Drum 268.00 13.40 80.140 46.56
Flake
Drums, Bulk 78.00
Solid, small lots 110.00 5.50 36:69 19.11
Soda Ash
Powder
Bulk 40.00 2.00 17.68 9.21
Bags 76.00 3.80 33.58 17.49
aAss ,es 100% reagent purity.
-------
Table 3. ANALYSES OF REAGENTS USED DURING OPERATING PROGRAM
Chemical Analyses (Dry Basis)
Weight Percent
Total Loss on CaO
Reagent Ca Mg Si0 2 Insolubles Ignition Equivalent
Hydrated limea 3084 0.25 1.42 1.48 25.33 53.65
Calcined lime 68.61 0.60 1.35 1.08 96.00
Grit from 46.03 -- 3.87 4.31 24.93
slaking calcined lime
Limestone 38.79 0.19 1.08 1.86 42.88 47.40
Hydrated dolomite 29.58 17.33 2.23 2.40 26.43
Calcined dolomite 38.77 23.57 2.62 2.68 2.28
Dolomite 20.71 12.36 3.48 4.72
apartially recarbonated from atmosphere while in storage bin.
-------
their use should be carefully evaluated, including consideration of
quality, uniformity, long range availability, etc.
Pulverized limestone and lime products are commonly transported in
bulk utilizing pneumatic handling. Loads range from 15 to 28 tons.
Air pollution by dust release during unloading can be controlled and
is to be avoided. 1 eagent and shipment costs by bags of up to 100
pounds capacity are higher and also increases subsequent reagent
handling costs. Pulverized stone products, hydrated lime and calcined
lime are conveniently stored in bins. Bin storage of hydrated lime
can result in extensive water and carbon dioxide absorption within
several weeks; thus long-term lime storage which can result from
excessively large bins is to be avoided. In addition to bin storage,
pneumatically-handled hydrated lime may be slurried (10-50 weight
percent) directly from the delivery tanker and stored as a suspension.
Larger particle sizes of limestone (> 3/8 inches) have the best
handleability, do not require protected storage facilities, and are
not affected by the weather. Some stone products having large amounts
of fines may handle differently when wet.
The total components comprising reagent costs in mine drainage
treatment extend beyond the fob unit cost (as per quantity, quality
and bid), to incorporate transportation, handling, storage, stoichi-
ometry, use efficiency, etc. The reagent cost may be expressed as
cents/l000 gallons water treated and compared with the theoretical
value. Overall one seeks the best system efficiency which is here
defined as the ratio of total treatment costs to reagent costs. For
the coal operator, a more meaningful expression may be cents/ton
of saleable coal produced; or for the ecologist, some broader
expression as $/mile of stream controlled.
Critical factors in reagent selection and use also include the location
of their origin, quality, mode and points of manufacture-distribution,
mode of transportation, and availability. Experience indicates a
disinterest and a neglect of these factors, perhaps even intentional
avoidance by researchers, government agencies, and engineers. A
proper understanding of the role of these aspects is important to
the correction of the coal mine drainage problem.
REAGENT REACTIVITY
The chemical characteristics of the reagents used to react with mine
drainage are well known. Each of the reagents is subject to ex-
tensive dissociation in aqueous solution, as are the reactive ionic
species in coal mine drainage. Consequently, the neutralization,
?cation exchange reactions occur essentially instantaneously when
the reactive species contact. The reaction rate is primarily physical
involving ionic species availability and contact. Reagent and
15
-------
product solubility and mechanical aspects (mixing) thus really become
the controlling factors. However, many different equilibria are in-
volved in these systems and they play a major role in defining the
direction and extent of the reactive system. The relative rate
variations between the reagents are quite extensive as illustrated
in Figure 1. Sodium hydroxide will be more responsive than hydrated
lime, while soda ash can be expected to respond somewhat more slowly
as controlled by the bicarbonate equilibria.
A review relating these reagents in the treatment of industrial
waste acids (Lewis 2 ) emphasizes the varied nature of the w4stes
(which incorporates coal mine drainage) and that they are not amen-
able to standardized treatment procedures. Lewis cites two con-
siderations: the pH range over which treatment is to take place and
the minimum time available for the reaction. For mine drainage, the
first is controlled by the water characteristics and governmental
regulations. The final pH is complicated by equilibria developed in
the reactions. The time available for reaction is a pragmatic
decision. Some reagents require very large reactors and power
inputs as reaction times exceed ten minutes. Figure 2 illustrates
the pH ranges of the treated waters which may be expected. The
upper pH limit in carbonate systems is controlled by mechanical
factors as particle size and conditions affecting the hydroxide-
bicarbonate-carbonate equilibria.
FERROUS IRON OXIDATION
This reaction in mine drainage waters occurs in natural situations
forming iron III as an unstable ferric sulfate. The subsequent
hydrolysis forms various amorphorous and crystalline species of in-
soluble iron oxyhydroxides. Lovell and Lachman 3 have detailed these
changes for a particular stream. The hydrolysis products, in wide
ranges of composition, appear in the bottom of streams carrying coal
mine drainage and are commonly referred to as yellowboy. These
deposits create the undesirable appearances of streams which identi-
fies this pollution loading. This natural oxidation is attributed
to slow air oxidation, to the action of indigenous autotropic bacteria,
and possibly to some catalytic effects.
In the treatment of coal mine drainage with alkalies, it is usually
considered desirable, if not essential, to convert the iron to the
ferric state during the process. There are two reasons for this
objective: (1) The solubility product of ferric hydroxide (1.1 x lO_ 6
at 18°C) is lower than that of ferrous hydroxide (1.64 x lO at
l80C) a. This means that the desired iron concentrations can be
met at lower pH levels requiring the use of less alkali; it also
provides for the potential use of limestone as a neutralization
reagent, and minimizes the potential for side reactions occurring
16
-------
Figure 1. Relative reaction rates for various alkali reagents
MiNUTES
Figure 2. pH range resulting from various alkali reagents
N
E
U
T
R
ACIDIC A - - ALKALINE
if
i 1 I
13 4
pH 0 1 2 3 4 5 6
SODIUM HYDROXIDE,
CALCIUM OXIDE
OR
HYDROXIDE
MAGNESIUM OXIDE
OR
HYDROXIDE
CALCIUM ________
MAGNESIUM CAR8ONATE __________
OR
17
-------
with aluminum and magnesium species in the water which are favored
at higher pH levels. (2) Iron II forms additional hydrogen ion upon
oxidation. Thus, whether iron II exists in treated water or a sludge
product, there is a potential for additional acid formation. Such
continuing reactions also lead to process control problems and un-
stable systems.
The oxidation of iron II may be accomplished by atmospheric Oxygen,
by other reagents, as ozone, chlorine, peroxides and potassium
permanganate, or by biochemical processes. A controlled process of
iron II oxidation by oxygen from air is nearly universally utilized
and has been the subject of considerable study. The process design
for this oxidation is controlled by the two stages of the process:
Cl) The rate of iron II oxidation with oxygen, and (2) The rate of
transfer of oxygen from air to the reacting system.
Rate of Iron II Oxidation with Air
The parameters, which have been established to govern this reaction
are: temperature, pH, iron II concentration and aluminum UI concen-
tration. The rate increases proportionately with the first three but
shows a reciprocal relationship to aluminum III concentration. There
probably are other factors involved. Catalytic agents, which may
increase oxidation rates, have not been shown to play a role in the
reaction unless specifically added. Singer and Stumm reported sul-
fate complexes may be inhibiting. Steinberg, et al.°° describe in-
creased rates when the reaction is carried out with high levels of
Co-GO gamma radiation. Obviously, sufficient oxygen, presumably in
the form of dissolved oxygen, must be available. This parameter is
excluded from the rate considerat 4 ins since it is assumed that the
aqueous system will be maintained in a saturated condition by con-
tinuous air transfer. The oxygen concentration may be considered
a constant at the prevailing temperature.
The overall reactions may be described by Equation 1 or as two steps
in Equations 2 and 3.
4Fe 42 + 02 + 101 -120 χ 4Fe(O}fl 3 + SF1 4 (1)
4FeS0 + 02 + 21-12 504 - 2Fe 2 (S0 4 ) 3 + 2H 2 0 (2)
Fe 2 (S0 14 ) 3 + 61 - 1 a 0 + 2Fe(OH) 3 + 3H 2 SOz, (3)
Equations 2 and 3 are probably over-simplicatioris of the actual
mechanistic steps which occur , It has been reported 5 that a faster
oxidation rate occurs if the initial iron I I form is in the solid state
Fe(01-L)z as opposed to a soluble ionic species. This may be related to
iron-sulfate complexes .
18
-------
Potentially one mole of sulfuric acid is formed per mole of iron II
oxidized and hydrolyzed. This acid product lowers the pH of the
system and reduces the oxidation rate. Thus the equilibrium pH which
develops in the oxidation stage becomes the parameter which controls
the oxidation rate. The basic stoichiometry as shown by Equation 1
appears to prevail and requires 0.143 pounds of available oxygen per
pound of iron II.
If the alkali is added prior to iron II oxidation, as in the yellow-
boy process, the pH level must be adequate to neutralize the acid
formed during oxidation and maintain an acceptable oxidation rate.
Should a process involve post-oxidation neutralization, the effect of
decreasing pH on the oxidation rate must be considered.
The experience with this oxidation reaction, as associated with various
alkaline reagents, has been reviewed. Stumm, et al. 6, 7 have studied
the reaction and reports the oxidation rate to be first order with re-
spect to iron II and independent of iron III. The strong effect of
pH is related to the role of the hydroxyl ion. Mihok 8 proposed a
broad oxidation rate schedule, as a design base for limestone treat-
ment systems:
Fe II Concentration Oxidation Rate
mg/l mg/i/hr
50 180
100 240
250 300
500 360
1000 450
1500 600
2000 750
pH range 6.5-7.5
Temperature 0-10°C
A detailed laboratory study of iron II oxidation rates, preliminary
to these studies, was made by Stauffer and Lovell 5 , under conditions
anticipated in the operating treatment plant. The aqueous system
was maintained saturated with oxygen from bubbling air, temperature
was controlled and pH was maintained at a constant level by the auto-
matic addition of sodium hydroxide solutions.
Stauffers data, based on synthetic solutions which cover but a
limited range of variables (pH from 4.6 to 5.7, iron 11 and aluminum
concentrations to 800 mg/i and 200 to 400 mg/i respectively, and
temperatures between 15 to 25°C), do not respond well to rigorous
mathematical treatment. Even multiple linear regression treatment
involving natural logarithms of temperature and concentration were
19
-------
non-significant. A reaction order can be established graphically.
With the reaction rate being a differential of iron concentration
with respect to time, a log-log plot yields a slope of 5/8 which can
be expressed in integrated form:
(Fe 2)3/8 = (Fe 2)3/ 8 kT (4)
where k = 3/8k (-r = k [ Fe+2] 5/8)
The levels of the constant k are dependent upon the cited parameters
and were related for Stauffers data through the following multiple
linear regression equation:
k = b 0 + bi(pH) + b 2 in l/T + b 3 (Fe 2)3/8 + b (i/Al) (5)
where the coefficients are given in Table 4.
Table 4. REGRESSION COEFFICIENTS OF THE LINEAR RESPONSE
SURFACE FOR THE RELATED RESPONSE OF IRON II
OXIDATION RATES
(Based on Stauffers Data)
Coefficient
k
b 0
-5.44892
b 1
0.170686
b 2
0.76211
b 3
0.013887
b
9.42902
Utilizing this constant k with the initial and desired final iron II
concentration, reactiOn retention time may be estimated as shown in
Equation 6.
( Fe 2)3/S - (Fe 2)3/ 8 (6)
tr= kt
where: tr = Oxidation time in minutes
Fe 2 Initial iron II concentration in mg/i
Fe 2 = Desired final iron II concentration in mg/i
k T = constant defined by Equation 5
T = temperature = 273 + degrees C°
p 1-1 = units
Al = initial concentration in mg/i
20
-------
This procedure was tested statistically. The hypothesis was that the
proposed model does not predict the rate constant. The fact that the
analysis gives a significant result indicates that the probability
of rejecting the hypothesis when it is true is less than 5 percent,
but greater than 1 percent.
This reaction time estimation is minimum and is considered to be
conservative due to the range limitations of Stauffers data. This
provides a safety factor when applied to design calculations. Table 3
in Appendix E lists various calculated rates for different waters.
It is not recommended that this estimate procedure be employed with
carbonate or bicarbonate systems due to the additional factors in-
volved with the changing equilibria associated with such systems.
The design engineer should be especially cognizant of the rapidly
decreasing oxidation rate with decreasing iron II concentrations.
This situation is especially complex in a dynamic system where the
final effluent iron II concentrations (which would be in equilibrium
with the oxidation tank iron II concentrations) must be at minimal
levels. The chemical air oxidation system is similar in this respect
to the biochemical oxidation system discussed in Section XIII.
PROCEDURE FOR PLANT OPERATION
As an experimental facility utilizing waters from abandoned mines,
there was no requirement to maintain a given effluent water specifi-
cation. However, the operating control principle was to meet the
water quality specifications of the Pennsylvania Sanitary Water Board.
The Facility was operated under Sanitary Water Board Water Permit
No. 267 MO 11 (April 28, 1967).
The alkali addition was controlled to maintain the required pH in
the plant effluent based upon Continuous measurement with a glass
electrode system. The electrical signal from the pH transducer was
utilized to operate an automatic flow control valve for the alkali
slurry addition. For yellowboy process, the pH probe was initially
located in the oxidation tanks which monitored the water near the end
of the processing operations. Although the lag time was relatively
long, the pH drift in the waters was not excessive. In latter tests
the probe was located near the effluent overflow of the flash mixer
to observe lag time response. Under these conditions there was a
definite trend to introduce local excess alkali which led to unwanted
reactions with aluminum and magnesium ions and consumes additional
alkali. See Figure 7 and in Appendix B, Figures 9, 13 for these
locations of the pH probe.
In the Densator operation, the pH probe was located external to the
reactor but received a continuous sample stream from the secondary
21
-------
reaction zone. See Appendix B, Figure 14. There was no provision
for automatic pH control with the limestone reactor although a pH
probe was located in the effluent sump for continuous monitoring of
the effluent.
The detailed plant operational program was based on a minimum of 48
hours operation under a given set of process conditions. It was
impossible to follow the initial schedule due to uncontrollable con-
ditions, including weather, stream flows, mechanical failures, and
personnel limitations. Similarly, experience with the processes
necessitated timing variations in some cases, since it was not feasible
to operate within the prescribed limiting period, as in the Densator
tests and biochemical iron oxidations.
The observations and data reported were developed over approximately
one years operating period. The operating personnel were usually
limited to one man per shift. In addition to control of the treat-
ment, they were responsible for sample collection, control analyses,
repairs, and facility maintenance including janitorial services,
snow removal and grass cutting.
22
-------
SECTION IV
DESCRIPTION OF PILOT PLANT
The concept of the Hollywood Facility was deve1oped by University
personnel to evaluate and compare different chemical neutralization
processes for treating coal mine drainage. Every effort was made to
incorporate the concepts set forth in the Mine Drainage Manual of the
Pennsylvania Sanitary Water Boardboa. Certain unit operations would
be less than total design capacity to reduce costs. Coal mine drainage
waters would be pumped to the Facility, while gravity flow through
the plant was sought with discharge of the effluent to natural drainage.
The variability of the water sources and processes precluded optimum
design for each situation. The basic unit operations would involve
alkali neutralization, iron II oxidation, and solid-fluid separation
of the impurity sludge. Provision included raw water storage conforming
to Pennsylvania Sanitary Water Board criteria oa and sludge dewatering
and disposal.
Capability would be included to utilize any alkaline reagent, in-
cluding limestone which would require a specially-designed reactor.
The thickener would involve a specially designed unit, the Infilco
Densator, to permit study of sludge recycle concepts. The settling
lagoon incorporated an original design for Continuous sludge handling.
Provision for the use of flocculant aids would be included. Auto-
matic plant control would be limited to pH response of alkali intro-
duction while instrumentation would permit data collection of several
process parameters, including water quality, temperature, and flow.
Power consumption would be measured on the larger motors. The
summarized design plan is given in Table S and Figure 3. The initial
design permitted evaluation of only one process at a time.
Based upon these criteria, Gannett, Fleming, Corddry and Carpenter,
Inc., an engineering firm, prepared a basis for detailed design as
related to the Hollywood site. The resulting data which formed the
basis of design are presented in Tables 6, 7, and 8. Appendix B
presents the detailed units and facilities planned. The estimated
construction cost, based on these data, was $805,000.00. The actual
costs are shown in Table 9. The power consumption observed for the
various unit operations is given in Appendix B, Table 3.
A photograph of the resulting Facility is shown in Figure 4. In Figure
5 is presented the hydraulic profile for the various unit operations.
Simplified drawings of the Facility are presented in Figures 6 and 7.
The unit operation locations are indicated in Figure 6; the exterior
piping in Figure 7. The thickener-Densator was constructed above
23
-------
Table 5. BASIS OF DESIGN OF THE HOLLYWOOD FACILITY
Mine Drainage Water Volume
gallons per day 500,000
gallons per hour 20,833
gallons per minute 347
Total Acidity - tons/day CaCO 3 2.08
Total Iron - tons/day Fe 0.21
Alkali Requirement
tons/day, hydrated lime (Ca(OH) 2 )C 1.62
Oxygen Requirement
for iron oxidation pounds/hourd 2.52
Estimated Air Requirement
ft 3 /min, 45 minute retention 22
Thickener or Settling Pond Sludge Resulting from Process
Solids content, percent by weight 1.0
gallons/l000 gallons water treated a. 50
b. 10
gallons/day a. 25,000
b. 5,000
gallons/hour a. 1,040
b. 208
Dry Solids in Sludge
pounds/bOO gallons water treated S
tonslday 1.25
pounds/hour 104
Densator-Thickerier Dimensions (Diam. K SWD) feet 25 x 20
Drum Filter
Solids loading pounds/square foot/hour a. 5.0
b. 15.0
1-6 foot diameter by 7 foot long rotary drum
belt vaccum filter Filter area, square feet 125
a. Based on sludge from normal settling.
b. Based on dense sludge.
c. Assumed to have 95% available CaO.
d. Assuming all iron in ferrous state.
24
-------
Figure 3.
Simplified schematic drawing of treatment plant
BIOLOGICAL
OXIDATION
A
BIbLOGICAL
OXIDATIO N
B
-t
SO LID FLUID
SEPARATION
LAGOON
TREATED SLUDGE
EIFLUENT
I SOLID FLUID1
SEPARATION
Lc JF R
TREATED
EFFLUENT
LK A I IN
AGENT
WATER SOURCES
[ B J C )
HOLDING
LAGOON
c ]
NEUTRALIZATION
REACTOR
LIMESTONE
1
LIMESTONE
REACTOR
1
AIR OXIDATION
R EACTOR
4.
DIRECT
DISPOSAL
________
DRYING FILTRATION
BASIN
25
-------
Table 6. SOURCES OF MINE DRAINAGE ESTiMATED VOLUMES,
CONSTITUENTS AND CHARACTERISTICS
Proctor
Source No. 2
Tyler
Run
Proctor
No. 1
Bennett
Branch
Volume - mgd
Maximum 0.063 0.600 0.288 not aVail-
able
Average 0.025 0.119 0.199 30
Minimum 0.009 0.003 0.127 3
pH
Maximum 3.81 4.30 3.49 4.31
Average 3.05 3.35 3.05 3.43
Minimum 2.70 2.91 2.60 2.80
Acidity (as CaCO 3 ) - mg/i
Maximum 8,112 420 1,271 523
Average 6,700 330 780 278
Minimum 5 343 150 316 39
Total Iron - mg/i
Maximum 3,604 90 199 105
Average 2,090 27 134 53
Minimum 1,516 2 99 11
Ferrous Iron - mg/i
Maximum 2,300 90 199 105
Average 1,780 26 133 53
Minimum 1,360 2 99 10
Aluminum - mg/i
Maximum 770 131 196 45
Average 725 100 141 30
Minimum 680 51 93 22
Ca],cium - mg/i
Maximum 252 108 144 S6
Average 150 77 102 40
Minimum 60 40 40 20
Magnesium - mg/i
Maximum 129 114 132 46
Average 80 83 80 42
Minimum 3 58 10 39
26
-------
Table 6 (continued). SOURCES OF MINE DRAINAGE ESTIMATED
VOLUMES, CONSTITUENTS AND
CHARACTER 1ST ICS
Source
Proctor
No. 2
Tyler
Run
Proctor
No. 1
Bennett
Branch
Sulfate - mg/i
Maximum
6,174
712
1,127
514
Average
5,150
640
1,080
385
Minimum
4,062
536
990
248
Temperature - °C
Maximum
26
25
21
28
Average
19
16
15
18
Minimum
8
3
4
3
Specific Gravity
1.11
1.09
1.10
1.10
Table 7. TREATMENT PLANT DESIGN CONDITIONS
Chemical
Treatment
Biochemical
Treatment
Daily Volume
- million gal
0.500
0.100
pH
2.9
2.9
Acidity (as
CaCO 3 ) - mg/i
1,000
1,000
Total Iron -
mg/i
100
100
Ferrous Iron
- mg/i
100
100
Sulfate - mg
/1
i 100
1,100
27
-------
Table 8. MAJOR TREATME1 T PLANS
Plan I
a) Lime neutralization in flash mixer
b) Mechanical oxidation
c) Flocculation and settling in earthen lagoon
d) Overflow to Tyler Run
e) Waste underfiow to vacuum filter, earthen sludge drying lagoon,
and tank truck for hauling to local strip pits
Plan II
a) Lime neutralization in flash mixer
b) Mechanical oxidation
c) Flocculation and settling in Densator unit operated as
clarifier or clarifierthickener
d) Overflow to Tyler Run
e) Waste underfiow to vacuum filter, earthen sludge drying lagoon,
and tank truck for hauling to local strip pits
Plan III
a) Lime neutralization in Densator unit
b) Flocculation, settling and thickening in Densator unit
operated on densating principle
c) Overflow to Tyler Run
d) Waste underfiow to vacuum filter, earthen sludge drying lagoon,
and tank truck for hauling to local strip pits
Plan IV
a) Biological oxidation by air diffusion, and turbine and surface
aerators
b) Flocculation, settling and thickening of biological floc in
Densator unit operated as clarifier or clarifier-thickener
c) Waste underfiow to vacuum filter, earthen sludge storage lagoon,
and tank truck for hauling to local strip pits
d) Limestone neutralization of Densator overflow
e) Flocculation and settling in earthen lagoon
f) Waste underfiow to vacuum filter, earthen sludge storage lagoon,
and tank truck for hauling to local strip pits
Plan V
a) Biological oxidation by rock filter
b) Flocculation, settling and thickening of biological floc in
Densator unit operated as clarifier or clarifier-thickener
28
-------
Table 8 (continued). MAJOR TR1 ATMENT PLANS
c)
Waste underf low to vacuum filter, earthen sludge
storage
lagoon,
and tank truck for hauling to local strip pits
d)
Limestone neutralization of Densator overflow
e)
Flocculation and settling in earthen lagoon
f)
Waste underfiow to vacuum filter, earthen sludge
storage
lagoon,
and tank truck for hauling to local strip pits
ground level to reduce installation costs and improve accessability
to the unit. This required the influent to be pumped and created
freezing problems by heat loss througl the top and sides of the
thickener.
The dimensions and volumes relating to the settling lagoon are given
in Appendix B, Table 4. The lagoon discharge sump was equipped to
record effluent temperature and flow. These flow rates were made
from water levels attained in a discharge sump equipped with a V-notch
weir. A float transducer converted water levels to current for
recording. This type of flow measurement was found to be the most
reliable and trouble-free of the several methpds utilized. The four-
inch, treated water return line to the plant water system also served
as a sample line to monitor the final effluent pH, turbidity and
conductivity.
In the rotary limestone reactor, horizontal lifters were included to
aid desired stone attrition. They were not originally included in
the fourth compartment to minimize wear to the exit port and screen.
Although this purpose had been served, the small stone which accuinu-
lates in this section did not receive adequate attrition. The
addition of four-inch wide lifters to this compartment improved
reactor performance. Dams were also placed in the reactor to control
stone distribution. This function was further aided by the addition
of rod screens of three-inch, two-inch and one-inch openings across
the apertures from influent to effluent ends of the reactor. The
reactor slope provides some control on retention time and stone
distribution. Other details are given in Appendix B.
To ensure minimal seepage from the raw water holding lagoon,
bentonite clay was worked into the lagoon bottom and the surface
compacted (see Appendix B, Figure 2).
In the design and operation of equipment to transfer air (thus oxygen)
and subsequent evaluation of process data in iron II oxidation, the
solubility of oxygen in water becomes important. Published data 11
29
-------
Table 9. CONSTRUCTION COSTS FOR HOLLYWOOD MINE DRAINAGE TREA]745hT FACILITY
Dollars
Equipment
Construction
Process
Piping
Electrical
Heating
Ventilation
General
Installation
Total
Installed
Cost
Pumping Station
Bennetts Branch
Proctor No. 1
Proctor No. 2
3,987
4,503
5,276
17,961
14,646
19,190
1,925
3,600
3,710
4,000
2,000
3,000
250
250
250
200
200
200
28,323
25,201
31,626
Force Mains
Bennetts Branch
Proctor No. 1
Proctor No. 2
3,850
11,600
4,800
5,600
15,450
4,800
5,600
Control Building
Laboratory
Plumbing
5,739
2,356
124,000
49,560
29,341
39,300
6,000
3,400
259,696
Service Building
18,150
1,500
2,800
22,450
Yard Equipment
16,196
Densator
37,527
14,400
10,000
7,000
68,927
Flash Mixer
4,440
9,000
13,440
Holding Lagoon
12,311
1,760
14,071
Ca(01-fl2 Handling System
20,989
3,000
1,900
12,100
39,989
CaD Handling System
21,489
3,700
1,600
12,100
38,889
CaCO 3 Reactor
29,500
5,300
4,350
4,350
43,500
Oxidation Tanks
11,793
15 .230
7,740
3,870
38,633
Surface Reactor
3,354
16,900
5,200
2,600
28,054
Settling Lagoon
30,220
8,586
4,293
45,099
Sludge Pumping Station
1,574
10,029
8,350
2,200
22,155
Sludge Drying Beds
17,8:9
3,000
20,829
Vacuum Filter System
41,142
36,700
77,842
Polymor System
2,212
6,400
8,612
instrumentation
35,322
5,295
TOTAL
40,ol :
39l,Sin
C
V)
-------
H
C
1
CD
rim
(D PI
p3 1 - i
f -to )
CD
rtW
0
P-tirt
030
noQ
H- 1
f__a 03
rt::r
0
.0
rt
CD
CD
Pt,
CD
1
H
B
CD
rt
03
a
H-
P t
03
H
P 3
e q
C D
1 -
-
-4
N
-
4- 4
- -
- r 1
-------
Figure 5.
Hydraulic profile the experimental mine drainage
t reatrncnt
facility
LIMESTONE
NEUTRALIZATION
UNIT
HOLDING LAGOON
CONTROL
El 1225.
WL 122300
FORCE MAtN FROM BENNETT BRANCH PUMP. STA
BLDG. FLASH El.
FORCE MAIN FROM PROCTOR 2 PUMP. STA.
FORCE MAIN FROM PROCTOR .1 PUMP. STA.
DENSATOR TRICKLING
W L 1225.00 FILTER
Den so tor
El
Top of Rock
-El 1217 00
T o Densator Feed Sump.
WL
1218 00
To Settling Lagoon
SETTLING LAGOON
1 121200
Io Densotor
Feed Sump
Densotor Overflow
To Tyler Run
SLUDGE
PUMPING SLUDGE DRYING
STATION BEDS
El 121200 -El 121000
WI. 1208.00
Overflow To Tyler Run
Settling Lagoon
TYLER RUN
Scale: Horiz. 1 5O
Vert. U 2O
-------
Figure 6. External unit operations of the experimental mine
drainage treatment facility
::
:7
= ==
f
LUDGL DRTING
-------
SLUDGE
DR YING
BE 1 )5
Figure 7. External piping arrangement of the experimental mine
drainage treatment facility
0
0
7 o
0
0
0
4
t _ _4 t
0
\ \
K
PAR
0
I
0
0
0
- ---- --
S TILING
LAGOON
I
-------
not only indicates the variation in solubility with pressure and
temperature, but that the oxygen-nitrogen ratio of air dissolved
in water differs from that in the atmosphere.
More recent data for the solubility of oxygen in water in contact
with air at one atmosphere pressure (air and water vapor) has been
determined by Truesdale, et al. 2 a. The solubility, C, may be cal-
culated with the following formula:
C(ppm 02) = 14.161 - O.3943t + 0.0077t 2 - 0.00006t 3
where t =
The literature indicates the solubility of oxygen in water is depressed
by approximately 10 percent from soluble compounds (salts, acids,
and bases) at concentrations of about O.1N (5,000 to 15,000 ppm). At
the salt and acid concentrations found in most coal mine drainage,
the suppression probably would not exceed 5 percent of calculated
values.
The flow response through and retention time in the oxidation tanks
was evaluated at 200 gpm by a single rapid addition of sodium car-
bonate solution to Tyler Run waters in the flash mixer. The solution
was directed to the oxidation tanks opened for parallel flow (see
Figure 7). The oxidation tank effluents were sampled and conductivity
measured as a function of time. The nominal retention time for the
nearly 10,000 gallon vessels under these conditions was almost 50
minutes. Oxidation Tank No. 2 received the influent somewhat more
rapidly and thus had a shorter retention time. The conductivity
data are plotted in Figure 8 including an integrated plot representing
tracer through-put. The first portion of tracer passed through Tank
No. 2 in about eight minutes but required nearly thirteen minutes
for the first tank. Half the tracer had passed the second tank in
38 minutes while this value for the first tank was 49 minutes. The
highest concentration of tracer was attained at the effluent of each
tank after about 18 minutes indicating this period to represent the
retention time of the greatest number of tracer molecules. Over 99
percent of the tracer had passed Tank No. 2 in 140 minutes with 160
minutes being the corresponding value for the first tank.
The flash mixer, utilized to mix raw water with the slurried alkali
reagent, was a rubber-lined steel vessel separated by baffles into
three equal volume compartments as shown in Figure 9 and in Appendix
B, Figure 9. The flash mixer had a minimum volume of 1,481 gallons.
The three-compartment sludge drying basin had overall dimensions of
103 ft. x 84 ft. Details of construction are shown in Figures 6, 7,
and Appendix B, Figure 12. Photographs in Figures 4, and 46 through
35
-------
Figure 8. Retention time studies - oxidation tanks
:1 500
J400
300
90
80
70
3
0.
-t
3
Zn 0
uu -c
I.-
S .
U
!
F-
50 °
S
a.
C
3
C
40 U
30
20
10
0
900
800
700
600
100
o Oxidation Tank No. 1
O Oxidation Tank No. 2
200
120
Time (Minutes)
0 60
180
-------
Fi:uxe 9 . Introduction of Lime Sltnry into Flas t i MIxer.
52 offer additional information. Basically, the bed was an earthen-
walled excavation having three sides with a 30° slope. The south
side, designed as an access ramp, had a 15° slope and was covered
with four inches of crushed stone upon which was laid a two-inch
bituminous surface,
The bed bottom was the prevailing clay base, carefully compacted.
Each bottom section contained two 4-inch vitrified clay pipe drain
lines of extra strength, bell and spigot-type with perforations
37
-------
restricted to the bottom half of the pipe. These were installed on
12 1/2foot centers. Each drain line was equipped on the up-stream
end with an air vent stand pipe six-feet high which also functioned
as a decant pipe to remove excess clear water. Each drain extended
through the south bed wail carrying the underf low to gravity drainage.
The drainage pipes were covered with a two-foot layer of red dog
(naturally-burned coal refuse) upon which was placed a four-inch
layer of coarse sand. The drainage rates of e originally-constructed
bed were excellent. During the heaviest rains, prior to the intro-
duction of sludge into the bed, there was no collection of water in
the drying bed with the water flowing freely through the underdrain.
The bed was divided into equal sections by concrete walls designed to
support a 6-mil polyethylene plastic canopy over the center section.
The plastic was supported by an alui iinum arch frame. The ends of
the canopy were open to maintain free air circulation. The precipi-
tation rejected from the central compartment by the plastic cover
was collected in run-off gutters leading to natural drainage.
Each bed compartment had a bottom filtration area of 1,270 ft 2 and
could accomodate a maximum of a three foot depth of sludge, thus
providing a working volume of approximately 4,700 ft 3 (35,000
gallons).
38
-------
SECTION V
OBJECTIVES
This study resulted from a request of the Pennsylvania Coal Research
Board to design, construct and operate a full scale plant to evaluate
treatment procedures for coal mine drainage waters. The study was
to provide year-round data covering operating and economic experience.
The Hollywood Facility was developed to provide maximum versatility
for the study of neutralization treatment processes and related pro-
cedures. Special emphasis was placed upon control and dewatering of
the sludge which represents the impurities removed from the contami-
nated waters.
This site permits the study of four waters of different qualities
as a process parameter. All basic stages proposed to treat these
waters by neutralization can be reproduced. In addition to the
defined program, the Facility provides the opportunity for detailed
study and observation of any specific unit operation. The available
basic systems include:
1. The Yellowboy Process A: Neutralization by an alkaline reagent
(hydrated lime, calcined lime, caustic soda, soda ash, limestone,
dolomite products, and waste lime products) in a flash mixer, air
oxidation of ferrous iron, and separation of the sludge by the
settling lagoon.
2. The ! Yellowboy Process B: Neutralization by an alkaline reagent,
air oxidation of ferrous iron, and separation of the sludge by
clarifier.
3. Infilco Densator Sludge Recycle Process: The entire treatment
proceeds in a specially designed reactor in which the raw drain-
age water is mixed with recycle sludge in the primary compartment,
alkali added and reacted in the secondary compartment, the sludge
separated in the bottom and subject tQ recycle or removal. Finally
the treated, clarified water is discharged from the top.
4. Biochemical Oxidation-Limestone Process A: Deep tank oxidation
of iron in the raw water, separation-recycle-removal of bio-
logically-active sludge in the clarifier, water neutralization
with limestone, and separation of sludge in the settling lagoon.
5. Biochemical Oxidation-Limestone Process B: Surface oxidation
of iron in the raw water, separatioi -recycle-removal of bio-
logically-active sludge in the clarifier, water neutralization
with limestone, and separation of sludge in the settling lagoon.
39
-------
6. Limestone Neutralization: Direct chemical oxidation and
neutralization in the limestone reactor and separation of
sludge in the settling lagoon.
40
-------
SECTION VI
TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME
The process most commonly utilized to treat coal mine drainage mixes
the water with calcium hydroxide, raises the pH and forms insoluble
iron and aluminum compounds. The reaction mixture is subsequently
oxygenated converting iron II to iron II I. The suspended solids are
then separated, discharging the treated water.
The studies, considering each of the four waters available, utilized
this procedure with two modifications of solids separation: settling
lagoon (Process 10) and thickener (Process 11). The process flowsheet
and test data for Process 10 are detailed in Appendix C, Tables 1
through 4. In these tests the controlling pH probe was located in
oxidation tank No. 2. The data for Process 11, where the pH probe
was located in the flash mixer, are given in Appendix C,, Tables S
through 9. The reagent was introduced as an aqueous slurry containing
between one and five percent calcium hydroxide as shown in Figure 9.
The system permitted the lime slurry to be controlled and introduced
either by a manual valve or by a proportioning valve which responded
to a setting of the pH probe. See Appendix B, Figure 9. A constant
manually-controlled slurry addition was utilized when treating large
volumes of the more highly mineralized water with the concurrent
addition of lime slurry through the proportioning automatic valve.
This procedure yielded minimal pH variation within the system. The
manual valve alone did not provide reliable, constant slurry flow.
The lime slurry feed system was effective in providing a continuous,
controlled flow completely automatic from lime storage bin. No
difficulties were experienced with lime scaling in the lines or
hang-up in the bins. Due to intermittant operation and prevailing
high humidity, the electronic controls of the feeder required frequent
servicing. Some carbonation of the lime slurry occurred in the lime
slurry sump, especially in those operations where the lime consump-
tion was low and the slurry was prepared at infrequent intervals.
This introduced an uncertainty in reagent measurements, especially
with short runs. The test program necessitated that some batches
of hydrated lime remain in the storage bin for weeks. Although no
handling or moisture absorption difficulties developed, there were
major levels of reagent carbonation which changed the character of
the alkali reagent, and its reaction mechanism. In operating treat-
ment facilities, the lime storage capacity should not exceed one
months requirements.
The addition of lime (or any of the reagents) as an aqueous slurry
avoids delays in wetting the solid. The slurry mixes with the water
more rapidly. The solubilized lime reacts on contact while those
solid particles remaining are more rapidly dispersed for solution
41
-------
and reaction. It appears desirable to add the reagent at the most
dilute level feasible. The total reagent requirements establish the
limits. This rapid mixing and reaction permits a more responsive
action of the controlling pH probe. Localized excesses of reagent
are less likely to occur thus avoiding unwanted side reactions.
Reagent addition control is most versatile when a slurry is employed.
Large clumps of unwet lime are more likely to exist with dry reagent
feeding and to remain together leading to possible coating and to
loss of unreacted reagent. Addition of solid reagents was not
utilized in these studies.
The lime slurry was prepared with either house water or treated
effluent. The preparation of a concentrated (10-20 percent) reagent
slurry with good quality water (as in the mixer-dissolver), to be
diluted with the water being treated, is indicated. The dilution
would become the first of several partial neutralization stages.
In the Hollywood design this could be accomplished in the lime slurry
sump. The level of dilution would vary with the acidity of the feed
water but the final slurry probably should never exceed 0.1 percent
by weight. This procedure would offer the best control to maintain
the desired uniform pH and avoid excess alkali consumption by un-
wanted reactions. An even more desirable reaction situation would
follow if the diluted reagent were added to the water in several
stages as follows:
1st Reaction
Stage
Hydrated )Feeder ______Dissolver Dilution _______ 2nd Reaction
Lime Mixer ) stage
I Slurry S 1 fP
Plant Water
Mi no
Water
Feed
20 gpm
10 gpo ) Stage 0 idation
pH control 100 gpm
6.08.0
I
Mine Water
Feed
40 gpo
A theoretical explanation for this recommendation, which was based
on these laboratory and plant observations, is offered by Whittemore 70 .
This mode of reaction also affects sludge behavior.
pH control
S.f
Mine Water
Feed
40 gpo
42
-------
In the Process 10 tests, the pH control was maintained at the lowest
pH possible, consistent with meeting final iron specifications. In
some cases the acidity requirements were not maintained unless signifi-
cant reagent excesses were utilized. In those cases (as Test No. 8-70)
the alkalinity developed in the effluent was less than that indicated
by the excess reagent. This suggests that other reactions, probably
involving aluminum and magnesium, were consuming reagent. This
observation is important in considering reagent consumption in re-
spect to theoretical requirements and treatment objectives. The
goal to reduce iron concentration and develop some alkalinity in-
volves aluminum and magnesium concentration secondarily. The reac-
tion of further lime with aluminum hydroxide beyone its minimum solu-
bility level (about pH 6.2) appears to lead to its resolubilization
as the aluminate. This means less removal of aluminum from the waters
and higher lime consumption. Despite theoretical considerations,
magnesium does appear to react at unexpected low pH levels, possibly
precipitating and reducing magnesium concentration in the effluent.
This consumes lime beyond the level related to established goals. The
desirability of these effects requires a management decision: use
of minimum reagent (reduced costs) versus less aluminum and more mag-
nesium removal.
This situation was more evident in Process ii tests, where the pH
probe was located in the flash mixer. Two probe locations in this
reactor were utilized. The pH variation was found to be greater
(more excess reagent utilized) when the probe was located near the
flash mixer overflow (Tests 66-71, 67-71, 68-71, 6971 and 71-71),
as opposed to a location in the same compartment where the alkali
was added (Tests 38-71, 56-71, 57-71 and 65-71). The excess of
reagent consumption tends to increase with the acidity of the raw
water and with smaller flows where reaction time was longer.
The Process 10 tests generally produced satisfactory effluent meeting
the specifications: pH > 6.0, slight excess alkalinity (several
tests resulted in low acidity values), iron < 4 mg/i, and a clear
effluent with suspended solids < 12 mg/i. The reagent consumption
decreased in efficiency with increasing acidity, ranging from 54 to
192 percent. The adjustment to levels near theoretical was feasible
but was difficult during these short duration runs. The sludge pro-
duction also varied with the total acidity of the feed water. The
settled sludge (two hour level) ranged from 0.14 to 0.52 weightJ
volume percent, the dry solids from 0.23 to 2.84 lbs/1000 gallons
and the sludge volume from 1.2 to 14.5 volume percent. The sludges
displayed Type I settling rate behavior (see Section XIV) which
tended to gel upon aging. Settling rate data are given in Figures
24 and 25, while summary sludge settling characteristics are given
in Table 24. The response to separation in the settling lagoon is
provided in Tables 39 to 42, while the data from a thickener are
43
-------
detailed in Tables 34 and 35. Chemical analyses of the sludges are
listed in Table 50. The iron II oxidation was satisfactory at the
prevailing retention times, ranging from 0.56 to 30 minutes/ppm
iron II.
The Process 11 tests were also satisfactory, with more emphasis on
higher acidity waters, but at lower flow rates due to the thickener
size. The effluent pH was maintained at higher levels than the
Process 10 (settling lagoon) tests, always exceeding 7.0. The alka-
linity values ranged to 345 but were generally near 14 mg/i. The
soluble iron was consistently less than 1 mg/i. The effluent was
sparkling clear without the use of flocculants. The reagent con-
sumption ranged to 283 percent of theoretical with control becoming
more difficult with the higher acidity waters. Adequate adjustment
was obviously feasible. In Proctor No. 2 waters the sulfate content
was reduced by 22 percent, as was magnesium by nearly 50 percent.
The two hour sludge content was similar to the previous tests (0.42
to 0.69 weight/volume percent) but increased to 2.6 percent after
10 hours. The dry solids were higher with Proctor No. 2 waters
ranging to 14.2 lbs/l000 gallons treated water as was sludge volume
extending to 26 volume percent. The oxidation retention time was
increased but proved to be adequate at levels as low as 0.37 minutes/
ppm iron II. The oxidation-reduction potential ranged between 162
and 350 millivolts. The minimum oxidation rate extended to 338 mg/l/hr
in these runs.
In contrasting the two unit operations for solid-fluid separation
(settling lagoon versus thickener), as applied with this reagent,
there are no obvious differences which affect the chemical treat-
ment. The plant experiences from these solid-fluid operations and
subsequent dewatering are presented in Section XIV.
44
-------
SECTION VII
TREATMENT OF COAL MINE DRAINAGE WITH CALCINED LIME
Tests with calcined lime followed the flowsheet (termed Process 10--
see Table 10, Appendix C) used with hydrated lime (Tables 1 through
4, Appendix C) with the replacement of a slaker for the dissolver-
mixer. Each of the four waters was treated in the flow range from
101 to 325 gpm. The slaked reagent was diluted to the one to five
percent level for pumping to the flash mixer. The average percent
concentration of Ca(OH)z for runs 6-71 through 11-71 were, re-
spectively, 2.8, 0.93, 1.70, 2.30, 1.31 and 1.18. The slurry addi-
tion was controlled as described with hydrated lime.
The pH probe controlling slurry addition was moved from the oxida-
tion tank to the first compartment of the flash mixer after Test 7-71.
For Test 15-71 it was moved to compartment No. 3 near the overflow
pipe. As with the hydrated lime tests, proximity of the control
probe to the slurry feed point tended to result in greater alkali
consumption. The difficulty of pH control (as evidenced by range
of pH observed) tended to increase as retention time decreased
as shown in Table 10. These values are extremes and do not reflect
the stable control attained which was normally less than ±0,5 pH
units. These variations with calcined lime were greater than hy-
drated lime, reflecting a slower level of reactivity. The feed
water was 8 to 17°F colder during these studies than the hydrated
lime tests which would also decrease reactivity. Otherwise there
were not any distinct operational differences observed. Calcined
lime is the most commonly used reagent other than hydrated lime.
The storage, conveying and slaking of calcined lime gave a satis-
factory response without mechanical difficulties. However, it re-
quired additional attention to remove the unreactable grit and to
maintain slaker operation. Consideration of known variations in
slaking characteristics with different calcined limes and attainment
of optimum slaking conditions was beyond the scope of this study.
The non-slaking grit represents unburned limestone, The four to
eight percent loss of grit, its handling, equipment maintenance as
well as its capital costs minimize the cost advantage of this rea-
gent. The grit products not only vary with the calcined lime pro-
duct but with the operation of the slaker. In test 12-71 the grit
production was 9.2 percent while in test 13-71 it was 5.2 percent.
The slaker must be kept at hydration reaction temperatures (>130°F);
thus it lacks the versatility of intermittant operation feasible
with the hydrated lime dissolver. The slaking temperature ranged
between 102 and 210°F with an average of 143°F. Failure to maintain
a proper slaking temperature tends to result in loss of alkalinity
45
-------
Table 10. SUMMARY OF pH CONTROL VARIATION
CALCINED LIME TREATMENT - PROCESS 10
Test no.
Oxidation tank
response time
minutes
Flow
rate
GPM
pH
Range
Flash mixer
Oxidation
tank
17-71
<5
325
8.0
7.7
771
10
197
6.8
1.9
10-71
minimal
196
1.6
2.5
12-71
10-15
199
1.4
2.3
15-71
15
200
1.9
2.3
14-71
15
199
4.0
4.1
13-71
15-20
195
1.0
1.2
971
20
107
3.2
2.8
8-71
<30
109
7.2
6.1
11-71
60
103
1.9
0.9
16-71
60
101
4.0
2.5
with the grit and a less reactive reagent. The lime consumption
must be at least adequate to maintain continuous operation of the
smallest slaker commercially available (about 300 lbs/hr and a pro-
ductiori of 12 to 25 lb/hr of grit). Only in a few tests was this
rate approached. Calcined lime is favored cost-wise as the reagent
requirements of the system increase.
The detailed test data are given in Appendix C, Tables 10 through
14. The observations during the hydrated lime tests were confirmed
regarding advantages of stepwise addition of the alkali to prevent
localized overtreatment and subsequent reactions with aluminum and
magnesium. These responses tend to consume excess reagent. There
was no indication of moisture or carbon dioxide absorption by the
calcined lime during storage. Calcined lime must be utilized as a
slurry to achieve an adequate utilization and reaction rate. A
recently developed, pulverized, calcined lime product is claimed to
react with sufficient rapidity that separate slaking is not necessary.
This special product was not available for these tests.
During these runs the pH control was adjusted to maintain a level in
excess of 7.0 to avoid any residual acidity. This control, believed
to be typical of industrial practice, leads to considerable excess
reagent consumption. The utilized excess increases with raw water
acidity and tends to increase at lower flow rates with longer reaction
times, ranging from 77 to 272 percent. These excesses were not
detectable as alkalinity in the effluents. Those tests showing very
high reagent consumption used a raw water having relatively large
concentrations of aluminum and magnesium.
46
-------
Despite the relatively high pH levels maintained and the excess
reagent utilized, the settling lagoon effluents tended to develop
small levels of acidity. The iron content was characteristically
1 mg/i or less. The effluents were sparkling clear with suspended
solids seldom exceeding 10 mg/i.
As with the hydrated lime tests) the sludge production increased
with total acidity of the feed water. The solids level of two hour
settled sludge was somewhat higher (0.20 to 0.95 weight/volume per-
cent) than with hydrated lime, as was dried solids production (0.55
to 19.1 ibs/l000 gallons). The settled sludge volume was consistent
with the hydrated lime tests) varying between 1.3 and 29.6 volume
percent. The settling behavior of the sludges was typical Type I
excepting those developed from Tyler Run and Bennetts Branch waters
which responded as Type II. These sludges tend to gel upon aging.
See Figures 24 and 25 for typical settling curves and Table 24 for
sludge settling data. Tables 39 through 42 show the response of
these sludges in the settling lagoon.
The iron II oxidation was adequate at the retention times employed
(0.19 to 196 minutes/ppm iron II). The largest minimum oxidation
rate observed was 416 mg/i/hr. The measured oxidation-reduction
potentials were somewhat higher (increased pH levels) than observed
with hydrated lime extending from 198 to 414 millivolts.
To summarize the responses shown by hydrated lime and calciried lime:
there are no obvious deviations. There is the very critical cost
difference between them but this appears advantageous only when the
reagent consumption is large enough to acconiodate the additional
equipment and maintenance necessary to slake the calcined lime.
47
-------
SECTION VIII
TREATMENT OF COAL MINE DRAINAGE TO PRODUCE
HIGH DENSITY SLUDGE BY RECYCLE
The complications resulting from the voluminous, low solids content
sludges which result when most alkalies are reacted with coal mine
drainage is cited throughout this study. The difficulties became
obvious when seeking daily disposal of 300,000 gallons of a slurry
containing less than one percent suspended s3lids resulting per
million gallons of coal mine drainage. The observed results from
secondary dewatering attained with the sludge drying basin (Section
XIV) indicate this approach to be feasible. The costs of some de-
watering operations can be exorbitant. However, this method as well
as other dewatering procedures, as multiple stage thickening, filtra-
tion and centrifugation, result in better efficiencies and lower
costs if the density of their feed can be increased. Some improve-
ment in settled sludge density can result from reagent selection
(especially limestone) or reaction procedures (pH control and stage
reagent addition). Polyelectrolytes were ineffective in increasing
sludge density, although at high dosages a balling effect was ob-
served which is worthy of further exploration. Seeking means to
maximize sludge density was a goal of this study.
The concept of sludge recycle has been applied successfully to
sewage slud es and was reported helpful when treating waste pickle
liquors 3 . The Infilco High-Density Solids-Contact Treatment
Process was investigated and incorporated into the Hollywood Facility
design. Preliminary bench scale studies were made by project per-
sonnel and at Infilcos Research Laboratories, utilizing mine water
from the Hollywood sources. A three-foot diameter Densator pilot
unit was operated at the Hollywood site and observed by industrial
and governmental representatives. Subsequently a similar full-
scale system has been constructed and operated . That operation
confirmed these preliminary studies attaining high sludge densities.
A number of Operating parameters were cited.
The physical Unit installed at Hollywood was a modified Infilco
uDensatorfl, High Density Solids Contact Treating Plant. Bottom
rakes were added to permit the operation of the unit as a conven-
tional clarifier-thickener, as described in Section XIV. A detailed
drawing of the unit is shown in Appendix B, Figure 14 with photo-
graphs in Figures 4 and S (Appendix B). The system provided for
controlled pumping of sludge from the bottom well to the Densator
feed well, to the sludge storage sump for subsequent dewatering,
or to the sludge drying basin. The recycle sludge rate could be
48
-------
monitored by a magnetic flow meter. These flows may be followed in
Figure 7 and Appendix B, Figure 9. Alkali slurry could be pumped to
the secondary reaction zone of the Densator in response to a pH
sensor. This probe sampled water from the secondary reaction zone.
It was also possible to add flocculant to this zone of the reactor.
The system permitted premixing either recycle sludge and raw water
or recycle sludge and alkali before further reaction. The above
ground installation was convenient for regional sampling of the
Densator.
The single unit was designed to provide neutralization, sludge
densification and clarification for a complete treatment facility.
The sludge densification capability was most important. The test
sought to demonstrate the potential of the Densator as a multi-
function unit producing a densified sludge with good dewatering
characteristics.
DENSATOR OPERATING CONDITIONS
The full-sized densification unit was not amenable to short time
evaluation. The test described covered about onehundred operating
days necessitating a modification of the planned project program.
Unfortunately, plant layout precluded simultaneous evaluation of
other processes while the Densator was in operation, although some
experience with biochemical iron II oxidation was carried out con-
currently. The biochemical effluent, as indicated in the flow-
sheet (Figure 10), formed a portion of the Densator feed water.
This concurrent study undesirably decreased the iron huron III
ratio, complicating the analysis and control. Accordingly, the
system was operated only with Proctor No. 2 water which was the
most severely mineralized. The experience gained, including the
laboratory and pilot tests, indicated the sludge recycle concept
holds its greatest advantage when applied to highly mineralized
drainage waters which characteristically form large volumes of
low density (0.2-1.0 percent) settled sludges. Lightly mineralized
waters, forming slow settling floc of negligible solids content,
are less amenable to the process. Such waters tend to have low
iron and aluminum content and normally produce less than one pound
dry solids per 1000 gallons.
The Densator operation was controlled to maintain effluent quality
consistent with Pennsylvanias regulations. This was accomplished.
The sludge recycle rate sought to maintain approximately 3.0 percent
solids in the primary reaction zone of the Densator. These solids
were composed of a mixture of freshly-precipitated and recycled
sludge. The Proctor No. 2 waters alone yielded about 1500 ppm
suspended solids upon neutralization. To maintain the control
49
-------
Figure 10. Flow sheet for producing high densitY Sludge
OXIDATION
PROCTOR 2 6 GPM TANKS
400 mg/I Fe 3
NG_ 4 oo:. (Blo -OXIDATION)
DEN SATOR
_____ 24GPM FEED
LAGOON
______ SUMP
9/I F. 2 ______ _____ _____
30 GPM RECYCLE SLURRY
320 mg/I Fe 2 (10.11% SOLIDS]
80 mg/I Fe 3 10 GPM
REACTOR
LEVEL - REACTOR OVERFLOW
SLUDGE BLANKET 2 ZONE 1 DENSATOR
TO
HYDRATED LINE SL
________________________ -ψZONE 2 I 1 LOCAL
POLYELECTROLYTE FLOCCULAN1 r3-4 SOLIDS) 1 F. 2 STREAM
(5 ppm)
HIGH
DENSITY
SLUDGE TO SLUDGE
p - DRYING BASIN
OR VACUUM
SLURRY FILTER
PUMP
50
-------
level of solids in the primary reaction zone, as was recommended by
Infilco, an optimum ratio of recycled to precipitated solids was
about 20:1. The feed water flow rate to the unit was established
in conjunction with the other variables, primarily by the settling
rate of the sludge blanket. The high density sludge concept func-
tions from a highly ferrous system and alkali addition to maintain
a pH level higher than conventional neutralization processes. A
4:1 iron Il/iron III ratio was maintained during most of the test,
although this ratio and total iron levels were decreased by rain
dilution near the end of the test. A decrease in solid content
of the underflow resulted from these concentration changes. No
provision is made for iron oxidation in the system. This repre-
sents a capital and operating cost savings. Air introduced into
the secondary reaction compartment interferes with sludge floccu-
lation and settling rate. The monthly analyses for Proctor No. 2
waters during this test period are shown in Appendix A.
Hydrated lime slurry was pumped into the secondary reaction zone in
the Densator to maintain a final pH of about 7.7. At this p 1-I level,
the iron concentration in the overflow was less than 1 mg/i. The
values for the operating parameters are presented in Figure 11. A
flow rate to the Densator in the range of 50 to 130 gpm permitted
proper continuous operation, with the lower rates offering better
overall response. The flow rate was modulated with the settling
characteristics of the prevailing sludge settling rate. The use of
a weakly anionic flocculant was desirable, if not necessary. The
dosage was in the range of 4-5 ppm based upon influent volume. It
appears that the flocculant dosages should be determined from the
solids loading in the secondary reaction zone and expressed in
terms of pounds of flocculant per ton of dry solids rather than On
a raw water volume basis. Under these conditions, high density
sludges ranging from 10 to 16 weight percent solids were produced
consistently. The initial increase in sludge density was quite
slow. The initially-formed sludges settled with typically low
solids contents. Accordingly large volumes of settled sludge were
recycled seeking to attain the desired three percent solids level
in the reaction sections. Over 50 operating days were expended
to meet the three percent criteria. As sludge density builds in
the unit during this period, the normal sludge free settling rate
is minimized by entering early stages of compression. It appears
start-up should utilize low raw water and recycle flow rates,
ignoring solids levels in the reaction sections. This would pro-
vide minimum rise velocities and permit greater sludge compression
to develop in the sludge collection well.
No sludge density improvement was indicated by premixing the lime
slurry with the recycle sludge before reaction with the raw water.
51
-------
Figure 11. Operational data froln densator reCYcle studies
OC
20-
15-
10
PPM
SO-
WATER TEMPERATURE - °C
i-.----H I -
SUPERFLOC 1.27
ATLASEP 1A1 ATLASEP IN
FLOCCULANT ADDITION RATE PPM
/ S.-
/
15- 0
/ 0
0
5 5
10.
5-
SLUDGE RE YCLE
TO SLASH MI R
0
. _0
0
10 20 30
AUGUST 1970 I SEPTEMBER
31
/
/ 0 5
0 5
0 - 0 -
000 0.- 5-
0 0 0 0
0 0 0 0
0_ 5
. 5 .
00
5- 0
0/ ,.- 0
5 -
t l ____________
SLUDGE RECYCLE TO DENSATOR FEED SUMP
II
70 80
I NOVEMBER
31
SOLIDS IN DENSATOR UNDERELOW
40
Doys)
60
OCTOBER
I I
-------
Figure U (continued)
Operational data from densator recycle studies
-w
30
T oe Doys)
6PM
So
0
GPM
100
50
0
GPM
l 50
103
50
DENSATOR UNDERPLOW RECYCLE RATE. GPM
6-
- 1
5
4.
2-
AUGUST 1970 1 SEPTEMBER
31
SUSPENDED SOL DS IN DENSATOR FEED
PRIMARY REACTOR ZONE
0 SECONDARY REACTOR ZONE
60 70 R0 90
OCTDEER NOVEMBER
31
53
-------
Individual underf low samples showed solids content to nearly the
thirty percent level. The samples were taken from a pumped stream.
The solids content values cited represent recycle sludge and not
that from additional settling. Recycle sludge densities varied
widely during adjacent samplings. This sampling difference was
associated with the position of the bottom rake, the depth of the
sludge compression zone, and sludge flowability. The recycle sludge
showed no tendency to rat-hole, as do gelled hydroxide sludges. Its
viscosity reduces flow rates. A photograph of the recycle sludge
(Figure 12) indicates viscosity in supporting a stirring rod upright.
These sludges were ferrous in character (dark green).
No direct evaluation was made of the pollution potential of these
ferrous recycle sludges after blow down. They are highly alkaline
in character and oxidized very rapidly at the sludge-air interface
when transferred to the sludge drying basin. This highly alkaline
nature should be adequate to neutralize any acid formed during
oxidation of the iron II content.
Water effluent quality was excellent during most of the test except
when the control pH was allowed to drop below 7.5. Even under these
conditions the overflow maintained an iron content consistently less
than one mg/l.
A summary of the operating parameters is given in Table 11. A wide
fluctuation in operating variables was experienced as the parameters
were modified during preliminary operation of the system. It was
difficult to maintain steady-state flow conditions because of the
slow settling rates of the high level sludge blanket in the Densator.
This sludge, which was well into the compression zone range, had an
average settling rate of less than 0.5 ft/hr, which made it difficult
to.operate at continuous water flow rates much greater than 50 gpm.
During these operations the level of the sludge blanket varied up to
about 10 inches of the overflow. The input flow rate was the vari-
able utilized to control the sludge blanket level. It was possible,
however, to determine optimum conditions for the 25-foot diameter
device. Based upon plant observations, optimum operating flow rates
for the Hollywood Densator under various recycle conditions have
been calculated and are presented graphically in Figure 13. These
calculations are based upon a maximum sludge blanket settling rate
of 0.5 ft/hr. The flow capacity of the Densator possibly can be
increased by using greater flocculant dosage.
The recycle sludge produced during the tests had the desired pro-
perties sought. Besides high density (thus significant sludge volume
reduction), the sludge had near ideal dewatering characteristics.
Upon transfer to the sludge drying basin, the recycle sludge drained
immediately, and never maintained any free standing supernatant. It
54
-------
; .re 1 . fl2 Ira na c, rc1e S)
also provided the best filtration characteristics of any sludge
tested. With a filter feed suspended solids level between 10 and
17 percent solids by weight, a filter cake in excess of 1/2-inch
could be maintained without blinding or build-up of filter vacuum.
The cake cut cleanly from the drum with the very minimum of precoat.
The filter cake had about 24.5 percent solids with excellent handling
characteristics. The solids filtration rate averaged about 312 lbs
dry solids/ft 2 /24 hrs. This sludge had a typical analysis as given
in Table 12.
The overall response of the recycle process may be illustrated by
comparing the product levels from the Densator with those resulting
from the Yellowboy process with sludge separation by a thickener.
A typical Proctor No. 2 water containing about 400 mg/l iron is
the base for the comparison.
55
-------
Table 11. DENSATOR RECYCLE TEST - OPERATING PARA V1ETERS
- -. Maximum Minimum Optimum
Raw Water
pf-f 3.4 2.9
Fe 2 - mg/i 530 100
Total Acidity - mg/i 1,700 1,500
Sulfate - mg/i 2,500 2,300
Temperature - °C 18 8
Densator Effluent
p11 6.3 9.1 7.7
Fe - mg/i 91 <1 <1
Densator Feed
Total Feed Flow Rate 160 50 40
Precipitable Solid (PS) - ppm 1,400 400
Recycled Solids (RS) - ppm 20,000 60,000 30,000
Ratio RS:PS 14 150 30
Total % Solids in Primary Zone 2 6 3-4
Floccuiant Addition - ppm 5 0 5-10
Recycle Slurry Flow Rate - gpm 60 10 10
Raw Water Feed Flow Rate - gpm 140 15 30
Ratio Fe 2 /Fe 3 4:1
Settling Rate - ft/hra 2.3 0.3 1.0
Densator Underflow
% Solids (average operation) 20 2 10-12
Densator Area required for optimum conditions -
ft 2 /ton PPTD solids/24hrs - 660
aAt solids concentrations of greator than 4%, the slurries are
already at compression zone conditions; hence, free settling
rates are not obtained.
56
-------
Figure 13.
Steady state densator flow rates for various recycle
operating conditions
NOTE CALCULATIONS BASED UPON OBSERVED AVERAGE DENSATOR
FEED SLURRY SETTLING RATE OF 0.5 FT. PER HOUR USING
4 5 PPM POLYELECTROLyTE FLOCCULANT
HOLLYWOOD DENSATOR
DIAMETER 2SFT.
. 3% SOLIDS FEED -
16% SOLIDS U F LOW
3% SOLIDS FEED -
10% SOLIDS UFLOW
SOLIDS FEED
SOLIDS UFLOW
MAXIMUM DENSATOR FEED FLOW RANGE
1
I I
50
40
V
U-
-o
30 -
0
w
E
0
2O -
I I
I I
10
0
0 25
50 75
Densator Influent Flow Rote GPM
100 125
-------
Table 12. DENSATOR RECYCLE SLUDGE ANALYSES
TREATING PROCTOR NO.. 2 DRAINAGE
Constituent
Weight% -
Calculated
omFonent
co position
Component
-
Weight
percent
Aluminum
6.65
CaSOL 4 2H 2 0
20.1
Iron
30.85
Al(O}-1) 3
21.2
Calcium
4.67
Fe(OH) 3 (30%
of
Fe)
17.8
Sodium
0.02
Fe(OH)2(70
of
Fe)
35.0
Sulfate
.
12.06
Total
94.1
Yellowboy
thickener Infilco densator
process recycle process 2
Plant Effluent
gal/l . .000 gal. feed 927 992.2
Sludge Production
gal/l O00 gal. feed 73 7.8
Sludge volume reduction, % - - 89.0
Solids content of dis- 2 16.0
charged sludge, %
Assuming the neutralized feed to the thickener contains 1,500 ppm
solids.
2 Assuming recycle ratios of 30:1 by weight (recycle sludge/new
sludge) and 1:4.25 by volume (recycle sludge/raw water).
Simple material balance calculations readily indicate the impracti-
cality of the recycle concept when applied to weakly mineralized
waters. The sludge recycle load to maintain 3-4 percent solids in
the reaction zone would be unrealistic. The recycle system is
susceptible to wide fluctuations of performance as feed water
characteristics change due to rain or events in the originating
mine.
CONCLUSIONS OF DENSATOR STUDiES
1. The Densator functions efficiently as a multipurpose unit to
treat coal mine drainage and to develop a high density sludge.
58
-------
4. The recycle treatment process functions most effectively when
treating highly mineralized waters having a high Fe 2 /Fe 3 ratio.
A decrease in this ratio and in total iron loading tends to de-
crease sludge density.
3. Sludge densities of 10 to 16 percent solids were obtained con-
sistently when waters with high ferrous iron ratios are processed.
Intermittent samples showed higher density levels.
4. The sludge produced has excellent dewatering characteristics as
related to either vacuum filtration or sludge drying bed operation.
The settling rate of the sludge found in the primary and secondary
zone of the Densator is relatively low (about 0.5 ft/hr) and
becomes the primary criterion for Densator sizing.
5 Under the conditions of this test, flocculants at high dosage
levels were effective and necessary to maintain reasonable flow
rates. Weakly anionic high molecular weight polymers were found
most effective.
6. Densator area requirements are about 500-600 ft 2 /ton precipitate
solids per 24 hours.
7. Difficulties in maintaining high density underfiows were experi-
enced when wide fluctuations in total and ferrous iron content
of the feed drainage occurred.
59
-------
SECTION IX
TREATMENT OF COAL MINE DRAINAGE WITH CAUSTIC SODA
The tests using sodium hydroxide to treat drainage waters were
carried out with the conventional Tt yellowboy procedure (Process 10,
see Table 15, Appendix C) . Each of the four waters was treated at
flow rates ranging from 87 to 292 gpm. As with the calcined lime
studies, the water temperatures were low, frequently near freezing.
The reagent was supplied as a 50 percent solution in 55 gallon
drums. The undiluted reagent was added to the flash mixer and
controlled manually through a variable-speed, positive displace-
ment pump to a head tank with return overflow. The head tank pro-
vided improved flow control at very low flow volumes.
This reagent is dangerous to handle, requiring gloves and eye pro-
tection. The drums utilized to ship sodium hydroxide are not
pressure tested. Safety considerations prohibit the use of air
pressure to discharge the solution from the drums. This reagent
freezes at relatively high temperatures (54°F), a factor which
frequently required the drums to be heated to allow transfer of
the reagent. Commonly, shipments were received in a solid condition.
The reagent drums tend to develop a concentrated slurry in the
bottom which can block the delivery system, create difficulties
in total recovery of the reagent from the drum and lead to some
deviation in concentration which complicates treatment control.
The reagent requirements were estimated from acidity and flow of
the raw water with operating control adjustments made manually
from pH measurements in the flash mixer and oxidation tanks. Auto-
matic addition of caustic soda solution is quite feasible. During
tests at large flow volumes (200 gpm) of highly mineralized waters,
the consumption of reagent was so great that a full time helper was
necessary to change drums. In operations utilizing more than one
drum per shift, the labor costs to handle the reagent supply become
excessive. In such instances large, expensive, storage facilities
would be required. These requirements (capital costs and labor)
for delivery, storage and handling of this reagent further add to
its high utilization costs. However, the use of a purchased,
aqueous reagent does minimize equipment and labor requirements for
reagent preparation.
The detailed results of these studies are given in Appendix C,
Tables 15 through 18. The reagent effectiveness and control capa
bility was satisfactory while treating the lower mineralized waters
(Tyler Run and Bennetts Branch). However, in the tests with the
more severely polluted waters the tendencies observed with hydrated
60
-------
and calcined lime toward excess alkali consumption resulting from
aluminum and magnesium reactions become pronounced. Utilizing manual
control, the pH variation in the oxidation tanks with flow and acidity
did not show the trend as did lime products under automatic control.
The pH deviations were smaller (0.77-4.27 - compare with Table 10).
Controlled, step-wise addition of caustic soda as described with
lime product reagents is indicated. The extreme difficulty of main-
taining a pH below 7.0, with acceptable iron and acidity levels in
the effluent, was evidenced by these tests. Although the ease of
control with sodium hydroxide is excellent, there is a danger of
overtreatment with subsequent damage to receiving streams. This
potential hazard requires constant personal monitoring or elaborate
automatic monitoring and control equipment.
The reagent consumption tended to be less than theoretical (66 to
125%) with the low acidity waters while maintaining a pH in the low
six levels. Under these conditions a small acidity value remained
although final iron concentrations were satisfactory. The consump-
tion excesses tended to increase with flow rates, contrary to the
experience with lime products. With the higher acidity waters, the
reagent consumption increased drastically (82 to 175%) in seeking
to maintain an alkalinity value in the effluents. These large
excesses beyond the measured alkalinity levels must represent side
reactions involving aluminum and magensium, In test 19-71, the
final effluent persisted with an acidity level (7 to 22 mg/i)
despite a reagent consumption of 131 percent. The iron content
of the effluent was consistently satisfactory. The effluents
were sparkling clear, ranging between 4 to 22 mg/i suspended solids
with turbidity levels between 2 and 10 jtu.
As with lime products, the sludge production increased with acidity
of the feed waters. The density (volume/weight percent) of settled
sludge (two hours) was somewhat lower than with lime products:
Tyler Run - 0.15, Bennetts Branch - 0.25 to 0.30, Proctor No. 1 -
0.19 to 0.28, and Proctor No. 2 - 0.61 to 0.76. The dry solids
production (lbs/l000 gallons) was definitely lower than with lime
produets: Tyler Run - 0.37 compared with 0.55, Bennett s Branch -
0.72 to 0.88 compared with 0.23 to 1.67, Proctor No. 1 - 2.16 to
3.28 compared with 2.7 to 4.9, and Proctor No. 2 - 10.1 to 12.8
compared with 11.7 to 19.1 The settled sludge volume (percent)
was greater than with lime products for the low acidity waters
(3.0 to 3.5) but less voluminous with the high acid waters (14.0
to 20.0), compared with a maximum of 29.6 with calcined lime.
The sludges displayed settling behavior described as Type I or
II depending on the raw water acidity. These sludges also tended
to gel on standing. Table 24 provides sludge settling data. The
shortest retention time utilized in the oxidation tanks, providing
0.28 minutes/ppm iron II, was adequate. The oxidation rates
-------
observed were similar to those with lime products. The oxidation-
reduction potentials measured in the oxidation tank were somewhat
higher than with lime reagents, especially in the lower acidity
waters (128540 millivolts).
The sulfate content of Tyler Run and Bennetts Branch effluents was
reduced to less than 25 percent of the feed water. The conductivity
of the effluents was doubled (171 to 258%) in the low acidity waters,
with a smaller increase of Proctor No. 1 effluent (113%).
The treatment of CMD with caustic soda is characterized by its rapid
reaction rate and quick responses, necessitating careful control.
It is an expensive reagent, dangerous to handle, which may require
special storage facilities. Its use eliminates reagent preparation
equipment. The weight of sludge produced is somewhat less than
formed with lime. This is attributed to the lack of precipitation of
calcium sulfate. The sodium hydroxide added to the several waters
as a neutralization reagent would result in a sodium concentration
between 45 and 742 mg/i (see Table 17, Appendix C). Tests 1-71
and 2-72 gave treated water effluent levels of 59 and 49 mg/l
respectively in contrast to sodium additions of 57 and 63 mg/i.
Sodium levels found in the raw waters ranged between 4 and 49 mg/I
(see Table 2, Appendix A). Otherwise the overall response is
similar to that observed with lime product reagents. its greatest
advantage lies with emergency use to prevent discharge of unaccept-
able quality waters.
62
-------
SECTION X
TREATMENT OF COAL MINE DRAINAGE WITH SODA ASU
The reactions with sodium carbonate were carried out with its addi
tion to the flash mixer as a dilute solution prepared in the dissol
ver (see Appendix B, Figures 8 and 9 and Table 20 in Appendix C).
The high solubility of soda ash (700 451000 gm/lOU ml) permits
concentration control. Although caution must be exerted, the hand-
ling danger is less than experienced with caustic soda. Unless fed
as a solid, dissolution equipment is necessary. Each of the four
waters was treated in flows ranging from 105 to 288 gpm. The water
temperatures ranged from 38-45°F, excepting Proctor No. 2 (50°F),
whose waters are collected in the mine. The system was controlled
by a pH probe located in the first compartment of the flash mixer.
Following iron oxidation, the solid-fluid separation was made in
either the settling lagoon (Process 10) or in the thickener (Process
11). The data are presented in Appendix C, Tables 20 through 24
for Process No. 10, and in Tables 25 through 29 for Process No. 11.
The reaction control, as evidenced by pH measured in the oxidation
tanks, was excellent with the extreme values occurring in less than
20 minute intervals. The improved uniformity results from the highly
buffered system. In maintaining these pH levels, the response of
the electrodes in the flash mixer led to the addition of excess
reagent, developing conditions favoring unwanted reactions. The
excess reagent did not remain available as free alkalinity. Even
with the very large reagent excesses employed, the buffering effect
did not allow the p 1- I to increase to unacceptable levels. Improper
dosages of soda ash can develop high pH levels, but these are less
likely to occur than with caustic soda. The pH ranges in the flash
mixer and oxidatjon:tank for these tests were:
pH Range
Test No. Flash Mixer Oxidation Tank
25-71 6.0 - 8.5 7.0 - 8.2
26-71 7.5 - 9.6 7.3 - 8.1
29-71 5.0 - 10.0 6.8 - 7.0
27-71 7.8 10.1 7.3 - 7.8
30-71 6.4 - 10.6 7.0 - 9.0
28-71 8.2 9.6 7.8 - 8.3
31-71 5.8 - 9.8 6.6 - 8.2
32-71 6.3 - 9.8 7.6 - 8.4
34-71 7.6 - 9.2 7.2 - 7.6
33-71 6.6 - 9.0 7.3 - 8.3
36-71 6.9 - 8.6 7.5 - 7.7
37-71 3.6 - 9.6 6.9 - 8.0
35-71 6.0 - 9.5 6.3 - 8.0
63
-------
The observed plant response was confirmed with a laboratory test using
Proctor No. 2 water (1,420 mg/i acidity). Four liters of this water
(requiring 6.0 grams of Na2CO3 based upon a neutralization equivalent
of 8.8377 x l0 lb/gal/mg/i acidity) was neutralized with dry, CP
laboratory reagent. Small increments of the solid reagent were
introduced over a 30 minute period until the solution reached a pH
of 8.3, simulating plant tests. The water temperature was 60°F.
Vigorous mechanical stirring served to introduce air into the
water for continuous oxidation. An addition of 10.7 grams sodium
carbonate was required. This represented an alkali-use efficiency
of about 56 percent.
In other laboratory tests, 400 ml of mine drainage (457 mg/i acidity)
was treated by the slow, drop-wise addition of 0.192 gm (theoretical)
of dry, CP, Na 2 CO 3 dissolved in 20 ml of distilled water. The sam-
ple, at 65°F, was subject to vigorous stirring. After 30 minutes
reaction time, the pH of the solution was 568. The addition of
50 percent more carbonate solution ( 96 mgm) increased the pH to
7.50, reflecting the buffering response. In another test with the
same water, a 20 percent excess of Na 2 CO 3 was reacted in the same
fashion. The solution was then boiled (disrupting the bicarbonate-
carbonate equilibria by driving off carbon dioxide) and cooled.
The final solution had a pH of 7.18, 31 mg/i alkalinity and con-
tained no detectable iron. Sodium carbonate will respond stoichio-
metrically with mine drainage when the carbonate equilibrium is
controlled. However, such response cannot be expected in industrial
practice. If used in mine drainage treatment, nearly 100 percent
excess reagent consumption should be anticipated.
Soda ash in the form of solid pellets, sometimes called ttprillsfl,
are used effectively in treating small water discharges during sur-
face mining by immersing the prills, contained in a wire basket, in
the draining stream. The slow rate of solubility from the prills
tends to make treatment self-regulating. However, with high iron
c oncentrations, the prills become coated with reaction products which
prevent solubilization. This practice during surface mining pro-
vides a responsive, handleable reagent and avoids the mechanical
difficulties associated with field use of powdered, hydrated lime.
Low reagent efficiency results during such applications.
The reagent consumption experienced during the plant tests ranged
from 146 to 313 percent. The excess increased with flow rates but
showed no particular trend with acidity levels, which probably
reflects the delayed response of the buffered system to pH changes.
The effluents did maintain an alkalinity value but the percentage
reagent utilized which was related to this contribution was small
and large excesses of reagent went into the treatment per se, as
shown below. The reagent representing effluent alkalinity decreases
with increasing acidity of the feed water.
64
-------
% Reagent Feed % Reagent
Representing Effluent Consumed for Treatment
Test No. Alkalinity Based on Theoretical
25-71 63.2 82.9
26-71 17.1 259.1
29-71 28.2 149.8
27-71 23.0 164.6
30-71 14.5 141.6
28-71 12.2 235.4
31-71 11.4 246.2
32-71 14.9 124.4
34-71 7..2 272.3
3371 4.0 256.1
The iron removal response was excellent with the effluents having
consistently less than 2 mg/l soluble iron. The effluents were
clear with suspended solids ranging between 3 and 26 mg/i and
turbidity levels between 2 and 20 jtu.
The sludge formed with soda ash settles to greater densities than
sludges resulting from lime products but does not approach densities
achieved with limestone. The two hour settled densities ranged from
0.37 to 2.1 weight/volume percent. The level of sludge production
is equivalent to that found with caustic soda (0.25-19.4 lbs dry
solids/l000 gal water treated) and less than the lime product sludges.
The corresponding sludge volume was decidedly less than caustic soda
and lime product sludges (0.3 to 11 volume percent) but was not as
favorable as the response from lii estone. The sludges may be de-
scribed as exhibiting Type I behavior but have better mobility and
are less inclined to gel. Typical settling rate curves are shown
in Figures 24 and 25 with other sludge data shown in Table 24.
The retention time provided for iron oxidation was excessive,
ranging from 0.78 to 72 minutes/mg/i iron II. The fastest oxida-
tion rate was at least 230 mg iron 11/1/hr. The oxidation-reduction
potentials were favorable within the limits of 84 to 383 millivolts.
The low iron levels in the effluents also reflect proper oxidation.
The use of sodium reagent in excess resulted in substantial increases
in dissolved solids as reflected by conductivity. The average in-
crease of about 450 micromhos was substantial (from 42 to 254 per-
cent). The sodium concentrations which would be developed in the
water with the reagent excesses employed ranged from 55 to 1713 mg/i.
This response was similar to that experienced with caustic soda
treatment.
The treatment response of Proctor No. 2 waters when using the thick-
ener (Process 11) for solid-fluid separation confirmed the other
studies using the settling lagoon. The thickener settling operation
65
-------
was equivalent to similar tests with lime product treatment. The
reagent consumption was excessive (223 to 292 percent). The ef-
fluents were alkaline with 1.4, 28 0, and 13.1 percent (Tests 36-71,
37-71, and 35-71, respectively) of reagent added occurring as alka-
Unity in the effluents. The actual excess reagent consumed in the
reaction was uniform for the three tests: 219.8, 210.0, and 194.8
percent. The consistent effluent pH levels were between 7.56 and
7.71 reflecting good reagent control response. The pH variation
during the three runs was 2.50, 2.86, and 0.90. The iron levels
were satisfactory with an unexplained exception in test 36-71.
The two hour sludge densities were less than the Process 10 tests
(0.56-0.75 weight/volume percent) as was the sludge production
(5.1 to 6.9 lbs dry sludge/bOO gallons) and sludge volume (1.1 per-
cent). This response is inconsistent since the raw water loadings
were similar.
The oxidation retention times were less than the Process 10 studies
(0.52, 0.49, and 0.39 minutes/mg/l iron II respectively for the
three tests) but were adequate.
In all of the tests utilizing soda ash with the settling lagoon, the
effluent specifications were achieved; however, the reagent con-
sumption was excessive. Of the excess alkali, only a portion
occurred as effluent alkalinity. The remainder apparently being
consumed in other reactions, favored by the buffered conditions.
The sludge characteristics are somewhat more favorable than those
resulting from lime products or caustic soda, especially in resulting
sludge volumes. The process drastically increases dissolved solids
in the effluents.
66
-------
SECTION XI
TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC REAGENTS
There were three groups of studies made with dolomitic reagents con-
sidering the hydroxide, the oxide anhydride and the carbonate. Some
process variations were also introduced. All four waters were
treated at flow rates from 31 to 139 gpm. The data for the hy-
drated dolomite studies are given in Appendix C, Tables 29 through
33. The prepared hydroxide slurry was reacted in the flash mixer
followed by air oxidation and solid-fluid separation in the thick-
ener. The effluent was discharged through the settling lagoon.
The ph control probe was located in the flash mixer with a set
point of 7.0. The slurry preparation did not differ from the use
of hydrated lime. The calcined dolomite values are given in Tables
34 through 39 of Appendix C. This reagent was utilized in the same
flowsheet as the hydroxide, replacing the dissolver-mixer with a
slaker. The calcined dolomite responded differently than the an-
hydride lime during slaking. The slaking temperature was difficult
to maintain; the solid feed resisted wetting. A solid crust tended
to develop in the slaker reaction chamber and formed a thick mix
which moved with difficulty to the grit separator. Much dilution
water was reRuired to separate the grit. The hydration may have
been inadequate, producing a less reactive reagent. This behavior
was related to the inert, insoluble MgO component, since only the
calcium portion of the compound slakes under atmospheric pressure.
The dolomite was fed as larger-sized particles (Grade 4) to the
rotary reactor. In some tests, it was supplemented with pulverized
limestone. The data are presented in Appendix C, Tables 37 through
42.
The dolomitic hydrate yielded a satisfactory effluent with the highly
mineralized Proctor No. 2 waters. The reagent consumption was
reasonable although in one test a 146 percent consumption occurred.
Even this level was less excessive than experienced with high cal-
cium reagents.
The p H control of the reaction was good, maintaining a level above
7.5. The effluents had an excess alkalinity, which accounted for
only 1.3 and 1.6 percent of the reagent fe c ! in tests 58-71 and 60-71,
respectively. When the reagent was introduced in large excesses,
it did not appear as water alkalinity. The iron removal and effluent
clarity were excellent. The settling lagoon effluents showed a re-
duced magnesium content, but the calcium levels rose nearly four-fold.
It appears that magnesium was being removed from solution.
The settled sludge densities (0.58 to 1.3 weight/volume percent)
were consistent with those resulting from hydrated lime treatment,
67
-------
as were the sludge production levels (11.4-16.2 lbs dry solids/l,000
gallons) . The sludge volumes were somewhat greater (to 33 volume per-
cent) but were subject to reduction under compression. Settling rate
curves are shown in Figure 25, while other sludge data are given in
Table 24. The settling behavior of these sludges respond as Type I,
tending to gel. The sludge composition (Table 50) had unusually
high calcium but low magnesium and sulfate content. Sludge drain-
age rates were also consistent with other lime product reagents as
shown in Table 61. As discussed in Section XII there is strong
evidence for magnesium-sulfate-complex association, indicating
possible advantages for magnesium reagent in treating high sulfate-
containing waters. Response of these sludges in the thickener and
to filtration is cited subsequently in reference to the calcined
dolomite tests.
The retention time utilized during iron oxidation (0.61-0.90
minutesJmg iron IT/I) was probably excessive. The oxidation rates
exceeded 263-297 mg/1/hr, while the oxidation-reduction potential
ranged between 202 and 229 millivolts. The settling lagoon maia-
tamed a pH consistent with the oxidation tank effluent.
The calcined dolomite also gave satisfactory responses with all
four waters using low flow rates. It appears that stoichiometric
dosages of calcined dolomite are feasible, assuming reaction with
only the calcium portion of the compound. The reagent consumption
levels were consistent, with near theoretical values being attained
for tests 61-71 (low acidity water) and 64-71 (high acidity water at
a low flow rate). However, tests 62-71 utilized 186 percent theoretical
reagent while test 63-71 was atypical (605 percent excess), being
unexplainably out-of-control. These last two responses must relate
to reaction time in the flash mixer, pH probe location or possibly
to unobserved mechanical problems.
In test 63-71, utilizing the control set point of 7.0, reaction
control difficulty was experienced throughout the test with the pH
in the oxidation tanks rising to 9.5 for over three hours. The
system was mechanically and chemically responsive. The control
point was lowered in increments to 5.3 with the oxidation tank
effluent pH slowly declining to 8.0 and finally to 7.5 near com-
pletion of the test. The reagent feed rate also approached the-
oretical with the control change. The test illustrated control
problems which could be costly in unattended plants and lead to
dangerous overtreatment with reactive reagents.
These tests showed greater variation in the alkalinity of the oxi-
dation tank effluent than with the hydrate reagent. Each test having
reagent excesses did not develop an alkalinity. The two (61-71 and
64-71) attaining an effluent alkalinity represented 15.6 and 4.2
68
-------
percent of the reagent fed. The iron reduction was satisfactory.
Reactions with these reagents continued throughout the process as
there was a rise of pH levels in the settling lagoon effluents.
The settled sludge density confirmed the values from the hydrated
dolomite tests, as did the sludge production level. The settled
sludge volumes were less than resulted with hydrated dolomite and
more consistent with lime product reagents. The sludges showed
Type I settling behavior and would gel. Other sludge settling data
including thickener observations are given in Table 34. The chemi
cal analyses of these sludges used in filtration tests are given in
Table 49. They are high in calcium and magnesium but generally
lower in sulfate content. The filtration data generally agrees
with that from lime product sludge (see Tables 51 and 55). Sludge
drainage rates (drainometer) are reported in Table 60 indicating
their bulk densities to be slightly lower than hydrated lime
sludges.
The iron oxidation levels were satisfactory at the retention times
utilized (0.76-82 minutes/mg iron 11/1), covering the range 3 to
303 mg iron/i/hr. The oxidation-reduction potential ranged between
197 and 293 millivolts.
The dolomite tests indicated this reagent to be less responsive than
high calcium limestone, but capable of developing higher pH levels
in the effluent. With the time allotted to study this reagent, it
was not possible to explore all the parameters of the rotary reactor
and stone properties (only one particle size was used) which control
the reaction. The reagent provided adequate treatment for Tyler Run
waters in the rotary reactor at a flow of 89 gpm. The feed water
had little iron (< 2 mg/I) which made iron oxidation unnecessary.
The reactor effluent maintained a pH above 7.0, a reduced iron con-
tent, and an excess alkalinity. The alkalinity added to the efflu-
ent represented 58 percent of the neutralizing requirement. The
estimated reagent consumption was excessive since equilibrium con-
ditions were not established. This unreacted reagent was not solu-
bilized but reported to the sludge. With appropriate control,
reasonable consumption could be expected. Under these conditions
reagent cost would not exceed 2 cents/bOO gallons.
The sludge behavior responded as those produced in limestone reactions
(Type III) maintaining a slight turbidity. The sludge settled
rapidly to a high density (9.1 weight/volume percent) and small
volume (0.4 percent). The sludge production was 3.0 lbs dry solids
per 1000 gallons of water. This sludge contained excess dolomite
fines. This sludge formed from a reactor effluent containing
600 mg/i suspended solids. The settling lagoon effluent did not
69
-------
change in quality from the reactor effluent, indicating litt1c con-
tinuing reaction. This response was different than observed with
high calcium limestone.
The reduced reactivity of dolomite was indicated in test 51-71 when
the flow rate was increased about 50 percent. Despite the available
reactive reagent surface experienced in test 50-71, effluent quality
was not maintained (pH 5.7) although 85 percent of the acidity was
neutralized. The suspended solids in the reactor effluent also de-
creased. Subsequent experience has indicated satisfactory treatment
is feasible by larger reactor loadings and control of reagent particle
size. In this test, an acceptable effluent was attained by the
inttoduction into the reactor of a dilute, pulverized, high--calcium,
limestone slurry--thus utilizing a mixed reagent. The effluent p11
increased to above seven and an alkalinity developed. The intro-
duced pulverized limestone accounted for 78 percent of the reagent
requirements, the remainder being developed from the dolomite,
showing further evidence of the relative reactivity between these
two reagents. The reactions continued in the settling lagoon, as
previously experienced with limestone products. The resulting
sludge production was reduced (1.1 lbs dry solids/l000 gallons),
as was settled sludge density (3.4 weight/volume percent) but the
sludge volume remained constant. Flocculant addition was successful
in improving effluent clarity.
Confirmation of these results using mixed reagents was attained in
tests 53-71 and 54-71 with Bennetts Branch drainage of slightly
different composition than Tyler Run. However, when a higher acidity
and iron II water (Proctor No. 1) was treated, the reactor effluent
pH was below seven and maintaining alkalinity was inconsistent.
Only about half the ferrous iron had oxidized and been removed from
solution. Additional oxidation time was necessary which was accom-
plished by transferring the reactor effluent to the oxidation tanks
to provide a long retention time (10.2 and 4.2 minutes/mg iron LI/i,
respectively for the two tests). This increased response time
reduced the iron levels satisfactorily. The mild agitation in the
oxidation tanks also favored continuing neutralization reaction
which allowed development of an alkalinity in the treated waters.
The level of agitation in the oxidation tanks allowed some sludge
deposition with a resulting decrease in suspended solids, as in test
54 from 1300 to 571 ppm.
These studies show the reduced neutralization reactivity level of
dolomite in respect to high calcium limestone. Waters with acidity
levels above 300 mg/i may be treated satisfactorily but only with
great increases in reaction time. As cited in Section XII the
reaction times necessary with relatively insoluble carbonate reagents
is extensively controlled by those parameters with establish contact
70
-------
between available reactive solid surface and diffusion levels
control carbon dioxide concentrations. The conditions maintained
for these tests utilizing the rotary reactor were inadequate when
utilizing dolomite but were satisfactory when applied with high cal-
ciuni limestone. The use of supplemental reagents, as pulverized
limestone, can be helpful in improving these reaction times.
The evaluation of dolomite-based reagents indicate their potential
use in treating coal mine drainage but point to major limitations.
In Pennsylvania, the higher cost of these reagents (mainly trans-
portation costs) would discourage their use; however, in locations
where the reagent supply sources are closer to the treatment plant,
their use may be indicated. Under good control, a near stoichio-
metric consumption of the hydrated and calcined products can be
expected based upon reaction of the calcium component. The more
costly pressure-hydrated product could be expected to achieve full
utilization but was not tested. The treatment effluent will meet
requirements but overall reaction control appeared more difficult
to attain. The effluents tended to maintain a slight turbidity
which can be controlled with flocculants. The natural dolomitic
carbonate was less reactive than high calcium limestone but did
permit attainment of higher pH levels. As with limestone, any
unreacted reagent reported to the sludge having minimum water
solubility. The use of mixed reagents offers advantages. Further
study should evaluate combined calcium-magnesium reagents to mini-
mize the gypsum deposition problems in high sulfate-containing
drainage.
71
-------
SECTION XII
TREATMENT OF COAL MINE DRAINAGE WITH LIMESTONE
The tendency to oversimplify processes to treat coal mine drainage
has led to overemphasis of the ratio of reagent cost to overall
treatment economics. Yet the selection of the reagent establishes
the system and to a significant extent controls total treatment
costs. The theoretical 1.92 cents per neutralization equivalent for
limestone, with its 2.56-fold advantage over quicklime and 3.83-fold
advantage over hydrated lime (see Table 2) has continued to be
viewed advantageously. Despite this and other advantages, pragmatic
process considerations have prevented the more extensive use of
limestone. These studies sought to develop a perspective of this
Situation.
Attempts have been made to use limestone. Tracy 6 17 in 1916
utilized -200 mesh pulverized stone seeking an economic product from
the sludge. Zurbuck 8 in 1963 developed a Water-powered drum reactor,
filled with limestone, to maintain alkaline quality of a remote forest
stream. Researchers at the U.S. Bureau of Mines 6 19. 20 subsequently
considered an electric powered, three-foot diameter, 24-foot long
rotary reactor using limestone for short duration tests. The necessity
for additional iron II oxidation and carbon dioxide removal was re-
ported. They favored using the reactor as an autogenous mill. Most
significantly, they reported the settled sludge volume to be one-
third that resulting from hydrated lime. Stone-use efficiency of
about fifty percent was suggested. The rotary reactor is not an
efficient conuninution device; the grinding response can be improved
by a ball mill operation 21 . The greater reaction efficiency from
using -400 mesh, high carbonate stone as well as response to other
stone properties have been reported 22 23 Such fine-sized stone is
not prepared commercially. The available pulverized product (-200
mesh) requires long reaction times. At some quarry locations, large
stocks of ungraded fines are available as waste by-products. To
attain desired reagent efficiency during CMD treatment, the recovery
and recycle of unused reagent is necessary. The low cost of lime-
stone has not suggested such recovery to be realistic.
In 1968 Calhoun 2t 25 described satisfactory results from the only
mine plant location using stone in a rotating reactor which was fed
with ferric waters. Stone hardness was reported to be the most
important parameter. Birch 26 reported combined use of stone and
hydrated lime. Holland, et al. 26 reported pilot plant studies Using
a tumbling mill to produce limestone fines for neutralization which
was followed with aeration. Hydrated lime was necessary to complete
the treatment. These authors conc1udeη aeration costs following
limestone neutralization are impractically high.
72
-------
More recent pilot studies by Wilmoth, et al. 29 utilized -200
mesh pulverized limestone to treat ferric waters, and further con-
firmed the continuing slow reactions. Excessive stone quantities
were required. Other valuable studies with limestone were reported
by Gehm ° using an u flow stone reactor system to treat pickle
liquor; iloak, et al. also treated pickle liquor with limestone.
Other authors studying limestone neutralization reactions include:
Khavskii et al. 32 , Jarrett and Lountz 3 , Yen 35 , Clifford and Snarley 37 ,
Armco Steel Corporation 38 , and Glover, et al .tfo, Lu, Li2
CHEMICAL ASPECTS OF LIMESTONE TREATMENT OF COAL MINE DRAINAGE
In the general equations written for limestone neutralization of
coal mine drainage, the water solubility relationships existing
for the reactant and products are too often ignored:
CaCO 3 + H2SOLu CaSO 4 + 11 2 C0 3 (7)
3CaCO 3 + Fe2(SOLu) 3 + 6H 2 0 3CaSOLu + 2Fe(OH) 3
+ I-1 2 CO 3 (8)
3CaCO 3 + Al 2 (SO ) 3 + 6H 2 0 - - 3CaSO 4 + 2Al(OH) 3
+ 3H 2 C0 3 (9)
The small amount of CaCO 3 (0.014 gm/i), which dissolves in water,
reacts rapidly with mine water. The continued neutralization
reactions are governed by the limestone solubility rate which varies
with the physical properties of the stone, the presence of dissolved
salts, and the carbon dioxide partial pressure. There is also some
reaction at the solid limestone surface which is governed by surface
area, surface reactivity, diffusion rates of reactants and products,
and reactant concentrations. Ohyama, et al .Lu 3 considered this
mechanism. As the pH rises above 4.5, the equilibrium which yields
an excess of hydroxyl ions become increasingly important. In this
same p1-I region, the solubility products of the hydrated ferric and
aluminum compounds are exceeded and their precipitation begins.
Ferrous compounds are not precipitated until higher pH levels develop.
Calcium sulfate, another reaction product, also has limiting solu-
bility and plays a unique role in this process. Gypsum tends to
deposit and grow, though its solubility may not be exceeded. Local-
ized concentrations of gypsum, especially on the limestone surface,
are undoubtedly high. Carbonic acid, subject to dissociation, acts
as a buffer to maintain a weakly acid environment. Carbon dioxide
can be stripped from such waters allowing the pH to rise.
These equilibria suggest overall slow, continuing responses,
especially when fine (near-colloidal) calcium carbonate particles
are present. They also indicate why mechanical effects tend to
become controlling.
73
-------
The overall chemistry of this system is complex, especially the
solution chemistry of iron between p 1 -I 2 and 10. Of concern are:
the solubility products of iron, calcium and magnesium salts (sulfates,
carbonates, bicarbonates and hydroxides) as well as carbon dioxide
arid oxygen; the interactions of three physical states (and the re-
sponse of partial pressures, complexes, ionic strength, particle
size, nucleation); and reaction conditions (temperature, contact
times, concentrations, oxidation state, equipment configurations)
Some appropriate solubility data are cited which indicate a wide
range for the Ca 2 and SO 2 ions as a function of carbon dioxide
partial pressure and the solid phases: calcite, dolomite and gyp-
sum. Shterning and Frolova, 194512, give solubilities of the
CaC0 3 -CaS0 -C0 2 -H 2 0 system, while Frear and Johnson, 192512, give
more detailed data at higher partial pressures of carbon dioxide.
Cameron and Seidell, 190612, indicate a negligible effect of cal-
cium hydroxide on the solubility of gypsum. Gypsum solubility is
not effected by low concentrationS (1000 ppm) of Ca(HCO 3 ) 2 , but may
be decreased at higher levels. Table 13 describes the C0 2 -CaCO 3
system. Calcite solubility is largely a function of concentrations
of HCO 3 - ion, and the partial pressure of carbon dioxide. Table 14
indicates the variation in gypsum solubility with temperature.
Table 15 shows the response of magnesium on calcium carbonate solu-
bility. Yanatevas (Table 15) and Garre1Ts 5 D data both suggest a
complexing effect between magnesium and sulfate ions which enhances
solubility of the latter. This effect suggests differences between
the use of dolomite and calcium carbonate. In particular, CaSO 4
precipitation may be significantly reduced when using dolomite.
This has significance when treating severely loaded CMD. These
waters often cause difficulty by the precipitation of large amounts
of gypsum which scales pipes, reaction vessels, agitation blades,
etc. Gypsum coating should not be anticipated with waters contain-
ing less than 1400 ppm SO 4 2 Ferric salt precipitation is inde-
pendent of calcium salt phenomena.
To produce a pH stable plant effluent in C1-ID treatment, the neutrali-
zation of free acid and hydrolyzable salts with their subsequent pre-
cipitation must be complete and the excess carbon dioxide eliminated.
This stability must include the sulfate and bisulfate complexes and
acid generated by the iron and aluminum salt hydrolysis. Ionic forms
of some iron compounds typical of CMD, and their equilibria formulae,
are cited by Pourbaix 44 , An adapted potential-pH equilibrium dia-
gram for the limestone-iron neutralization system is shown in
Figure 14. By proper control of p1-I and Eh, it is possible to form
ferrous hydroxide, ferric hydroxide, ferrous carbonate or mixtures
thereof. The region designated A in Figure 14 represents typical
treatment conditions, but the condition could be extended to include
region B while using hydroxide reagents. With limestone systems a
74
-------
Table 13. SOLUBILITY OF CALCIUM CARBONATE IN WATER IN
CONTACT WITH AIR (16°C)
(Combined data of Schiossing, Engel, and
Johnston-Seidelj and Linke) L2
Partial
Total
pressure
Total
Total
Total
CaCO 3
of CO 2 in
calcium
(HCQ 3 ) 2
Ca(HCO 3 ) 2
Equivalent
atmospheres
ppm
ppm
ppm
ppm
0.0005 30 89 121 75
0.0008 34 136 160 85
0.0033 55 166 221 137
0.0139 90 271 361 223
0.0282 119 361 480 296
0.0500 144 439 583 360
0.1422 214 648 862 533
0.2538 265 808 1,073 663
0.4167 313 960 1,273 782
0.5533 354 1,080 1,434 885
0.7297 391 1,190 1,581 997
1.0 432 1,325 1,957 1,086
2.0 566 1,725 2291 1,411
4.0 734 2,240 2,974 1,834
6.0 855 2,600 3,455 2,139
Range of Ionic Strength (0.0005 to 1.0 Atm CU 2 ) = 0.0022
to 0.0324.
Increase in Calcium Solubility (0.0005 to 1.0 Atm C0 2 ) =
432/30 + 14.4 X.
The usual partial pressure of carbon dioxide in the atmos-
phere is 0.00034.
75
-------
Table 14. SC)LUBILITY OF CALCIUM SULFATE DI-IJYDRATE
(CaSO+ H O GYPSUM) IN WATER
(Hulett and Allen, 1902)12
Temperature
°C
CaSO
ppm
Ca
equivalent
ppm
equ
SO
ivalent
,242
0
1,759
517
1
10
1,928
568
1
,360
18
2,016
595
1
,421
25
2,080
612
1
,468
30
2,090
615
1
,475
35
2,096
617
1
,479
55
100
2,009
1,619
502
47
1
1
,4l7
,142
Table 15. SOLUBILITIES IN TUE QUATERNARY
CaCO 3 + MgSO + CaSO + MgCO 3
(Yanateva, l95S)
SYSTEM
(at
Solubility of
0°C and 1.0
pressure
ions in ppm
atm partial
of C0 2 )
Solid
phases
present
Ca
Mg
(HCO 3 ) 2 S0 1 +
600
0
1,860
---
Calcite (CaCO 3 )
214
130
1,330
---
Dolomite (CaCO 3 1gC0 3 )
545
32
1,850
---
Calcite and Dolomite
---
540
2,790
---
MgS0 7H 2 O(M)
136
295
1,940
Dolomite + M
970
---
1,660
1,040
Calcite + Gypsum (CaS0 4 2I1 2 O)
1,040
39
1,900
1,170
Dolomite + Calcite + Gypsum
825
172
1,810
1,370
Dolomite + Gypsum
760
252
1,890
1,360
Dolomite + Gypsum
800
276
1,940
1,530
Dolomite + Gypsum
790
320
2,000
1,560
Dolomite + Gypsum + M
610
---
1,480
Gypsum
76
-------
Figure 14. A limestone-iron neutralization system
+0 8
+0.6
+0.4
1 -0.2
0.0
0.2
0.4
--0.6
- 0.8
1.0
44+
Fe oq
χff
oq
r n 1
o
o 0
U
E F.(OH) 3
CARBONIC ACID
DOMAIN
- 1.0
HCO
DOMAIN
2
4
6
8
12
14
77
-------
p t -I of 8 is seldom exceeded. However, the location of the siderite
field has potential of precipitating a more dense form of iron with-
out aeration. The area of the siderite field can be increased with
dissolved carbonate concentration and partial pressure of carbon
dioxide.
Freshly precipitated siderite oxidizes more rapidly than ferrous
ion in solution. If siderite were precipitated in the absence of
air in a slightly reducing system, then aerated, a more desirable
reaction sequence could develop. The rapid oxidation of insoluble
ferrous compounds agrees with the observations of Lovell and
Stauffer 6 .
Similar equilibria data for the sulfur-water systemL +a and iron-
water system (non_carbonate) Th are available.
Considerations of the characteristics of the impurity sludges are
presented in Section XIV, including: solid-fluid separation, secon-
dary dewatering, and disposal. The sludge characteristics are deter-
mined by the reagent and processes used in the treatment. Limestone
treatment appears to result in an optimum situation forming a sludge
which is rapidly settled, dense, and readily dewatered. The treated
water tends to maintain a persistent turbidity after solid-fluid
separation which clears very slowly. This haze is attributed to
the presence of near-colloidal limestone particles, with the particles
eventually settling or solubilizing. It is also feasible that part
of the turbidity originates from colloidal silica in the limestone.
The mass of sludge produced during treatment relates to the pollution
loading of the raw waters, primarily the iron and aluminum content.
In some high sulfate (> 1400 ppm) waters, gypsum may also form.
Other sludge contents result from insoluble impurities in the reagent
and from excess reagent. The inefficient use of limestone increases
the mass of sludge to be handled and total process costs.
There is no reason to anticipate limestone requirements in excess
of the stoichiometric value: 8.3453 x l06 lbs of calcium carbonate
per mg/liter acidity per gallon of water treated. This value must
be adjusted to consider the stone purity and the efficiency with
which it is utilized. The later is dependent upon a myriad of
parameters to be indicated from these studies.
CONSIDERATIONS IN THE USE OF LIMESTONE
The chemical factors of concern include: reactant solubility rates,
precipitation and neutralization rates, iron oxidation rates (either
pre- or post-neutralization), and composition of the sludge, which
determines suspended solids settling rates, and sludge dewatering
78
-------
rates. These factors determine equiment selection and sizing, hence
capital costs. The equipment basically establishes reagent efficiency
and operating costs.
Anong the advantages in the use of limestone as a reagent are:
1. Lowest cost neutralization reagent - about O.002q per part
acidity per thousand gallons.
2. Minimal sludge volume is developed - usually less than one-third
volume from more soluble reagents.
3. The limestoneproduced sludges have high solids content.
4. The sludge tends to be granular and crystalline, making dewater-
ing operations less costly and more efficient.
5. It is impossible to overtreat the water, leading to minimum con-
trol problems.
6. Reagent availability is extensive over a wide geographical
distribution.
7. Limestone storage is favorable. The larger sizes do not require
protective enclosures. Pulverized stone, which requires bin
storage, handles well and is non-hygroscopic.
8. Simplicity of plant design.
9. Reduced plant maintenance and operation requirements.
10. Reduced safety hazards relative to reactive reagents.
11. Reaction capability indicate stoichiometric reagent requirements
need not exceed 20 percent theoretical.
There are limitations and disadvantage in the use of limestone.
Among them are:
1. Use is limited by the ferrous iron concentration in the feed
water. The limiting pH attainable, near 8.3, establishes
maximum iron oxidation rate.
2. Reactivity is lower than more soluble reagents. Surface coating
of the reagent by products reduces reactivity.
3. Settled sludge has clay-like character which may create handling
difficulties.
79
-------
4. The actual sludge weight produced (on a dry basis) may be greater
than with lime products since stone tends to be more impure.
Slower reactivity and lower use efficiency may introduce un-
reacted reagent to the sludge.
5. Near-colloidal-sized particles of stone and silica create tur-
bidity in plant effluents.
6. The limestone-produced sludge decreases in volume very little
under compression.
7. The carbon dioxide formed during reaction must be removed to
provide a stable pH.
8. Rotating reactors are noisy. They require power to turn the
drum and are inefficient in comminution.
The response of mine drainage waters to limestone treatment suggests
three types of feed waters, grouped by ferrous iron content. They
are listed in Table 16. The ferric iron acidity concentrations of
CMD do not appear to be limiting criteria. In other tests, not de-
tailed here, it was established that mine drainage containing up to
about 100 mg FeII/l could be satisfactorily treated in the rotary
limestone reactor at maximum flow without a separate iron oxidation
stage. The stoichiometric oxygen requirements for the oxidation of
this level of iron concentrations are closely met by the air satura-
tion level in water, assuming no replacement by additional oxygen to
the solution. However, at higher Fell levels, acceptable iron con-
centrations in the settling lagoon are not attained.
Various limestone reactor modes have been considered by Glover 2 .
Two modes of reacting limestone with CMD have been utilized which
differ primarily in feed stone size- -pulverized or larger sizes
(+ 1/4 inch). The pulverized reagent is mixed with the CMD in a
flash mixer in near stoichiometric quantities. Long reaction times
are required with vigorous agitation to maintain particle suspension.
The equipment should be housed. This mode tends to result in poor
reagent utilization.
Larger particle-sized reagents are introduced, in excess, to auto-
genous, attrition reactors. Only rotary reactor designs seem feasible.
The reaction time is short. The reactors provide vigorous movement
of stone and water but do not require weather protection or housing.
Much of the water movement results from gravity flow. Reactor
capabilities must permit desired reaction levels but avoid excessive
loss of unreacted stone.
80
-------
Table 16. POTENTIALITIES FOR LIMESTONE TREATMENT OF
MINE DRAINAGE WATERS
Ferrous
concentration
mg/i Response to limestone treatment
0-100 Effective treatment may be achieved with-
out pre- or post-neutralization iron
oxidation.
100-500 May be effectively treated but requires
post-neutralization aeration and signifi-
cant reaction-retention time. Pre-
neutralization oxidation can reduce iron
II concentrations.
>500 Potential treatment is uncertain with
experience to date, unless combined with
pre-neutralization ferrous iron oxidation
to achieve the above ferrous levels.
Reagent cost favors the use of larger-sized stone. Although the
reactivity increases with decreasing particle size, the extent each
particle responds lessens. This phenomenon is associated with
particle surface coatings by reaction products and, the ease in
maintaining the surface active. With the coinmercially-available
pulverized stone sizes, reaction times in excess of twenty minutes
become essential to minimize loss of unreacted stone.
The power requirements between the two modes are utilized differently.
With the pulverized stone, the power required to maintain particle
suspension must also move large volumes of water which serves mini-
mal useful functions. With the larger stone reactor systems, the
power is more effectively utilized to create and maintain reactive
surfaces.
The rotary limestone reactors perform two principal functions, the
relative extent of each has not been defined: a) maintain freshly
exposed, reactive surfaces by attrition to remove insoluble reaction
products; and b) achieve an autogenous comininution of the stone to
produce fines of great reactivity. Additional functions also occur:
c) contacting the neutralization reactants to enhance reaction rates
at the solid-liquid interfaces which includes wetting of solid sur-
faces; d) contacting the oxidation reactants (dissolved iron II and
oxygen); e) contacting air (oxygen) and water to maintain an oxidizing
81
-------
environment; f) removing carbon dioxide from the water; g) con-
trolling the reaction-retention time; h) transporting the water through
the reactor, i) providing a suitable transport and distribution of
the stone in the reactor to optimize the other functions.
The capability with limestone treatment to achieve an effluent with
an alkalinity value is the main control requirement. The occurrences
which do not achieve an alkaline effluent or tend toward inadequate
reagent utilization may be modulated, but tend to have counteracting
responses. They may be enumerated:
A. In pulverized stone systems:
1. Increase reagent addition.
2. Decrease stone particle size.
3. Use stone of greater purity and reactivity.
4. Increase reaction time.
B. In larger-sized stone systems:
1. Increase rotation speed.
2. Increase stone loading of reactor.
3. Add attrition devices to reactor.
4. Use smaller top size stone.
5. Use softer stone.
6. Use stone with higher alkali equivalent.
Should these steps fail to be adequate, supplemental additions of a
more reactive reagent may be employed after the limestone reaction.
A greater reagent efficiency may result by achieving only partial
neutralization with limestone and completing the treatment with
hydrated lime slurry.
Even with an adequate effluent alkalinity, a depressed pH may result
from carbon dioxide solution. This gas should be stripped from the
effluent by subsequent mixing or aeration. During this gas strip-
ping, the alkalinity and pH of the waters may decrease but an excess
alkalinity level must be maintained. Should an acidity develop,
inadequate available alkalinity has been introduced to the water.
Should an excess ferrous iron concentration persist, the following
actions may be considered:
1. Increase oxygen transfer rate. This dissolved oxygen level in
the effluent should be maintained at least 5 ppm.
2. Increase reaction-retention time. Iron II oxidation rates are
relatively slow at the pH levels prevailing under these conditions.
3. Increase pH of the system by the addition of further limestone
or other alkali supplement.
82
-------
4. Add a chemical oxidant such as chlorine gas or calcium hypo-
chlorite.
5. Decrease ferrous iron concentration of feed water by dilution
or pre-neutralization oxidation.
With the attainment of suitable water quality after separation of the
sludge, effluent turbidity may persist. This can be controlled with poly-
meric flocculants if necessary, but it isusually non-deleterious.
PLANT OPERATING OBSERVATIONS
Limestone neutralization studies were made with the rotary reactor
employing feed stone with varied top sizes from 1/4-inch to three-
inches (see Appendix B, Figure 16). In some tests pulverized lime-
stone was also added to the reactor as a slurry. In other studies
pulverized stone slurry was reacted in the flash mixer and the
Densator. Tests were made with high calcium as well as dolomitic
stone from several geographical origins and were applied to the
several waters available.
The rotary reactor was designed to meet the multiple functions cited
(see Appendix B, Figures 6, 7, and 16). Vigorous movement of water
and stone was achieved within the reactor by horizontal lifters
which enhance autogenous comininution, mixing, and gas exchange.
The rotational movement of the drum carried the Stone and water
load to about ten oclock before discharge.
The 2 mm wedge wire screen, covering the discharge aperture, was
effective in preventing loss of all but the finest size stone
particles. The most excessive loss of unreacted stone, as well
as potential sanding problems in pipes, was prevented by this screen.
Some of the fine particles which do escape were trapped in the efflu-
ent sump while others are carried with the effluent flow, providing
additional reaction time while reaching the settling lagoon. Physi-
cal characteristics of limestone and dolomite sludges found in the
bottom of the settling lagoon, as the water level was dropped, are
given in Table 17.
The sample stations may be located in Figure 11, Appendix B. The
larger particles would be expected to settle in the first region of
the lagoon, with decreasing diameters as distance from the weir in-
creased. The particle sizes were small and their distribution rela-
tively uniform. The bulk density and solids content were surprisingly
high - showing a significant density increase under compression in
comparison with 24-hour settled sludge (Table 24). X-ray diffrac-
tion patterns of these sludges showed calcite, dolomite, quartz, and
83
-------
Table 17. CHARACTERISTICS OF LIMESTONE SLUDGE
FROM SETTLING LAGOON
Sample 1
Sam 1 le 2
Sample
3
Sample location
Inches from weir
96
104
124
Bulk density - gm/mi
1.167
1.139
1.171
Solids Content
Wt. % @ 105°C
35.4
30.2
37.9
Size
Content
Direct
Weight
Percent
Size (mesh)
Sample 1
Sample 2
Sample
3
passed retained
100
2.2
1.4
4.7
100 200
23.2
39.1
26.3
200 325
24.5
47.9
25.2
325 400
19.2
8.4
27.4
400
30.9
3.2
16.4
and opal (c cristobalite), in addition to ferric oxyhydroxides. No
difficulty was experienced from the build-up of non-reactive inerts
in the reactor.
The systems used with pulverized stone permit no control over loss
of unreacted reagent; the stone reacts, settles in the reactor or
is carried with the water to the next unit operation. In two years
of intermittant operation of the rotary reactor, there was no
evidence of internal wear and corrosion. Rubber, ceramic, or other
abrasion-resistant liner material would further extend reactor life.
The control anticipated by varying the rotary reactor slope was in-
effective due to inadequate design of the feed end of the reactor.
The open design, which permits the extension of a vibrating
feeder and the influent pipe into the reactor, also permitted ex-
cessive splashing from the reactor at low slope angles and high
flows. The result was potential corrosion, and limitations in
the useful limit of fluid flow and horizontal reactor angle. The
horizontal reactor angle effects stone movement within the reactor,
but has negligible control over water retention time. The angles
available favored excessive movement of the stone to the exit end
of the reactor. The reactive capacity of the device is optimized
only if the stone charge maintains a uniform distribution throughout
84
-------
the length of the unit. The experience attained with this unit
suggests a horizontal, or even a shallow reverse angle to be pre-
ferrable. The slope variation capability provided in this research
reactor is not needed in an industrial unit. A minor design change
to a closed feed system would circumvent these difficulties and
enhance reactor performance.
The retention time within the rotary reactor is indicated in Figure
15 as measured by batch introduction of a soluble salt and monitor-
ing of the effluent conductivity as a function of time. The median
retention time appears at about 150 seconds regardless of rotation
speed, flow rate (up to 250 gpm), or reactor angle (3-4°). The re-
tention time in the pulverized limestone reactors is a function of
the reactor size and flow rate.
The belt drive of the reactor created difficulties in training the
belt. Changes in temperature, humidity, precipitation and load
caused the belt to creep, resulting in unreliability and undue
operator attention, thus limiting the testing program. A chain
drive is preferrable.
The power requirements for the rotary reactor are distributed be-
tween those required to turn the base reactor, to move the water,
and to abrade the stone. These are detailed in Figure 16. The
power requirements associated with the water movement do not appear
to vary appreciably with flow. They increase with stone loading
and change rapidly with reactor rotation speed between 9.6 and 16
rpm. With this unit, the power costs range between 6 and 14 cents/
hour or between 0.4 and 2.3 cents/1,000 gallons of water for the
conditions cited. Power consumption by the reactor reflects current
(amperes) levels drawn. These levels were monitored to protect
the motor from overload and also served as a rough indicator of
reactor stone loading. The range observed was from 12 to 24 amperes.
More effective power transmissions would be expected with a chain
drive.
In contrast, the two small mixers used in the flash mixer unit with
pulverized stone were each powered by a 1/2 HP motor and would re-
quire about 1.1 KW/hr or between 0.18 and 0.32 cents/1,000 gallons
of water. However, these mixers were designed to produce homogenous
solutions from two liquids (mine water and alkali slurries) and did
not provide adequate agitation to maintain pulverized limestone in
suspension. The flash mixer had a median retention time between
2 and 15 minutes which was inadequate for pulverized limestone reac-
tions. It is probable that the power requirements of a properly
selected agitator in a flash mixer providing adequate retention time
for use with pulverized stone would be comparable or exceed those of
the rotary reactor since large volumes of water would be continuously
moved at high velocities.
85
-------
Figure 15.
Limestone reactor retention time
1 2 3 4 7
Time Differential (Minutes) after Introduction of Salt at Influent (0 Minutes)
WATER CURVE
F LOW
(gpni)
250
BENNETTS 0
BRANCH
PROCTOR 1 AA
ROTATION
(RPM)
9.6
PROCTOR i O-O
120 16
ANGLE
30
40
40
:1
1900
120 10
NOTE: For Bennetts Branch, 1000 micromhos
were added to the actual test values
in order to fit this curve on the same
graph.
1800
Sn
a
E
a
I
1100
0
1600
9
1500
1400.
1300
I
/
1200
I
/
0
o
8
86
-------
0.53
0.56
1.05
1.13
45. 6
33. b
1.57
0.56
1.31
8.96
0.75
1.49
9.47
0.19
1.59
10.01
0.83
1,67
11.33
0.94
1.89
Figure 16. Distribution of power requirements for limestone reactor
12.69 14.08 hr. . 1.7CKW - 1500 lb Stone
1,06 1.17 1000 gal.- 200 gpm
2.12 2.34 2 l 00 0gal. . lOCgpm
__ [ 1 Water
__ Stone
___ ___ Reactor
I I 43 Power Distribut on
:o 0 ;η
45 o ____
327 309
16.0
10.0
0
7y1
11.0
:
:
:
2°
29 6
:
- -
°°
Q.o
°
28 7
0
k
: .:
U
H
42.1
29.3
41.7
42
29
30.9
RPM
8.0 . 1
7.0
6.0
5.0
3.0-
2.0
1
-------
The effective capacity of the rotary reactor is related to achieve-
mont of each of the necessary functions in response to the water
nd tone characteristics Most critical is the reactor capability
to prov ide adequate reactive stone surface which is a function of
reactor stone load, of stone size distribution, of stone physical
ittrition, of stone distribution within the reactor, and of the
stone consumption. The reactor must operate with excess stone
rrescnt, but at an optimum and constant load. The power capability
rf this reactor will carry a load in excess of 1.5 tons of stone.
Since power consumption varies with stone load, the optimum load
nust be limited to process requirements. The stone feed rate must
balance the required stone consumption and must relate to the stone
feeding system. Stone consumption results from reaction and the
discharge of unreacted fines, with the latter determining reagent
utilization efficiency.
With the introduction of a maximum load of stone to the reactor,
the stone will distribute itself rapidly (within minutes) through-
out the reactor length, governed by movement of different size
particles through the three screens. The stone tends to move with
the water flow toward the effluent end. Reactor slope control
assists in establishing this rate; but appears to be effective only
at angles less than 3° from the horizontal. Other configurations
were not available for study as previously described.
The potential stone distribution within the reactor has extremes
ranging from complete loading in section 1 (with + threeinch stone)
to complete loading in section 4 (with - one-inch stone) . The
reactor design did not provide for exclusive use of pulverized
stone due to the very low retention time (less than three minutes)
which would result in excessive losses of reagent. In the first
condition, attrition and colnnhinution will be maximal in the first
compartment, but stone surface area will be minimal, with three-
fourths of the reactor being ineffective. With rotation time, the
rapid attrition of the sharp edges of the large stone permit particle
transfer through the screens to subsequent reactor compartments with
the particles assuming smooth surfaces and round edges. A gradual,
oven distribUtiOn of stone throughout the reactor is approached.
Larger total surface area of the stone is then achieved. However,
the attritiOfl -COTllIfliflUtiofl rate decreases as the initial stones
abrade tO smooth surfaces and redistribute themselves to a reduced
volume level throughout the reactor. With batch loading, it is
during this initial period that conditions permit the greatest loss
of unreacted stone. The attrition rate is also a function of particle
size since the mass of larger particles falling will offer the greater
impact for breakage. Thus, with increasing distribution and de-
creasing compartment loading in the reactor, the loss of unused
stone decreases. Subject to the water flow and acidity loading,
88
-------
treatment will remain satisfactory with increasing stone surface
area until the steadily decreasing attrition-comininution rate
reaches a level that the total available reagent surface becomes
inadequate. The production of a reactorstone consumption level
in excessof three pounds/minute was attainable. This would meet
the requirements of CMD water having 2,000 ppm acidity at a flow
rate of 300 gpm. This rate is beyond the design and hydraulic
capacity of the reactor with the existing end-feed design.
The other loading extreme (minus one-inch stone which moves rapidly
to reactor section four) would also limit reactor volume use and,
due to greatly reduced attrition, would limit production of reactive
stone surface despite large surface area. The proximity of this
section to the effluent port would provide functional retention
time of seconds, leading to inadequate reaction and excessive stone
loss.
The optimum stone loading operation would require a uniform stone
distribution throughout the reactor, controlled size distribution
in each reactor section, and a continuous, uniform stone feed rate
of specified size distribution. Ideally, the feed rate would pre-
cisely conform to the theoretical reagent requirements of the feed
water. These conditions will vary with stone properties, especially
hardness. Design of a reliable reactor feed system is feasible and
would limit labor requirements to filling the stone feed hopper and
maintenance.
Stone was introduced in the rotary reactor via a vibrating feeder
from an 1,800 lb hopper. The hopper was loaded by a mobile convey-
or belt. The stone feed rate to the reactor was adjustable by
varying the amount of vibration and the size of the hopper diaphram.
The reliability of feed control varied extensively with stone size
distribution and moisture content. Mixed sizes (- three-inch) were
least reliable, while narrow size ranges (especially 1/4- to one-
inch) were most satisfactory. Limited moisture content of the stone
tended to resist flow while operations during rain gave enhanced
flow. Some difficulties were experienced from freezing during inter-
mittant operation in cold weather. The stone feed equipment design
is optimum for a given product with a limited range of feed rates,
and will have decreased reliability with deviation from these speci
fications. The wide range of stone feeds and rates considered in-
creased the control difficulties.
The E-lollywood unit has been applied satisfactorily to CMD volumes
in excess of 200 gpm, with higher levels limited by the reactor
feed design. Evidence suggests a capacity of at least 400 gpm is
entirely feasible for this size of vessel. It would be recommended
that treatment capacity be increased by enlarging the drum diameter
and reaction response by increasing drum length.
89
-------
The economic, operating, and design criteria basic to the selection
of the limestone treatment in contrast to other reagents for a given
water is most complex. Figure 17 illustrates the alternate routes
for the limestonewater reaction.
The more significant design parameters for limestone treatment
given by unit operation follow:
Reagent Control
1. Consumption rate - lbs/hr.
2. Specifications for limestone grade, size, and hardness.
3. Type of storage (bins or piles).
4. Feeders and hopper design.
5. All weather provisions.
Limestone Neutralization Unit
1. Mode of operation (mixer or rotary mill).
2. Working volume and excess capacity allowance.
3. Weight and volume of limestone charge.
4. Pulp retention time vs. ferrous iron content.
5. Rotation or mixing requirements.
6. Gas exchange capability (oxidation and CO 2 removal).
7. Supplementary reagents required, e.g. Ca(OH) 2 .
8. Materials of construction.
9. Cold weather provisions.
Iron Oxidation Unit (CO 2 Elimination)
1. Type (tank or pond).
2. Function - pre- or post-neutralization oxidation, aeration
or bacterial.
3. Aeration capacity required by iron II concentration.
4. Retention time required.
Settling Lagoon(s)
1. Settling rate of solids.
2. Loading rate (feed flow rate and percent solids).
3. Range of allowable rising velocities (gpm/ft 2 ).
4. Sludge bulk density.
5. Sludge storage capacity.
6. Provision for sludge removal.
7. Baffles.
8. Multi-pond systems.
9. Cold-weather provisions.
Thickener
1. Settling rate of feed slurry (including recycle).
2. Optimum feed solids content.
90
-------
Figure 17.
Alternate unit operations for limestone neutralization
POST-N F TRALIZATION IRON OXIDATION
TREATED WATER
- SLUDGE
SOLID-FLUID SEPARATION
CO 2
02
- SLUDGE
PRE-NEUTRALIZATION IRON OXIDATION
(BACTERIA-CHEMICAL-CATALYTIC)
LARGE SIZED
L
ROTARY
LIMESTONE REACTOR
I .
FLASH MIXER REACTOR
02
* CO 2
91
-------
3. Maximum underf low solids contnet.
4. Allowable rising velocity range - gpm/ft 2 .
5. Ratio of recycle volume.
6. Operating characteristics - w/wo flocculant.
7. Feed flow rate.
8. Temperature effects.
Filter
1. Solids filtration rate - lbs/24 hrs/ft 2 filter area.
2. Liquid filtration rate - gpm/ft 2 filter area.
3. Type filter, filter area, filter medium.
4. Feed load - lbs solids/24 hrs/l,0 00 gallons CMD/ppm acidity.
5. Filter aid (pre-coat) requirement.
6. Flocculant requirement - lbs/l,00 0 gallons feed slurry.
7. Percent solids in filter feed.
8. Percent solids jn filter cake.
9. Materials of construction.
10. Vacuum requirements.
Sludge Drying Basin
1. Bed design including porosity and drainage rates, area and
depth.
2. Percent solids in feed slurry and dried solids.
3. Settling rate of feed slurry.
4. Mass of dried solids per 1,000 gallons feed water.
5. Precipitation and evaporation rate of the immediate area.
ROTARY LIMESTONE REACTOR DATA
The capability of the limestone neutralization process was demon-.
strated under severe conditions utilizing highly ferrous Proctor
No. 2 waters. Water quality specifications were attained. Reason-
able reagent use efficiency was experienced. Any coal mine drain-
age of lesser pollution loading will also respond satisfactorily.
The iron ii content was biochemically oxidized prior to neutrali-
zation and the waters were neutralized in the limestone reactor.
The overall flowsheet is detailed in Figure 18 and the supporting
conditions and analyses given in Table 18. The bacteria inocula-
tion is used only to initiate the biochemical oxidation system and
need not be maintained. The weight of limestone was evaluated
volumetricallY. The weight of given bulk volume of reagent was
established. Stone was added through a vibrating feeder with nearly
constant manual attention. The suspended solids in the effluent
of the limestone reactor for this test were higher than normal,
the value usually being closer 2,000/mg/i.
The limestone reactor sludge solids settled rapidly, with one-third
final sludge volume being achieved within three minutes and nearly
92
-------
Figure 18.
Biochemical iron oxidation - limestone neutralization
system
PROCTOR NO. 2 PUMP
FEED PUMP ROCK FILTER
LIMESTONE
RAW WATER
TREATED EFFLUENT
DEWATER SLUDGE TO LANDFILL
PUMP
0
INOCULATION TANKS
FACE BIOLOGICAL REACTOR
F LOCCULANT
SLUDGE DRYING BASIN
-------
Pump
Pump
Sludge Slurry
1.1 gp.
3.0 Ibs/ p, d y ioIid
Sludge 1
15% solids
Sludge Drying Basin
H 1!!
t
- -
S Ti Treated Wafer
- - - - S AIk lini 84
Settling Lagoon
Figure 18 (continued). Schematic drawing og biological iron-oxidation
limestone neutrali:ation system
48 gpm
60 gpm
Holding Lagoon
Fe(ll) 355
16
Surface Biological Reactor
Fe(Il) 155
,,,,Air
Mine
p !1 3.0 .DI) 43! Acidity 1584
a
262
I .
ION QIIOLZING IACILLI
Pump
12 gpm
P.(IIJ 431
Oxidation Tanks
6.5
F. Limestone Reactor
C .
-------
Table 18. NEUTRALIZATION OF CMD BY LIMESTONE IN THE ROTARY REACTOR
Neutralization of CMD by Limestone in the Rotary Reactor
Test Period 9/210/71
Reactor Conditions
Speed - 13 rpm; Power consumed - 7.98 KWH/Hr. 0 19-23 amperes: Slope - 4 5
Limestone Reagent
Hardness
Los Angeles Rattler Test - B Grading
100 revolutions 5.6% passing
500 revolutions - 25,7% passing
Size Anal sis :
Screen ize osrect cumulative Wt.%
Passed Retaine Wt.% Wetained Passing
1/2 inch 39,0 39.0 100.0
3/8 inch 10.0 49.0 61.0
8 mesh 29.5 78.5 51.0
60 mesh 16.3 94.8 21.5
200 mesh 2.7 97.5 5.2
2.5 100.0 2.5
Source - Proctor So. 2
Flow - 60 gallons/minute
Process Temperature 18 °C
Theoretical limestone requirements - 0.0127 lb/gal.
Theoretical requirement of stone reagent utilized 0.0152 lb/gal. or
Flocculant addition S ppm Atlasept lAl as 0.2% wt. solution.
Control :
Assuming transfer of about 2,600 gallons of sludge at 9.5% wt./vol. solids daily from
settling lagoon to sludge drying basin over period of one hour, would add about 3 inches
of sludge depth (assuming 1400 ft 2 area in a compartment). With the three compartment
sludge drying basin, the transfer cycle would be: daily transfer of three inches of
sludge for two consecutive days to each compartment.
Day No.: 1 2 3 4 5 6 7 8 9 10 11 12
Compartment No. : 1 1 2 2 3 3 1 1 2 2 3 3
Experience shows this would result in 1.50 inches/6 days of sludge level at solids con-
tent between 15 and 45 wt. % solids depending on drainage rate and weather: i.e., upon
dewatering the settled solids reduced to about 1/4 their original volume. This sludge
accumulation rate would require emoval of sludge from the drying basin ahout every 144
days representing about 600 yd . 4 or 450 ton (about 110 ib./ft 3 ) . The sludge drying
basin was emptied with a 2 yd. bucket on a model 922 caterpiller unit on cats. Two
trucks with a total capacity of 15 ton carried the dewaterad sludge about 1/2 mile for
landfill. The sludge drying and volume reduction continues, reaching a final volume
between 25-33 percent of the dewatered sludge and a water content related to tho ambiailt
conditions (110%).
a 0035 lb. dry sludge/gallon or 2.12 11./mis.
6 5ettled sludge volume - 3.0 vol. % in 10 minutes.
0.03 gal. sludge/gal. process influent or 1.8 gal./minute.
Composite sample during sludge transfer, value represents dilution with transfer
water. Samples taken from lagoon bottom have exceeded 30 wt. %.
5 Modified procedure (ph 7.3) 1,433 mg./l.
Seam - Valentine (south western end)
Location Oak l)all, Centre County, Pa.
Neidigh Bros. Limestone Co.
Chemical Analysis
Moisture 2.0%
Dry Basis:
C eO = 50.0%
CaCO3 85.0%
MgCO 3 1.5%
R 2 0 3 = 0.5%
Feed mate 60 lbs./hr
continuous feed
10% excess theoretical
Cost
Stone $1.33/ton
Transportation 1.95/ton)65 mi.@ 30/ton/mi.)
1/2 inch
3/8 inch
8 mesh
60 mesh
200 mesh
Process Flow Analyses
0.90 lb./minute.
Feed Water
p 1 1
2.83.2
Acidity Alkalinity
mg/i CaCO3 mg/i CaCO 3
l,S 2 O
Fe I I
mg/i
374
Fe II I
mg/i
56
Al
L!
230
SO4
2346
Cond.
nόos,
2490
Sus. Solids
mg/ i
Limestone Reactor
Influent
Efflue nt
Settled Solids
2.93.2
6.26.6
1.500
128
26
20
183
35
42 32 a
19.6% wt./vol.
Settling Lagoon
Influent
Effluent
Settled Solids
6.40
6.48
78
6
<1
1
<4
B
9.5% wt./vol.
95
-------
all of it within ten minutes, without the use of flocculants. These
limestone-produced sludges (Type III settling behavior) do not ex-
hibit a clear settling interface and their settling rates can best
be observed by measuring sludge build-up. The supernatant water
tends to be cloudy, but its clarification rate and level can be
enhanced by the use of high molecular weight flocculants. A very
high flocculant dosage was employed during this test, seeking indi-
cations of sludge balling with carbonate sludges, as was attained
with hydroxide sludge. Positive results were attained which suggests
that a bottom feed-type of solid-fluid separation operation would be
especially effective for this type of sludge. In such an approach,
the settling sludge blanket serves to collect and filter the floccu-
lating colloidal particles resulting in an extremely clear effluent.
The bottom fed weir (see Figure 11, Appendix B) to the settling
lagoon provided sufficient retention time to settle the flocculated
sludge and allowed a very clear effluent to enter the settling
lagoon. This operation was effective to the degree that, had a
proper pump been available to remove the settled sludge from the
bottom of the weir, the settling lagoon would have been completely
superfluous. Such a procedure would have major advantages in both
capital and operating costs.
Further details of the overall limestone process from the Hollywood
Facility have been presented elsewhere 98 . The response to
other limestones was satisfactory but it varied according to par-
ticle size, purity and hardness as described previously in this
report. Dolomite also responded satisfactorily but was more diffi-
cult to control having reduced reactivity and greater hardness thus
limiting maintenance of adequate available reactive surface. The
use of pulverized limestone (-200 mesh rock dust) was helpful as a
supplemental feed to the rotary limestone reactor to complete the
reactions and to add excess alkalinity to the effluent. However,
beyond requiring a separate feeding system, control of its addition
must be exceptionally tight to prevent use of excess reagent.
The use of pulverized limestone in the flash mixer and Densator
are not detailed here. These units were not designed for this
application and did not provide sufficient agitation velocity
to maintain stone suspension. Although a satisfactory effluent
could be maintained, the resulting sanding of the unreacted reagent
required the introduction of excessive quantities of pulverized
limestone.
PROCESS COST ANALYSIS
A process cost analysis is given in Table 19. The total cost, as
calculated, is based on charges experienced at the Hollywood Facil-
ity and includes all phases of the treatment process except taxes
96
-------
Table 19. ESTIMATION OF COSTS - LIMESTONE TREATMENT PROCESS
(basis: cents/l,000 gallons)
(process conditions: table)
5 ortization based on diroct cost for ten year life.
bLabor based at $3.00/hour.
Assumes $1,000.00 equipment cost, $100.00/year supplies and $155.00/year labor.
Lubrication 15 minutes/day - 0.13; replacement parts 0.06.
ppm NH 4 H 2 PO 4 @ O.IOC/lb. - 0.40; filter cleaning N d - 1.27.
1100 man hours/ year.
g 16 7 lb/l,000 gal. @ 0.164 /lb.
l.7Q/KWUr.
1 Fill limestone hopper - 15 minutes/day.
Lubrication 15 minutes/day - 0.13; replacement parts - 1.16.
Flocculant 1 ppm 5 $1.50/lb.
1 Prepare flocculant 19 minutes/day - 0.56; sludge transfer - 1 hr/day - 2.91.
tm Remove sludge from drying basin.
ncharges for caterpiller, two trucks and labor - $0.67/Ton.
°Alkali costs only 3.6%.
PAlkali costs only 11.6%.
Totals
Excluding
Amorti zation
3.44
unit Operation
Power 1
Reagents
Laborb Maintenance
Amortization 5
Inoculatiozf
Contract°
Totals
%
%
Water Collection
3.25
0 1 9 d
5.01
8.45
11.12
14.55
%
38.5
2.2
59.3
Transfer Plant
1.56
1.56
2.05
%
100
Holding Lagoon
4.46
0.58
5.04
6.63
0.58
2.45
%
Iron Oxidation
1 67 e
f
0.95
88.5
9.10
11.5
0.58
12.30
16.19
3.20
13.54
%
13.6
7.7
74.0
4.7
Pump to Reactor
0.56
0.56
0.74
0.56
2.37
%
Neutralization
100.0
3.72
2.74
.
0.872
.
1.29
11.92
20.54
27.04
8.62
34.46
%
SolidFluid Sep.
18.1
12.4 k
1.25
4.2 1
3.47
6.3
0.10
58.0
13.67
18.49
24.34
4.82
20.39
%
Sludge Dewater.
6.8
18.8
0.5
73.9
6.61
2.4 1
9.03
11.89
2.42
10.24
. 3
%
73.2
26.8
Totals
7.53
5.66°
5.29
1.58
52.33
1.16
2.42
75.97
100.00
23.64
100.00
%
9.91
7.45
6.96
2.08
68.88
1.53
3.19
100.00
Totals
7.53
S.66
5.29
1.58
1.16
2.42
23.64
(exci. amor)
%
31.85
23.94
22.38
6.68
4.91
10.24
100.00
NOTES:
-------
and administration overhead. The values listed were derived
directly from pilot plant costs and cannot be realistically related
to industrial operation. Obviously, they are high. The costs are
also shown without the inclusion of amortization values. This com-
pilation offers realistic expression of the factors which must be
included in cost considerations. The low percentage of total costs
related to the reagent is noteworthy.
Reasons for High Costs
Most of the equipment was not being utilized at optimum capacity.
The equipment would permit handling a slow of at least 300 gpm
which would reduce amortization charges by 80 percent. The exist-
ing conditions of the limestone reactor and particular limestone
reagent were the limiting process flow factors. The surface bio-
logical reactor was shown to perform satisfactorily with these
waters at levels at least two times those utilized in this test.
Increased flows would incur no additional labor costs and very
little increased power costs. The reagent and sludge control
costs are closely proportional to flow levels. The amortization
costs are very high since they involve research equipment designs.
A ten-year depreciation period was employed. The equipment can be
expected to provide useful service for a longer period.
98
-------
SECTION XIII
BIOCHEMICAL OXIDATION OF IRON II
The use of limestone to neutralize coal mine drainage has several
limitations. Limestone has a siowerreaction rate which is controlled
by its limited solubility. The limestone particles are subject to
surface coating with reaction products. Further, the maximum pH
which can be developed in the aqueous system with limestone cannot
exceed about 8.3 due to several chemical equilibria, including the
presence of soluble carbon dioxide. In fact, only under extremely
vigorous agitation and with the use of very finely-divided lime-
stone particles can the pH of the treated effluent rise above 7.0.
Although this situation prevents overtreatment of CMD, the lower
pH levels place restrictions upon oxidation of ferrous iron, which
normally follows neutralization and is necessary to complete the
treatment 2 The practical limitations of limestone found in
these evaluations have also been cited in Travers 8 and by
Birch 50 51 who combined lime with limestone to meet water specifi-
cations.
As described in the limestone neutralization section (Table 16), our
experience indicates a practical limitation of the use of this desir-
able reagent to waters containing less than about 100 mg/i iron II.
This arbitrary value appears to be related to the oxygen solubility
attainable. Waters with somewhat higher levels of iron II may be
treated satisfactorily if the limestone neutralization reaction
is followed by lengthy air oxidation periods in the presence of
large excesses of pulverized limestone 52 . During this oxidation
period the limestone must be maintained in suspension.
To permit the effective use of limestone neutralization in ferrous
waters above the 100 mg/l level, it becomes desirable to oxidize
iron II in an acid system. Oxidation by air (oxygen) under these
lower pH conditions is impractically slow. Stronger chemical oxidants
such as peroxides, permanganate, and chlorine may be used but are
deemed too expensive--at least to the degree that their use would
increase the costs to the level which would preclude the advantages
of limestone.
It is generally accepted that autotrophic or chemolithotrophic
bacteria, such as Ferrobacillua ferroxidana and Ferrobacillu6 thio-
oxidans play a role in the development of mineralized mine drain-
age from pyrites 53 . It is also known that these bacteria will
oxidize soluble ferrous iron in an acid solution 5 . This response
by the bacteria may be considered a catalysis of the reaction be-
tween atmospheric oxygen and iron II ions, which is thermodyna icaily-
99
-------
favored, but apparently kinetically-hindered. Glover has reported
utilizing this biochemical iron oxidation as part of a total coal
mine drainage treatment rocess in a small pilot plant ° L 1 2
Huddleston and Whitesell have reported laboratory studies of
this oxidation reaction. Silvermann 57 , Lundgren 58 , and Unz 59 , among
others, studied the nature of these bacteria.
The Hollywood Facility was designed to permit the biochemical oxi-
dation of iron II. In addition to the vattype of oxidation vessel,
somewhat similar to Glovers design but simpler and with surface
aerator, Lovell included a trickling filter-type of unit opera-
tion at Hollywood to carry out this biochemical oxidation. Bench
scale studies made in 1966 established the feasibility of this
latter concept, including its combination with a rotary kiln-type
limestone reactor.
This trickling filter-type of unit operation, termed a surface
biochemical reactor, has several inherent advantages: 1) No
energy is required to control the water distribution or to intro-
duce oxygen and carbon dioxide into the reacting water. The pressure
supplying the water to the reactor, whether achieved by pump or
gravity pressure head, should be adequate to distribute the water.
Porosity through the media bed permits distribution of oxygen and
carbon dioxide from the air. 2) The reactor provides reaction sur-
face area which may be controlled by selection of the media particle
configuration. 3) Capital cost of the unit is low. Designs less
complex and costly than the Hollywood reactor may be feasible based
on this experience. Glover 52 cited as a disadvantage of the
percolating filter concept a tendency to build up insoluble iron
oxidation products. Although such deposition occurs, it is the high
surface area associated with the deposition which supports the bac
teria and provides interphase transfer of nutrients. The cement-
itious nature of the deposit which builds on the surface of the media
in the reactor is indicated in Figure 19. This reaction area creates
the superior oxidation rates (300 to 400-fold higher) reported in
this study. The build-up in the surface reactor at the Hollywood
Facility did not reach a level, in over one years operation, detri-
mental to flow through the filter. Continual deposit growth will
lead to channelling and a reduction in efficiency. Preliminary
tests indicate several approaches to remove the sludge: partial
solubilization and water scrubbing in situ , or after drying, the
deposit will spall from the media surface with minor vibration.
Chemical control to prevent iron III hydrolysis is possible but
does not appear to be feasible as a solution to the deposition
problem.
The two controlling factors in this biochemical process include the
growth rate of the bacteria and the rate at which the bacterial
100
-------
system oxidizes the ferrous ion. In a continuous treatment system,
the retention time in the reactor is the critical design factor.
Parameters which establish the bacterial growth rate include:
inoculation level, temperature, iron II concentration, bacterial
nutritional requirements (as oxygen, carbon dioxide, phosphate,
calcium, nitrogen, and trace elements), agitation, and the unique
characteristics of a given bacterial strain, including tolerable
population levels. These parameters determine the required bac-
terial population. Evidence suggests minimal cell counts of 5 x 106
cells/mi are necessary, whereas higher levels are more desirable.
Cell levels in excess of i0 7 cells/mi have been maintained in con-
tinuous flow systems, whereas counts approaching io cells/mi have
been achieved in laboratory cultures. Glover 52 reports the t acti-
vated sludge serves to maintain an appropriate concentration of
bacteria--a conclusion not wholly supported by the current studies.
The bacterial iron oxidation rate is dependent upon maintenance
of a viable cell population (growth rate, cell viability time,
system cell loss), temperature, iron II concentration, pil, and
at least one inadequately-defined parameter which is related to
liquid-solid interface surface areas. This latter factor(s) may
be involved in Glovers activated sludge, and may explain the
high level of oxidation activity at the surface of natural stream
yellowboy. Glover reports the bacterial sludge to have a low
cell count, which was confirmed in these studies. It is the liquid-
solid interface area which is believed to be responsible for the
much higher oxidation rate found in the surface biochemical reactor
during these studies in contrast to those reported in vat-type
systems 56 .
Figure 19. Deposit on Surface Biochemical Reactor
101
-------
Associated with the effect of pH upon the biochemical oxidation
of iron is another related equilibrium: the hydrolysis of the
ferric ionic species (presumably as some sulfate complex) which
results in the insoluble hydrous ferric oxide deposit. When the
iron II oxidation is carried out chemically with oxygen at higher
pH levels, this hydrolysis proceeds very rapidly, reducing the
soluble iron species to less than one mg/i. As the oxidation is
carried out at low pH levels (pH 2.9-3.0), the hydrolysis reaction
is slow and the resulting iron III species t. d to remain in solu-
tion. Under acid conditions, the biochemical oxidation system
effluent remains relatively free of suspended iron oxide, but does
deepen its amber-like color as the concentration of the ferric
ion species increases. The sludge deposition appears to grow from
nucleation sites rather than direct precipitation containing ferric
ion. A slight increase iii pH (by 0.1-0.3 units from addition of
lime slurry) of the water slowly develops a colloidal haze but
the formation of distinct particles is retarded. higher p11 levels
from minor lime additions to waters are apparently detrimental
to the bacteria metabolism. Other bacterial s ecies may prefer
higher pH levels. Glover s activated sludge 2 may play a role
in this hydrolysis reaction, perhaps beyond that as a carrier of
bacterial cells. The recycled activated sludge may serve as
a physical growth nucleus favorable to this reaction. This phenomenon
may be a key to the further enhancement of this biochemical oxida-
tion process and may be the undefined parameter which is related
to surface area. In these studies, the merit of separating and re-
cycling the sludge formed after biochemical iron oxidation could
not be supported, and was discontinued.
The chemical analysis of the sludge developed by biochemical oxi-
dation is given in Table 20. It is characterized by high iron and
sulfate concentrations as well as low aluminum concentrations, when
contrasted to sludge from alkali neutralization (see Table 50).
Apparently between 15 and 16 percent of the iron present is asso-
ciated with the sulfate. The high sulfate content suggests another
advantage of biochemical iron oxidation in achieving some sulfate
concentration reduction. The potential to enhance this sulfate
reduction is of great interest.
The iron oxidizing bacteria were named Ferrobacillus ferrooxidans
by Leathen et al.bD but are regarded by research workers as closely
related to members of the Thiobacillus genus 61 since many of them
can oxidize sulfide compounds as well as iron. These bacteria
are variously referred to as Ferrobacillus ferrooxidans, Ferro
bacillus tlijooxjdans and Thiobacillus ferrooxidane. In waters
containing Iron II which is associated with sulfide, there no doubt
exist several strains of these bacteria that vary slightly in
appearance and in their biochemical activities.
102
-------
Table 20. CHEMICAL ANALYSIS OF SLUDGE FROM BIOCHEMICAL
OXIDATION OF COAL MINE DRAINAGE WEIGHT PERCENT
Settled Settled
sludge from sludge from
oxidation tank thickener
Fe 42.54 40.34
SO 4 17.39 15.68
Loss on ignition 180°C 14.16 15.44
P 2 0 5 2.44 2.25
Si0 2 0.29 1.13
Ca 0.21 1.95
MG 0.12 0.24
Al nil 0 .98
Members of this group of microorganisms are nonspore forming, gram-
negative rods varying from 1.0 to 2.0 microns in length and having
polar flagella. Figures 20 and 21 show one such isolate, Strain Z,
that has slightly curved cells. They are approximately 1.5 microns in
length and have a single polar flagellum. Figure 20 results from
optical magnification through a phase contrast microscope, while
Figure 21 is an electron microphotograph at 128,000 magnification.
The organisms can be observed in vivo with a phase contrast micro-
scope and appear to be straight or slightly curved rods, actively
motile and varying in diameter as well as length.
The organisms derive their basic metabolic carbon, hydrogen, and oxygen
from carbon dioxide, water, and air. They grow slowly compared to
common heterotrophic gram-negative rods, having generation times of
about five or six hours at their optimum temperature of 25 to 28°C.
Temperatures higher than 35°C are harmful to most natural isolates.
Glover 52 reports 40°C to be immediately fatal. The organisms grow at
colder temperatures, although their growth rate decreases with
temperature. Laboratory experiments with several strains have indicated
that at 10°C, the bacteria require from 24 to 30 hours to double their
103
-------
Figure 20. Microphotograph of Culture of Ferrobacillus Ferrooxidans
Enrployed. to Oxidize Iron II.
S
S
S
I
1
.
p
0
S
S
S
ii I
4$
0S .κ
..
0
r p
O
S
I
I,
I ,
a
I
S
.
I
0*
jQL4.
-------
Figure 21. Electron Microphoto rapn. of Ferrobacillus Cell
105
-------
numbers. They achieve a low cell weight (1 g, dry weight) from the
much larger amounts of iron (180 grams) 52 .
A growth lag period of 10 to 20 days was reported by Clover 52 . High
levels of inoculant encourages initial growth. If the initial plant
system has cell levels of the order of 102 cells/mi, a much longer
incubation period can be anticipated. The permissible pH range appears
to be between 1.5 and 4.0, with an optimum near 3.0. The pH levels
which might be specified in this biological process illustrate the
complex nature of the reactions which occur. There are at least two
optima which should be satisfied: the pH condition which relates to
the growth levels of the bacteria, and that p 1 -I condition which relates
to the Eh level most favorable to the iron II oxidation. The latter
level increases with rising pH, thus developing conditions more con-
ducive to chemical oxidation. Efforts to enhance the oxidation rates
by maintaining higher pH levels in the biochemical oxidation tanks
(by the addition of lime slurry) have not been advantageous__suggesting
that the metabolic levels of the bacteria decrease in these higher pH
ranges. This observation confirms a distinct role of the life
process in this biochemical oxidation. Further study of the mechanisms
involved is strongly indicated.
Under continuous flow conditions, Giover 2 reported iron II oxidation
rates of nearly ii mg/i/hr when maintaining an equilibrium concentra-
tion of 50 mg/i iron II. Rates nearly three times greater were
reported in ten-day-old laboratory cultures. Air was introduced by
sparger into Clovers oxidation reactors. In subsequent, larger scale
studies, inoculation of the water and slow development of an activated
sludge, the rate increased to 120 mg/i/hr while maintaining a tempera-
ture of 15-20°C. The suspended solids were maintained near the one
percent level. At near-freezing temperatures, the rate dropped to
12 mg/i/hr. The activated sludge was recycled and became very
tacky and cohesive. He employed retention times of one hour at 20°C
and up to five hours at a temperature near 0°C. Huddleston et al. 55
in 1969 reported much higher rates in laboratory cultures with synthet-
ic solutions but were unable to sustain them in continuous flow with
actual mine water (at 20-26.5°C) in a small pilot plant 56 . These
authors found approximately 0.1 volume of air/volume of water to be
optimum. They also utilized about 20 mg/i each of ammonia and phos-
phoric acid as nutritional supplements.
PLANT OPERATIONAL DATA
Development of Bacterial Cultures
The bacteria employed were isolated from samples of Hollywood waters
collected in sterile containers, and, in one instance (Strain Z), from
a Central Pennsylvania coal refuse product derived from Lower Kittanning
106
-------
seam coal. Cultures were obtained by an enrichment technique employ-
ing either Silvermans 9K medium 57 or American Type Culture Collec-
tion Thiobacillus medium No. 5562 These media are highly selective
for members of Ferrobacillus and Thiobacillus. The composition and
preparation of the media is described in Table 21. During the culture
enrichment, the media were inoculated at the ten percent levels,
incubated at 26°C on a reciprocal shaker until the characteristic
yellow-brown ferric precipitate developed in three to five days. The
cultures were purified by tube dilution with the final tube being
transferred to fresh, sterile media. Repeated transfers in the highly
selective medium were assumed to result in a pure culture. Stock
culture is maintained in cold storage for subsequent study and the
development of inoculum. A sample of Ferrobacilius ferroo.xidans ATCC
l 661 was grown concurrently and exhibited behavior similar to
cultures collected at the plant site.
Although the changing iron II levels during oxidation can be employed
as evidence of the extent of bacterial action (and probably adequate
as a control procedure in an operating plant), direct cell assays are
helpful since cell population is obviously an oxidation rate parameter.
Biological assays are too slow for control purposes, requiring from
48 hours to 14 days to yield a measurement. In addition, these species
do not respond well to the plate-type counting assays commonly employ-
ed for coliform and other common types of bacteria. Huddleston
et al. 55 56 reported some cell population data. Direct cell counts
are more practical.
Direct Micros p c Count Assays
Direct microscopic counts were made with a Petroff-Hauser bacterial
counting chamber from observations with a phase contrast microscope.
The microscope factor of this chamber was 2.0 x per square. A
minimum of 20 squares were evaluated. Significant results were limited
to samples containing at least 1-5 x 106 cells/ml. Care was taken to
pipette a representative sample of the suspension of microorganisms.
The accompanying yellow precipitate caused some difficulty in filling
the chamber and required care in observing the cells. In active
cultures, the high motility of the cells made counting difficult. The
addition of one-tenth volume of 95 percent ethanol served as a
tranquilizer to slow cell action.
Tube Dilution Assays
Most probable number of cells was also established by a tube dilution
procedure based upon Woodwards table of MPN 63 . A triplicate series
of decimal dilutions comprised each test involving 10 to l0 ml for
mine water samples and from 102 to i0 ml to estimate laboratory
107
-------
Table 21. CULTURE MEDIA FOR IRON OXIDIZING
BACTERIA
9K Medium
Salt Solution
(NHk)2S0 3.0 g
KC1 0.1 g
K 2 HPO 0.5 g
MgS0 .7H 2 0 0.5 g
Ca(N0 3 ) 2 0.01 g
10N H 2 SO to pH 2.8 approx. 1.0 ml
Distilled water 700 ml
Ferrous Sulfate Solution
FeS0 .7Ii 2 0 44.2 g
Distilled water 300 ml
Adjusted pH 2.8 with }4 2 S0
Autoclave solutions separately. When cool, decant ferrous sulfate
solution from slight precipitate and combine aseptically with salt
solution. (Silverman employed the medium at a pH 3.0-3.6; in these
studies it was routinely adjusted to pH 2.8).
Medium No. 64
Solution A Solution B
(NHi+) 2 S0 0.8 g FeS0 .7H 2 0 20.0 g
KH 2 PO 0.4 g Distilled water 200 ml
MgS0 .7H 2 0 016 g pH 2.8 with 1 N H 2 S0 1
Distilled water 800 ml
p1! 2.8 with 1 N H 2 S0
Autoclave separately, cool, decant as with Medium 9K and combine.
cultures. The dilutions were made using sterile, distilled water
buffered with KH 2 P0 (0.4 g/l) and Fl 2 S0 to the pH range of 2.8-3.0.
The culture tubes were incubated at 28°C for 10-14 days and were
inspected daily. The formation of a yellow color and orange-brown
precipitate was taken as the end point. The results with laboratory
cultures using Medium No. 64 compared favorably with direct counts,
Usually being within a factor of two. However the results on natural
mine water samples were less precise. The influence of other
organisms in these waters may be responsible for this deviation. The
tube dilution technique serves as a check for direct count measure-
ments and provides a means of evaluating counts for samples containing
less than 106 cells/mI. Cell counts on the four Hollywood water
Sources made in August 1969 were:
108
-------
Fe II mg/i Cell/mi
Proctor No. 2 Borehole 600-650 90-250
Proctor No. 1 30-70 2-25
Bennett s Branch 20-2 40-240
Tyler Run <5 1-50
Bacterial Growth Factors
Growth rates are definitely lower at pH 3.6 in comparison with the
2.8-3.0 range. No evidence has been developed of growth inhibition
at increased oxygen levels. Flask agitation stimulates growth in
comparison with stationary cultures. The addition of organic growth
stimulants such as 0.005 o yeast extract does not enhance and may even
retard cell growth. The addition of calcium ion (as CaC1 2 or CaCO 3 )
in the range of 50-200 ppm to the No. 64 Medium stimulated growth of
some of the isolated strains. This enhanced response was related to
the calcium ion rather than the carbonate ion. The Hollywood waters
all contain calcium concentrations in this range so that further
calci im supplement should not be necessary. The response to additions
of N E - I 4 and P0 3 ions is indicated in Table 22. Only Proctor No. 2
waters showed significant stimulation which suggests that the oxida-
tion rate of highly ferrous waters may benefit from such additions.
Phosphate alone was as effective as the ammonium ion or perhaps the
combination of both.
Preparation of Inoculants
A 10 ml portion of stock culture, stored at 5°C, was utilized to
inoculate 100 ml of No. 64 Medium in shaker flasks. Growth reached
the 108 cells/mi in about 48 hours. A combined culture, totalling
500 ml front five shaker flasks, was used to inoculate a five gallon
carboy of Medium No. 64 used to seed the tanks at the plant. The
medium in the carboy was aerated by a simple glass sparger.
Initiation of the Biochemical System
Experience indicated the advantage of starting these systems from high
cell count, pure, laboratory cultures rather than seeking to increase
the population from a low cell coui t, natural water or attempt to
introduce a natural culture from the surface of the yellowboy
deposit in a stream. A series of batch steps is desirable in which a
minimum of ten percent inoculant is utilized in each step. Typical
batch oxidation rates are illustrated in Figure 22 while the condi-
tions utilized are detailed in Table 23. In stage A (Figure 22), a
150 gallon tub was employed as the reactor with the matured growth
being transferred to the sludge sump for stage B and subsequently to
one of the concrete oxidation tanks (see Appendix B, Figure 13) for
109
-------
Table 22. GROWTH SIMULATION OF IRON OXIDIZING BACTERIA
IN NATURAL MINE DRAINAGE
IRON II CONTENT (mg/i)
Bennetts Branch
Blank
NH
i +1
P411 t
PO ; 3
?sTTT+ j
lNfl 4
nr - 3
rJl.f
23.5
15.4
<5
23,5
15.4
<5
23.3
11.0
<5
23.3
8.8
<5
23.1
<
<5
stage C. In proceeding to stage D, the third stage product was equally
distributed between the two oxidation tanks which were then filled
with raw wqter. A final iron II concentration of about 10 mg/i was
reached within 24 hours. Little iron III is hydrolyzed and precipi-
tated from solution during these batch oxidations. The amount of
suspended solids near the end of stage A was about 150 mg/i which
increased to about 300 mg/l in Oxidation Tank 1 toward the end of
stage D. With the maturation of the culture representing two-lO,000
gallon stages, the batch operation was converted to a dynamic multi-
stage system by the continuous introduction of raw water.
In addition to the water loading rate to the surface reactor (>0.16
gpm/ft 2 surface), the retention time in the reactor is critical. The
detailed response of this reactor, especially the media configuration,
was not included in the present study but is being pursued. The
retention time of the surface reactor was found to range between 4.7
and 5.3 minutes in the flow range considered. This parameter will not
Water source
Proctor No. 2
Incubation time
Initial 1 Day 2 Days 4 Days
Blank
NH
NH
P0 3
NHt 1 }
P0i 3
1
10
10
10
10
mg/i
mg/i
mg/i
mg/i
nigh
665
665
658
658
652
529
520
4741
1232
208
<5
<5
<5
<5
<5
Proctor No. 1
Original
cell
count
90
cells/mi
1
cell
count
2,400
cells/mi
2
cell
count
4,000
cells/mi
3
cell
count
40,000
cells/mi
25
No observed change from control in two days.
<5
1 mg/i
10 mg/.l
10 mg/i
10 mg/i
10 mg/i
110
-------
Figure 22. iron II oxidation rates with batch inoculation of mine
drainage
40 60
TIME.HOURS
400
300
200
-I
0
I
I-
z
U
U
x
0
U
z
0
(A)
0
20
80
00
111
-------
Table 23. DETAILED CONDITION OF BATCFI INOCULATION
Reactor volume Inoculation level Overall oxidation
Stage gallons volume % rate mgFe/L/FIr.
A 150 3.2 3.6
B 1,257 10.8 7.9
C 9,930 12.4 7.6
li-i 9,930 50.0 9.8
D-2 9,930 50.0 11.5
Stage A Stage C
Nutrient supplement: NH 1.6 mg/i 1.0 mg/i
(as NII 4 H 2 PO ) P0T 3 3,4 mg/i 5.1 mg/i
Aeration: Air sparger in stages A and B at approximately 1/10 vol.
air/min./volume water.
Surface aeration in stages C-D-E
Agitation: Slow mechanical stirrer by surface aerator.
24 RPM in stages C and D
Water temperature: 61°F
Culture: Strain Z, 93.5 hours old
Raw Water Analyses: Iron (II): 340-358 mg./1.
Iron (Total): 410-420 mg./l.
Sulfate: 1,950 mg./1.
pH: 3.0 - 3.3
Acidity: 1,3501,500 nig./1.
deviate extensively with flow changes until a flooding condition
would develop which probably would not permit sufficient oxidation
time. This retention value was attained by batch introduction of a
concentrated tracer (sulfuric acid and sodium sulfate were both help-
fully used) to the surface of the media and the conductivity of the
effluent measured as a function of time. The first increase in con-
ductivity was observed after one minute. The response of the tracer
had passed in about 12 minutes.
The overall oxidation performance of the reactor is shown i -n Figure
23, indicating the extent of iron II reacted varied between 60 and 94
percent.
In comparing deep vat and surface reactor biochemical iron II oxida-
tion, three pragmatic conclusions are made which favor the surface
type reactor:
112
-------
Figure 23. Surface biochemical reactor performance
Ioo
90-
60-
70-
60
50-
40-
30-
20-
10-
I-
U-
0
0
Average Influent-Effluent Water Temperature
c nHuent Iron Deposited in Reactor
o Influent Iron Oxidi2ed in Recctor
T - I I r - - I I 1 fr I
2 9 16 23 301 6 13 2b 2 4 U 18 25 ,1 8 15 22 291 6 13 20 27 3 10 17 24 3
AUGUST 1971 SEPTEMBER 1971 OCTOBER 1971 NOVEMBER 1971 DECEMBER 1971 JANUARY 1972
-------
1. The surface reactor is least sensitive to disruptive influences,
i.e., change of iron II concentration, metabolic inhibitants,
flow rates, etc.
2. The surface reactor permits much higher reaction rates than the
deep vat systems under continuous flow operation conditions: at
least 6000 mg/l/hr versus 50 mg/i/hr.
3. The surface reactor has no power requirements to control water
flow, introduce oxygen and carbon dioxide.
The control and removal of the deposit of ferric oxyhydroxides in the
surface reactor must be further studied. In the years operation of
the surface unit at the Hollywood Facility, hydraulic flow rates have
not been significantly reduced. Water temperature response during
winter operation needs further evaluation. Present experience extended
into sub-freezing weather permitting feed water to freeze and prevent
flow. In studies now in progress, a simple translucent, plastic film
cover over the reactor appears satisfactory to permit continued
operation in winter weather thus avoiding the cost of introducing any
heat to the water.
CALCULATION OF OBSERVED IRON OXIDATION RATES IN PLANT OPERATION
The estimation of the biochemical oxidation rate for ferrous iron in
a continuous vat system involves two parameters: (1) iron II concen-
trations in the influent and effluent, and (2) the available retention
time. The latter is controlled by reactor size and flow rate. Other
parameters, as pH, nutrients, temperature, agitation, etc., are
assumed to be constant and to be maintained at optimum levels. Start-
ing with a vat filled with inoculated raw mine water as a batch
operation, the iron II content will decrease as the cell population
increases. When continuous flow is initiated, should no further
iron oxidation occur, the iron concentration will rise according to
the relative concentrations as it approaches the influent concentra-
tion. The rate of approach will vary with flow rate and reactor
volume. However, with oxidation, a new equilibirium iron II concen-
tration will be established, based on the oxidation rate. A computer
program to establish this rate was developed and is listed in
Appendix D along with an illustrative calculation. The program pro-
vides a time-step integration, which begins with a constant reactant
volume having a given initial iron II concentration. Utilizing a
specified time interval (as 1, 2 or 5 minutes), the iron input is
designated (flow and iron II concentration) and assumes instantaneous
mixing. The resulting concentration of iron II will change with the
existing oxidation rate, accompanied by the effluent loss correspond-
ing to the stated flow rate. Based on a defined oxidation period
such as 30 or 60 minutes, effluent iron II concentration, and an
114
-------
initial oxidation rate of zero, the program will read out the iron
content per time interval (in pounds and mg/i), the change in iron
content per interval and the cumulative change during this defined
period. The program subsequently calculates a first approximation
oxidation rate to maintain the defined initial reactor concentration.
In a second ioop of estimations, the computer utilizes the approxi-
mation oxidation rate and, printing out the above interval data,
establishes the adequacy of the rate for the defined oxidation
period (one to two hours at five minute intervals is usually an
adequate simulation period). Should the initial rate estimate not
conform within one percent of the equilibiriuin concentration, a new
oxidation rate is estimated and the calculation repeated.
This program, and minor variations thereof, serves as a simulation
model for design purposes. It also provides an insight into the
capabilities of multi-stage systems. Two important factors must be
considered in such an evaluation: (1) the specified or acceptable
iron II level of the effluent of the oxidation system, and (2) the
iron II level of the raw water influent. In general, the former
concentration becomes the equilibrium concentration of the system
and, since the oxidation rate is distinctly a function of iron II
concentrations, the oxidation rate can become a severely limiting
factor for the system. Although this model was designed for the
biochemical vat process, it is applicable to the chemical-air oxida-
tion procedure.
The limitations of oxidation rate defined by the equilibrium iron II
concentration appears to account for the limitations of the vat oxi-
dation approach and explains the superior results of the surface
reactor system. In the latter case, there is no dilution and the
oxidation proceeds at the highest prevailing iron concentration as
the water contacts the media surface.
The implications of biochemical iron oxidation were included in
Section XII, describing the limestone treatment system.
115
-------
SECTION XIV
SLUDGE CONTROL IN COAL MINE DRAINAGE TREATMENT
INTRODUCTION
Any treatment removing the undesirable ions from mineralized waters
creates a by-product which is troublesome and costly to control. In
neutralization processes applied to coal mine drainage, a sludge is
formed from the separated water-insoluble compounds. Lovell in
l97O6 reviewed the sludge control philosophy basic in these studies.
This detail should provide a positive base for developing an optimum
sludge control system.
Lovell noted the need in sludge control planning to coordinate the
nature and volume of the water to be treated with the site, treat-
ment process, and treatment period anticipated. The solids separa-
tion approach must offer control to prevent accidental discharge
of inadequate quality waters. A final polishing pond enhances this
capability.
Sludge control may incorporate three stages: solid-fluid separation
of the sludge, sludge dewatering, and sludge disposal. Two techniques
may be satisfactorily employed in the solid-fluid separation: settling
lagoons and mechanical clarifier-thickeners. Lagoons are most commonly
employed. Although the choice is usually determined by land avail-
ability and process economics, there are other more subtle differ-
ences which should not be ignored. The process economics do not con-
sistently favor either practice; the guiding factors which define
costs will be indicated here.
Solid-fluid separation in a settling lagoon may provide limited
sludge storage or permit permanent disposal. Limited storage re-
quires sludge transfer but uses less land. Settling and disposal
in the same lagoon necessitates large land areas, but involves mini-
mal design and construction. The method can be quite satisfactory.
This approach is the least acceptable aesthetically. Other detailed
design, construction, and operation considerations are also necessary.
In contrast, mechanica1 clarifier-thickener systems require the least
land area and provide the least sludge storage. They are more versa-
tile in process control by permitting continuous, automatic sludge
removal. Their mechanical nature does not necessarily imply higher
Costs.
Further dewatering of the settled sludge prior to disposal may be
indicated by its voluminous nature, low percentage of solids, and
116
-------
tendency to gel. It may involve some form of in situ evaporative
drying or direct water removal by a drying basin, by a filter, or by
a centrifuge. The settled sludge usually may be pumped through pipes
to the disposal area or to a tanker truck for haulage to the disposal
site requiring multiple handling. In either case, handling large
volumes of water increases costs. However, after dewatering the
solids may be moved by dump truck. Ultimate disposal is usually
made into abandoned mines (deep and surface) or to land fill. No
broad, feasible use for the impurity solids has been developed.
The settled sludge is difficult to handle and dewater. When collected
by pump suction, the gelled or flocculated solid particles do not
readily disperse. They move at slower rates than the supernatant
water which leads to rat holing. This phenomenon makes removal of
settled sludge from a lagoon bottom difficult and costly. Some
sludges tend to gel, while others achieve a plastic clay-like
character which resists any flow.
The important physical properties of the sludge are: settling rate,
settleability (an arbitrary concept combining settling rate and final
settled volume which also incorporates compression phenomena), bulk
density, and flowability. All relate to the hydrated character of
the sludge. The sludges settle to a solids content between 0.5 and
14 percent by weight Most are in the lower range. When dried or
frozen, the highly hydrated character of the sludge is changed.
These hydration properties are not restored upon rewetting. The
physical properties of sludges are governed by chemical composition
and precipitation processes which tend to be reasonably uniform for
a given drainage water and treatment process. However the roperties
vary widely with the neutralization reagent employed. Hoak ,
Sanmarful and Kountz 66 , Duehl 67 , and G1over 2 have cited some detail
of sludge property variation with different neutralization reagents.
Limestone shows the greatest deviation froni the other reagents, pro-
ducing a rapidly settling sludge.
The volume of the settled sludge increases with iron-aluminum con-
centration of the raw water extending to more than 30 volume percent.
The sludge volume level obviously relates to disposal problems and
costs; there are advantages in achieving higher density sludges.
In addition to control by the initial precipitation reactions, a
greater sludge density may be achieved by recycling a portion of
the settled sludge to the initial reaction chamber. This approach
is utilized in municipal water and sewage treatment and is cited
for ap 1ication to coal mine drainage treatment by Gustafson 3 ,
Lovell , Birch 5 ° and Kostenbader, et al. 15 . A very dense, rapid
settling magnetic sludge has been described [ Streeter and Lovell 68 ,
and Stauffer and Lovell 69 ] when the iron is converted to a magnetite
structure. Although of extensive potential, this procedure has not
be developed in a feasible system.
117
-------
The factors which relate density, flowability, and surface character
of sludges to their composition and formation, have not been adequately
elucidated. Some of these aspects have been considered by Langmuir
and Whittemore 70 71 from a geochemical viewpoint. Obviously the
crystal composition, size, surface area, and the nature of its sur-
face charges all affect the sludge behavior. Sludges formed in the p11
range of 6.4 to 7.2 appear to settle most rapidly. When feasible, this
pH range was sought throughout these studies. Following this princi-
pie also permits use of minimum reagent levels. These researchers
find the sludges to be primarily iron oxyhydroxides. The natural
? yel1owboyt is poorly crystallized goethite (orthorhombic Fe 2 O2H2O)
and amorphorous ferric oxyhydroxides. Sludges prepared by synthetic
neutralization yield a mixture of poorly crystallized goethite and
lepidocrocite. It is probable that aluminum and other ions are
detrimental to crystallization. In pure chemical systems, superior
crystallization results. When using sodium bicarbonate to neutralize
these solutions, it appears lepidocrite initially forms then reverts
to goethite. When seeding is utilized, a reversal of forms occurs
with goethite appearing first but a mixture still results. Sludges
resulting from hydrolysis after biochemical iron oxidation of drain-
age waters are poorly crystallized, amorphorous, ferric oxyhydroxides.
SLUDGE SETTLING RATES
Three types of settling rate phenomena have been observed for the
CMD Slurries developed at Hollywood.
Type I - The sludge settles rapidly and uniformly with a distinct
interface and a clear supernatant liquor. This response
results from reaction with strong alkalies (hydrated and
pebble lime, sodium hydroxide, and sodium carbonate) when
the mine water has a typical or high iron and aluminum
content. The settling rate curve has a classical shape,
i.e. the sludge volume decreases sharply during the first
10-30 minutes, asymptotically approaching a relatively
constant terminal volume within 24 hours. See Figures
24 and 25.
Type ii - This behavior occurs when very low iron- and aluminum-
containing waters are treated with strong alkalies.
Tyler Run waters containing 1 ppm iron and 30 ppm alumi-
num are illustrative. The amount of precipitate formed
is quite small and resists flocculation. The flocs tend
to remain diffused and form a nearly stable suspension.
Complete settling may not occur for several days. Even
then the settled sludge might not exceed 0.1 weight per-
cent solids. The water may retain a slight turbidity.
Rates are difficult to evaluate.
118
-------
Figure 24. Sludge settling rate - solid-fluid separation influent
proctor no. 1 water
0
10 Hours
-. 120 Hours
1000
-J
; . w
-J
-J
o Ca(OH) 2
Na 2 CO 3
200
o 20 40 60 20 40 60
SETTLING TIME (MINUTES)
&
-------
Figure 25. Sludge settling rate - solid-fluid separation influent
proctor no. 2 water
b l8Hotgs
0
C
1000
1000
900
800
o CaO
Ca(OH) 2
Na CO 3
t Ca(OH) 2 MgO
600
I
uJ
-j
0
>
uJ
0
-j
(I ,
500
400
300
200
100
0 20 40 60 20 40 60 20 40
SETTLING TIME (MINUTES)
4 Hours
-------
Type III - A two-phase solids system is developed. The major por-
tion (to 90 percent) of the solids settle very rapidly
and can best be measured as they collect on the bottom of
the test device. The particles are not gelatinous flocs
and do not appear to grow significantly in size. This
build up is an asymptotic reciprocal of Type I behavior.
The bottom collection usually approaches its maximum
level within 10 minutes. This volume is used as the
sludge volume. This build-up rate can be measured but
with less certainty. The values reported are conserva-
tive. The supernatant liquid is turbid. This settling
behavior appears to be independent of pollution loading
of the feed drainage. Sometimes an interface will form
slowly to yield a clear supernatant. This behavior is
similar to Type II slurries, requiring about 24 hours
to settle. In other cases the solution will decrease
in turbidity but remain cloudy. This behavior results
with limestone neutralization. The residual turbidity
is attributed to the presence of near-colloidal solids.
The behavior has also been observed with sodium carbonate
and sodium hydroxide when continuing reactions (both
neutralization and iron oxidation) are experienced.
When observed with these reagents, the behavior is a
symptom of improper reaction control.
Solids separation from all three types of settling behavior can be
improved by the use of flocculants, by sludge recycling to increase
the total solids content of the suspension, or by filtering the
floc through a deep sludge blanket such as can be maintained in a
thickener. The sludge blanket in a thickener can be improved by
weighting it with loading agents such as pulverized limestone, fine
coal and silica.
The standard settling rate test utilized was performed by filling a
1,000 ml. graduate with a freshly prepared slurry. After gentle
mixing (not stirring) to insure uniform distribution of the particles
and to permit floc growth to be initiated, the interface volume be-
tween the sludge and clear supernate was recorded as a function of
time. The settling rate was determined from the free settling portion
of the data (i.e.., prior to the start of sludge compression).
A summary of sludge settling data from various plant tests are pre-
sented in Table 24. No data resulted from the sludges exhibiting
Type II behavior. The observed settling rates ranged from 1.2 to
5.4 ft/hr. The higher values resulted with limestone treatment but
yielded a turbid supernatant liquor. Plant data did not indicate
rate trends among the other reagents. The sludge settling rate
seemed to approach a maximum with increasing iron-aluminum levels
121
-------
Table 24. SuMMARY OF SLUDGE SEfl LING DATA FROM HOLLYWOOD PLANT TESTS
Suspended Solids
Sludge Solids in Sludge in Calculated
Total Settling Neutralized Settled Settled Solids in
++
Acidity Fe Al Final Alkali Rate Water Volume Sludge Settled Siudae
Water Source mg/I rug/i mg/i P 11 Used ft/hr rng/l (aver.) % of original wt. % (T.A. + 2Fe )
Tyler Run 114 4 15 6.26 NaOH - 50 2.2 0.23
BennettS Branch 120 29 15 6.51 NaOFI -- 91 -
Bennetts Branch 206 35 17 6.95 CaO 96 45 + T 0.21 0.61
Bennetts Branch 206 35 17 6.95 CaO - 96 2.5 + 5 0.38 1.10
Proctor No. 1 502 80 80 8.30 CaO 349 350 10.0 0.35 0.66
Proctor No. 2 1,707 420 210 7.46 CaO 1.02 1,365 29.6 0. 6
Tyler Run 78 1 15 7.00 Soda Ash - - -
BennettS Branch 86 10 5 8.03 Soda Ash - 1.0 0.37 1.06
c.,l
Proctor No. 1 397 50 50 7.38 Soda Ash 2.96 290 5.0 0.58 1.00
BenflettS Branch 79 18 10 8.00 Soda Ash -- 55 0 + T --
Proctor No. 1 328 57 57 7.60 Soda Ash 3.95 252 3.0 0.84 1.47
Proctor No. 1 302 52 75 7.78 Soda Ash 1.87 178 30 0.59 1.35
Proctor No. 2 1,398 241 205 7.31 Soda Ash 1.78 2,318 11.0 2.11 1.71
Proctor No. 2 1,174 118 59 7.38 Soda Ash 3.00 8.0 - 1.76
Proctor No. 1 405 69 62 7.25 Ca (OH) 2 4.72 848 8.0 0.06 0.68
Proctor No. 1 1,380 317 191 6.75 Limestone s 4 o T 1,692 3.0 5.63 6.69
Tyler Run 104 <1 12 7.75 Pul.Liinestone 4 68 +T 104 0.2 5.2 5.2
Proctor No. 2 1,401 322 191 7.75 Pui.Limestone 410 +T 1,356 2.0 6.78 10.3
Proctor No. 2 1,305 250 191 7.1 Ca(OH) 2 1.47 1,370 18.0 0.76 1.0
Proctor No. 2 1,360 266 191 8.2 Do1.Hydrate 1.95 1,990 18.0 1.11 1.05
NOTE:
Indicates supernatant liquor was turbid.
-------
in the 100-200 ppm range then decreased. At low metal levels floc
growth was slow while at levels above 200 ppm the growing flocs
approached compression and settled more slowly. There was no dis-
tinct trend measured during plant tests in settling rates with pH
(6.3-8.3) variation, if the neutralization-oxidation reactions were
complete. However, laboratory tests indicate maximum rates near
pH 6.8. With prior reports of this dependence, the relationship
between settling rates and p1-I precipitation levels might be con-
sidered further. Continuing reactions literally muddied the
water. The density of the settled sludge does not directly reflect
either metals content of the raw water or settling rates of the sludge.
USE OF POLYELECTROLYTES AS SETTLING AIDS
Polyelectrolytes (usually polyacrylamide polymers of various molecular
weights and structures) have found universal application in enhancing
the settling rates of all types of slurries, including most coal
mine water sludges. They enhance floc growth and decrease turbidity.
They are available in a range of ionic charges, molecular weights,
and other characteristics which govern their performance. There is
no evidence that these polymers increase final sludge densities
either in bench or plant tests. This was evident in sludge from
Tyler Run waters which have low suspended solids containing pre-
dominantly aluminum hydroxide At very high polymer dosage, a
balling phenomenon occurs with these sludges which may have the
practical response of increasing density. The polymers do not de-
crease residual soluble iron levels in treated waters as has been
suggested 72 . Since their usage does aid uniform plant operation and
reduce the size of equipment, an evaluation of several types of poly-
electrolytes was made. Previous pilot studies 75 involving several
preparations found non ionic Superfloc 84 and Nalcolyte 67 effective
at dosages of about 10 ppm.
Representatives of several polyelectrolyte producers carried out
evaluation of their flocculants on fresh sludge at the pilot plant.
These individuals, being knowledgeable regarding specific formula-
tions, were most qualified to make preliminary screening tests.
Although the use of these formulations is simple, the techniques
must be carefully established. Considered were: selection of
the most effective reagent, preparation of the reagent solution,
and reaction parameters as dosage, dilution and mixing procedures.
These polymers consist of long length molecules. Their response is
governed by the reaction of the polymer and the particle surface,
thus contacting methods between the polymer and the suspended particles
are critical. Excessive agitation (as passing a solution through a
high speed mixer or pump impeller) can shear polymer molecules and
destroy their effectiveness.
123
-------
The data presented are not intended to compare products nor imply
that other products may be unsatisfactory. High molecular weight
polymers can be effective flocculants at dosage levels of 0.5 to
2.0 ppm of neutralized mine water. Those polymers designated non
ionic or weakly anionic were most responsive. The settling rates
with the polymers ranged from 4 to 16 feet per hour. The optimum
dosage level increased with the suspended solids level. In recycle
sludge systems, flocculant dosage levels should be based on the
ppm polymer/lb of dry solids. It cannot be assumed that polymers
associated with settled recycle particles remain effective. Ex-
cessive agitation can destroy polymer-formed flocs. The best floc-
culant and dosage for any particular sludge must be established
experimentally, although the data presented here should limit the
evaluation. No obvious sensitivity of the polymers to pH was
indicated over the range 6.0-9.0 normally experienced.
In testing American Cyanamid reagents, the diluted flocculant was
added to 1,000 ml slurry in a graduated cylinder during the first
three of fifteen slow, mixing strokes using a perforated disk
plunger. Timing was initiated at the completion of mixing. The
sludge interface location was estimated for Bennetts Branch and
Tyler Run water due to its indefinite nature. Supernatant turbidity
was established with a Helige Tubidimeter.
The evaluations with different waters and neutralization reagents
are detailed in Tables 25 through 31. Those recommended by their
respective companies were: Superfloc 127, Decolyte 930 and Atlasep
lAl.
Some polymers were utilized in plant tests involving the thickener,
Densator, settling lagoon, filter and centrifuge operations. Their
response confirmed bench scale testing and are described in the
appropriate sections of this report. Somewhat higher dosages than
those indicated by bench tests were desirable during the plant tests.
The ability to increase the capacity several fold of a solid-fluid
separation unit by using polymers was demonstrated.
PRIMARY SOLID-FLUID SEPARATION OF TREATED COAL MINE DRAINAGE
Thickeners
Gravity settling of suspended solid particles is divided into clarifi-
cation and thickening 73 . The former term is applied to dilute pulps
as illustrated by Type II mine drainage sludges. The devices utilized
are characterized as to means for removal of the settled solids.
Unit Sizing is the primary consideration. The size is established
empirically, based on influent volume and particle levels and charac-
teristics. Particle settling rates establish the surface area required
124
-------
Table 25.
ANALYSES OF MINE DRAINAGE WATERS EVALUATED WITH SUPPERFLOC* REAGENTS
Reagents* Evaluateda
SF-16 SIJPERFLOC 16 Fiocculant
SF-127 SUPERFLOC 127 Flocculant
AF-550 AEROFLOC 550 Reagent
SF-202 SUPERFLOC 202 Flocculant
Z SF-210 SUPERFLOC 210 Flocculant
U i SF-212 SUPERFLOC 212 Flocculant
SF -2l4 SUPERFLOC 214 Flocculant
S-3645 Reagent S-3645 Flocculant
* erican Cyanamid Company, Inc.
as 01% by wt. aqueous solutions.
to 30 ml before reacting with the test slurry.
Water Acidity
designation Initial pH (mg/i CaCO 3 )
Iron (II)
mg/i
Aluminum
mg/i
Slurry
PHa
Tyler Run
3.6
63
3
20
6.6
Bennetts Branch
3.7
230
38
30
6.7
Proctor No.
1
3.4
463
64
74
6.6
Proctor No.
2
2.9
1765
320
213
6.7
aFollowing
neutralization
with Ca(OH) 2 . Flocculation
studies were
made at this
pH level.
SF-310 SUPERFLOC 310 Fiocculant
S-3698
Reagent
S-3698 Flocculant
S-3700
Reagent
S-3700 Floccuiant
GPG
AEROSOL
GPG Surface Active
Agent
S-3665
Reagent
S-3665 Surfactant
S-3699
Reagent
S-3699 Floccuiant
The required volume of this solution was diluted
-------
Table 26. SLUDGE SETTLING RATES WITH SUPERPLOC POLIELECTROLYTES.
WATER ORIGIN - BENNETTS BRANCH
Flocculant
Dosage
(ppm)
Floe Size
Settling Rate
Flocs
after
Suspended
15 minutes
Control
-
V.
Small
V.
Slow
75%
SF 127 0.5 Large Fast 0%
0.25 Large Fast 0%
0.1 Medium Large M diun 20%
AF 550 1.0 Small Slow 40%
0.5 Small Slow 40%
SF 202 0.5 Large Fast 5%
0.25 Large Fast 15%
0.1 Medium Slow 60%
SF 210 0.5 Large Fast 5%
0.25 Large Fast 15%
0.1 Medium Slow 50%
SF 214 0.5 Large Fast 5%
0.25 Large Fast 15%
0.1 Medium Slow 50%
SF 310 2.0 V. Small V. Slow 60%
3.0 Small Slow 40%
5 3 6 86 2.0 V. Small V. Slow 75
3.0 Small Slow 50c
5-3698 2.0 V. Small V. Slow 75%
S3699 2.0 V. Small V. Slow 75 %
SF 16 0.25 Medium Medium 40%
SF 212 0.25 Large Fast 15%
0.1 Medium Slow 50%
S-3645 0.25 Large Fast 5%
0.1 CeIiwa Slo w C
aAmorican Cyanamid Company
-------
Table 27. SLUDGE SETTLING RATES WITH SUPERFLOC POLYELECTROLYTES.
WATER ORIGIN - PROCTOR NO. 1
t J
-4
Flocculant
Control
SF 16
SF 127
AF 550
SF 202
SF 210
SF 214
364 5
SI? 310
S 36 9 9
SF 212
S 36 86
S 3698
S370 0
Dosage Time to 200 ml
(ppm) Sludge Volume
(seconds)
772
670
0.15 503
0.05 580
0.15 220
0.05 410
0.05 342
0.3 903
0.10 606
0.15 428
0.05 525
0.15 268
0.05 420
0.15 398
0.05 458
0.15 340
0.05 475
0.6 545
0.20 677
0.6 597
0.20 611
0.05 444
0.20 600
0.20 702
0.20 699
Pulp Volume
after 33
mm (ml.)
105
103
102
70
89
90
115
110
100
105
90
98
100
100
88
90
102
100
102
100
99
100
100
100
Thrbidity of Fffluent
after 33 in or 15
min* (ppm)
40
225*
22
125*
22
40 -
50*
22
105
22
70
22
50 *
22
70 *
22
90
30
240 *
30
205
50
125
l85
165
Caic. Set-
tfin9 F atc
ft/ir)
4.67
3.37
7.14
6.20
16.14
8.76
10.50
3.98
5.33
8.10
6.18
13.4
7.85
9.00
7.08
10.60
6. 84
6.62
4.79
6.00
5.30
7.26
5.98
5.14
5.15
a flerjcafl Cyanamid Company
-------
Table 28.
SLUDGE SETTLING RATES WITH SUPERFLOC POLYELECTROLYTES.
WATER ORIGIN - PROCTOR NO. 2
Flocculant Dosage
(ppm)
Time
to
Sludge Volume, secor s Pulp
ml 500 ml 400 m l 300 ml Volui ie
after 40
miri (ml)
Overflow
Turbidity
after 40
mm (ppm)
Caic,
t1±n
(from
ml-ft/hr.)
Set-
Rate
500
800
ml
600
Control
514
963
1313
1810
343
50
1.71
SF 16
0.5
283
650
1044
1456
300
30
2.16
SF 127
0.5
0.25
126
315
322
710
515
997
819
1479
1372
235
290
30
30
4.37
2.26
AF 350
1.0
2.0
350
504
892
898
1231
1267
1863
2023
310
365
30
30
1.37
1.78
SF 202
3.5
385
824
1183
1752
350
30
1.93
SF 210
0.5
0.25
217
354
516
801
880
1167
1221
1722
2192
295
330
30
30
2.06
1.93
SF 214
0.5
0.25
324
387
734
857
1078
1240
1576
1734
335
340
30
30
2.09
1.22
S3645
0.5
0.25
232
343
569
765
851
1106
1291
1614
2090
290
320
30
30
2.65
2.04
SF 310
4.0
2.0
350
450
641
807
874
1097
1341
1581
320
345
50
50
2.57
2.06
5 3699
2.0
494
899
1220
1737
350
50
1.80
S 212
0.25
361
820
1212
1761
335
30
1.86
S3586
4.0
396
697
943
1381
320
50
2.39
S3698
4.0
521
935
1254
1768
345
50
1.80
S3700
4.0
338
620
845
1303
320
50
2.65
a] noriccn Cvanam±d Company
-------
Table 29. SLUDGE SETTLING RATES WITH DECOLYTE POLYELECTROLYTES
Water Source: Tyler Run - neutralized to about pP 9.0 with hydrate lime slurry
Typical Analysis:
pH - 3.42
Acidity - 126.7
Iron (II) <1 mg/i
Iron (T) - 3.4 mg/i
Aluminum - 19 mg/i
SO 4 - 259 mg/i
Evaluation of Flocculant Type
0
DECOLYTE
DECOLYTE
DECOLYTE
DECOLYTE
DECOLYTE
Blank
710
720
930
940
950
Dosage, ppm
0
1
1
1
1
1
Polymer Ionic Character
+
+
-
-
-
Fioc Size
Settling Rate
Small
Slow
Small
Slow
Small
Slow
Large
Fast
Large
Fast
Medium
Medium
Relative Performance
6
5
4
1
2
3
Rating (1 = Best)
Evaluation of Fiocculant Dosage
(DECOLYTE
930)
Floc Size
Floc Settling 1
Relative Settling
Performance ( 1 = Best)
6
o
0.25
0.50
0.75
1.5
2.0
ppj
ppm
ppm
ppm
ppm
Small
Large
Large
Large
Large
Large
Slow
Fast
Fast
Fast
Fast
Fast
5
4
3
The level and nature of floc produced from this water does not permit interface
evaluation.
2
1
-------
Table 29 (continued). SLUDGE SETTLING RATES WITH
DECOLYTE POLYELECTROLYTES
Water Source: Proctor No. 2 - neutralized with hydrated limo slurry to pH 8.3
Typical Analysis:
ph 2.81 Iron (Ii ), ppm - 440 AlumInum, ppm - 182
iici ditv, ppm 1689 Iron CT), ppm 498 SO 4 , ppm 2510
Flocculant Dosage: 0.75 ppm
Sludge Interface Level
(ml. 0.0017 ft/mi)
DECOLYTE DECOLYTE DECOLYTL DECOLYTE DECOLYTE
( mine.) Blank 710 720 930 940 950
0 1000 lOGO 1000 1000 1000 1000
5 970 960 970 950 950 970
8 925 930 930 900 930 940
10 880 880 830 840 890 910
Slurry Origin: Oxidation Tank effluent using Proctor No. 2 water treated ijth
calcined lime S
iater Analysis: pU >9.0
Iron CT) <1.0 mg/i
Sludge Interface Level
DECOLYTLa93O Dosage, ppm
Time
(nina.) Blank 0.25 0.50 0.75 1.0 1.25 1.50 1.75 2.0
0 500 ml 500 ml 500 ml 500 ml 500 ml 500 ml 500 ml 500 ml 500 ml
1 480 440 350 400 390 320 300 270 250
2 455 330 260 280 280 230 220 205 180
3 420 260 210 230 230 200 190 180 163
4 370 220 185 190 190 170 160 155 150
S 320 200 170 170 170 160 155 150 145
aDiamond Shamrock Corporation
-------
Table 30. SCREENING TESTS WITH ATLASEP FLOCCIJLANTS
Polymer T Control iN 1AI 2A2 3A3 105C
Weakly Moderately Strongly
_______ Monionic Anionic Anionic Anionic Cationic
Relative Response
TylerRun-pH8.7 3 2 1
Proctor No. 1 - pH 6.9 3 1 1
Proctor No. 1 pH 7 9 d 1 1 2 3
Proctor No. 2 pH 7.5 5 1 1 2 4 3
Proctor No. 2 pH 7 8 e 1 2 3 4
Proctor No. 2 pH 8 3 e 2 1
Table 31
Sludge Settling Rates with ATLASEP IN* Flocculant
Water Source: Proctor No. 2 neutralized with hydrated lime slurry
Dosage, ppm Settling Rate, ft/hr .
o 0.85
0.25 2.25
0.50 4.10
1.0 8.OI
1.5 11.7
aAtias Chemical Industries, InC. , ultra-high molecular weight polyelectroiytes.
bjar tests in which fiocculant was added as 0.01% so1u ticn and mixed for 15 seconds.
O0 ml. in 1.000 ml. beaker. Dosage - 1 ppm.
CBased on floc formation characteristics, sludge settling rate and supernatant clarity.
dNeutralized with limestone, all others with hydrated lime slurry.
eDensator r cyc1e sludge, diluted to about 10,000 mg/i suspended solids. Ferrous sludge.
-------
to provide the desired water rise velocity (ft/hr or gal/min/ft 2 )
The clarified effluent rise velocity must be less than the settling
rate of the slowest moving suspended particle. A rise velocity,
which is much lower than the particle settling rate, may be required
with some slurries to provide adequate retention time for floe growth.
The depth of the thickener must comprise four zones: the clear
supernatant, the feed region, and the transition and the compression
zones. The clarified zone contributes a safety factor against dis-
charge of turbid waters (usually at least two feet) while the com-
pression zone aids in producing a more dense sludge. Observations
during these studies indicate the advisability of introducing the
feed slurry from some mine waters into the transition zone to gain
a filtration effect of the settling sludge blanket. At the bottom
of the compression zone, a slow moving rake moves the sludge to a
removal port and prevents non-uniform, excessively dense sludge
build up. Typical unit areas for many suspensions range between
50 and 150 (ft 2 /ton/day) for influent levels containing one or two
percent suspended solids 73 . These values are significantly less
than found by Dorr Oliver (lO9-4000) and in these studies for
mine drainage sludges.
Hollywood Thickener- -
The device utilized is atypical in its design as a thickener since
in addition to this a plication, studies of the sludge recycle con-
cept were to be made 1 with this unit. The studies with this latter
mode of operation are described in Section VIII. The unit is rug-
gedly constructed and was not prone to mechanical problems. It
performed both types of functions satisfactorily; provided means
for good control, sampling and observation capability. The cal-
culated flow rate parameters are given in Table 32, while related
sludge parameters are given in Table 33.
In this thickener, the primary influent zone serves as a feed well,
regulating flow surges and providing additional reaction time. The
secondary zone permits flocculant addition and provides slow agitation
for floc formation. This design permits the introduction of the
feed slurry below the sludge blanket into the transition zone, if
desired.
These atypical aspects of this thickeners design reduces the over-
all capacity. In the solids separation zone (feed solids and tran-
sition regions) the downflow settling of the floc proceeds along with
the upflow rise velocity of the clarified water. The feed well ex-
tends ten and onehalf feet below the overflow level. This provision
permits an upflow sludge filtration region of eight feet when the
sludge blanket is maintained two feet below the overflow. When the
132
-------
Table 32. THICKENER FLOW RATE PARAMETERS
I -.
Zone
Zone
capacity
gallons
Retention
minutes
time Flow velocity
feet per hour
Gallons/minute!
ft 2
100 200
gpm gpm
350 100 200 350
gpm gprn gpm gpm
100
gpm
200 350
gpm gprn
Primary
(Downf I ow)
3,730
37.3 18.6
10.7 8.46 16.9 29.4
1.05
2.10 3.68
Secondary
(Downf 1 ow)
3,730
37.3 18.6
10.7 8.46 16.9 29.4
1.05
2.10 3.68
Compression
Zone
13,550
a
1,130
See Table 33
.
for Compression Zone Data
Solids
Separation
(Upf 1 ow)
29,200
292.0 146.0
83.2 2.06 4.12 7.2
0.253
0.506 0.885
Total
51,340
513.0 257.0
147.0
asludge Collection Zone
-------
Table 33. SLUDGE RATE PARAMETERS FOR
25-FOOT THiCKENER
Influent Solids Level mg/i
1000 - 2000
100 200 350
Flow Rate - gpm
100 200 350 100
500
200 330
Solids in
Und rf low
2
4
6
a
Retention Time-Compression Zone hoursa
98
49
28
49
25
14
25
12
7
196
98
56
98
49
28
49
25
14
295
147
84
147
73
42
73
37
21
392
196
112
196
98
56
98
49
28
Sludge Generation (Equilibrium
Blow Down)
Rat p
2
2.5
5.0
8.75
5.0
10.0
17.5
10.0
20.0
35.0
4
1.25
2.5
4.38
2.5
5.0
8.7
5.0
10.0
17.5
6
0.83
1.66
2.91
1.67
3.33
5.8
3.3
6.7
11.7
8
0.63
1.25
2.19
1.25
2.50
4.4
2.5
5.0
8.8
lbs/day
605
1210
2102
1210
2420
4204
2420
4840 8408
lbs/ft 2 /
1.23
2.47
4.29
2.47
4.94
8.58
4.94
9,88 17.16
24 hrs
acalculated from compression zone volume of 14,700 gallons and bottom
area of 490 ft 2 . Sludge depth from bottom of secondary zone -
5 feet.
134
-------
sludge level is carried above the bottom of the secondary zone wall,
a desirable upflow filtration of fines through the blanket occurs
and can function to remove near-colloidal material from the rising
water column. The thickener was satisfactorily operated under these
conditions during several limestone neutralization runs resulting in
good effluent clarity. This actual performance deviated signifi-
cantly from bench scale settling tests with the same feed which
failed to yield clear effluents, even with some flocculant additions.
The use of thickener designs which include deep feed wells thus per-
mitting upflow sludge blanket clarification is deemed to be an im-
portant discovery.
The upf low velocities of this thickener ranged from 2.06 ft/hr at
100 gpm to 7.2 ft/hr at 350 gpm. The measured settling rates of
freshly precipitated sludges ranged between one and six ft/hr with
a maximum of 16 ft/hr with flocculants. More typical sludge settling
rates were nearer two ft/hr. This size thickener would function
satisfactorily with flows near 100 gpin; however, when flocculants
were used the rates could be expanded about three times and still
maintain a clear effluent. The functioning of the primary and secon
dary zones were quite adequate for the maximum flow (350 gpm) to
which the unit was subjected. In some tests, with Type II, slow
settling sludges, and with the use of pulverized limestone as a
reagent, excessive turbidity in the effluent did occur.
Densator Performance as a Thickener- -
The Densator was operated as a thickener. Typical performance data
for the various conditions are shown in Tables 34 and 35. All data
utilized the following flowsheet. Limestone reactions introduced
a pulverized slurry into the flash mixer since the rotary limestone
reactor effluent could not be introduced to the thickener.
Raw Water -+ Flash Mixer - Oxidation Tanks + Thickener
Alkali Slurry Dewater χ Underfiow Effluent
+
Settling Lagoon
of Stream
Good clarification was maintained at flow rates less than 125 gpm
when 1 to 2 ppm of flocculant were used. Flocculant additions were
essential. Higher flow rates presented difficulty in keeping the
sludge blanket level below a 2-foot depth from the overflow. This
suggests rising velocities greater than about 3 ft/hr should not
be exceeded.
Sludge thickening was erratic with underfiow slurries ranging from
0.7 to 7 percent solids. The lower levels resulted from low acidity
135
-------
Table 34. TYPICAL ThICKENER PERFORMANCE DATA
IJNDERFLOW
2 Solids, vt/vol
0.8 1.6 1.96
Water
Source
Tyler Run Bennetts Branch
Neutralization
Proctor No. 1 Proctor No. 2
Reagent
Pul.
Pu!.
Pul.
Pu!.
Ca0 Mg0 CaCO 3 Ca(OH) 2
Cao .MgO CaCO 3 Ca(0H) 2
Ca0 Mg0 CaCO 3 Ca(OR) 2
CaO NgO CacO 3
IN FLUENT
Flow Rate gpm
62
100
140
62
90
100
85
100
106
257
120
Suspended Solids mg/i
64
75
190
160
780
675
750
1500
1860
1600
1870
pH
7.7
7.6
8.6
7.4
7.5
8.3
7.5
7.0
7.1
8.0
8.3
Ferrous Iron irig/1
1
1
1
1
1
4
2
10
2
9
1
Alkalinity mg/i
26
32
23
15
15
9
17
10
24
914
84
Temperature F
50
49
51
51
52
50
49
45
60
45
51
Settling Rates ft/hr a
Volume of Settled Sludge 2
0.5
20
*
1.2
4.5
1.5
0.54.0
*
3.5
12
1.9
5
*
2.2
20
1.8
10
1.9
20
FiOAUfltDS ) mg/i
12
12
12
12
12
12
12
12
12
NONE
12
OVERFLOW EFFLUENT
Suspended Solids mg/i
15
30
5
10
27
18
19
683
51
Soluble Ferrous Iron mg/i
1
1
1
1
1
4
1
1
1
Alkalinity mg/i
20
32
22
85
5
65
1632
66
pH
7.5
7.5
8.3
8.2
8.5
7.1
7.2
7.2
8.4
4 4
I I
10 16
8.2 7.5
Settled sludge volume is percentage of original volume of infiuent after two hours settling time.
An average value of 2.0 ft/sec which will yield a moderately clear effluent. Probably acceptable ecologically
caused by CaCO 3 fines. A conservative value of 20 ft/sec assumed for settling rate of bulk of precipitate.
2.1 2.5 6.7 2.5 4.8 2.7 1.3 7.19
since the turbidity would be
-------
Feed DIlUSIOfla
Final D1lU Ofla
Maximum - Feed
Retention Tine - hrs.
j Compression Zone - Sludge
Retention Time - hrs.
Unit Areab
ft T/day
Sludge Production
lbs/ft 2 /daj
Table 35 OBSERVED THICKENER PERFORNANCE - SIZING DATA
a ight ratio of liquid to solids based upon suspended solids determination
l.333(Feed DijutionWt. liquid Final Dilution liquid )
wt. solids wt. solids
Suspended solids settling rate ft/hr
Water Source
Tyler Run Bennetts Branch
Proctor No. 1 Proctor No. 2
1 eutralization
Reagent
Ca(OH) 2
Pul.
CeO MgO CaCO 3
CaO MgO
Pul.
OaCO 3
2a(OH) 2
CaD MgO
Put.
CaCD 3
Ca(OH) 2
CaD 3 gO
Pal.
CaC3 3
15625 13333 5263
6250
1282
1481
1333
667
538
625
535
125.0 62.525.0 51.036.8
47.631.3
40.0
l 1.9
40.3
20.8
37.0
67.3
13.3
11.7 8.6 6.2
11.7
9.5
8.6
9.5
8.6
2.1
3.4
7.3
494 5231307 180249
518789
87
243
96
78
34
8
79
41,230 8,845_88L4 5,790
5,512
82883
558
907
43043
303
412
35637
0.05 0.23 0.34
0.36
2.4
3.6
2.2
4.6
6.6
4.8
bu 0 tt Area
-------
waters at high flow rates. A typical underfiow solids content was
about 2 percent. The more dense sludges developed when pulverized
limestone was used as the neutralizing agent. This experience was
consistent with the results of bench settling tests.
Other observations from the operation of the Densato-r as a thickener
are indicated in part by the data in Tables 34 and 35:
1. Pulverized limestone, as a neutralizing agent, often resulted
in turbid overflows which were difficult to control. Pulverized
limestone produced a greater quantity of neutralized solids in
the influent than with a similar alkaline equivalent of hydrated
lime.
2. Reagents with slower neutralizing reactions rates, as dolomitic
hydrate and limestone, continued both neutralizing and oxidizing
reactions in the thickener. These continuing reactions were
apparent from decreases in ferrous iron concentration and in-
creases in pH in the thickener effluent. A thickener can be ex-
pected to provide important secondary reaction time to insure
specification effluents, but with possible turbidity problems.
3. A sludge level maintained above the bottom of the feed well pro-
vides for particle entrapment during upf low through the sludge
blanket. This entrapment improves effluent clarity when diffi-
cult dispersions occur, as in limestone neutralization.
4. The thickener bottom rake action produces a dewatering action
on the settled sludge. Some underfiows attained a solid con-
tent of five to seven percent after 24 hours of settling time,
whereas the same slurry, settled under quiescent conditions,
never exceeded two percent solids after 48 hours settling time.
5. Waters having low levels of pollutant loading, produce low settled
solids density and require large unit areas. In plant design,
serious consideration should be given to the use of a second
stage thickener possibly with flocculant addition, to attain a
more dense sludge product.
6. Sludge removal control followed either slow Continuous pumping
or timed-interval blow-down, utilizing automatic pneumatic
valves in the system. The test schedule was too short to es-
tablish optimum sludge storage and blow-down considerations.
Recommended Design Parameters for Thickeners--
The thickener studies provided information for establishing the
design criteria for effective clairfying and thickening of mine
138
-------
slurries. An optimum design would provide the following features:
1. Clarity of overflow with all types of slurries.
2. An uz-iderflow of high density solids suitable for secondary de-
watering.
3. Provision for sludge recycling to achieve underflow solids con-
tent of at least five percent.
4. Provision for second stage thickening if necessary.
5. Provision for upf low of suspensions through sludge blanket if
colloidal solids are present.
6. Provision for settled sludge surge capacity to permit economical
sizing and operation of secondary dewatering devices or disposal.
7. Provision for flocculant addition to improve and control sludge
settling rates.
Design parameters required to attain the above conditions were
established and are listed in Table 36.
Estimated Thickener Operation Costs - -
Capital costs of thickening equipment can vary appreciably. For pre-
liniinary estimation purposes, the installed equipment costs for
various sizes of thickeners are presented in Table 37. The thickener
sizes selected are based upon a minimum sludge settling rate of 1.5
ftlhr and a rising velocity of about 0.2 gpm/ft 2 . The settling rate
largely controls the size of the thickener required. The effect of
sludge settling rate on equipment costs is shown in Figure 26. The
natural settling rate of most slurries can be increased several fold
with polyelectrolyte flocculants. Figure 26 permits estimation of
the potential of flocculant usage in determining thickener sizing and
costs.
Estimates of thickening costs for three sizes of plants are shown in
Table 35. These are based upon a labor requirement estimate of one
man-hour per day and a flocculant dosage of 1 ppm of thickener influ-
ent.
These estimates are very significant in demonstrating low processing
costs for large volume plants. These costs, combined with reduced
area requirements, superior water quality control and the positive
solids handling which a thickener provides, makes their use in large
plants more attractive than a system of large settling lagoons.
139
-------
Table 36. SUGGESTED OPTIMUM DESIGN PARAMETERS
FOR THICKENER OPERATION
1. Rising Water Velocity - gprnlft 2 0.2 maximum
2. Solids Settling Rate - ft/hr 2.0 minimum
3. Unit area ft 2 /T/day 300 minimum
(i nay exceed 10,000)
4. % Solids in Underfiow 5.0 minimum
5. Sludge Recycle Provision - S to 30
% Circulating Load
6. % Solids to Feed Well (Combined 1 minimum
raw feed plus recycle sludge)
7. Flocculant Dosage - ppm influent 1 - 2
8. Depth of Feed Well Below Sludge Level - ft 2 minimum
9. Sludge Depth Below Feed Well - ft 6 minimum
10. Settled Sludge Retention Time - hrs 72 minimum
Table 37. APPROXIMATE THICKENER EQUIPMENT COSTS
Annual
.
deprec-
iation
Daily
Capital
CMD plant
Thickener
Thickener
(10 yr
deprec-
cost
capacity,
size diam.,
cost,
basis),
iation
( /
gpd
ft
$
$
$
1000 gal)
100,000
25
55,000
5,500
15
15.0
250,000
40
72,000
7,200
20
8.0
500,000
50
80,000
8,000
22
4.4
1,000,000
100
100,000
10,000
27
1.0
5,000,000
150
150,000
15,000
41
0.8
10,000,000
200
250,000
25,000
69
0.7
140
-------
Figure 26. Sizing and thickener cost estimation data
AMORTIZATION COST 4/1000 GAL. FOR 500,000 G.P.D. CAPACITY
1.0 2.0 3.0 4.0 5.0 6.0
I I I I I
4
INSTALLE
D COST
OF
THICKENER
$50,000
I I I I
i
I
i
I I
$100,000
I
I
I
100
200
300 400
500
WATER
FLOW RATE.G.P.M.
.
100,000
-
700,000
200:000 300000 400:000 500:000 600000
MINE WATER TREATMENT PLANT CAPACITY-G.P.D.
141
0
/
1/
60
50
40
uJ
U-
a
w
a
30
UJ
z
w
U
2O
10
4
S..
,_1-
,
/
/
/
-------
Table 38. DAILY THICKENER OPERATION COSTS
CMD plant capacity - gpd 250,000 - 1,000,000 5,000,000
Thickener Diameter - ft
Equipment Cost - $
Annual Depreciation (10 year basis) - $
Operating Costs
Direct Labor - @ $3.50/hr - $
Supervision (25% Direct Labor)
Utilities
Electrical - 0 l.5 Jkwh - $
Flocculants (I ppm) - 0 $1.25/lb - $
Maintenance (1% equipment cost/yr) - $
Total Operating Cost - $
Amortization - $
Total Unit Operation Cost $/day
Cost of Thickening - cents/1000 gal CMD
25.00
72,000.00
7,200.00
3.50
0.88
1.20
2 .63
1.97
10.18
19.73
29.91
12.01
100.00
100, 000.00
10,000.00
3.50
0.88
4.80
10.52
2.74
22.44
27 .40
49.80
S.0
150. 00
150,000.00
15,000.00
7.00
1.75
24.00
52.00
4.11
89.46
41.10
130.56
2.6
-------
Summary and Conclusions for Thickener Operation- -
Pilot plant experience confirmed engineering predictions that a
thickener is the key primary dewatering process for any plant in-
volving the treatment of flows greater than 750,000 gpd with a
neutralized suspended solids content of 1,000 mg/l or greater.
The precipitated solids handling and disposal problems arising
from such waters require the superior process control that only a
thickener can provide. This is an essential requirement for an
effective, low-cost treatment process and is the logical alterna-
tive for settling lagoons. The great advantage of a thickener
over the settling lagoon is the greater efficiency and the less
costly sludge handling by making secondary dewatering feasible.
Settling Lagoons
The objectives of the settling lagoon studies were to observe:
lagoon performance during year-round operation as a water clarifying
process; flow and turbulence within the lagoon; secondary water
quality changes within the lagoon; and the response to atmospheric
factors as temperature, precipitation, and wind. Further, informa-
tion was desired on the nature of sludge accumulation and changes
in sludge properties which control its handling by gravity flow and
by pumping.
Simplicity, convention and apparent low cost have encouraged the use
of this unit operation. However, maintained and reliable performance
of lagoons is found to involve many complex functions, all of which
are seldom optimized. Among the functions frequently sought, in
addition to solid-fluid separation, are: uniform effluent quality
based on buffering response of large volumes, providing treated
water for other uses, effluent quality monitoring, retention time
to complete reactions, and the control to densify, dewater, store,
and even dry the sludge. The particular functions required will
determine the design. The lagoon system types may be grouped:
clarification of low level solids requiring minimum sludge storage
(Calhoun 24 ), clarifier-thickener providing for temporary sludge
storage and intermittant removal of sludge (most common approach),
total sludge control without sludge removal (requiring very large
multiple lagoons (Mihok 8 ), and secondary sludge storage including
permanent disposal (Holland 28 ). Providing capability for inter-
mittant sludge removal usually limits the lagoon size. The process
factors which determine lagoon design are: flow rate, unit sludge
volume, slurry settling rate, sludge characteristics, and require-
ments for secondary dewatering of the sludge. Weather conditions,
including heavy rainfall, wind, temperature layering and freezing,
can drastically affect operation of a settling lagoon but these
effects may be partially counteracted by design.
143
-------
Lagoon operation based on continual sludge deposition is not only
subject to decreasing retention time but currents and particle
suspension which lead to particle discharge in the effluent.
deVilliers 6 reports on design factors contributing to these diffi-
culties. However, such trends in lagoon design modification in-
crease costs and lead to mechanical sludge removal and peripheral
effluent discharge- -just a step removed from a thickener.
Basis for the Hollywood Settling Lagoon Design- -
Seeking compliance with Sanitary Water Board guidelines]oa and versa-
tility with minimal construction cost, the unique lagoon design
incorporated a bottom, gravity flow sludge removal capability. It
is illustrated in Figure 4 and in Appendix B, Figures 4, 5 and 11.
Some of its major design features include:
1. Three-compartment, sectionalized construction permitting opera-
tion of the influent section as a thickener, with the following
sections acting as clarifying and polishing operations.
2. Full lagoon-width influent weir minimizing velocity and pro-
viding a uniform feed stream at the surface leading to effec-
tive utilization of the entire lagoon. This greatly reduced
turbulence and bottom agitation at the feed end. This action
and a typical sludge settling pattern is shown in Figure 27.
.
L ______
Figure 27. Settling Lagoon Weir Overflow Showing Sludge Settling
Pattern,
144
-------
3. Transverse baffles separating compartments but also minimizing
currents, surface windage effects, and short circuiting. The
bottom elevation of the baffles defined the design sludge storage
volume.
4. Tapered sides and bottom with transverse sludge channels (in first
two compartments) allowing bottom sludge removal with minimal
disruption of the settling functions or development of turbidity.
The slopes assisted the sludge flow to the outlet.
5. Smooth, impervious, bituminous-covered lagoon bottom assisting
movement of the settled sludge along slopes.
6. Sludge pump station (Figures 4, 6, 7 and Appendix B, Figures 5
and 10) transferring the settled sludge to secondary dewatering
operations.
7. Lagoon effluent structure (Figures 4, 6, 7 and Appendix B, Figures
4, 5, 10 and 11) releasing clear water from surface of lagoon,
providing for monitoring flow and quality, permitting transfer
of treated water for plant use and allowing variation in water
depth.
The lagoon retention times at different flow rates are plotted in
Figure 28. The changes in retention time with filled sludge storage
areas are also shown. Major changes in retention time with flow
rate occur at low flows while small retention differences result at
high flow rates.
Settling Lagoon Operating Observations- -
The settling lagoon was essentially in continuous use during the
study as the final settling and water polishing unit operation.
This provided observation of long-term operating performance and
operating information, especially during cold weather.
Water Handling--Very few water handling problems arose with lagoon
except during severe winter weather. Under normal operating condi-
tions, a minimum clear water depth of five feet was maintained in
the feed compartment while handling maximum plant flow of 350 gpm
containing up to 2,000 ppm of neutralized solids. Retention time
was more than adequate even with sludge storage exceeding design
capacity. Turbid effluents occurred only when situations unrelated
to settling phenomena existed. The baffle and lagoon bottom design
controlled short circuiting difficulties. No sludge inversions or
floating islands of sludge were experienced. The clarity of lagoon
effluents under these conditions was satisfactory for discharge. A
major increase in turbidity of the lagoon effluent was associated with
145
-------
Figure 28. Settling lagoon calculated retention time
Totol
150
80
60
j
0
\
Total
Excluding Live Sludge Storage
\
1
0
-c
S
E
I
J 40
Compartment 3
/
Compartment 1
20
10
0
50
100
Flow Rote (gpm)
-------
heavy rains. Lesser increases in turbidity occurred during rapid
sludge removal periods. The heat loss from the quiescent water
during severe winter weather permitted ice cake formation on the
lagoon surface. Under these conditions, a limited water flow was
formed as a narrow channel between the ice, growing down, and the
sludge, building up. See illustration in Figure 29. This situa-
tion impeded clarification. Under these conditions, the water
velocity increased, while the retention time and sludge settling
rates decreased. Sludge accumulation was gradually extended into
the final compartment and the effluent turbidity increased. Con-
tinuous or frequent sludge removal from the bottom minimized this
problem, although adequate sludge storage should be provided. If
ice prevents surface removal of the sludge, the increased compression,
aging, and gelling of the sludge further degrades its flow charact-
istics.
The adjustable height overflow pipes discharged only the clearest
effluent in the lagoon. They were maintained nearly vertical to
maximize retention time. There was no apparent advantage from
changing their position from center of the effluent structure. The
pipes were moved only to lower the water level prior to sludge clean-
Out operations.
The full-width influent weir compartment was found to be excessively
deep which permitted settling of larger or rapid settling particles
in the weir. However, the design objective to achieve a low in-
fluent velocity and initiate rapid sludge settling functioned well,
even during freezing temperatures.
The only effluent clarity difficulties occurred during the use of
carbonate reagents or when the preceding neutralization or oxidation
reactions were incomplete. The Surface Scatter Turbidimeter evalua-
ting the effluent responded well indicating even minor trends toward
an unsatisfactory condition.
Sludge Handling--Bottom sludge removal from the lagoon was at times
troublesome. The observations regarding sludge handling character-
istics included:
1. The sludge moved satisfactorily from the bottom channels of the
settling lagoon to the sludge drying basin by gravity, prior to
its gellation, although it tended to resist flow from the sloped
bottoms to the channel. The positive displacement pump in the
sludge pumping stations was used only during major lagoon clean-
ing and in transferring sludge to the Control Building for fil-
tration. The limestone-produced sludge is most difficult to move
due to its density. The plant operators were reluctant to trans-
fer sludge from the lagoon to the drying basins at controlled
147
-------
Figure 29. Sludge deposition in settling lagoon
Influent Wejr
2 Sludge Depth
NORMAL OPERATION
Sludge Channel
Effluent Weir__ 7
Ice Layer
Flow Channel
Ice
u i Proposed Reserve Sludge
J Storage Capacity Required
for Winter Operation
Ice
-4
-
-------
slow rates, seeming to prefer more rapid flows. This psychological
reluctance, unexplained, extended beyond the press of other duties.
Low velocity flows are preferable since the tendency toward rat
holing increases with flow rate.
2. The mobility of an aged, settled sludge, when covered by sev-
eral feet of water was poor. It would not flow freely from the
sloped bottom into the sludge channels and hence through the
effluent pipes by gravity flow or from the pump. This indi-
cates that the angle of repose of the submerged sludge is greater
than 300, and slopes approaching 450 are preferable. A slow
moving screw with deep blades within the bottom channel would
be helpful in directing the sludge toward the exit ports to main-
tain maximum sludge density.
3. Mobility of aged sludge on the sloped surface of a drained lagoon
was satisfactory when movement was initiated by a squeegee or
hose water pressure. It tends to move as a large block. On
quiescent compression and aging, the sludge forms a gel which
has thixotropic tendencies. The gellation behavior develops
within 48 hours under sludge and water pressure compression.
There was no evidence of this gellation phenomena in the settled
sludge in the thickener. Apparently the slow movement of the
sludge by the thickener rake prevented this gel formation. A
detailed understanding of this aging and gel formation is needed
to assist lagoon design and operation. A two- and three-foot
thick layer of gelled sludge in one Section of the lagoon could
be moved into the sludge channel by two men in 4 to 5 hours
using squeegees and low pressure hoses. Blocks of sludge were
cut with shovels and would slide down the sloping bottom to the
sludge channels, provided the bottom was kept thoroughly wetted
by a small flow of plant water from a hose. Such settled sludge
would normally range between five and 15 percent solids. This
procedure was rapid and minimized sludge dilution. There was
no evidence that the bituminous bottoms significantly enhanced
sludge movement, but they did maintain stable dimensions and
resist erosion or mud formation during sludge removal. The
side walls of the lagoon were earthen. Although no significant
erosion problems developed with the compacted clay lagoon walls,
sludge removal from these areas was more difficult. An occasional
small stone from the lagoon walls would block the check valve of
the sludge pump. Total bituminous side and bottom surfaces are
definitely advantageous and recommended. Under no conditions
should a stone rip-rap be employed below water levels in a
lagoon.
4. Transfer of gelled sludge from the settling lagoon to the drying
beds by gravity flow was unsatisfactory. This sludge form tends
149
-------
to rat-hole quickly, allowing excessive amounts of clarified
supernatant to move with the sludge and complicate subsequent
operations.
5. The sludge had adequate flow characteristics in lines and through
pumps when movement began. The sludge transfer lines were flushed
with clear water after ceasing sludge transfer to prevent scaling
or further gellation. No line blockage was experienced.
6. Sludges produced with hydroxide-type alkalies were the least
dense and did not adhere very tightly to the lagoon bottom. In
contrast, settled limestone slurries were dense, sticky, and
clay-like in character and adhered tenaciously to the bottom
surface.
7. The distribution of sludge among the three compartments of the
settling lagoon could be related to the sludge settling rates.
Sludges settling at rates in excess of 2 ft/hr seldom passed
into the second compartment. when the sludge was not removed
and the first compartment approached its sludge capacity level
(bottom of baffles), the settled sludge was carried into the
second compartment. During one three-month period without
sludge removal, some sludge collected in the final polishing
compartment.
Slower settling, Type II sludges, tended to drop more uniformly
throughout compartments one and two with little deposition in
the final compartment. The very rapidly settling sludges, such
as those from limestone treatment, accumulated on the sloped
bottom of the first compartment, a few feet from the weir. See
rapid settling shown in Figure 27. Only as the accumulation in
this region increased, did the sludge storage extend beyond the
sludge removal channel. The sludge removal practice sought to
maintain low sludge storage levels in the first compartment.
There were distinct differences in the sludges settled in the
three compartments. That settling in the second compartment
was less dense and tacky with smaller, lighter floe. These
sludges showed lesser tendencies toward gelling, rat-holing
and were easier to move. They maintained a smaller angle of
repose.
Settling Lagoon Performance Data- -
The settling lagoon functioned well as a clarifier in dewatering.
The turbidity values of the effluent ranged between 1 and 5 jtu with
suspended solids between 2 and 13 ppm. The ditch returning the plant
effluent to Tyler Run showed no staining after the years program.
Satisfactory water quality was indicated by several generations of
150
-------
animal and plant growth in the third lagoon compartment. These in-
cluded toads, frogs, salamanders, and insects. Trout and bass trans-
ferred to this section of the lagoon found ways of accumulating ex-
cessive-levels of sludge in their gills, although they survived well
in the treated waters transferred from the lagoon. Deer and bear
frequently drank from the lagoon. Operating data for the four water
sources and several reagents are given in Tables 39 and 42. The slow
settling floc of Type II sludges gave the highest effluent solids
level, although not necessarily the highest turbidity. The suspended
solids and iron levels of these effluent waters met specifications
at the highest flows hydraulically possible through the plant. The
lagoon was operated once continuously without sludge removal for a
sufficient period of time to consider its liveu sludge storage
capacity. The longest service (over a period of three months) was
equivalent to 42 days of continuous operating time at an average
influent flow rate of 175 gpm (250,000 GPD). Assuming the formation
of 12,500 GPD of settled sludge (5 percent of original volume) and
a live sludge storage volume of 77,800 gal., the capacity of the
lagoon would be only six days of settled sludge production. At the
conclusion of long operating period, sludge had accumulated nearly
to the influent weir with about three feet of clear water in the
first two compartments and about two feet of settled sludge in the
third or polishing compartment. The clarity remained satisfactory.
This observation suggests that sludge compression had occurred and
a live sludge load capacity, based upon 48-hour settling volume, is
a conservative but realistic estimate for design purposes. A con-
siderable safety factor beyond the one-month sludge storage recom-
mended by the Pennsylvania guidelinesboa is achieved. Due to diffi-
culties which can develop during sub-freezing weather, a two-month
sludge storage volume seems preferable.
Time limitations of the test program did not permit further consid-
eration of the settling lagoon as a static sludge storage vessel as
is commonly being used by many mining companies. It is a practice
which should be avoided by more frequent sludge removal.
The Continuous or frequent interval removal of sludge from a settling
lagoon is definitely advantageous. When sludge was permitted to
accumulate in the lagoon and its removal was carried out, many diffi-
culties were experienced. Removal would have been most difficult
during freezing winter conditions. Labor requirements to manually
assist sludge removal, even with described bottom design, was high.
It appears that a settling lagoon is not an optimum unit operation
in a mine water treatment flowsheet except for plants with small
capacity or when treating waters with low solids loading.
151
-------
Table 39.
TYPICAL SETTLING LAGOON PERFORMANCE -
BENNETS BRANCH WATER
Table 40. TYPICAL SETTLING LAGOON PERFORMANCE -
TYLER RUN WATER
Pebble
Lime
Pebble
Lime
Hydrated
Lime
Tnt luent
Flow Rate gpm
100
200
307
Suspended Solids - ppm
105
95
pH
5.49.8
5.19.3
6.16.3
Ferrous Iron - ppm
1-4
15
46
Alkalinity ppm
46
114
43
Separated Sludge
Sludge Volume, %
4.5
4.5
4.5
Settled Sludge Solids, wt. %
0.2
0.2
0.2
Sludge Production, gpd
6,500
13,000
19,800
Dry Solids, lbs/day
Estimated Operational Days
108
216
330
without sludge removal
Compartment No. 1
6
3
2
Compartments No. 1 and 2
Feed Dilution b
Final Dilution
12
9,537
500
6
10,500
500
4
25,100
500
Unit Area, 1t 2 /t/dayC
24,093
26,660
65,582
Effluent
pH
7.08.9
6.69.6
6.3
Ferrous Iron ppm
01
14
24
Suspended Solids - ppm
6
8
7
Temperature °F
32
32
57
Alkalinity ppm
64
31
15
abased on lagoon sludge storage capacity and live sludge
volume.
bBdsed on liquid to solid weight ratios.
Pebble
Lime
Pebble
Lime
Hydrated
Lime
Influent
Flow Rate - gpm
Suspended Solids - ppm
pH
Ferrous Iron ppm
Alkalinity
100
70
8.49.6
01
120
200
55
7.612.0
0-1
15
31 4
65
5.69.8
Separated Sludge
Sludge Volume, % 2.0
Settled Sludge Solids, WI. % 0.2
Sludge Production, gpd 2,880
Dry Solids, lbs/day 48
Estimated Operational Daysa
without sludge removal
Compartment No. 1 14
Compartments No. 1 and 2 27
Feed Oilutiorib 13,330
Final Dilutjonb 500
Unit Area, ft 2 /t/dayC 68,408
2.0
0.2
5,760
96
7
14
18,200
5 Q
94,376
2.0
0.2
9,907
165
4
8
15,400
500
79,L 44
Effluent
pH
Ferrous Iron ppm
Suspended Solids - ppm
Temperature °F
Alkalinity ppm
6.59.6
0-1
25
38
30
8.910.6
0-1
15
42
20
6.67.5
01
18
65
13
3 Based on lagoon sludge storage capacit/ and live sludge
volume.
bBased on liquid to solid weight ratios.
N
L()
-4
CSe settling rates in thickener section.
CSee settling rates in thickener section.
-------
Table 41. TYPICAL SETTLING LAGOON PERFORMANCE -
PROCTOR NO. 2 WATER
Pebble
Lime
Pebble
Lime
Pulverized
Limestone
Influent
Flow Rate gpm
Suspended Solids ppm
pH
Ferrous Iron - ppm
Alkalinity ppm
100
11905830
7.39.0
20
60
200
2289
7.510.0
01
50
133
1355
7.58.1
0-1
83
Separated Sludge
Sludge Volume, %
Settled Sludge Solids, wt. %
Sludge Production, gpd
Dry Solids, lbs/day
Estimated Operational Days a
without sludge removal
Compartment No. 1
Compartments No. 1 and 2
Feed Oilutionb
Final Dilutionb
Unit Area, ft 2 /t/dayc
29.6
0.146
42,624
1,636
0.9
1.8
454
20
579
29.6
0.146
85,248
3,273
0.5
0.9
1455
20
581
2.0
6.8
3,830
2,173
11
20
773
15
246
Effluent
pH
8.09.5
914
7.2
Ferrous Iron - ppm
0-4
0-1
2
Suspended Solids -
ppm
4
3
10
Alkalinity ppm
20
50
65
Temperature - 03
140
140
45
a Based on lagoon sludge storage capacity and live sludge
volume.
bBased on liquid to solid weight ratios
Table 42. TYPICAL SETTLING LAGOON PERFORMANCE -
PROCTOR NO. 1 WATER
Pebble
Lime
Pebble
Lime
Hydrated
Lime
I nf lu ent
Flow Rate - gpm
Suspended Solids - ppm
pH
Ferrous Iron ppm
Alkalinity - ppm
100
5.78.4
111
99
200
360
5.49.5
16
160
237
380
6.36.6
1-3
-
Separated Sludge
Sludge Volume, % 10.0
Settled Sludge Solids, wt. % 0.35
Sludge Production, gpd 114,400
Dry Solids, lbs/day 1421
Estimated Operational Daysa
without sludge removal
Compartment No. 1 3
Compartments No. 1 and 2 5
Feed Dilutior 2,878
Final Dilutionb 286
Unit Area, ft 2 ft/day 987
10.0
0.35
28,803
841
1.14
2.7
2,878
286
987
8.0
0.10
27,302
228
1.5
2.8
2,782
1,000
505
Effluent
pH
Ferrous Iron ppm
Suspended Solids - ppm
Temperature - °F
Alkalinity - ppm
7.49.3
13
10
35
-
6.89.4
01
5
38
35
6.46.5
26
6
-
45
aBased on lagoon sludge storage capacity and live sludge
volume.
bBased on liquid to solid weight ratios.
v i
cJ
C See settling rates in thickener section.
°See settling rates in thickener section.
-------
Cost of Settling Lagoons--
In addition to sludge settling rates, the sludge storage and/or
removal capability must be established for design purposes. Live
sludge storage capacity must be adequate for the service time re-
quired of the lagoon. This capacity is defined as the 48-hour
settled sludge volume. With highly loaded waters, the 48-hour
settled sludge volume will fall within the range of 5 to 20 percent
of the treated water volume.
Assuming lagoon operation by sludge sedimentation, temporary sludge
storage, and interval surface sludge removal, cost estimates for
several lagoon sizes are given in Table 44. The plant capacities,
sludge development levels, and provision for two-months sludge
storage are developed in Table 43. The costs include the elementary
excavation of lagoons 12 feet deep under fair soil conditions. Soil
compaction, complex shapes and bituminous-covered surfaces will in-
crease costs. Construction must include compaction and erosion
control. The one-month sludge storage capacity recommendedboa is
generally inadequate to insure good operation in subfreezing weather
in northern Appalachia, while the five-foot minimal clear water
depth may be too conservative. The 48-hour sludge volume basis may
be too conservative as Hollywood data and observation by others 28
indicate some sludge volume reduction (up to 50 percent) while under
long term compression; however the basis appears to incorporate a
highly desirable safety factor.
The simple pond concept used for the estimates represents the in-
efficient, troublesome units now in widespread use. The lagoon
sizes calculated for plants handling greater than one million GPD
are larger than those presently in operation in the same flow capacity
bracket. Many of these lagoons perform satisfactorily for only a
brief period. Difficulties are experienced from charging excessive
amounts of suspended solids and in sludge removal. The costs of
these lagoons are high. They have minimal versatility and no in-
trinsic or salvage value. For large lagoons, the land area require-
ments are significant and costly. Difficulties in sludge removal
increase with size. At some locations, existing topographic features,
including abandoned surface mines, may reduce these costs.
Comparative sizes and costs for thickeners and simple settling
lagoons are shown in Table 45. With 500,000 GPD capacity, the costs
are about equal, while for larger systems, equipment costs and space
requirements are significantly lower for thickener installations.
Since a thickener is the ideal settling lagoon, careful evaluation
and comparison between a thickener and settling lagoon should be
considered for the primary sludge removal of systems treating large
volumes of neutralized coal mine water. Settling lagoons should be
154
-------
Table 43. SETTLING LAGOON AREA AND VOLUME REQUIREMENTS
Basis: Lagoon providing 10 foot depth of live settled sludge working volume
CMD Plant
Capacity
gpd
48-hr
sludge
vol.
raw
Settled
volume
% of
GMD
Daily
settled
ga
volume
sludge
is.
Area
sludge
required for l0-foo
depth accumulation
t
for
One day
ft 2
60
f
Days
t 2
One
year
ft
Acres
100,000 2 2,000 27 1,600 9,600 0.3
5 5,000 67 4,000 24,000 0.7
10 10,000 134 8,000 48,000 1.4
250,000 2 5,000 67 4,000 24,000 0.7
5 12,500 167 10,000 60,000 1.4
10 25,000 334 20,000 120,000 2.8
I-
500,000 2 10,000 134 8,000 48,000 1.4
5 25,000 334 20,000 120,000 2.8
10 50,000 668 40,000 240,000 5.5
1,000,000 2 20,000 267 16,000 96,000 2.8
5 50,000 668 40,000 240,000 5.5
10 100,000 1,336 80,000 480,000 11.0
5,000,000 2 100,000 1,336 80,000 480,000 13.8
5 250,000 3,340 200,000 1,200,000 27.5
10 500,000 6,680 400,000 2,400,000 55.0
10,000,000 2 200,000 2,672 160,000 960,000 27.5
5 500,000 6,680 400,000 2,400,000 55.0
10 1,000,000 13,360 800,000 4,800,000 110.0
-------
Table 44. SETTLING LAGOON EXCAVATION COST ESTIMATES
Settling Lagoon Excavation Cost Estimates
Basis: 10 ft depth of live sludge bed plus 2 ft depth of clear surface water.
Costs: $1.70/yd 3 (< 100,000 yd 3 ), $l.501yd 3 (100,000 - 1,000,000 yd 3 ), $l.00/yd 3 (> 1,000,000 yd 3 )
Plant
Capacity
gpd
Settled
Sludge
Volume
%
Excavation Requirements for
12
Depth
Excavation
Cost
$
60
-Day Service
1Year
Service
Surface
Area Volume
acres Cu. yds
Surface
Area
Volume
cu. yds
60-Days
Service
1Year
Service
ft
ftz
acres
100,000 2 1,600 0.04 710 9,600 0.22 4,250 1,200 7,230
5 4,000 0.09 1,780 24,000 0.55 10,600 3,020 18,000
10 8,000 0.18 3,550 +8,000 1.10 21,200 6,040 36,000
250,000 2 4,000 0.09 1,780 2L ,000 0.55 10,600 3,020 18,000
5 10,000 0.23 - ,4l40 60,000 1.37 26,600 7,550 45,200
10 20,000 0.46 8,880 120,000 2.75 53,200 15,100 90,400
500,000 2 8,000 0.18 3,550 48,000 1.10 21,200 6,0 1+0 36,000
5 20,000 0.46 8,880 120,000 2.75 53,200 15,100 90,400
10 40,000 0.92 17,800 240,000 5.50 106, +O O 30,200 160,000
1,000,000 2 16,000 0.37 7,100 96,000 2.20 42,400 12,000 72,000
5 40,000 0.92 17,800 240,000 5.50 106,000 30,200 160,000
10 80,000 1.84 35,600 480,000 11.00 212,000 60,400 320,000
5,000,000 2 80,000 1.84 35,600 480,000 11.00 212,000 60,000 320,000
5 200,000 4 60 88,800 1,200,000 27.50 532,000 150,000 798,000
10 400,000 9.20 177,600 2,400,000 55.00 1,064,000 266,000 1,064,000
10,000,000 2 160,000 3.68 71,000 960,000 22.00 +24,000 120,000 636,000
5 400,000 9.20 177,600 2,400,000 55.00 1,064,000 266,000 1,064,000
10 800,000 18.40 355,200 4,800,000 110.00 2,128,000 532,000 2,128,000
-------
Table 45. COMPARISON OF THICKENER AND SETTLING LAGOON
SIZES AND COSTS FOR VARIOUS CMD CAPACITIES
Settling lagoon
for 1 year
Settled
service
CMD plant
capacity
mm gpd
volume
of sludge
%
Thickener
required
(12 ft deep)
size cost
acres $
diam.-ft cost-$
0.1 2 25 55,000 0.22 7,230
5 -- -- 0.55 18,000
0.25 2 40 72,000 0.55 18,000
5 -- -- 1.37 45,200
0.5 2 50 80,000 1.10 36,000
5 -- -- 2.75 90,400
1.0 2 100 100,000 2.20 72,000
5 -- -- 5.50 160,000
5.0 2 150 150,000 11.0 320,000
5 -- - - 27.5 798,000
10.0 2 200 250,00 22.0 636,000
5 - -- 55.0 1,064,000
considered only for small systems, secondary clarifying units, or
temporary emergency application.
Lagoon maintenance should be minimal, involving precautions against
undesirable seepage, side-wall erosion, and corrosion of structures
from the highdissolved-solids-content waters.
SECONDARY DEWATERING OF SLUDGES FROM COAL MINE DRAINAGE TREATMENT
The solids content of any feed slurry to a mechanical dewatering
operation is important. In general, feasibility increases and costs
decrease with increased feed density 7 . The physical and chemical
nature of the sludge and its solids content are determined by the
reagents and processes which precede the secondary dewatering.
Thus any dewatering process evaluation must involve a coordinated
process flowsheet.
157
-------
Continuous vacuum filtration is a principal industrial dewatering
unit operation. The filtration tests, all made on thickener under-
flow products, sought to establish technical and economic feasibilities
for settled mine water sludge.
The major objectives were:
1. To develop filtration techniques attaining solids filtration
rates greater than 100 lbs./ft 2 /24 hours;
2. To establish relative costs for filtration in respect to other
dewatering approaches;
3. To show whether filter cake has advantages in the handling and
disposal (or recovery) of sludges;
4. To determine variation in the filtration characteristics of
settled solids from various mine waters and the alkalies used
to neutralize them;
5. To study the effects and costs of chemical additives (primarily
flocculants) on filtration rates.
Technical and Economic Feasibility of Filtration
The technical feasibility of filtration with hydrous iron oxide
sludges has been demonstrated in the rocessing of waste pickle
liquors in the steel industry 9 76, and with sewage sludges 73 a
79, Typical filtration rates for several types of sludges are
shown in Table 46.
Important properties of mine water sludges affecting filtration are:
1. The particle size and the zeta potential of the solids.
2. The pH and temperature of the slurries.
3. The floc size, as enhanced by the use of flocculants or degraded
by agitation of high-shear pumps.
4. The solids content of the feed slurry.
5. The chemical composition of the solids, including the presence
of impurities as excess neutralizing agents.
Filtration operational techniques and economics must be related to
subsequent handling and disposal of the sludge filter cake and the
filtrate. Among these factors are:
158
-------
Table 46. FILTRATION RATES FOR SEVERAL VARIETIES OF SLUDGES
Reagent
Sludge used
%
Solids
in feed
Filtration
Rates
Type
of
Filter
Liq uid
gpm/ft
Solid
lb/ft 2 /24 hrs
Mine Water Sludgekl Limestone 1.2 0.12 120 Rotary Drum
String
Discharge
Pickle Liquors Hydrated Lime >500 Disk Filter
(carbide sludge)
Pickle Liquor 78 High Calcium 40 0.75-1.13 2600-3800 Pre-coat
Lime Test Leaf
.0
Pickle Liquor 77 Limestone 5 110 Vacuum Drum
Raw Sewage Sludge 79 10 240 Vacuum Drum
Raw Sewage Sludge 8 ° FeC1 3 -CaO 89-240 Vacuum Drum
Sewage Sludge 73 a 25-250 Vacuum Drum
-------
1. Clarity of the filtrate.
2. Moisture content of the filter cake, which will determine its
bulk handling characteristics, the equipment required for trans-
ferring, costs of transportation and ultimate disposal.
Plant Operation - Ec [ uipment Used- -
The filtration tests used a precoat rotary drum filter with diato-
maceous earth (CeJ.ite 545) as a filter aid. The EIMCO filter was
6 feet in diameter by 6 feet in length, having 113 ft 2 of filter
area (see Appendix B, Figure 9, and its operation in Figures 30 and
31). The filter is of acid-proof construction and has a variable
speed drum drive, a variable speed precoat knife-cutting mechanism,
a variable rate slurry feed pump, and a slurry flocculant conditioning
system.
Bench tests utilized a 0.1 ft 2 test leaf. Some leaf tests were made
with strongly flocculated slurries but no precoat with quite promising
results. These tests could not be confirmed with the Continuous
filter because internal filter modifications would have been necessary
for its conversion to a non-precoat configuration permitting a blow-
back cake discharge.
Filtration - Operating Parameter Limits- -
The parameter operational ranges of the filtration unit are listed
in Table 47. The values shown under Maximum Operation Range Ex-
perienced indicate the parameter spread observed while the Typical
Range values could be maintained under steady-state condition
without mechanical operating difficulties.
Filtration - Bench Scale Filtration Testing- -
Sufficiently large volumes of thickener sludge were not available
to make continuous filtration tests on all types of neutralized
sludges produced. In providing the desired additional, supplementary
data, a bench scale vacuum filtration test station was operated.
The conventional equipment used in the bench testing is shown in
Figure 32. The 0.1 ft 2 EIMCO test leaf was covered with the same
filter cloth as the large filter and was provided with a one inch
deep brass guard ring which permitted the deposition of about 1/2-inch
of Celite 545 on the filter cloth. Excess Celite was removed to
provide a precoat surface level with the guard ring thus permitting
the filter cake to be removed.
The filtration cycle was varied. The filter cake was carefully
lifted from the filter medium with minimum medium contamination.
160
-------
Figure 31. Cutting Sludge Cake from Vacuum Filter
A
Figure 30. I line Water Sludge Coating on Precoat Drum Filter
161
-------
Table 47.
OPERATIONAL PARAMETERS FOR FILTRATION UNIT
OPERATION HOLLYWOOD PILOT PLANT
1. Filter Slurry Feed Rate - gpm
2. Percent Solids in Feed
3. Dry Solids Feed Rate - lbs/mm
4. Filter Drum Speed - rpm
5. % Submergence
6. Vacuum - in. Hg. a
7. Dry Solids Filtration Rate
(lbs/ft 2 filter area/24 hours)
8. Liquid Filtration Rate - gpm/ft 2
9. Temperature of Feed - °F b
10. Percent Solids in Filter Cake
11. Filter Cake Thickness - inches
12. Flocculant DosageC - lbs/ton solids
(Atlasep lAl)
13. Filter Aid Used - lbs/hr
(Celite 545)
14. Knife Advance Rate - mils/rev.
15. Filter Cloth Used - Eimco No. P0-801
ment 1 X 1 plain
count per sq. in.
flow = 30 cfm/ft 2 .
NOTES:
0.6-7.2 4-5
HF - polyethylene monofila-
weave - 112 x 48 thread
- 11 oz. per sq. yd. - air
alhese rates are for nine drainage sludge only, and do not include
weight of the Celite filter aid. Weight percent of Celite in the
filter cake ranges from 10 to 40%.
bFilter cake is a combination of mine drainage sludge and Celite filter
aid. Solids in Celite 545 filter medium wet cake run about 43%.
small amount of flocculant was present in most filter feeds, having
been added to the thickener at dosages of 1 to 3 ppm of neutralized
water.
Maximum
operating
range
Typical
Parameter experienced
range
5-40
0.8-15
0. 35-50
0. 22-0. 85
18-23
5-25
10- 300
0.1-0.5
40-75
15-27
1/32-1
0-9.9
5-20
1-3
1.26-4.7
0. 29-0. 73
2023
1822
1025
0. 1-0. 35
60-70
18-24
1/16-1/4
0
50-65
162
-------
J,)
Test leaf filter
for top loading
process
Test leaf filtration
equipment
Figure 34. Sleeve for top loading
filtration
Figure 32.
Figure 33.
163
-------
The wet cake was weighed immediately, dried in an oven at 105°C
overnight, and then weighed to evaluate its moisture content. The
filter medium remaining on the dried sludge cake was brushed off.
This separation can be made easily, permitting the weighing of the
sludge directly and computing of the actual filtration rate. A
standard filtration test data sheet showing the method of handling
the data is given in Table 48.
Bench Scale Filtration Test Data--
Cycle Time Selection--The filtration cycle sequences were based upon
a range of values related to the plant filter at various operating
speeds. These cycle sequences are shown in Table 49. Tests made
to determine the effect of filter drum speed on solids filtration
rate utilized hydroxide neutralized sludge from two waters at three
levels of solids content. The results are summarized graphically in
Figure 35. Theoretical curves were constructed to illustrate the
relationship of liquid filtration flow rate to solids content of
the filter feed at two levels of constant solids filtration rates
(25.2 and 120 lbs/ft 2/24 hours) . These solids rate values illustrate
a spread of filter capacities from the economically unattractive to
feasible. These theoretical curves are shown in Figures 36 and 37.
The liquid filtration rate (i .e., rate of water permeability through
the sludge filter cake) will be the chief factor in determining fil-
ter capacity. The range of liquid filtration rates which could be
attained under reasonable operating conditions, and the corresponding
rate of solids build-up were determined experimentally.
Bench Scale Sludge Filtration Tests--A series of sludges were pre-
pared in the plant utilizing different waters and neutralizing rea-
gents. The settled sludges were then subjected to filter leaf
and drainage rate tests. They were produced in the plant using the
following flow sheet:
(pH - 9-11)
Raw Water - Flash Mixer - Oxidation Tank - Thickener + Overflow
150-200 gpm 4 (p11 8-9)
Sample for Bench - Sludge Sump
Alkali Slurry
Filtration Test
Preparation
Before each run the system was drained and washed thoroughly to
insure a representative sludge. The treatment tests were continued
under uniform conditions to develop several feet of settled sludge
in the thickener. These runs did not permit attainment of an equi-
librium condition in the thickener with continuous blow down. How-
ever, sludge densities attained were typical. Sludge sample densi-
ties for testing were controlled when necessary by additional settling.
164
-------
Table 48. FILTRATION TEST DATA
0
(11
Sludge
Alkali
Source:
Used:
Proctor No. 2
from Densator
Vacuum: 23 Hg
% Solids in Feed: 7.19
Filter Aid: Celite 545 (1/2 )
Test Leaf Area: 0.1 ft 2
Temperature: 70 F
Flocculant Used: Atlas 1A1 (1.5 ppm in
Densator)
Test
no.
Drsim
speed
settinga
Filtrate
volume
ml
F
Filtered Solids Weights - Grains
Solids
in
wet
cake-i
N
100K/J
Filtration rates
Pan +
wet
cake
G
Pan +
dry Wet Dry Filter
cake Pan cake cake aid
H I J K L
C-I H-I
Dry
sludge
M
K-L
solids
Liquid
lb/ft/hr 1b/ft /24hrs
0 P
ExN/45.4 24x0
GPM/ft
Q
0.16F/A
1
1
100
34.4
11.77 4.57 29.83 7.20 2.40
4.81
24.1
1.26 30.2
0.053
2
1
105
32.0
11.20 4.58 27.42 6.62 1.77
4.87
24.1
1.27 30.5
0.055
3
4
75
22.7
8.96 4.55 18.15 4.41 1.25
3.13
24.3
2.34 56.1
0.13
4
4
75
23.2
8.98 4.55 18.65 4.33 1.18
3.00
23.2
2.24 53.7
0.13
5
4.3
65
20.15
8.32 4.60 15.55 3.72 1.15
2.57
24.0
2.37 61.6
0.134
6
4.3
56
22.04
8.66 4.73 17.31 3.93 1.27
2.63
22.7
2.63 63.0
0.115
aSee Table 49 for Filtration Time Cycle.
Date: May 25, 1971
-------
Table 49. HOLLYWOOD PLANT VACUUM FILTER CYCLE CALIBRATIONS
Drum
Submergence
Drying
Discharge
Total
Cycles
Speed
Time
Time
Time
sec.
%
Cycle
Time-sec.
per
Hour
Drum
RPM
Speed
RPH
Setting
sec.
%
sec.
A=ca B=ab C=bc D E
3600/D
1 60
23
1 3 16
20 192 63 52 17
25 55 61 13 i tt
21 50 6+ 12 15
30tt
92
78
11.85
33.91
+6. 10
0.20
0.65
0.77
12
314
+6
.0
0
-------
Figure 35. Solids filtration rate as a function of drum speed
0 ,
N
IN
p ..
U.
UJ
0
0
-J
0 1
3
-j
U
I-
z
0
I -.
-J
U-
Ifl
3
.J
120
100
80
60
0.40
0.30
0.20
0.10
0
IN
I-
II .
0
U
z
0
I-
I -.
U-
0
a
-J
(FILTER TEST LEAF DATA
20
0
0
0
10
30
20 40 SO
FILTER DRUM SPEED. RPM
SOLIDS FILTRATION RATE AS A FUNCTION OF DRUM SPEED
167
-------
1.40
1 ,20
1.00
0.30
0.60
0.40
Figure 36.
Variation of filtrate flow rate vs. % solids in filter
feed at constant solids filtration rate
% SOL.1D5 N PILT#R PUD
BASIS FILTER CAKE CONTAINS 20% SOLIDS, A TYPICAL VALUE
FOR MD SLUDGE. NO FILTER AID USED CONSTANT
SOLID LOADING RATE AS INDICATED THEORETICAL
MATERIAL BALANCE BASIS.
4
I
4
U i
I .-
-J
U.
I-
0 1 2 3 4 5
168
-------
Figure 37.
Relationship of corresponding solids and liquid filtration
rates at various levels of solids content in the filter
feed slurry
1
1/
/
/
/
NOTE; ALL CALCULATIONS BASED UPON
FILTER CAKE ACTUALLY CON.
TAINING 30% SOLIDS BY WT
/
S
U I
I .,
z
4
UI
4
z
0
0-
4
I-.
-J
U-
a
-j
U I
-J
4
0
a-
I-
1)
0
2 4
SOLIDS FILTRATION RATE. LBS/ HR/F l 2 - DRY BASIS
0.50
0.40
0.30
0.20
0.10
4
UI
4
UI
I-
-I
U.
N
I .-
U-
UI
a-
0.
0
0
UI
a
UI
UI
I
4
z
0
I-
4
I-
-J
U-
0
a
-J
/
------4
/
0
/
0
5
169
-------
The chemical analyses of the sludges are shown in Table 50. In cal-
culating the compound composition, the loss in weight at 180°C was
assumed to include waters of hydration of gypsum and basic magnesium
carbonate but not from decomposition of metal hydroxides. The sodium
content was less than 0.5 percent and not included. The probable
compound analyses were based on solubilities of the most likely-
occurring compound form in sludges produced at pH 7.0-8.5 and ex-
posed wet to the atmosphere for several days which would convert
excess hydroxides to carbonates. All of the sulfate was assumed to
be gypsum (CaSOk 2H2O). The calcium in excess of that related
stoichiometrically to the sulfate was expressed as CaCO 3 , since any
Ca(OH) 2 would have been converted by carbon dioxide absorption. The
iron and aluminum were considered to be hydroxides which do not de-
compose under 300°C. Further, based on solubility and formation con-
ditions, basic magnesium carbonate was assumed to be present. Its
reported structure (5Mg0 4C0 2 6H 2 O) is that of the natural mineral,
hydromagnesite (3MgC0 3 Mg(0H)2 3H20). The solubility of this com-
pound at 25°C in pure water is 0.25 gil. This compound loses one
mole of H 2 0 at 100°C. The exact form is uncertain but this basic
carbonate formula appears to give the most reasonable total balance.
Other sludge compositions have been indicated 6 .
Summary of Filtration Test Data
The thickener-settled sludges were subject to the filter test leaf,
utilizing the previously described procedure and the following fil-
ter time cycle.
Submergence 16 seconds 21%
Drying Time 50 seconds 64%
Discharge Time 12 seconds 15%
Total Cycle Time 78 seconds 100%
Cycles/hr - 46.1
Corresponding Filter Drum Speed - 0.77 rpm.
The resulting filter test leaf data are presented in Tables 51 and
52. Only limestone and the aged lime containing CaCO 3 showed a major
difference in the solids filtration rates or chemical composition.
This confirms evidence from many sources (both coal mine water and
pickling liquor reports) that calcium carbonate produces sludges with
superior settling and filtration characteristics.
Filter feed slurries containing less than two percent solids con-
sistently produce the lowest solids filtration rates (in the range
of 15 to 30 lbs/ft 2 /24hrs). However, at five percent solids or
higher, filtration rates of 125 to 331 lbs/ft 2 /24 hrs were obtained,
which makes vacuum filtration a reasonable cost unit operation.
170
-------
FILTRATION
Table 50.
CHEMICAL ANALYSES OF SLUDGES USED IN
AND DRAINOMETER STUDIES
Weight, % (dry basis 105°24 hrs)
Alkali Used
Mine Water
Component
Al
Hydrated
a
Lime
-
Air
Oxidation
Hydrated Lime
Bio-Oxidation
Hydrated
Dolomite
Calcined
Dolomite
Bennetts
Branch
Proctor
1
Proctor
2
Proctor
2
Proctor 2
Tyler Proctor 1
Proctor 2
3.8
4.7
3.1
8.0
2.8
5.5
4.5
4.8
19.5
17.7
23.1
24.3
13.0
7.4
13.5
23.2
- J
Ca
6.9
5.8
5.2
4.8
17.2
10.7
6.7
5.2
Mg
6.6
4.3
5.1
1.3
3.8
11.8
9.8
5.8
so
5.7
6.8
5.8
11.5
4.4
1.6
2.3
5.5
H 2 0 at 180°C
12.5
15.8
14.8
10.2
8.7
11.7
14.7
Al(OH) 3
11.1
13.7
8.9
Compound Composition
8.2
16.0
13.2
13.9
23.1
Fe(OH)3
37.6
34.0
44.3
46.6
24.9
14.2
25.9
44.7
CaCO3
11.4
7.5
7.0
0.0
38.3
25.0
14.4
7.2
MgCO3
21.0
12.0
6.1
3MgCO 3 Mg(OH) 2 3H 2 O
25.2
16.4
19.4
5.1
14.4
22.4
24.6
15.6
CaSOk2H2O
10.7
12.7
10.9
20.7
8.4
3.1
4.4
10.4
Total
96.0
84.3
90.5
95.5
94.2
101.7
94.5
97.9
awarner Company fresh batch
-------
Table 51. MINE WATER NEUTRALIZED SLUDGE SOLIDS FILTRATION RATES
Mine water source
% Solids
in filter
feed
slurry
Fresh
hydrated
Ume
Old
hydrated
limea
Pebble Dolomitic.
dolomitic hydrated
lime lime
Ca(OH)z
Sludge
Ca(OH) 2 CaOMgO Ca(OH)2-MgO
filtration rates - lbs dry solids/ft 2 /Z4 hrs
Proctor No. 1
1.62
264 b
27.8
Proctor No. 1
2.53
54.0
Proctor No. 2
2.70
62.7
72.0
53.2
53.8
Prcctor No. 2
7.19
131.0
137.0
63.0
64.0
Proctor No. 2 (bio)
1.48
39.6
50.3
Proctor No. 2 (bio)
3.26
82.5
81.4
Proctor No. 2 (bio)
7.18
331.0
Bennetts Branch
1.96
33.1
38 . 6 c
Bennetts Branch
2.72
40.1
42.9
Tyler Run
1.58
15.6
21.2
Tyler Run
5.03
128.5
125.3
a 1 this case, hydrated lime stored in the plant lime storage silo for a period of 3 months,
during which time it converted to about 50% CaCO 3 .
bThe double values reported for most tests represent duplicate determinations made on each
sludge slurry.
CNQ precoat material employed.
-------
Table 52. VACUUM FILTER LEAF TESTS OF SLUDGE FROM LIMESTONE
NEUTRALIZED COAL MINE DRAINAGE
Temperature 70°F Filter Aid: Celite 5145
Sludge Source: Settling Flocculant Used: ATLASEP 1A1
Lagoon 2-3ppm in original
neutralized slurry
Water Source: Proctor No. 2
Biochemically
Oxidized
Vacuum
Test inches
No. Hg
Cycles
per
hr.
Filtration Rates
%
Moisture
in
Filter
Cake
Filter
Aid
Condi
tion
Filter
Cake
Thick-
ness
inches
Solid
lbslft/
24 hrs
Liquid
GPM/
ft 2
Waters neutralized with Centre County, Pa. limestone
(Neidigh Brothers Quarry). Free Solids: 13% by weight
1 24 45 381 0.16 66.0 Normal 3/16
Precoat
2 24 4 .6 451 0.18 65.6 Normal 3/16
Precoat
3 16 46 352 0.12 67.0 Normal 3/16
Precoat
4. 16 1+6 331 0.12 68.8 Normal 3/16
Precoat
5 2 4 46 394 0.15 65,3 Normal 3/16
Plus 0.1%
Body Feed
6 2 1 + 46 372 0.14 65.6 Normal 3/16
7 214 46 3140 0.l4 71.0 None 3/16
8 24 23 297 0.21 71.0 None 1/14
Top1 oad
9 24 +6 796 0.42 81.0 None 3/4
Topload
10 24 46 995 0.34 78.6 None 3/4
To pio ad
Waters neutralized with York County, Pa. limestone
(Thornasville Limestone Company). Feed Solids: 27.2% by
weight.
11 24 +6 990 0.16 55.5 Normal 3/16
Precoat
12 24 146 1,200 0.16 51.5 Normal 3/16
Precoat
173
-------
Most of the sludge from. limestone neutralization resulted from a
flowsheet involving biochemical oxidation and the rotary limestone
reactor (see Figure 16, Appendix B), but some had been reacted with
pulverized stone in the Densator. Such sludges have rapid settling
rates, high solids content in thickener underfiows, and excellent
filtration characteristics. The filter leaf data given in Table 52
clearly show the excellent filtration rates attainable with these
slurries.
The replicate tests using normal precoat vacuum filtration at vacuum
levels between 16 and 24 inches Hg yielded greater than 350 lbs/ft 2 /
24 hrs. Rates in excess of 1000 lbs/ft 2 /24 hrs were measured with
the higher solids slurry from Thomasville stone. The moisture con-
tent of 66 percent is a great improvement over the more typical 80
percent observed with hydroxide slurries. Test seven suggests pre-
coat media may not be necessary. With slurries of this nature, fil-
tration becomes a realistic low-cost method for handling mine water
solids.
Filter leaf tests eight, nine, and 10 were made utilizing top load-
ing with a more open cloth and no filter aid was used. Even higher
filtration rates were observed by this technique as shown in Table
52.
In this procedure, the vacuum primarily dries the cake and is not
needed to load the media. The gravity force will aid rather than
hinder the operation. In making these tests a sleeve was added to
the 0.1 ft 2 test leaf to permit a deep cake to accumulate as shown
in Figures 33 and 34. A three- to four-fold increase in solids
filtration rate over submerged-type loading resulted. These tests
suggest a top-loading type filter would be most effective.
With flocculated slurries, some individual flocs were so large and
cohesive they appeared filterable directly on a filter cloth without
the use of precoat. Several tests (Table 57) were made and confirmed
this possibility. The results indicate that strongly flocculated
puips do not require filter aid and produce filtrates with good clarity.
The elimination of filter aid would remove the highest operating fil-
tration cost factor.
The response on filtration of dilute slurries to a body feed of
filter media (one percent of Celite 545) is reported in Tests 20 and
21 of Table 56. This technique greatly improved the filtration of
this 1.96 percent solids slurry.
Filtration of High Density Recycle Sludge
In the start up operations of the plant filter, the heavy (> 15 per-
cent solids) recycle sludge was selected anticipating good filtration
174
-------
properties. The response was excellent, in fact the best observed,
attaining filter cakes with thickness in excess of 3/4-inches. The
cake was dry and readily handleable. The filtration rates were so
high that the total Densator storage supply of recycle sludge was
consumed in the qualitative startup procedures. It could not be
replaced during the available test period.
Limited data were collected from filter leaf tests with a less ade-
quate, specially prepared sludge. The limited recycle sludge was
prepared, batch-wise, with gentle agitation in a 230 gallon reactor.
Beginning with 50 gallon Proctor No. 2 water, four pounds of aged,
hydrate lime was used to raise the pH to at least 12. Fresh drain-
age was added to lower the p11 to 9 with 90 minutes reaction time.
After settling and decantation, the procedure was repeated through
six cycles.
The final settled solids were filtered and the data summarized in
Table 53 was developed. A solids filtration rate in excess of
130 lbs/ft 2 /24 hrs offers confirmation of the qualitative tests.
Table 53. FILTRATION RATE TEST DATA FOR RECYCLE SLUDGE
Solids (%) in Settled Sludge 5.0
Solids Filtration Rate 131.0
(triplicate determinations) 138.0
lbs/ft 2 /24 hrs 134.0
Liquid Filtration Rate 0.17
(triplicate) gpmJft 2 0.19
0.18
Analysis of Raw Water Used
Total Acidity, ppm 1722
Ferrous Iron, ppm 378
Total Iron, ppm 391
Temperature of Sludge 70°F
The filter leaf test data of settled sludges permit the following
conclusions:
1. A realistic operational liquid filtration rate range is
0.1 to 0.3 GPM/ft 2 .
2. The highest liquid rates are obtained at the fastest drum speeds
(i .e., shortest submergence time).
175
-------
3. The highest solid filtration rates are obtained at the fastest
drum speeds.
4. The highest solids rates are obtained from feed slurries having
the higher solids contents and may exceed 1000 lbs/ft 2 /24 hrs.
5. There is an excellent correlation of the 1.6 percent and 5 per-
cent solids slurries with the theoretical curves in Figure 37
but the 7 percent slurry curve in Figure 35 does not conform.
6. Solids filtration rates of 120 lbs/ft 2 /24 hrs can be attained
without the use of flocculants when using filter feeds containing
5 percent solids. Assuming a limiting liquid rate of 0.2 GPM/ft 2 ,
the theoretical curve shows that a solid rate of 120 lbs/ft 2 /24 hrs
is possible using a 4 percent solids feed.
7. Top loading filter techniques appear to be advantageous.
8. Media body feed is helpful for dilute slurries.
9. Under some conditions, filter media may not be necessary for this
type of sludge dewatering.
Pilot Plant Filtration Testing
Filtration testing with the Eimco drum filter was carried out with
slurries pumped from the settling lagoon or from the thickener under-
£ low. In many cases, insufficient sludge at constant solids content
was available to permit smooth, steady-state operation. The plant
data are presented in Table 54. The range of liquid and solids fil-
tration rates are similar to those established in the test leaf fil-
ter study. Under normal, full-scale operation, filtration rates
greater than 100 lbs/ft 2 /24 hrs should be possible with a properly
designed flowsheet. The chemical composition of these filter cakes
is presented in Table 55.
Man-Power Requirements in Filtration
The frequent attention required by the cake cutting knife mechanism
is a major labor consideration. Frequent adjustment of the knife was
necessary to maintain a clean surface (thus maximum solids capacity)
due to changes in compaction of the 3- to 4-inch starting precoat
layer. Although continuous operation with a uniform feed minimized
this variable, it appears cost estimates should include a significant
allowance for operating labor. Taking all phases of this operation
into consideration (e.g. flocculant preparation, precoat slurry
preparation and application, knife-cutting adjustment, and filter
cake disposal supervision), a manpower allowance of about 1/3 man
per shift appears realistic.
176
-------
Table 54. SUMMARY OF TESTS ON PILOT PLANT ROTARY VACUUM FILTER
inches Hg
3-4 inch layer
Celite 545
% Solids
in
Filter
Feed
aThese solids rates include the weight of the filter aid,
which will constitute 10 to 40% of the weight over the
range of 6100 lbs/ft 2 /24 hrs.
bFeed slurries in all tests were mixtures from several
types of mine water treated with partially carbonated
hydrated lime.
CAithough the % solids in these feed slurries were high,
the filtration rates were poor. This sludge came from
compartment two of the settling lagoon where a very fluffy,
slow settling component of the original sludge settled.
Vacuum: 22
Filter Aid:
Test
No.
Sludge
Source
Temperature: 65° F
Drum Speed: 46 RPH
% Moisture
in Filter
Cake
Filtration
Rates
Solids
lbs/ft /24
hrs
Liquid
GPM/ft 2
1
Lagoon
4 _ 6 c
83.7
62 a
0.07
2
Lagoon
46
83.0
7.4
0.06
3
Lagoon
46
81.5
11.9
0.19
4
Lagoon
46
86.7
13.6
0.19
5
Lagoon
46
83.3
17.5
0.12
6
Lagoon
46
88.9
17.2
0.10
7
Densator
1.6
71.2
35.2
Underf low
8
Densator
1.5
72.5
21.0
0.34
Unde rf low
9
Lagoon
60.2
10.4
0.30
10
Densator
45.7
42.0
-
Underfiow
11
Densator
4.5
63.8
98.5
0.17
Underf low
12
Lagoon
81.4
21.5
0.11
13
Lagoon
80.0
10.4
0.54
14
Lagoon
80.0
5.2
0.24
15
Densator
3.7
81.7
68.3
0.24
Notes
177
-------
Table 55. CHEMICAL ANALYSES OF FILTER CAKE
FROM CMD SLUDGE
(weight % (dried at l05 C))
Test
no.
Iron
Aluminum
Calcium
Sodium
Sulfate
1
18.1
5.4
3.2
0.88
3.7
2
20.1
5.9
3.1
0.91
2.1
3
20.4
6.7
3.1
0.82
2.0
4
28.7
4.6
5.0
0.34
4.5
5
21.4
4.7
4.4
0.76
3.9
6
26.2
5.5
5.9
0.24
5.5
7
22.0
4.5
4.3
1.05
2.2
8
9.8
3.6
2.3
1.69
0.7
12
23.9
6.0
2.7
0.51
5.6
13
4.8
3.9
16.9
0.62
1.2
14
17.5
6.3
2.8
0.84
3.1
Precoat Requirements
Plant experience indicated a precoat usage rate of 50 lbs/hr for
the 113 ft 2 drum, or 0.4 lbs/ft 2 /hr. Continuous, uniform operation
may reduce this requirement to 0.3 lbs/ft 2 /hr.
A loss of 100 to 200 pounds of precoat may result from residual
suspension in the filter tank after precoating, Provision should
permit its recovery in the precoat slurry feed tank for use in the
next precoating cycle. Another precoat loss can result at the
completion of a precoat loading cycle. When a precoat layer of about
3/4-inch remains, further removal with the cutting knife could re-
sult in damage to the filter cloth. It is difficult to add another
layer evenly directly on this residual layer. It is preferable to
drop this residual layer into the filter tank and provide for its
inclusion with the new precoat slurry.
Filter Cake Handling
The filter cake prepared with the plant filter had a moisture content
of about 80 percent. This cake fell freely from the cutting knife
on to a moving rubber conveyor belt for discharge. No difficulties
were experienced with the filter cake sticking to the belt. In sub-
freezing weather, the filter cake tended to freeze quickly creating
handling problems in truck transportation.
178
-------
Filtration Using Flocculants
The effect of flocculants on the filtration characteristics of several
sludges were made in bench scale and plant tests. The filter test
leaf data are presented in Table 56. At flocculant levels of 20 to
40 ppm (2 to 3 lbs/ton dry solids), an extensive agglomerating effect
caused a very rapid separation of liquid and solids in the treated
slurry. Formation of these ball-like flocs should produce much higher
filtration rates. There were higher liquid rates but no improvement
in solids filtration rates. The low solids filtration rate occurred
when large portions of the filter cake fell from the filter media
surface as it emerged from the liquid feed bed. In continuous prac-
tice with the high liquid filtration rate, the solids in the filter
bed would soon reach an unmanageably high level. The flocculants
created a slippery, non-cohesive sludge flow. The cohesiveness de-
creases with higher flocculant dosage (see Tests 5, 6, 12, 13, and
29 in Table 56)
Plant Tests with Strongly Flocculated Feed Slurries
The data from the full-size precoat filter using strongly floccu-
lated pulps are presented in Table 57. The runs at each dosage
level were short, and operating difficulties prevented any tests
at the highest drum speeds. The results were in agreement with
those from the bench scale test leaf. The most significant obser-
vations made were:
1. Strongly flocculated filter feeds do not respond as well as
weakly flocculated slurries with a conventional precoat filter,
even when pulp suspension is maintained.
2. At high flocculant dosages, the solids settle very fast to the
bottom of the filter tank, providing little contact of the drum
with the solids. In a separate operation, slurries could be
dewatered in this manner to attain a heavier and more homo-
geneous pulp for filter feed slurry.
3. In contrast to the test leaf results, high liquid filtration
rates were not attained while using strongly flocculated slurries.
Correlation of Bench Scale and Plant Operations with Design Criteria
In general, the test leaf results could be extrapolated directly to
predict performance of the full scale vacuum filter. The major f li-
ter design criteria for a low cost vacuum filter operation, which
were developed and confirmed in both bench and plant, are summarized
in Table 58.
179
-------
Table 56. SUMMARY OF FILTRATION LEAF TESTS USING FLOCCULANTS
All tests made with an Eiinco 0.1 ft 2 test leaf.
Temperature: 700 F Filter Cycle: 16 sec submergence
Vacuum: 24 inches Hg 50 sec drying
- a Typeof
Flocculant Used Slurry
Filtration
Type Dosage Loading
Test lbs/ton Filter Aid to Filter Saiius
No.
dry solids Usede Leaf lbs/ft 2 /24 hrs
Rates
c-----
Liquid
Proctor No. 2 - 2.71% Solids Feed - Dolomitic Pebble Lime
1 None (Controls) Precoat NormaiC 53.2
2 None (Controls) Precoat Normal 53.8
3 1A1 20 1.47 Precoat Normal 52.0
4 1A1 40 2.94 Precoat Normal 53.2
5 1A1 80 5.88 Precoat Normal 33.8
6 lAl 80 5.88 None Normal 25.2
7 lId 80 5.88 None Top Loadd 100
8 None (Control) None Top Load 34.6
0.15
0.15
0.16
0.16
0.27
0.19
0.30
0.12
Proctor No. 1 - 2.53% Solid Feed
9 None (Controls) Precoat Normal 54.0
10 None (Controls) Precoat Normal 53.0
11 lId 20 1.58 Precoat Normal 49.5
12 lAl 40 3.16 Precoat Normal 40.3
13 lAl 40 3.16 None Normal 33.4
14 lAl 40 3.16 None Top Load 118.3
0.18
0.18
0.18
0.23
0.30
0.36
Bennettg Branch - 1.96% Solids Feed Fresh Lime Hydrate
15 None (Control) Precoat Normal 33.1
16 IA1 10 1.02 Precoat Normal 40.5
17 lAl 20 2.04 Precoat Normal 41.7
18 lAl 10 1.02 None Top Load 51.1
19 lAl 20 2.04 None Top Load 75.9
20 1A1 20 2.04 Precoat + Normal 68.2
1% Body Feed
21 IA1 20 2.04 Precoat + Top Load 94.0
- 1% Body Feed
42 None - None Normal 38.6
0.16
0.30
0.53
0.24
0.42
0.42
0.42
0.16
Proctor No. 2 - 7.18% Solids - Old Lime Hydrate - Biochemical Oxjd j 0
23 None (Control) Precoat Normal 331.0
24 1A1 20 0.56 Precoat NormalB.T 400.0
25 1A1 20 0.56 Precoat NormalB.r. 326.0
26 1A1 20 0.56 Precoat NorlnalB.T. 290.0
27 1A1 20 0.56 None Normal 264.0
28 lAl 20 0.56 None Top Load 370.0
29 lId 30 0.84 None Normal 92.5
30 lAl 30 0.84 None Top Load 232.0
0.33
0.37
0.43
0.33
0.37
0.38
0.69
0.57
Notes:
aAT SEP
bFlocculant dosage indicates ppm per slurry volume (e.g., 20 ppm = 20 mg/liter
of slurry. To convert to lbs/ton dry solids:
lbs flocculant/ton dry solids Dosa e (ppm )
5 x % solids in slurry
CNormal loading is the conventional cake pik up obtained by dipping the
inverted test leaf into the slurry.
doTop Load is the technique of pouring slurry into the upright filter leaf,
simulating a Buchner funnel test.
ecelite 545.
This alkali was high calcium hydrated lime containing 50% CaCO 3 from
CO 2 absorption.
gmis water was neutralized in the Densator.
hThe settled slurry in the bottom of the settler is more dense with high
flocculant doses. B.T. = bottom of tank T.T. = top of tank.
180
-------
Table 57.
00 806 33.6 3.71
I - .
PILOT PLANT PRECOAT ROTARY VACUUM FILTER TESTS WITH
STRONGLY FLOCCULATED FEED SLURRIES
410 17.1 0.84 1.41
Vacuum - 22 Hg. 2
Filter Area - 113 ft
Filter Aid Celite 545 (2 cake)
Filter Aid Consumption 50-60 lbs/hr.
Filtrate Quality - Clear
Note:
20.8 2.5
18.3 2.8
19.2 3.6
19.0 0.6
Flocculant Used - ATLASEP 1-A-i
Sludge Origin - Proctor No. 2 water subject
to biochemical iron oxidation,
hydrate lime neutralization.
aThe flocculant level used in Test 4 settled sludge to the bottom of the filter
vat, preventing pickup by the filter drum.
1
2
3
Flocculant
Test
Drum Speed %
Cycles/day RPH in
Solids
Feed
Dosage
lbs/ton dry
feed solids
% Solids
Wet Cake
in
Filtration Rates
lbs/ft 2
er hr.
lbs/ft 2
per day
Liquid
GPM/ft 2
No.
806
806
806
33.6 7.53
33.6 3.71
33.6 3.71
5
0.95
0
3.03
9.9
20.2
59.7
68.3
86.1
1.5 36.21
0.21
0.21
0.21
0.23
0.36
14.6
-------
Table 58. RECOMMENDED VACUUM FILTER DESIGN PARAMETERS
1. Optimum solids content of filter feed - 3% solids minimum.
2. Liquid filtration rates - 0.35 gpm/ft 2 - maximum (70°F)
0.150.20 - average (70°F).
3, Solids filtration rates - 15 to 25 lbs/ft 2 /24 hrs. @ 1% solids
feed.
120 lbs/ft 2 /24 hrs. @ 4% solids feed.
(all values for 70°F)
A graph relating the entire range of filtration rates vs. solids
content in the filter feed is presented in Figure 38.
4. Filter drum speed - 0.8 rpm minimum.
5. Drum submergence - 25%.
6. Precoat consumption - 40-50 lbs/hr for 113 ft 2 filter area.
0.4 lbs/hr/ft 2 filter area.
40% of dry sludge weight @ 24 lbs/ft 2 /24 hrs.
8% of dry sludge weight @ 120 lbs/ft 2 /24 hrs.
7. Knife advance - 4 mils/drum revolution.
8, Moisture in filter cake - 76 to 82%.
Cakes of this moisture content could be handled satisfactorily on
a conveyor belt.
9. Specific gravity of filter feed - 1.03 (@ 4% solids).
10. Specific gravity of filter cake - 1.13 (71 lbs/ft 3 ).
11. Type of filter cloth - EIMCO P.O. - 801 HF or equivalent.
12. Flocculant requirement - 2 to 4 ppm added to Densator feed.
(will vary with flocculant)
13. Filter aid type - Celite 545 or equivalent.
182
-------
The recommended design parameters should produce satisfactory,
economical filter operation if a precoat filter is to be considered.
However, the preliminary test work without filter aid with strongly
flocculated puips suggested more study seeking to eliminate the use
of filter aid and its high labor requirements.
In sizing a precoat filter, the capability to store sludge and the
frequency and extent of filter operation requirements must be re-
lated to sludge volume and filtration rates. Once precoated, a drum
filter must be kept under vacuum with continual moisture conditions
to maintain the filter media on the drum until it is all consumed.
The power requirements for maintaining the vacuum represent a major
portion of filtration costs, comparable only to the costs for pre-
coat material.
Filtration Costs
Filtration costs for coal mine drainage slurries are determined by
the following factors:
1. The maximum liquid filtration rate attainable, which was found
to be 0.3 GPH/ft 2 .
2. The maximum solids filtration rate attainable at maximum liquid
filtration rate (a function of the solids content of the filter
feed , which filter leaf tests indicated could be as high as
300 ibs/ft 2 /24 hrs with hydroxide-produced sludges.
3. The solids content of the filter feed, which both pilot plant
and bench scale tests indicated, could be as high as 15% with
carbonate-produced sludges.
4. The size of the filters required, and their construction materials.
5. The amount of filter aid required, which was established on the
pilot plant filter to be about 0.4 lbs/hr/ft 2 filter area.
Using the operating range as established in these evaluations, the
costs of filtration at various solid filtration rates were calculated
and presented in Table 59. Filtration equipment cost variation with
sludge product rates and filter area may be derived from Figures 38
and 39.
The cost estimates of Table 59 indicate that reduction in filtration
costs is attainable with higher feed solids in contrast to the so-
called normal conditions (l o solids feed or lower) reported by
others 72 . For cost comparisons with other secondary sludge
183
-------
a This use rate might be decreased to 0.3 Lbs/Ft 2 /Hr.
bBased upon a minimum carload lot of 60,000 Lbs @ $84.00/Ton F.O.B.
California, plus $33.50 freight to Pennsylvania.
24
!20
1
4
240
8
Table 59. ESTIMATED FILTRATION COSTS FOR VARIOUS LEVELS OF SOLIDS
FILTRATION RATES
Basis: 1,000,000 GPD mine water containing 1,000 ppm neutralized
solids
(4.15 tons!
day of dry sludge). 365 operating days per year.
A
Solids Filtration Rate - Lbs/Ft 2 /24 Hrs
B
C
- -
1. % Solids in Filter Feed
2. Filter Area Required - Ft 2 340
68
34
3. Installed Filter Cost (mild steel) $140,000
$35,000
$20,000
4. Annual Amortization (10% per year) $ 14,000
$ 3,500
$ 2,000
5. Filter Aid Consumption (0.4 Lbs/ft 2 /Hr)_Lbsa 3,260
650
325
6. Daily Costs
a Amortization 2 $ 38.00
b. Electric Power (l2 /Ft /Day @ 1.5 KWH) 41.00
$ 10.00
8.00
$ 6.00
4.00
c. Filter Aid (@ $117/ton)b 191.00
38.00
19.00
d. Flocculant (@ $l.20/Lb)
5.00
10.00
e. Operating Labor (1 manl/3 time @ $3/hr) 24.00
24.00
24.00
f. Maintenance (5% installed cost/yr) 19.00
5.00
4.00
g. TOTAL OPERATING COST - DAILY $ 313.00
$ 90.00
$ 67.00
7. Cost per 1,000 gal. CND treated 31.3
9.0
6.7
8. Cost per ton dry solids filtered $ 75.50
$ 21.70
$ 16.10
9. Cost of Filter Aid per 1,000 gal. treated 19.l
3.8
1.9
Cost of Filter Aid per ton solids treated $ 46.00
$ 9.15
$ 4.60
03
Notes:
-------
Figure 38. Sludge production rates and filter area requirements
44
>-
4 1 ,
-J
0
UI
I.-
U
UI
( I )
-I
UI
N
-I
1-
UI
z
70,000
60,000
50,000
+
+
FILTER AREA REQUIRED - FT 2
I
1200
30.
20,000
0
f
2 3
PLANT CMD CAPACITY - MILLION GPD
SLUDGE PRODUCTION RATES AND FILTER AREA REQUIREMENTS
185
-------
Figure 39. Approximate installed costs of vacuum drum precoat
filters (1971)
FILTER AREA REQUIRED. FT
APPROXIMATE INSTALLED COSTS OF VACUUM DRUM PRECOAT FILTERS (1971)
INSTALLED COST INCLUDES ALL tUXILLIARY EQUIPMENT SUCH AS
THE VACUUM SYSTEM, PEED PUMPS. AND PRECOAT SLURRY TANKS.
NOTE: COST FOR PLASTIC CONSTRUCTION
IS INTERMEDIATE BETWEEN STAINLESS
AND MILD STEEL.
-j
-j
0
0
0
0
z
In
0
I
I .-
In
0
U
0
-j
-j
I-
In
I
600
500
400
300
200
100
0
0 200 400 600 800 1000 1200
2
186
-------
dewatering methods, the cost per ton of dry solids processed are
plotted iii Figure 40. Present treatment activity indicates a dry
solids production range of 1 to 25 tons/day.
A study of filtration costs permits the following conclusions:
1. Filtration cost for sludges can range from $15/ton of dry solids
to $90/ton, using a conventional vacuum drum precoat filter.
2. The conventional precoat filtration cost would probably never be
less than $10/ton dry solids filtered.
3. Filter aid is the major cost item, constituting about 50% of
the total cost.
4. If, with flocculants, filtration rates of at least 240 lbs/ft 2 /24hrs
could be attained on top-loading type filters with the eliinina
tion of filter aid, filtration costs could probably be reduced
to about $7/ton dry solids.
Centrifuge Tests
Performance sludge dewatering experience was attained with a Dorr-
Oliver Merco Bowl Centrifuge (Model Z-l-L). It has a maximum bowl
diameter of nine inches and an inside length of 24 inches. This is
a horizontal, continuous flow machine. The solids settle and com-
pact against the walls under centrifugal force (up to 5200 gravities).
They are conveyed from the liquid pool for further drying under
centrifugal force. The clarified centrate moves counter current for
pump discharge at 30 psi.
The sludge dried (primarily of limestone reaction origin) was pumped
from the bottom of the settling lagoon to a storage feed tank and
maintained by slow stirring as a uniform feed to the centrifuge.
Prior experience 72 75 assisted selection of operating conditions
and maintenance of reasonable clarity in the centrate. The test
conditions and results are given in Table 60. These data are in
agreement with previous results with the same machine 72 . At
higher flows the turbidity of the centrate increased markedly.
Should this clarity be a problem the relatively small volume could
be recycled through the process thickener or settling lagoon. The
centrifuge cake had satisfactory handling characteristics although
it showed a tendency to flow and was more subject to a loss of
physical integrity than the filter cake.
The device functions satisfactorily and holds great advantage in
ease of operation with a minimum of manpower requirements. As a
high speed mechanical device, it is subject to high maintenance
187
-------
Figure 40. Costs of vacuum drum precoat filtration as a function of
solids filtration rate capacity
4O 100
BASIS: CALCULATIONS FOR A 1,000,000 GPD PLANT
WITH WATER CONTAINING 1,000 PPM NEUTRALIZED
SOLIDS(4.1S TONS/DAY DRY BASIS)
z
APPROXIMATE PRICES AS OF SEPTEMBER, 1971 80 ...
U 0
30
0
u. I 0
I-.
4 U
0
I- 0
60
co
U w
i2O
4
0
a
a
2 40
- -- -1
I L .
I 0
a-
z
I- 0
I -
0
010
I 20 ILl
a.
t
I-
0
U
0. , 4 ,
-0
0 100 200 300
SOLIDS FILTRATION RATE - LBS/FT 2/24 HRS
COSTS OF VACUUM DRUM PRECOAT FILTRATION
AS A FUNCTION OF SOLIDS FILTRATION RATE CAPACITY
-------
Table 60. OPERATING PARAMETERS FOR THE MERCO BOWL CENTRIFUGE
Run #1 Run #2
% Solids in Feed Slurry 0.9 11.5
Feed Flow Rate - gpm 1.71 1.76
Centrate - gpm 1.71 1.76
Centrate Solids - ppm 57 31
Cake - % Solids 19.3 21.0
Cake Discharge Rate - lbs/mm 0.125 0.775
Flocculant (Atlasep 1A1) Dosage 5 5
Rate - ppm
Anperes 11 11
Power Consumption -
kwh/l,000 gpm Centrate 27.2 28.0
kwh/Ton Dry Cake 3,840 562
Centrifuge Speed - rpm 4,620 4,620
Centrifuge Differential - rpm 20 20
Centrifuge Pool Level - inches 5.9 5.9
ANALYSIS OF CENTRIFUGE FEED SOLIDS
Element Weight% Compound Weight %
Al 7.68 A1(OH) 3 22.2
Fe 24.10 Fe(OH) 3 46.4
Ca 7.46 Gypsum 8.7
Na 0.12 CaCO 3 13.6
S0 4.86 ____
Total 90.9
189
-------
and power costs. As with other secondary dewatering devices, its
response improves with feed solids content.
Sludge Drying Bed
The sludge drying bed is another approach for secondary dewatering
settled sludge to improve their handling characteristics and reduce
volume. Water is removed from the solids by decantation from the
surface of a newly-filled bed, by percolation through the bottom
of the bed, and by evaporation from the surface.
Sludge drying beds have been applied to many types of industrial
waste and sewage sludges. It had not been previously considered
for CMD sludges. The operating experience reported for sewage
sludge dewatering can be related. Of concern is quantitative in-
formation regarding mechanisms of dewatering, drainage rates, soil
seepage rates, evaporation rates, and regional rainfall.
Jeffrey 83 reported sewage sludge drainage rates and described a
procedure for calculating the size of a drainage bed. These drying
beds have low solids handling capacity and are normally used on a
two year cycle. 1-us drainage rate data are summarized in Figure 41,
using an influent sludge slurry with 8 percent solids with several
different sand filter bottoms. It was concluded: Initial drainage
rates are erratic while the starting layer of sludge is compacting,
but the highest rates did not exceed 0.07 GPH/ft 2 ; The drainage rates
stabilized at a level less than 0.01 GPH/ft 2 after 12 weeks opera-
ting time; Stabilized drainage rates were proportional to hydrostatic
head; The ultimate practical bed loading rate, when drainage is the
only dewatering force, is 2.4 lbs of dry solids per ft. 2 per 30 days,
of drainage; A practical limit to drying by drainage is 75 to 80
percent moisture; Drainage alone produces a sludge mass of uniform
moisture content which is independent of sludge depth; The dewatering
bed must be located upon a supporting medium with adequate water
permeability rates; Evaporation definitely reduces moisture content
of the sludge, but the practical effects of evaporation are limited
to the top six to 12 inches of sludge; and Since evaporation is
necessary to produce a workable sludge, the depth of a sludge bed
must be limited to depths at which evaporation can be effective.
These data and actual operation of the drying basin suggest the
following factors as to the design and operation of a basin for
mine drainage sludge:
1. The decantation stage can be expected to remove much of the
water but may require most operator attention to prevent dis
charge of particulate matter.
190
-------
Figure 41. Sewage sludge drainage rates
70
60
0
I ,
U
a
0
s-a
30
.06
.05
.04
I
C :,
.03
.02
.01
0
50
40
20
0c
0a
.
10
0
SLUDGE DEPTH SUPPORTING
INCHES MATERIAL
22 SAND
52 SAND
47 DISTURBED SOIL
76 SAND
106 SAND
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60
June 13 Sept. Dec. Mardi June
Time, Weeks
-------
2. The drainage function is essentially a sand filtering operation
without backwash capability. The bed capacity is a function of
drainage rates through the sludge and the bottom of the bed.
This requires a permeable bottom with a controllable water trans
fer capability.
3. The drainage rate limitation should be that developed by the
sludge rather than by the supporting bed. Although drainage
rates of supporting soils might be adequate, a specially-
constructed sand filter bottom with underlying drainage lines
to provide positive, high capacity drainage is preferable.
Such a design must consider ground water elevations, direction
of drainage flow from the bed, and possible interference from
and to natural drainage. Coal refuse banks should be avoided.
4. Evaporation controls the moisture content of the sludge in the
final stages. Evaporation rates versus precipitation levels
for the bed site are thus critical.
Soil seepage and permeability rate data reported by Todd 8 show
drainage rate equivalents ranging from 0.03 to 1.66 gph/ft 2
for sand-clay mixtures to gravelly loam. The rate for coarse
sand is 120 gph/ft 2 . Figure 42 gives a relationship between de-
creases in bed depth to bottom drainage rates. The drainage rates
for several types of soil which fall within the range of the graph
have been indicated.
Data for rainfall and evaporation rates for western Pennsylvania-
eastern Ohio are summarized in Figure 438k as a 12-month plot. The
interaction of these two weather factors is shown as the difference
in level of an impervious pond. The graph clearly indicates that
appreciable evaporation can be expected during the summer months,
but that over the year, they are roughly equivalent. However, in
well-drained drying beds the continuous percolation continues during
precipitation periods so that rainfall retards only the evaporative
process. Obviously a bed cover would minimize these effects and was
considered in these studies. Perhaps the more significant weather
factor results from sub-freezing conditions which can nearly block
drainage. This effect simply requires excess holding capacity for
sub-freezing winter operating periods, which seldom exceed two
months in this region.
Sludge Drainage Rates - Bench Scale Drainometer Determination- -
Only minimal data were available for the design of the bed. The type
of information sought for CMD sludge drying bed design and for cost
estimation purposes are:
192
-------
Figure 42. Pond liquid level dropping rate vs. pond bottom drainage
rate capacity
I-
>
-J
C-,
I-
-J
z
z
I-
I
>-
-J
C-)
-J
2 4 6 8 10 12
SEEPAGE RATE OR POND SURFACE DROPPING RATE-INCHES PER DAY
C
-J
.30
.25
Note: Seepage rates for various soils within this
range are indicated. Seepage rate for
coarse clean sand is 385.0 ft/day.
I-
U-
=
C!,
LU
I
C
C
I-
0
C
z
C
0
.15
0
193
-------
/
MARCH MAY JULY SEPT. NOV.
BASIS: EVAPORATION RATES-MEAN MONTHLY VALUES
RAINFALL DATA-30-YEAR AVERAGES
Figure 43. Precipitation and reservoir evaporation data for Western
Pennsylvania and Eastern Ohio
DFFFERENCE BETWEEN PRECIPITATION AND EVAPORATION-INCHES OF WATER
U i
=
C-,
Ui
>
U i
-J
z
0
U i
=
C.,
U i
I.-
Li
0
C -,
Ui
C-)
χ2
+1
0
-1
-2-
5.0
4.0
3.0
2.0
1.0
0
PRECIPITATION
/
/
-.
/
--- - \
S
C- I
S
S
S
-
--
-- -
-.-1 ----
JAN.
-------
1. Drainage rates of various sludges versus time.
2. Permeability rates of water through consolidated sludge.
3. The effect of multiple sludge layers (with intermittent drying)
on drainage rates.
4. Effects of sludge surface cracking on drainage rates.
5. Bulk densities and moisture content of the completely drained
sludges.
Since the test schedule did not permit production of sufficient sludge
from one flowsheet for direct evaluation, a small-scale test unit
was developed. A drainometer was constructed from a 100-gallon tank,
as detailed in Figure 44.
The bottom of the drainometer was filled with 1/2-inch by 1-inch
limestone to a depth of four inches upon which was placed a six-
inch layer of coarse sand. This simulates an ideal drainage bed
medium with the larger stone providing a drainage bed while the
sand would trap the sludge particles. Before loading with sludge,
the sand bed was backwashed with water to remove any entrapped
particles. The clean bed had a drainage rate with water of about
two gpm/ft 2 which is comparable for a conventional sand filter.
With this design, it was possible to load the drainometer to a sludge
depth of about two feet.
In each test, the drainometer was carefully filled with slurry to
avoid disturbing the sand bed. The slurry was allowed to drain for
a period of about 48 hours, at which time the top of the cake was
void of free-standing water and cracks were beginning to form from
the decreasing volume of the hydrated particles. The initial drain-
age rates were measured at hourly intervals and subsequently at
longer intervals up to 24 hours total elapsed time. Final rates,
which were essentially constant, were measured at eight-hour intervals.
In two tests, a second and third layer of sludge was added to the
dewatered layer to simulate the cyclical drying bed loading that
was practiced in the full scale bed. Cyclical operations have been
cited by the Bureau of Mines 8 . The drainometer seemed to simulate
the full scale bed, including the cracking phenomenon on the sur-
face of the dewatered cake. Even the cake, which drained only to
six percent solids, cracked as it further air-dried and reduced in
volume. The final drained sludges had sufficient integrity to be
handled by mechanical bucket.
Each of the sludges tested was produced in the plant equipment as
with the filter leaf tests (see the Filtration Section). The results
195
-------
Figure 44.
Drainometer test unit
0
0 0
/ x1 LIMESTONE PEBBLES e
000 000 °o° 0 Q
a 0
22.4
SLUDGE LEVEL t BACKWASH
OVERFLOW
AREA 2.73 FT 2
45
FINAL SLUDGE SURFACE
ND
DRAIN
196
-------
of sludge handling characteristics for filtration and drying bed
may be compared for the same sludge. About 100 gallons of settled
sludge was withdrawn from the thickener and stored for several days
prior to transfer into the drainometer. During this 48- to 96-hour
queiscent period under typical conditions experienced in a settling
lagoon, none of the sludges (about two feet deep) compressed to a
density greater than 2.8 percent solids. This indicated that settling
techniques (such as would occur in a settling lagoon) only slowly
produce any significant change in terminal sludge density. Chemical
analyses of these sludges are presented in Table 50, in the Filtra-
tion Section. The drainometer test results are summarized in Table
61. The variations in the drainage rates obtained for lime neutralized
sludge from three different waters were limited to the initial part
of the cycle and approached nearly identical rates after about 20
hours drainage time. This common rate of about 0.05 gph/ft 2 at one
foot hydrostatic head is indicated graphically in Figure 45. The
permeability rate of water through 20 inches of 12 percent solids
sludge was found to be 0.06 gph/ft 2 at one foot hydrostatic head.
The plot of drainage rates for CMD sludges are similar to those
reported for sewage sludge (see Figure 41), but were initially more
rapid and approached the limiting values more quickly. The drying
bed size is influenced by both influent solids content and drainage
rates as illustrated in Table 62.
The major conclusions from these tests are:
1. The drainage rate for mine water sludges produced by lime
neutralization attains a stable level after 24 hours of about
0.05 gph/ft 2 . The drainage rate for limestone treatment sludges
is nearly double (0.10 gph/ft 2 ). Both values are for one foot
of hydrostatic head.
2. The water permeability rate of consolidated sludge on a drain-
age bed bottom is about 0.02 gph/ft 2 at one foot of hydrostatic
head.
3. Multiple layers of sludge (with intermittent drying) maintain a
drainage rate of 0.03 to 005 gph/ft 2 .
4. Drainage rates through a cracked dry bed were high initially
(about 0.20 gph), but gradually decreased to the 0.05 gph/ft
level as the cracks filled with new sludge.
5. Solids content of dewatered sludge after several days of draining
ranged from 12-15 percent for hydroxide produced sludges, and
20-30 percent for limestone-neutralized sludges.
6. Shortly after drainage began, the underfiow liquids were clear.
197
-------
Table 61. SUMMARY OF DRAINOMETER TESTS
Temperature: 65°F
Test
No.
Raw water
Source
Alkali Used
Solids
in
Feed
Slurry
Initial
Sludge
Depth
inches
Dral
Rate
nage
After
% Solids
in Final
Drained
Cake
Bulk Density
of Final
Sludge
lbs/ft 3
24 hrs
gph/ft 2
48 hrs
gph/ft 2
1
Proctor No. 1
Hydrated
Lime
2.5
25
0.05
0.05
6.0
63.1
2
Tyler
Dolornitic
1.5
12
0.03
0.02
3
4
S
Proctor No. 2
Proctor Nc. 2
Proctor No. 2
Pebble Lime
Dolomitic
Pebble Lime
Dolomitic
Hydrate
Hydrated
Lime
2.7
2.7
2.8
20
25
20
0.05
0.05
0.05
0.05
0.05
0.05
114
15.2 @ Top
21.3 @ Bottom
21.2
647 b
71.1
70 5
Proctor No. 2 a
Limestone
9.3
20
0.10
0.08
30.7
71.2
aThIS water subject to biochemical iron oxidation prior to neutralization in rotary
reactor. All other slurries were produced by neutralization in flash mixer and
subseciuent air oxidation.
b . . 3
Bulk density of bone dry material is D 2 lbs/ft
-------
Figure 45. Summary of sludge-drainometer dewatering rates
HYDRATED LIME
HYDRATED LIME
DOLOMITE HYDRATE
£ TYLER - PEBBLE DOLOMITE
PROCTOR NO. 2 - PEBBLE DOLOMITE
PROCTOR NO. 2 - BlO-OXIDATION - LIMESTONE - 10% SOLIDS
imestone iuage
20
BASIN DRAINAGE
3.0
0
0
PROCTOR NO. 1 -
PROCTOR NO. 2 -
PROCTOR NO. 2 -
(N
U-
0
LU
F-
LU
0.
0
1
10
30
TIME-HOURS
199
-------
Table 62.
EFFECT OF DRAINAGE RATE AND INFLUENT SOLIDS CONTENT
ON DRYING BED SIZE
Basis: 820 ft 2 drainage area reqd/lOOO GPD bed influent at
2% solids and drainage rate 0.05 GPH/ft 2
C ompar-
ative
Set Condi
No. tions
% Solids
in
Influent
Drainage
Rate
GPI-1/ft 2
Drainage
Area
Required
ft 2 /
1000 GPD
Decrease in
Required
Drainage Area
ft /
1000 GPD % of Area
Effect of Increased Influent % Solids at Constant Drainage Rate
1
A
B
0.
2.
5
0
0.05
0.05
820
720
100
12.2
2
A
B
2.
5.
0
0
0.05
0.05
720
550
170
23.6
3
A
B
5.
8.
0
0
0.05
0.05
550
375
175
31.8
A
B
2.0
8.0
0.05
0.05
720
375
79
3L 5
Effect of Higher Drainage Rates at Constant % Influent Solids
5
A
B
0.5
0.5
0.
0.
05
25
820
150
670
81.
7
6
A
B
2.0
2.0
0.
0.
05
25
720
135
585
81.
2
7
A
B
5.0
5.0
0.
0.
05
25
550
107
43
80.
5
8
A
B
8.0
8.0
0.
0.
05
25
375
76
299
79.
7
of Increased Solids
Rate
Combined Effect
9
10
Content and High Drainage
AB
A
B
0.5
2.0
0.05
0.25
820
135
685
83.5
A
B
2.0
5.0
0.05
0.25
720
107
613
86.0
2.0
8.0
0.05
0.25
720
76
6L L
89.4
200
-------
7. Penetration of sludge into the sand bed was less than 1/16 inch.
8. The drainometer used could not determine the effect of hydraulic
heads greater than 12 inches on drainage rate. However, the
sewage sludge data (Figure 41) indicate that drainage rates are
directly proportional to hydraulic head, hence faster drainage
rates may be attainable by operating a drying bed at a constant
operating head of several feet.
The design of the sludge drying basins sought: Capability for de-
cantation from settled sludge; A sand filter bottom for maximum
drainage capacity; Elevation to permit drainage from the sub-bed
drain lines to suitable discharge; Multiple compartments to provide
for continuous operation by allowing drying to proceed in several
compartments simultaneously; Access for sludge removal equipment;
A bed cover to minimize effect of precipitation; Splash pads under
influent lines to prevent erosion; and Provision for a maximum de-
watered sludge level of three feet.
Operating Objectives of Sludge Drying Basin- -
With design and operating concepts cited, the drying bed utilization
sought further evaluation to compare secondary dewatering approaches,
including: Realistic design drainage rates; Identification of factors
that control drainage rates (e.g., percentage of solids in influent,
methods of addition of sludge to the bed, treatment processes, hy-
draulic head in bed, etc.); Materials handling experience with the
dewatered sludge in removing it from the bed (including equipment);
Determining bulk densities and water content of the sludges in bed
at various stages of dewatering; Determining bed drying times;
Effects of weather on drying bed performance; and Operating and
maintenance labor requirements for bed operation.
Pilot Plant Parameters- -
The range of operating variables are indicated in Table 63. The
test program permitted only limited production of several types of
sludges which were introduced into the drying basin. The laboratory
drainoineter data provided a comparison of response for different
sludge types (Table 61).
Sludge Drying Bed Operational Observations--
The sludge was introduced into a single compartment of the bed by
gravity flow or by pump, as shown in Figure 48, from the settling
lagoon or the thickener. Most sludges tended to settle rapidly when
contacting the bottom with the supernatant water selectively moving
faster over the previously dewatered sludge. This clear water per-
colated rapidly to the under bed and drain providing a very high
201
-------
Figure lj.6.
a,
p
Sludge Drying Basin
I
.
Figure L .7. Newly Constructed
Sludge Drying Basin
Showing Sludge Effluent
-
C
p - .4
T.
. - . , I
r .
o * S .
& 4 4
Figure L.3 1
Introduction of
Settled Sludge into
Drying Basin
_ S
S.
S .
.z.
-------
Figure 11.9,
Sludge in Dewatering
Process
I
Figure 50. Dewatered-Drying
Sludge
Dried Sludge in Covered
Drying Basin
Figure 51.
203
-------
Table 63. OPERATING PARAMETERS LIMITS OF DRYING BED
Average
Maximum working
range range
1. Slurry working volume - 4,700
ft 3 /section
Slurry working volume - 35,000
gal ions/section
Drainage area/section - ft 2 1,270
2. Slurry feed rates - GPM 0-127 25-70
3. % solids in feed slurry 2-15 2-10
4. Drainage rates - GPN/section 1-50 1-4
GPH/ft 2 /bottom area 0.03-0.05 0.03-0.06
5. % solids in dewatered sludge 11-100 13-15
6. Drainage time for 3 ft depth 3-10 5-7
of feed slurry - days
7. Working sludge depth - ft 0-3 1-2
8. Weather protection - center section provided with plastic
canopy.
9. Sludge removal facilities - access for front end loader driven
onto drainage bed surface.
10. Sludge removal rates (with front end loader) - 2 foot depth
of sludge in one section removed in three hours.
Equipment:
1 front end loader - 2 cu. yd.
2 trucks - 6 ton and 9 ton
3 men as equipment operators
11. Anti-freeze provisions - none.
12. Dried solids disposal system - truck haulage to adjacent land-
fill area.
204
-------
initial dewatering rate. This rate varied with the depth and
moisture content of the previous sludge layer. A dried sludge
layer yielded high initial drainage rates. During the evaporative
months, the sludge dried to less than 10 percent moisture (see
Figure 51). Under these conditions, the gel-like sludge structure
had been broken and rapid percolation followed. This type response
favored a slow continuous introduction of the sludge, with longer
drying periods.
The optimum sludge depth to be transferred in a given interval was
difficult to quantify:
1. A transfer of sh.dge to the drying bed creating a fresh depth
greater than six inches leads to conditions favorable for de-
cantation (the most rapid and effective dewatering procedure)
and produces a grea.ter hydraulic head which is conducive to
higher percolation rates.
2. The longer interval possible for one filling cycle in one com-
partment permits longer percolation and evaporative intervals
for the other compartments.
3. Evaporation of water from the sludge decreases rapidly with
sludge depth after the first six inches. Thus, excessive
sludge depths tend to minimize and limit final sludge moisture
content.
4. Dry, non-gelled sludge permits rapid percolation rates.
The following qualitative conclusions are indicated regarding
the optimum sludge depth:
1. The desirable depth of s . ludge in the drying bed is a reciprocal
function of influent sludge density; i.e. A low density feed
sludge may be allowed to accumulate to greater depths since
the potential for decantation is greater, the effect of hy-
draulic head on the drainage rate greater, and yet the final
sludge depth will be the least.
2. The dewatered sludge depth desirable is proportional to the
seasonal evaporative level; i.e. during periods of high temper-
ature, low humidity, and high wind velocities, the sludge
depth may be increased.
3. Some minimum moisture content of dewatered sludge should be
attained before a second layer of sludge is introduced.
Decantation of water in the drying basin was practiced to a limited
degree by attaching a hose to a board floated on the water surface.
205
-------
By initiating a siphon flow, the water was transferred to the under-
drain air vent. A small pump, also floated, would increase the
transfer rate. The supporting board served as an automatic level
limitation to prevent the discharge of sludge. Under any conditions,
constant manual attention was necessary.
Percolation through the bottom of the bed proceeded rapidly during
sludge introduction and slowed as the settled sludge became drainage
rate controlling. After completing manual decantation, the drain-
age continued with a decrease in volume and bed depth (as illustrated
in Figure 49) until no free standing water remained. This time in-
terval seldom exceeded 48 hours regardless of sludge depth intro-
duced. Drainage continued at these slower rates, and surface evapor-
ation began. The effect of evaporation first became evident in the
shallow sludge layers at the periphery of the bed. The continual
decrease in sludge volume increased the growth of shrinkage cracks
(see Figure 50) and continued as illustrated in Figure 51 with about
20-inches of dewatered sludge containing 15-30 percent solids.
Further decrease in moisture content proceeded very slowly and the
effect of sludge depth upon drying became obvious.
It was during the dewatering stages illustrated by Figures 49 and
50 that the plastic bed cover was most effective. After the sludge
had dried to the level shown in Figure 51, the cover gave little
advantage. Physically, the cover was quite stable and no special
problems were experienced. The thin plastic is subject to loss
of strength from ultraviolet radiation during the spring and summer
months. The weakness develops at any crease or fold and extends
upon pressures as from high winds as shown in Figure 4. Replace-
ment appears necessary at six-month intervals. Higher temperatures
developed under the canopy were evidenced year around.
The excellent ability of the sand-surfaced bed to retain sludge
was evidenced by a very clear underflow and by direct observations
of the bed. Figure 53 is a photograph of the bed containing two
to three inches of sludge which had reached nearly a bone dry condi-
tion. A portion of the sludge and sand was removed. It appeared
that the sludge had penetrated less than 1/4 inch into the sand.
Unfortunately, upon each removal of the dewatered sludge by
Capillar Model 922 High Lift, the sand was disturbed and the under-
flow developed some turbidity. This had been anticipated and could
be readily avoided by replacing the sand. This operation was not
attempted since the sludge removal equipment available was some-
what large and difficult to manipulate in these relatively small
beds.
When the dewatered sludge reached depths of 24 to 36 inches having
at least 15 percent solids, they resisted flow and could be hauled
by dump truck without leakage. Rubber-tired equipment did not have
206
-------
sufficient traction to manipulate in the sludge bed, unless the bed
was allowed to achieve much higher solids content. Equipment with
treads performed well although the 15° access slope seemed excessive
when covered with sludge. The use of heavy equipment further dis-
rupted the bed.
The weight of the sludge removal equipment demonstrated an inadequacy
of low-cost red dog as a porous bed support. This material did not
have sufficient physical stability and was compressed during sludge
removal reducing bed drainage rates. The compressed material did
protect the clay drain tiles from breakage. Beds probably should
utilize physically-stable gravel (such as grade 3A) for the bottom
12 inches followed by a second foot of one-inch stone to support
the sand filter layer.
Quantitative Drying Basin Data- -
With the basin containing about six inches of dewatered sludge, on
August 20, 1971, a sludge settled from limestone treatment of
Proctor No. 2 waters was drained by gravity from compartment No. 1
of the settling lagoon to drying basin No. 2. A composite sample
contained 8.8 percent solids by weight. Over 13,000 gallons of
sludge were transferred in four hours raising the bed depth about
15 inches. This represented a rapid transfer rate, which exceeded
sludge production levels. In 20 hours, the bed had dropped nearly
13 inches, more than 0.6 in/hr (about 0.4 g/hr/ft 2 ). After drain-
ing and evaporating ten days, an additional transfer of similar
sludge was made to the same compartment which was allowed to de-
water for 72 hours when a third transfer of similar sludge was
made. The dewatering response of the drying basin under these
conditions is detailed in Table 64 and Figure 58. The residual
volume and solids content of the dewatered sludge in the drying
basin was quite different from that indicated by a sludge settling
test (see Figure 59 for the sludge settling data). In the drying
basin the residual volume was 24.6 percent, in contrast to the
63.1 percent shown by the settling test. The drainometer tests
previsously described provides a satisfactory laboratory approach
to the acquisition of such bed drainage data, although giving
conservative data.
In further drying operations, following five days of dewatering
as described in Table 64, 10,005 gallons of similar sludge (10.6
percent solids) were transferred to drying bed compartment No. 2
which showed a drainage rate of 0.10 g/hr/ft 2 in the next 40 hours
and a volume reduction of 65 percent in the next four days. Nearly
simultaneously with the above dewatering operation, about 14,000
gallons of the same sludge were transferred to the adjacent drying
bed No. 3 which gave a drainage rate of 0.15 gphJft 2 in the first
40 hours.
207
-------
Figure 53. Bone-Dry S1ud e
Showing Mini!nal
Penetration of Sand.
Figure 52. Removing Dewatered
Sludge from Drying Basin
Figure 5I . Dewatered Sludge Drying
in Land Disposal Area
208
-------
Figure 55.
Cubic Foot of
Dewatered Sludge
Frozen Dewat erecj.
Sludge
Figure 57.
Dried Sludge After
Freezing
Figure 56.
4.
209
-------
Table 64. SLUDGE DRYING BASIN EVALUATION
Initial bed conditions: Drying Basin No. 2 Covered
Depth of Sludge in Bed - Approximately 12 in.
Solids Content (Surface) - 15.8% by wt.
Drainage period since previous transfer -
72 hours
Weather - Temperature - 59-84°F
Overcast, humid with rain on
fourth day of dewatering
Sludge transfer: Origin - sludge from settling lagoon, developed
from limestone neutralization of Proctor No. 2
water after biochemical oxidation. Pumped from
settling lagoon compartment no. 1 to sludge drying
basin no. 2.
Suspended Solids - 9.5% by wt. (composite)
Transfer Rate (average) - 73 gpm for 170 minutes
Total Volume Sludge - 12,396 gallons.
11,775 gal. water, 10,315 lb dry solids
(using estimated density - 1.05)
Bed Level Increase - 14.25 inches
Drainage rate: Averaged 0.19 g/hr/ft 2 over first 40 hours (see
Figure 58 for details)
Sludge drying period: 5 days (121 hours), 97% of total volume
reduction in 40 hours
Final bed condition: Sludge Bed Depth Increase - 3.50 inches
Solids Content (surface) - 44.8% by wt.
Final Volume Sludge Added - 3,045 gallons
(407 ft 3 )
(Using estimated density - 1.10) -
1,850 gal water, 12,522 lb dry solids
Water rejection: 84%
Operating time: 1 man hour during initial sludge transfer
210
-------
Table 64 (continued). SLUDGE DRYING BASIN EVALUATION
Sludge dry bed capacitya: 80 days
(3 beds - 36 inches 94,000 gallons dewatered sludge
of sludge) 176 tons dry sludge - 390 tons dewatered
sludge
12,500 ft 3 = 463 yd 3
587,000 gallons settled sludge
59,000,000 gallons coal drainage
(47% excess capacity of plant design)
Cost to remove dewatered sludge: $231.50 ($0.50/yd 3 )
0.31 cents/l,000 gal.
aBased on addition of 4 inches dewatered sludge to alternate beds
every 3 days, thus a nine day drying cycle. Proctor No. 2 with
bhmestofle, 1% settled sludge volume, 350 gpm.
Transfer to land fill - 1/4 mile.
Response of Frozen Sludge -
During periods of sub-freezing temperature, the sludge in the dry-
ing basin and the dewatered sludge placed for disposal as land fill
will freeze and retard dewatering. As the water in the sludge
changes physical state, the sludge structure is modified which leads
upon thawing to major reductions in moisture content and volume.
Although freezing of the dewatering sludge requires planning in
basin design and operation to provide adequate storage volume, the
overall response to sludge freezing is favorable. To quantify this
response, a sludge freeze-thaw test was made.
A cubic foot of dewatered sludge was cut from the drying basin.
The physical instability of the sample led to a bulk volume of
2.05 ft 3 upon transfer to a wire mesh freezing basket (Figures 55
and 56). The latter figure shows the sample after being maintained
at 0°C for 118 hours. The sample was allowed to thaw at about 20°C,
while observing the changes in weight and water drained from the
sample. The resulting data are plotted in Figure 60. The water
ceased to drain from the sludge after 143 hours when the original
one ft 3 volume had reduced to 0.73 ft 3 and the material was trans-
ferred to a tray to prevent sludge loss. At that time the total
weight loss was 54.4 percent of which 52.2 percent was from drain-
age water, and the balance from evaporation. After an additional
28 days drying, the moisture content was less than six percent and
211
-------
Figure 58. Sludge volume and drainage rates sludge drying bed
o
0
0
0.9
0.8
4.
4.
U
C
4.
4.
-J
4 ,
a
C
0
C
0
0
0
10
0.7
-c
0
a
4.
0
4 ,
a
0
a
0
. -1
0
3
6
9 12 15
30
45 60
0
Time (minutes x 102)
-------
E
U
a
700
Figure 59. Settling rate of limestone-produced sludge removed from
bottom of settling lagoon
950
900
800
0
500
1500 2000
5000
Settling Time (minutes)
-------
Figure 60. Dewatering rates from frozen sludge
0
-J
-c
0
4,
20
10
18
16
14
12
In
41
-J
10 -
CI
I
C I
0
U
0
o
41
C .
a
C
a
6
-4
80
/
2161 Hr.
30
0
2% Evaporative Loss During Freezing
4
3
2
0
100
Tkawing Time (Hours)
150 250 350 450 550 650 750 850 950
-------
the final volume was 0.25 ft or one-fourth the original dewatered
sludge volume (Figure 57) - Freezing assists in final dewatering
and drying.
Sludge Drying Bed Design Criteria--
The data obtained from drainometer tests and actual drainage bed
operation recommended design parameters presented in Table 65.
Design graphs for determining bed sizes are contained in Figures
61, 62 and 63.
The drying bed size and number of beds must be established with a
selected operating cycle time. Typical bed sizing calculations
with their corresponding cycle times for a three foot accumulation
of 15 percent sludge are shown in Table 66. Bed size selection must
also be dictated by the requirements of the sludge removal system.
It appears bed lengths should be limited to 300 feet and widths to
about 100 feet , For large bed area requirements use multiple ponds.
Sludge Removal Facilities- -
Since sludge removal requires most labor, efficient and low-cost
methods of sludge handling should be employed. The sludge must be
removed by a system that will not unduly disrupt the sand filter
bottom. An option may be the use of a special dragline. The bed
size and location should be compatible with a particular dragline
configuration.
Most plants would require at least two drying beds for continuous
operation. In order for two or more beds to be serviced by a single
dragline, a radial pie-segment arrangement is suggested 85 . A po-
tential layout is shown in Figure 64. This system provides for the
servicing of two drying beds by a dragline tower located at the
center of the quadrant sector and the stacking of the dried sludge
around the periphery of the sludge bed area. This arrangement
should permit rapid, low-cost removal of the sludge, with dragline
power equipment provided at intervals as needed.
The underbed drainage lines are located for servicing using a common
sump for control discharge.
p4, osal of Dewatered Sludge- -
The dewatered sludge attained adequate consistency for truck trans-
portation without leakage, whether dewatered by the drying basin,
filter or centrifuge. In all cases only the minimum volume and
weight had to be transported. The relative efficiency of handling
is better when removing sludge from the drying beds since the truck
was used constantly only for the day or two required. In contrast,
215
-------
Table 65. SLUDGE DRYING BED DESIGN PARAMETERS
1. Bed Drainage Rate Capacitya
2. Percentage of Solids in Feed Slurry
3. Percentage of Solids in Dewatered Sludge
14. Final Sludge Depthd
5. Bone-Dry-Solids Capacity
(a) for hydroxide slurries
(b) for limestone slurries
6. Specific Gravity, Feed Slurry
7. Specific Gravity, Dewatered. Sludge
8. Bulk Density, Dewatered Sludge
9. Bed Drying TimeCafter influerit cutoff)
10. Total Bed Depthb
11. Bed Holding Capacity
12. Drainage Area Required
13. Average Bed Drainage Rate Through
Sludge
l4. Number of Beds Required
15. Influent Feed Distribution
16. Bed Length
17. Bed Width
1 . Evaporation LOSSeSe
Notes:
0.5 GPH/ft 2 minimum
High as possible
1530%
3 ft
8-12 lbs/ft 3
1320 lbs/ft 3
1.031.014
1.071.13
67 to 71 lbs/ft 3
10 days minimum
6 ft minimum
30 day production mm.
see Figures 61,62,63
3.03 to 0.05 GPFI/ft 2
2 - minimum
Even distributed over
bed area.
300 ft maximum
See Note C
Varies with weather
aThe bed should contain at least 4 inches of clean coarse
sand laid on 8 to 12 inches of 112 by 1-inch stone sup-
ported by 12inches 1/14 by 3-inch stone with at least two
perforated under-drain tiles running the length of the
bed. The 12foot spacing utilized at Hollywood was adequate.
b 1 freezing climates, extra bed depth should be available
to provide sufficient bed volume to store 2 to 4 months
sludge production during the winter months when drainage
rates may decrease drastically.
CBed width should be chosen to be compatible with the type
of sludge removal system to be used. A dragline appears
desirable for a large bed, hence the bed width should not
exceed twice the casting radius of the dragline.
dGreater sludge depths may be used, but drying times may
be excessive for the sludge to attain handling consistency
for belt or truck haulage.
eEVdPOPdtIOfl will aid the final drying of the top 6 to 12
inches of sludge.
216
-------
Figure 61.
Calculator for determining drying basin loading rates
and sludge volumes
SOLIDS
IN
BED INFLUENT
WI. %
16
DRAINAGE AREA REQUIRED
FT 2 /1000 GPD AMD/1000 PPM
FOR BED DRAINAGE RATES OF-
0.010 0.025 0.050 0.100
14
700
(GALLONS/HR PER FT 2 BED AREA)
390
180
108
76
48
156
74
45
30
20
/
12
78
37
22
15
10
39
17
11
7.6
4.8
60O
BASIS: FINAL SLUDGE CONTAINS 15% SOLIDS
10
PROPERTIES OF FINAL SLUDGE
/
500
I - ,
uJ
I-
LU
LU
Ll
0
I-
J
C.)
8
1. SPECIFIC GRAVITY 1.11
2. WI. PER CU. FT. = 69.3 LBS/FT 3
3. WT. PER CU. YD. 1,870 LBS/YD 3
4. DRY SOLIDS CONTENT: 15% SOLIDS
10.4 LBS DRY SOLIDS PER FT 3 SLUDGE
6
LU
0
LU
I-
I
400
LU
0
0
C.)
LU
-A
LU
0
-J
0
-.
4.
300
200
1
2 3
% SOLIDS IN DRAINAGE BED INFLUENT
4
0
217
-------
50
I
Figure 62. Drainage basin area calculator
DRAINAGE BASIN AREA REQUIREMENT-FT 2 PER 1000 GPO CMV TREATED
100 300
150
/
/
/
350 400
I
I / /
/ /
>1
F .,
5 /
4
I-
125 .
0 .
C
C- )
r
C
-J
C-
a
C.,
C
C-
/
/
/
/
/
6
/
/
A
/
00
-4
IN
0 1 ) 2% 3)0 400 500 6% 700 800 900 1000
PPM NEUTRALIZED SOLIDS IN BED INFLLIENT
-------
I-
Lu
-J
-J
C,
8
CD
Lu
iL l
>-
I
C-)
a-
C-)
Lu
900
BOO
700
600
500
400
300
200
100
uJ
J
L I-
-j
8
-J
>-
-J
Figure 63. Sludge drying basin area calculator
DRYING BED AREA REQUIRED-FT 2 /1000 GPD INFLUENT
§ §
Deuteting Referetice Line
0.1 0.2 0.5
10 20
2 3 5 2.0 3.0 4.0 5.0 7.0 10.0
% SOLIDS IN INFLUENT % SOLIDS IN INFLUENI
1.0
0
-------
Table 66. TYPICAL SLUDGE DRYING BED OPERATIONAL PARAMETERS
A CMD Plant Capacity - GPD x l0 .
B Drainage Bed Area Required - ft 2 x l0 3 (A/l0OO x
C z Dry Weight o Solid Precipitated -
lbs/day x 10 (A x 8.34 x 2000) 1106.
0 = Volume of 15% Sludge Generated - GPD x i0 3 =
C/0.15 x 9.25.
5, F, C Volume of Sludge in bed 3 ft deep.
E Gallons x lO 7.4SF.
F - ft 3 x l0 3B
C yd 3 x l0 Ff27
H, I, J, K = Operating Schedule
H Fi1lTime -DaysrEiD
I Drain Time - Days = E12 x 1.2B
J = Clean-out Time - Hours C/l ao TPH
1 < Down Time - Days 1 + J
L Excess Settling Basin Capacity Required-
gallons x 106 A x Kilo
1. A/ bOO x 42
2. A x 8.34 x 1000/106
3. Weight of 1 gallon of 15% sludge
E34 x 1.11 (5.6) = 9.25 lbs.
A
B C D E F C H I J
1<
L ,
Basis:
2,000 ppm neutralized suspended solids settled to 2% solids slurry. 72
area required/bOO CPD raw MD at sludge drainage rate of 0.05 GPH/ft2.
ft 2
drainage bed
500.
36. 8.3 6. 808. 108. 4. 134 10 40
15
0.75
1,000.
72. 16.6 12. 1,616. 216. 8. 134 10 80
20
2.0
2,000.
144. 33.2 2k. 3,232. 432. 16. 134 10 150 (80)
20
4.0
5,000.
360 , 83.0 60. 8,080. 1,080. +0. 134 10 400 (200)
35
17.5
Basis:
500.
1,000 ppm neutralized solids settled to 2% solids. 42 ft 2 drainage bed
GPD raw CMD - 0.05 GPHift 2 drainage rate.
21. 9? 472. 63. 2.3 158 10 24
area
13
required/lOGO
0.65
1,000.
42. 8.3 6. 944. 126. 4.7 158 10 48
15
1.60
2,000.
84. 16.6 12. 1,880. 252. 9.4 158 10 96 (48)
16
3.20
5,000.
210. 41.5 30. 4,720. 630. 23.4 158 10 240 (120)
25
12.50
72)
0
C 4
-------
Figure 64. Suggested layout for dual sludge drainage basin system
TRAVELING TAIL TOWER
NOT DRAWN TO SCALE
221
-------
the truck had to be available continuously in handling sludge from
the filter or centrifuge or a temporary storage developed which re-
quired multiple handling.
The dewatered sludge was trucked to an abandoned borrow pit
(several acres) adjacent to the Treatment Facility which would have
capacity for years of plant opcration at design capacity. The dry-
ing and volume reduction of the dewatered sludge continued slowly in
the disposal area by percolation and evaporation whether in warm or
sub-freezing weather. The volume reduction as a result of freezing
was previously detailed (about 75 percent). The volume reduction
from drying is illustrated in Figure 65 for a dewatered limestone
sludge. The small cylinder (8.95 in 3 ) was utilized to separate a
dewatered specimen (left) which was dried in the laboratory. The
final cylinder volume was 1.76 in 3 (right) for a reduction of about
80 percent.
After the dumped sludge reached a moisture content near that of the
surrourding soil, it was graded, compacted, and seeded with common
grasses, which germinated and grew. The seeding prevents possible
windage of the dried sludge.
1i ure 65. Volume Reduction in Dewatered. S1udr e Άi en Dried
iL
222
-------
SECTION XV
REFE RE NCES
1. Hill, R. D. Mine Drainage Treatment - State of the Art and Research
Needs. U.S.D.I., Federal Water Pollution Control Administration,
Cincinnati, Ohio, December 1968.
la. Hill. Ihid, p. 7-8.
2. Lewis, C. J. Some Practical Suggestions on Waste Acid Treatment.
Rock Products. 149:117-9, June 1949.
3. Loveil, H. L., and R. I. Lachman. An Investigation of the Natural
Beneficiation of Coal Mine Drainage. Coal Research Section.
The Pennsylvania State University, University Park, Pa. Special
Report-76, May 1970.
4. Singer, P. C., and W. Stumm. Kinetics of the Oxidation of Ferrous
Iron. In: Second Symp or Coal Mine Drainage Research. Pitts-
burgh, Pa., 1968. p. 12.
5. Stauffer, T. E., and H. L. Lovell. The Oxygenation of Iron (II) -
Relationship to Coal Mine Drainage. Coal Research Section. The
Pennsylvania State University, University Park, Pa. Special
Report-69, November 1968.
Also: Preprints of paper, Am Chem Soc, Div Fuel Chem. 13:2,88,
1969.
6. Stumm, W., and G. F. Lee. Oxygenation of Ferrous Iron. md Engr
Chem. 53:143, February 1961.
7. Stumm, W. Oxygenation of Ferrous Iron - Properties of Aqueous Iron
as Related to Mine Drainage Pollution. In: Symp Acid Mine
Drainage Research. Pittsburgh, 1965,. p. Si.
8. Mihok, E. A. Mine Water Research. U.S. Bureau of Mines, Report of
Investigation 7368, April 1970.
9. Heise, L. W., and M. Johnson. Practical Aspects of Waste Pickle
Liquor Disposal. In: Proc Thirteenth md Waste Conf. Purdue
University Engr Ext Dept. 1958. Series No. 96, p. 140.
10. Mine Drainage Manual, Pennsylvania Department of Health, Division of
Sanitary Engineering, Sanitary Water Board, Harrisburg, Pa.
Publication No. 12, 2nd Edition. Revised December 1966.
223
-------
lOa. Mine Drainage Manual, Section III, Design Criteria for Treatment
Facilities.
11. Hodginan, D. C., Editor. Handbook of Chemistry and Physics. Cleve-
land, Chem Rubber Pub Comp. 34th Edition, 1952-3, P. 1533.
ha. Hodgman. Ibid, p. 1558.
12. Seidιll, A., and W. F. Linke. Solubilities - Inorganic and Metal
Organic Compounds. Washington, D. C., Am Chem Soc, Vol I, 4th
Edition, 1958. P. 495-693.
13. Gustafson, H. C. Process and Apparatus of Producing Dense Sludges.
U.S. Patent No. 3,247,105. April 1966.
14. Evans, R. R. Precipitation of High Density Metallic Hydroxides for
Recovery or Disposal. tn: Proc Twenty-first md Waste Conf.
Purdue University, Engr Ext Dept. 1966.
15. Kostenbader, P. D., and E. F. Haines. High-Density Sludge Process
for Treating Acid Mine Drainage. In: Third Symp Coal Mine
Drainage Research. Pittsburgh, 1970. p. 12-26.
16. Tracy, L. T. Six Tons of Ferric Hydrate Secured Daily from Water
at a Connellsville Mine. Coal Age. 18:13-6, 1920.
17. Tracy, L. T. Mine-Water Neutralizing Plant at Calumet Mine.
Trans Amer Inst Mm Engrs. 66:609-23, 1921.
18. Zurbuck, P. E. Dissolving Limestone from Revolving Drums in
Flowing Water. Trans Amer Fisheries Soc. 93(2):173-8,
1963.
19. Mihok, E. A., and C. E. Chamberlain. Factors in Neutralizing Acid
Mine Wastes with Limestone. In: Proc Second Symp on Coal
Mine Drainage. Pittsburgh, p. 265-73. 1968.
20. Mihok, B. A., M. Deul, C. E. Chamberlain, and J. G. Selmeczi.
Mine Water Research - The Limestone Neutralization Process.
U.S. Bureau of Mines, Report of Investigation 7191, September
1968.
21. Anonymous. Flow Sheet for Neutralization. Deco Trefoil. Winter
Issue, Denver Equipment Co. p. 13-4, 1970-71.
22. Ford, C. 1. Selection of Limestones as Neutralizing Agents for
Coal Mine Wastes. In: Proc Third Symp on Coal Mine Drainage.
Pittsburgh, 1970. p. 27-51.
224
-------
23. Bituminous Coal Research, Inc. Studies on Limestone Treatment
of Acid Mine Drainage. Federal Water Quality Administration.
Water Pollution Control Research Series, DAST-30, 14O1OEIZ.
24. Calhoun, F. P. Treatment of Mine Drainage with Limestone. In:
Proc Second Symp on Coal Mine Drainage. Pittsburgh, 1968.
p. 386-91.
25. Anonymous. Acid Mine Water + Limestone Clean Stream. Coal
Age. 74:(2) p. 112-4. 1969.
26. Birch, J. J. Coal Research Board Report. Commonwealth of Pennsyl-
vania.
27. Birch, J. J. Application of Mine Drainage Control Methods. In:
Proc Second Symp on Coal Mine Drainage. Pittsburgh, 1968.
p. 372-5.
28. Holland, C. T., R. C. Berkshire, and D. F. Golden. An Experimen-
tal Investigation of the Treatment of Acid Mine Water Contain-
ing High Concentrations of Ferrous Iron with Limestone. In:
Proc Third Symp on Coal Mine Drainage. Pittsburgh, 1970.
p. 52-65.
29. Wilmoth, R. C., and R. B. Scott. Neutralization of High Ferric
Iron Acid Mine Drainage. In: Proc Third Symp on Coal Mine
Drainage. Pittsburgh, 1970. p. 66-93.
30. Gehm, H. W. Neutralization of Acid Waste Waters with an Up-flow
Expanded Limestone Bed. Sewage Works J. 16:104-20, 1944.
31. Hoak, R. D., C. J. Lewis, and W. W. Hodge. Treatment of Spent
Pickling Liquors with Limestone and Lime. md Engr Chem.
37: (6),553.-9, 1945.
32. Khavskii, N. N., I. A. Yakubovich, and N. A. PruobrazheflSkii:
Neutralization of Sulfuric Acid Solution with Limestone in
the Presence of Ultrasound. Sb Mosk Inst. Stali Splavov,
53:11720, 1969
33. Hill, R. D.,, and R. C. Wilmoth. Preprint 14010, Soc of Mining
Engrs. St. Louis, Missouri. October 1970.
34. Jarrett, R. E., and R. R. Kountz. The Pennsylvania State Univer-
sity, University Park, Pa. Water Resources Research Publi-
cation No. 2-66, June 1966.
35. Yen, A. F. Design of Limestone Barriers in Acid Mine Water
Streams. M.S. Thesis in Civil Engineering. The Pennsylvania
State University, University Park, Pa. September 1969.
225
-------
36. Braley, S. A. Summary Report to Commonwealth of Pennsylvania.
Numbers 1-7. Mellon Institute, Pittsburgh, Pa. February
1954.
37. Clifford, J. E., and C. A. Snarley. Battelle Memorial Institute,
Columbus, Ohio. 1954.
38. Armco Steel Corporation. Limestone Treatment of Rinse Waters
from Hydrochloric Acid Pickling of Steel. Environmental
Protection Agency, Report 12010 DUL. February 1971.
39. Wilmoth, IL 0., and R. D. Hill. Neutralization of High Ferric
Iron Acid Mine Drainage. Federal Water Quality Administration,
Report 14OIOETV. August 1970.
40. Clover, H. G., J. W. Hunt, and W. C. Kenyon. Process for the
Bacteriological Oxidation of Ferrous Salts in Acidic Solution.
U.S. Patent No. 3,218,252 , 1965.
41. Clover, H. C., .1. W. Hunt, and W. C. Kenyori. Process for the
Bacteriological Oxidation of Ferrous Salts in Acidic Solution.
British Patent No. 1,056,262. 1967.
42. Clover, H. C. The Control of Acid Mine Drainage Pollution by
Biochemical Oxidation and Limestone Neutralization Treatment.
In: Proc Twenty-second md Waste Conf. Purdue University
Engr Ext Dept. 1967. Part I, Series No. 129, p. 823.
43. Ohyuma, T., .1. Shimorzuka, and S. Usni. The Sedimentation
Characteristic of Calcium Carbonate Neutralization Method
in Mine Water Disposal. Tohoku Kozan 4:21, 1957.
44. Pourbaix, M. Atlas of Electrochemical Equilibria in Aqueous
Solutions. Translation by J. A. Franklin. New York,
Pergamon Press. 1966.
44a. Pourbaix, Ibid. p. 308-9.
44b. Pourbaix, Ibid. p. 313.
45. Carrels, R. 14., and L. L. Christ. Minerals, Solutions, and
Equilibria. New York, Harper and Row. 1965.
45a. Carrels, et al. Ibid. p. 107.
46. Lovell, H. L., and T. Stauffer. The Oxidation of Iron (II) -
Relationship to Coal Mine Drainage Treatment. The Pennsyl-
vania State University, University Park, Pa. Coal Research
Section. Special Report SR-69. 1968.
226
-------
47. Johnson, F. C. The Treatment and Filtration of the Underground
Mine Waters of the Wilwaterstand. J S Afr Inst Mech Engrs
91:117, 1952.
48. Travers, J. T. Process and Apparatus for Treating Polluted Acid
Waters. U.S. Patent No. 1,685,300. 1928.
49. Travers, J. T. Treatment of Polluted Wastes. U.S. Patent No.
1,685,301. 1928.
50. Birch, J. J. Application of Mine Drainage Control Methods. In:
Second Symp on Coal Mine Drainage Research. Pittsburgh, 1968.
p. 372.
51. Birch. J. J. Process of Treating Acid Mine Water. U.S. Patent
Application No. 3,516,931. 1968.
52. Glover, G. H. Personal Communication. Autotrophic Acidophilic
Iron Oxidizing Bacteria. The Pennsylvania State University,
April 1969.
53. Lovell, H. L., M. L. Miskovsky, P. T. Luckie, and T. S. Spicer.
Regarding the Water Pollution Potential from Coal Refuse
Piles. Fifth Middle Atlantic Reg Meeting, Am Chem Soc.
Wilmington, Delaware. April 1970.
54. Kuznetsov, S. I., M. V. Ivanov, and N. N. Lyalikova. Introduction
to Geological Microbiology. New York, McGraw-Hill, 1963.
55. Whitesell, L. B., R. L. Huddleston, and R. C. Allred. The Micro-
biological Oxidation Products of Ferrous Iron in Mine Drainage
Water. Preprints, Div Fuel Chem, Am Chem Soc. 13:2:79 1969.
56. Whitesell, L. B., R. L. Huddleston, Z. V. Koswski. The Micro-
biological Oxidation of Ferrous Iron in Mine Drainage Water
Pilot Plant Studies. Preprints, Div Fuel Chem, Am Chem Soc.
15:3 1971.
57. Silverman, M. P., and D. G. Lundgren. Studies on the ChemoaUtO
trophic Iron Bacterium Ferrobacillus ferroxidans . J Bact
77:642, 1959.
58. Lundgren, D. G., and C. A. Schnaitinan. The Iron Oxidizing Bac-
teria. In: Symp on Acid Mine Drainage Research. Pittsburgh,
1965.
59. Unz, R. F., and D. C. Lundgren. A Comparative Nutritional Study
of Three Chemoautotrophic Bacteria: Perrobacillus ferroxida !,
Thiobacillus ferroxidans and Thiobacillus thiooxidans . Soil
Sd 92:302, 1961.
227
-------
60. Leathen, W. W., H. Kensel, and S. A. Braley. Ferrobacillus
ferrooxidans: A Chemosynthetic Autotrophic Bacterium.
.1 Bact,72:700-4, 1956.
61. Hutchinson, M., K. I. Johnstone, and D. White. Taxonomy of the
Acidophilic Thiobacilli. J Gen Microbiology, 44:373, 1966.
62. American Type Culture Collection Catalog of Strains. 8th Edition.
p. 134, 1968.
63. Woodward, R. L. How Probable is the Most Probable Number? J Am
Water Works Assoc. 49:1060, 1957.
64. Lovell, H. L. The Control and Properties of Sludge Produced from
the Treatment of Coal Mine Drainage Water by Neutralization
Processes. In: Third Syrnp on Coal Mine Drainage Research.
Pittsburgh, 1970. p 1-11.
65. Hoak, R. D., C. L. Lewis, C. J. Sindlinger, and B. Klein. Pickle
Liquor Neutralization. md Engr Chem 40:11,2062, 1948.
66. Sanmarful, I. deC. Evaluation of Common Alkalies in Nuetralizing
Acid Mine Water. M.S. Thesis in Civil Engineering, R. R.
Kountz, Advisor. The Pennsylvania State University, Univer-
sity Park, Pa. 1969.
67. Deul, M., and E. A. Mihok. Mine Water Research Neutralization.
U.S. Bureau of Mines, Report of Investigation 6987, 1967.
68. Streeter, R. D., D. C. McLean, and H. L. Lovell. Reduction of
Hydrous Ferric Oxide to a Magnetic Form with Sodium Dithionite;
Implications for Coal Mine Drainage Treatment. Reprints, Ani
Chem Soc, Div Fuel Chem. 15:2,13-25, 1971: and Ibid:
Streeter, R. D. Ph.D. Thesis. Mm Prep. The Pennsylvania
State University, 1971.
69. Stauffer, T. E., and H. L. Lovell. The Oxygenation of Iron (II) -
Relationship to Coal Mine Drainage Treatment. Special Report-
69, Coal Research Section. The Pennsylvania State University,
1968; and Ibid: M. S. Thesis, Mm Prep 1968.
70. Whittemore, D. 0. The Chemistry and Mineralogy of Ferric Oxyhydro-
xides Precipitated in Sulfate Solutions. Ph.D. Thesis,
Geochemistry, The Pennsylvania State University, University
Park, Pa. 1973.
228
-------
71. Whittemore, D. 0., and D. Langmuir. Variations in the Stability
of Precipitated Ferric Oxyhydroxides. Nonequilibrium Systems
and Process in Natural Water Chemistry. Washington, D. C.,
1971. Am Chem Soc Advances in Chem Symp. Robert F. Gould,
Editor.
72. Dorr-Oliver, Inc. Operation Yellowboy - Mine Drainage Treatment -
Plans and Cost Evaluation. Pennsylvania Coal Research Board,
Department of Mines and Mineral Industries, Harrisburg, Pa.
June 1966.
73. Perry, R. H. et al. Chemical Engineers Handbook. New York,
McGraw-Hill, 1963. 4th Edition, Chapter 19, p. 42.
73a. Perry, Ibid. p.19-86.
74. Trubrick, E. H. Vacuum Filtration of Raw Sludge. J Water and
Sewage Works. 107:R-287-9l, October 1960.
75. Dorr-Oliver, Inc. Operation Yellowboy - Mine Drainage Plan,
Bethlehem Mines Corporation, Mariana Mine No. 58. Pennsyl-
vania Coal Research Board. Department of Mines and Mineral
Industries, Harrisburg, Pa. January 1966.
76. deVilliers, J. W. An Investigation into the Design of Underground
Settlers. J S Afr Inst Mng and Met. p. 501-21, June 1961.
77. Galloway, R. E., and J. F. Colville. Treatment of Spent Pickling
Plant Liquors. Management of Water in the Iron and Steel
Industry. Iron and Steel Institute, London. p. 128, 1970.
78. Hoak, R. D., and C. J. Sindlinger. New Technique for Waste Pickle
Liquor Neutralization. md Engr Chem 41:1,65-70, 1949.
79. Brown, J. M. Vacuum Filtration of Digested Sludge. W W S.
p. R324-7, Reference No. 1960.
80. Trubnick, E. H. Vacuum Filtration of Raw Sludge. W W S, p. R287-
91, Reference No. 1960.
81. Rudolfs, W. Lime-Handling, Application and Storage in Treatment
Process. First Edition, 1949; Second Edition, 1971; National
Lime Association.
82. Parsons, W. A. Chemical Treatment of Sewage and Industrial Waters.
Bulletin 215, National Lime Association, Washington, D. C.
1965.
82a. Parsons, Ibid. p. 73-6.
229
-------
83. Jeffrey, E. A. A Laboratory Study of Dewatering Rates for
Digested Sludge in Lagoon. In: Proc Fourteenth md Waste
Conf. Purdue University, 1959. Series No. 104, p. 359-84,
1959.
84. Todd, 0. K., Editor. The Water Encyclopedia. Water Information
Center, Port Washington. 1970.
85. Sauerman Brothers, Inc. Private Communication. October 1971.
86. Standard Methods for the Examination of Water and Waste Water,
New York, 12th Edition, 1965. American Public Health
Association.
86a. Standard, Ibid. p. 46.
86b. Standard, Ibid. p. 48.
86c. Standard, Ibid. p. 156.
86d. Standard, Ibid. p. 225.
86e. Standard, Ibid. p. 287.
86f. Standard, Ibid. p. 37-8.
86g. Standard, Ibid. p. 175.
87. Kupiec, A. R. Analytical Methods Review and Sampling Procedures as
Related to a State Control Agency. In: Second Symp on Coal
Mine Drainage Research. Pittsburgh, 1968. p. 39.
88. Anonymous. FWPCA Methods for Chemical Analysis of Water and Wastes,
U.S. Dept of Interior, Federal Water Pollution Control Adminis-
tration, November 1969.
89. Anonymous. Methods for Chemical Analysis of Water and Wastes,
Environmental Protection Agency. Water Quality Office,
Cincinnati, Ohio. No. 16020.
90. Lovell, H. L., and E. M. Heilman. Procedures for the Analysis of
Coal Mine Drainage Waters. Special Report, The Pennsylvania
State University, Mine Drainage Research Section. In Press.
1971.
91. Salotto, B. V., E. E. Barth, M. B. Ettinger, and W. E. Tolliver.
Determination of Mine Waste Acidity. U.S.D.I. Federal Water
Pollution Control Administration. Cincinnati, Ohio. 154th
National Meeting of the Am Chem Soc. Chicago, Illinois.
230
-------
92. Payne, D. A., and T. E. Yeates. The Effects of Magnesium on
Acidity Determinations of Mine Drainage. In: Proc Third
Symp on Coal Mine Drainage Research. Pittsburgh, 1970.
p. 200-26.
93. Medlin, J. H., N. 11. Suhr, and J. B. Bodkin. Atomic Absorption
Analysis of Silicates Employing L1BO 2 Fusion. Atomic Absor-
ption Newsletter 8:2,25-9, 1969.
94. ONeil, R. L., and N. H. Suhr. Determination of Trace Elements
in Lignite Ashes. Applied Spect 14:2,45-50, 1960.
95. Anonymous. Caustic Soda Handbook. Diamond Alkali Company,
1967.
96. Kirk, R., and D. Othmer. Encyclopedia Chem Tech. New York,
John Wiley and Sons, 5, 1967, 2nd Edition, p. 711-15.
97. Johns-Manville Products Corporation. Rotary Precoat Filtration
of Sludge from Acid Mine Drainage Neutralization. Environ-
mental Protection Agency. 14010Dil 05/71. May 1971.
98. Lovell, H. L., and D. E. Kaelin. Operating Experience with
Biochemical Iron Oxidation-Limestone Neutralization Treatment
of Coal Mine Drainage. Reprint No. 72-F-l04, AIME-SME
Annual Meeting, San Francisco, California. February 1972.
99. Lovell, I!. L. Experience with Biochemical-Iron-Oxidation Limestone-
Neutralization Process. In: Fourth Symp on Coal Mine Drainage
Research. Pittsburgh, 1972. p. 292-1 to 292-14.
100. Steinberg, M., J. Pruzansky, L. R. Jefferson, and B. Manowitz.
Removal of Iron from Acid Mine Drainage Waste with the Aid
of High Energy Radiation. In: Second Symp on Coal Mine
Drainage Research. Pittsburgh, 1968. p. 291-318.
231
-------
SECTION XVI
GLOSSARY
Air Oxidation - A process in which a substance, as ferrous sulfate,
is oxidized to a higher oxidation number by reaction with oxygen
contained in air.
Attrition - An autogenous process which maintains surfaces of solids
clean and free of foreign substances, as reaction products. It re-
suits from frictional action upon contact of two particles.
Autotrophic - Needing only inorganic compounds for nutrition.
Biochemical Oxidation - A process in which a substance, as ferrous,
sulfate, is oxidized to a higher oxidation stage by some mechanism
in which the metabolic life process predominates.
Calcined - A material which has been subjected to very high tempera-
tures which changes its composition. As calcined lime, prepared by
heating limestone with the subsequent removal of carbon dioxide.
Cation Exchange Reaction - The replacement of undesirable cation com-
ponents in coal mine drainage by more acceptable soluble ions by
solubility principles.
Clarifier - A mechanical device used in solid-fluid separation of
suspensions designed to provide a clear, particle-free liquid product.
Coal Mine Drainage (CMD ) - A water transferred from a coal mine environ-
ment. Such waters usually have enhanced dissolved solids and poor
quality.
Compression - A condition which develops during quiescent particle sed-
imentation in which the particle-liquid structure becomes firm and
develops compressive strength. The layer of solids supports layers
above it, reducing subsidence by mechanical support propagated upward
from the bottom of the containing vessel.
Densator - A proprietary type of thickener-reactor.
Dewaterable Sludge - An aqueous suspension of sparingly insoluble
materials from which a major portion of the water may be removed by
physical processes.
Dynamic System - One which is continuous by the addition of reactants
and removal of products.
232
-------
Effluent - A water product leaving a treatment process or system.
Feed or Raw Water - A water entering a treatment process of system.
It is usually applied to those waters entering the first stage of a
treatment system.
Gel - A process in which particulate matter and associated water mole-
cules interact to form a physical state which resists flow.
Fleterotrophic - Obtaining nourishment primarily from organic matter.
Influent - A water entering a treatment process.
Mineralized Waters - Waters which contain soluble mineral substances
acquired by the water as it passed through strata of the earth.
Neutralization - A chemical reaction or processes which decreases the
hydrogen ion content of coal mine drainage.
Neutralization Equivalent - The quantity of a chemical needed to react
completely and without excess in a neutralization reaction. The quan-.
tity is consistent with that defined stoichiometrically by a balanced
chemical equation.
Polluted Water - Waters containing substances inimical to normal util.-
ization and which are not indigenous to most waters. The term as used
is not restricted to biological contamination.
Rat Holing - A phenomena observed in settled sludges during removal of
sludge by a vacuum line. The sludge tends to move more slowly than
the clarified water and does not readily disperse. An interface
envelope forms such that water, not sludge 1 is removed by the pump.
Reactivity - The extent of response, usually chemical, between two or
more materials.
Red Dog - A non-volatile combustion product of the oxidation of coal
or coal refuse. Most commonly applied to material resulting from
in situ, uncontrolled burning of coal or coal refuse piles. It is
similar to coal ash.
Secondary Sludge Dewatering - A subsequent process to solid-fluid
separation in which more water is removed from the solids.
Settleability - An arbitrary concept for the behavior of solids sedi-
menting in a fluid which seeks to describe a variety of settling char-
acteristics.
233
-------
Slaking - A hydration process which converts calcined lime to hydrated
lime.
Sludge - A thick, aqueous suspension of sparingly insoluble materials.
Usually, but not necessarily, composed of valueless, waste substances.
Sludge Blanket - An interface between clear supernatant and settling
solids in a clarifier or thickener. It generally represents the
transition zone.
Slurry - A dilute aqueous suspension of a sparingly insoluble material.
pernatant - The clear liquid which occurs above a sludge in a sedi
mentat ion operation.
Surface Biochemical Reactor - A device providing means by which bio-
chemical reactions are carried out on the surface of an inert substrate.
Thickener - A mechanical device used in solid-fluid separations of sus-
pensions designed to concentrate or thicken the solids.
Thixotropic - A rheological property of gelled sludges describing an
increased tendency to flow after stirring or mixing.
Transition Zone - A region of sedimentation process in which free set-
tling particles approach and contact each other to exist in a constrained
structure so that all settle at the same rate.
Unwanted Reactions - A secondary or side reaction which consumes reagents
or results in products beyond those desired, or planned, by an arbitary
definition.
Vat Reactor System - A means by which chemical or biochemical reactions
are carried out submerged in a liquid held in a large volume container.
Yellowboy Process - A conventional method of treating coal mine drainage
Eaddition of hydrated lime, followed by ferrous iron oxidation by
Oxygen in air, and the separation of insoluble impurities by sedimentation.
234
-------
SECTION XVII
APPENDICES
Page
A. Description of Water Characteristics at Plant Site 240
Tables
1. Summary Stream Analyses at Hollywood Site 241
2. Detailed Stream Analyses at Hollywood Site 244
Figures
1. Seasonal Variation in Water Quality at Hollywood 246
Facility - Proctor No. 2
2. Seasonal Variation in Water Quality at Hollywood 248
Facility - Proctor No. 2 Bore Hole
3. Seasonal Variation in Water Quality at Hollywood 250
Facility - Proctor No. 1
4. Seasonal Variation in Water Quality at Hollywood 252
Facility - Bennetts Branch
5. Seasonal Variation in Water Quality at Hollywood 254
Facility - Tyler Run
B. Plant Design 256
Tables
1. Units and Facilities 256
2. Specifications for Surface Aerators 273
3. Power Consumption by Various Unit Operations 274
4. Dimensions of Settling Lagoon Compart nents 277
Figures
1. Photograph of Control Building 278
2. Application of Bentonite Clay to Holding Lagoon 279
235
-------
APPENDICES (continued)
Page
3. Filling Holding Lagoon 279
4. External Unit Operations 280
5. The Settling Lagoon 281
6. Limestone Reactor Showing InflUent Detail 282
7. Limestone Reactor Showing Effluent Detail and Drive 283
Mechanism
8. Alkali Slurry Preparation Equipment 284
9. Internal Piping and Unit Operations 285
10. Sump Designs 286
11. Settling Lagoon Design 287
12. Sludge Drying Basin Design 288
13. Oxidation Tanks with Surface Aerator 289
14. Infilco Densator Design 290
15. Surface Biochemical Oxidation Reactor 291
16. Schematic of Limestone Reactor 292
C. Detailed Plant Data 293
Tables
1. Treatment of Coal Mine Drainage with Hydrated Lime - 293
Process Conditions - Process No. 10
2. Treatment of Coal Mine Drainage with Hydrated Lime - 294
Raw Water Analyses
3. Treatment of Coal Mine Drainage with Hydrated Lime - 295
Process and Reagent Requirements
4. Treatment of Coal Mine Drainage with Hydrated Lime - 297
Oxidation Tank and Settling Lagoon Analyses - Process
No. 10
236
-------
APPENDICES (continued)
Page
5. Treatment of Coal Mine Drainage with Hydrated Lime - 298
Process Conditions - Process No. U
6. Treatment of Coal Mine Drainage with Hydrated Lime - 299
Raw Water Analyses
7. Treatment of Coal Mine Drainage with Hydrated Lime - 300
Process and Reagent Requirements
8. Treatment of Coal Mine Drainage with Hydrated Lime - 302
Oxidation Tank Analyses - Process No. 11
9. Treatment of Coal Mine Drainage with Calcined Lime 303
Process Conditions - Process No. 10
10. Treatment of Coal Mine Drainage with Calcined Lime - 304
Raw Water Analyses
11. Treatment of Coal Mine Drainage with Caicined Lime - 305
Process and Reagent Requirements
12. Treatment of Coal Mine Drainage with Calcined Lime - 306
Oxidation Tank Analyese - Process No. 10
13. Treatment of Coal Mine Drainage with Calcined Lime 307
Settling Lagoon Effluent Analyses - Process No. 10
14. Treatment of Coal Mine Drainage with Sodium Hydroxide 308
Process Conditions
15. Treatment of Coal Mine Drainage with Sodium Hydroxide 309
Raw Water Analyses - Process No. 10
16. Treatment of Coal Mine Drainage with Sodium Hydroxide - 310
Oxidation Tank Analyses - Process No. 10
17. Treatment of Coal Mine Drainage with Sodium Hydroxide 311
Process and Reagent Requirements
18. Treatment of Coal Mine Drainage with Sodium Hydroxide - 313
Settling Lagoon Effluent Analyses - Process No. 10
19. Treatment of Coal Mine Drainage with Sodium Carbonate - 314
Process Conditions - Process No. 10
237
-------
APPENDICES (continued)
Page
20. Treatment of Coal Mine Drainage with Sodium Carbonate - 315
Raw Water Analyses - Process No. 10
21. Treatment of Coal Mine Drainage with Sodium Carbonate - 316
Prucess and Reagent Requirements
22. Treatment of Coal Mine Drainage with Sodium Carbonate 318
Oxidation Tank Analyses - Process No. 10
23. Treatment of Coal Mine Drainage with Sodium Carbonate - 319
Settling Lagoon Effluent Analyses - Process No. 10
24. Treatment of Coal Mine Drainage with Sodium Carbonate 320
Process Conditions - Process No. 11
25. Treatment of Coal Mine Drainage with Sodium Carbonate 321
Raw Water Analyses Process No. 11
26. Treatment of Coal Mine Drainage with Sodium Carbonate - 322
Process and Reagent Requirements
27. Treatment of Coal Mine Drainage with Sodium Carbonate 323
Oxidation Tank Analyses - Process No. 11
28. Treatment of Coal Mine Drainage with Sodium Carbonate 323
Thickener Effluent Analyses - Process No. 11
29. Treatment of Coal Mine Drainage with hydrated Dolomite - 324
Process Conditions
30. Treatment of Coal Mine Drainage with Hydrated Dolomite - 325
Raw Water Analyses
31. Treatment of Coal Mine Drainage with Hydrated Dolomite - 326
Process and Reagent Requirements
32. Treatment of Coal Mine Drainage with Hydrated Dolomite - 327
Oxidation Tank Analyses
33. Treatment of Coal Mine Drainage with Hydrated Dolomite 327
Settling Lagoon Effluent Analyses
34. Treatment of Coal Mine Drainage with Pebbled Dolomite - 328
Raw Water Analyses
238
-------
APPENDICES (continued)
Page
35. Treatment of Coal Mine Drainage with Pebbled Dolomite - 329
Process and Reagent Requirements
36. Treatment of Coal Mine Drainage with Pebbled Dolomite - 330
Oxidation Tank and Settling Lagoon Effluent Analyses -
Process No. 11
37. Treatment of Coal Mine Drainage with Dolomitic 331
Limestone - Process Conditions
38. Treatment of Coal Mine Drainage with Dolomitic 332
Limestone - Raw Water Analyses
39. Treatment of Coal Mine Drainage with Dolomitic 333
Limestone - Process and Reagent Requirements
40. Treatment of Coal Mine Drainage with Dolomitic 335
Limestone - Limestone Reactor Effluent Analyses
41. Treatment of Coal Mine Drainage with Dolomitic 336
Limestone - Oxidation Tank Analyses
42. Treatment of Coal Mine Drainage with Dolomitic 337
Limestone - Settling Lagoon Effluent Analyses
D. Procedures for the Analyses of Coal Mine Drainage Waters 338
Figure
1. Correlation Between Conductivity and Sulfate Concen- 340
tration in Mine Drainage Waters
E. Estimating Iron Oxidation Rates 345
1. Computer Program for Estimating Biochemical Iron 345
Oxidation Rates
2. Example of the Estimation of Biochemical - Iron II 346
Oxidation Rates
3. Example of the Estimation of Air - Iron II Oxidation 347
Rates
239
-------
SECTION XVII
APPENDIX A
DESCRIPTION OF WATER CHARACTERISTICS AT PLANT SITE
The site of the Hollywood Facility was selected through the former
Department of Mines and Mineral Industries and other cooperating
Pennsylvania State agencies. After a review of several potential
locations, the New Shawmut Mining Company (Mr. A. Palumbo, President),
St. Marys., Pennsylvania, made the 40 acre tract available. It is
in Huston Township, Clearfield County, near the Village of Hollywood.
This site, adjacent to Pennsylvania Route 255, includes coal mine
drainage flows from two non-working deep mines, Proctor No. 1 and
Proctor No. 2, from a small stream (Tyler Run) related to these mines
and associated abandoned surface mines and from Bennetts Branch.
The latter is a stream normally 60 feet wide, which receives drainage
from several outfalls upstream. Bennetts Branch would insure a
reliable water source for the Facility through all seasons. These
streams pass through the site and flow via Bennetts Branch and the
Sinnemahoning to the West Branch of the Susquehanna River.
The water quality of these streams available at the Facility have
been monitored monthly between the springs of 1966 and 1971. The
analyses are summarized in Table 1. A more detailed analysis of the
waters in January 1968 is given in Table 2. The monthly samples were
taken at stream locations where the waters are collected for transfer
to the Treatment Facility. In addition, during the active portion of
the study, two further points down stream from the Facility, referred
to as Stream Combinations No. 1 and No. 2, were sampled. The former
combination would include the flow from Proctor No. 2, Proctor No. 1,
Tyler Run and also the treated effluent from the Facility. The
second combination was further downstream in Bennetts Branch, and
would represent the combined water of all four sources, including any
plant effluent. There was a variable overflow through a borehole,
about one mile northwest of the Facility originating in the Proctor
No. 2 mine. This source was also monitored.
The seasonal variation in the water quality of these streams is shown
in Figures 1 through 5. All the waters reflect change from precipi-
tation and subsequent run-off. The seasonal changes are most pro-
nounced in the stream samples.
240
-------
Table 1. SUMMARY STREAM ANALYSIS AT HOLLYWOOD SITE
June, 1966 - June, 1971
Proctor
No. 2
Proctor
No. 2
Bore Hole
Proctor
No. I
Tyler
Run
Bennetts
Branch
Stream
No. 1
Combinations
No. 2
Average 2.98 3.20 3.34 3.43 3.88 3.26 3.36
High 3.81 3.41 4.40 4.30 5.10 3.71 4.10
Low 2.69 2.85 2.60 2.91 2.80 2.79 2.88
ACIDITY (mg/L Ca00 3 )
Average 3302 2485 489 160 164 479 288
High 8112 4121 1271 420 523 1138 584
Low 478 1718 48 54 15 123 58
IRON II (mg/L)
Average 548 804 83 4 29 84 38
High 2150 1579 199 90 105 290 116
Low 23 509 1]. <1 2 1 5
IRON (TOTAL ) mg/L)
Average 780 809 161 40 52 86 39
High 5728 1579 2538 1269 1340 295 117
Low 23 543 10 1 5 8 10
-------
Table 1 (continued). SUMMARY STREAM ANALYSIS AT HOLLYWOOD SITE
June, 1966 - June, 1971
Proctor
No. 2
Proctor
No. 2
Bore Hole
Proctor
No. 1
Tyler
Run
Bennetts
Branch
Stream Combinations
No. 1 No. 2
ALUMINUM (mg/L)
Average 245 232 87 32 25 100 47
High 861 393 229 131 87 528 115
Low 78 184 13 1 0 21 10
CALCIUM (mg/L)
Average 201 96 59 36
High 368 144 108 56
Low 60 40 20 20
MAGNESIUM (mg/L)
Average 97 76 64 40
High 145 132 114 46
Low 3 10 28 33
MANGANESE (mg/L)
Average 8.2 9.5 2.6 3.0 3.5 4.0 2.9
High 9.0 12.0 3.5 6.9 9.4 7.3 5.7
Low 6.8 6.7 1.3 1.2 1.2 1.8 1.3
SULFATE (SOTf 2 ) (ing/L)
Average 2587 3295 809 301 260 1156 424
High 7491 5276 1188 712 747 2018 1009
Low 815 2650 107 118 48 229 101
-------
Table 1 (continued). SUMMARY STREAM ANALYSIS AT HOLLYWOOD SITE
June, 1966 - June, 1971
Proctor
Proctor
No. 2
No. 2
Bore Hole
Proctor
No. 1
Tyler
Run
Bennetts
Branch
Stream
No. 1
Combinations
No. 2
CONDUCTIVITY
(MICROMHOS)
Average
2666
3311
1223
602
519
1150
704
High
3090
5420
1620
1070
1335
2320
1550
Low
2348
2580
194
270
132
5
5
DENSITY
(20°C)
Average
1.00505
1.0058
1.00305
1.0021
1.0027
1.0031
1.0032
High
1.0060
1.0080
1.0040
1.0031
1.0041
1.0050
1.0050
Low
1.0041
1.0045
1.0020
1.0000
1.0010
1.0010
1.0012
t -)
-------
Table 2. DETAILED STREAM ANALYSES AT HOLLYWOOD SITE
COLLECTION DATE JANUARY 11, 1968
Proctor
Constituent No. 2
Proctor
No. 2
Bore Hole
Proctor
No. 1
Tyler
Run
Bennetts
Branch
pH 3.06 3.40 3.50 3.58 3.80
Acidity, mg/i CaCO 478 2810 469 124 210
Dissolved solids, mg/i 1324 5892 1342 429 545
Fe (total), mg/i 23 911 84 <1 49
Fe (II), mg/i 23 911 83 <1 49
Al, mg/i 78 229 91 33 36
Ca, mg/lb 185 -
Mg, mg/i - 122 - -
S0 , mg/i 815 3457 827 266 322
Na, mg/la 49 34 19 4 7
K, mg/la 0.8 7 2 0.7 1
Mn, mg/i 10.3 10.6 2.1 1.5 3.3
Ni, mg/lb 0.7 2.9 0.6 0.1 0.2
V, mg/lb 0.009 0.2 0.005 0.002 0.02
Cu, mg/lb 0.1 0.3 0.04 0.03 0.2
Mo, mg/lb ND ND ND ND ND
Zn, mg/lb 5.9
Si, mg/lb 55
Sr, mg/lb 3.4
Ba, mg/lb 0.6
Ti, mg/lb 2.0
Hg, ppjjC <50
Total dissolved solids 978 5032 1026 308 419
defined, mg/i
-------
Table 2 (continued). DETAILED STREAM ANALYSES AT HOLLYWOOD SITE
COLLECTION DATE JANUARY 11, 1968
Proctor
Constituent No. 2
Proctor
No. 2
Bore Hole
Proctor
No. 1
Tyler
Run
Bennetts
Branch
aFI e photometer, on raw water.
bEmission spectroscopy from evaporated residue (105°C).
CAtomic absorption after concentration on gold, N.H. Suhr, Pennsylvania State University,
University Park, Pa.
t )
U
-------
Figure 1. Seasonal variation in water quality at Hollywood Facility -
Proctor No. 2
PROCTOR NO. 2
TOTAL Fe
SULFATES (SO 4 )
/
/
2 O 4S O
ieo
1600 4000
1400
1200
i 00
000
600
400
204
a
U-
3200
2800
2400
2000
1600
1200
o . ., 0 0 Q 0
-------
Figure I (continued).
Seasonal variation in water quality at
Hollywood Facility - Proctor No. 2
S S
h i
a
4
PROCTOR NO. 2
ALUMINUM
ACIDITY
t e
ax
1W
2W
0
-------
Figure 2.
Seasonal variation in water quality at Hollywood Facility
Proctor No. 2 bore hole
PROCTOR NO. 2, BOREHOLE
TOTAl. Fe
SULFATES (504)
A
I\
I
(I
I
I
4 .
Th 0 0
1400 40W
1700 3800
1000 3400
4003400
600 3200
4003800
200 2000
0
U-
U,
\/\!
I 1 I I I I 1 I I T T 11 T I I I I I I I I I I r I r r-1 i I I I I I I I I I I I I I I
-------
Figure 2 (continued). Seasonal variation in water quality at
Hollywood Facility - Proctor No. 2 bore hole
PROCTOR NO. 2, BOREHOLE
ALUMINUM
ACIDITY
I\
I
\ )
400
250
.O
/
200
150
Sc 1000
C P r .. .o ia. ,e .c.
C
-------
Figure 3.
Seasonal variation in water quality at Hollywood Facility -
Proctor No. 1
(111$)
A
/\i
310 1100
110 1000
110
140 000
131100
PROCTOR NO. 1
(
(
I
I
I
I
I
I
I
II
II
II
/\
I \
TOTAL Fe
SULFATES (SOs)
t
A
00
500
(
I
40
V
0
Lf)
( 4
100
0
I .
.,
-------
Figure 3 (continued). Seasonal variation in water quality at
Hollywood Facility - Proctor No. .
U N
240 UN
1*
II
p. .)
Ill
PROCTOR NO. 1
ALUMINUM
ACIDiTY
N
0 0
.4
-------
Figure 4.
Seasonal variation in water quality at Hollywood Facility -
Bennetts Branch
700
/
/
500
BENNETTS BRANCH
TOTAL Fe
SULFATES (SOs)
1 \
I
I
I I
I
I
k
100
I0
60 0
40200
20 100
0
U-
I-
A
I
(N
LI
(N
I
/
U., b 9 f, , . f ) D 2
-------
Figure 4 (continued) -
Seasonal variation
Hollywood Facility
in water quality at
- Bennetts Branch
BENNETTS BRANCH
ALUMINUM
ACIDITY
is
N
61
tn
0
_ =
1 1 .i 2
-------
Figure 5. Seasonal variation in water quality at Hollywood Facility -
Tyler Run
/\
\
U
700
\
600
500
TYLER RUN
16
TOTAL Fe
I
A
SULFATES (504)
12
/
/
I
I
I
I
I
I
I
I
I
200
4 100
\I
/
\/
0
0
L I )
a
I
-------
Figure 5 (continued).
Seasonal variation in water quality of
Hollywood Facility - Tyler Run
TYLER RUN
ALUMINUM
ACIDITY
,1
p )
(11
(J1
120000
1*0
I. S
*0 300
*
I
30
100
00
$3
i..
-------
APPENDIX B
PLANT DESIGN
Table 1. UNITS AND FACILITIES FOR COLLECTING AND
TREATING MINE DRAINAGE DISCHARGES - PRE-
LIMINARY BASIS OF DESIGN
Major Collection and Treatment Units
Three (3) pumping stations
Holding lagoon
Flash mixer
Oxidation unit
Rock filter
Densator unit
Settling lagoon
Drum sludge vacuum filter
Limestone neutralization unit
Lime, sodium hydroxide and sodium carbonate feed units
Polyelectrolyte feed unit
Sludge drying lagoon
Control building
Garage and storage unit
Control Building
Construction
Reinforced concrete sub-structure, concrete block walls,
steel frame, concrete plank roof, and rear loading platform
Units and facilities
Electrical controls
Pipe gallery
Rate-of-flow control
Flow, pH, temperature, iron, turbidity and DO indicators
and recorders
Flash mixer
Two (2) blowers and appurtenant equipment
(1) Based on information furnished by the University; subject
to change as additional information is made available
Densator unit influent suinp and pumps, sludge pumps, and
appurtenant equipment
Drum sludge vacuum filter, cake conveyor, coal storage and
feed facilities, conditioning tank, and appurtenant equipment
256
-------
Table 1 (continued) UNITS AND PACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Hydrated lime storage bin) hydrated lime feed facilities, and
apurtenant equipment
Quicklime storage bin, slaker and other feed facilities, and
appurtenant equipment
Sodium hydroxide storage and feed facilities, and appurtenant
equipment
Sodium carbonate storage and feed facilities, and appurtenant
equipment
Polyelectrolyte storage and feed facilities, and appurtenant
equipment
Plant water pumps and piping
Office
Laboratory
Workshop
Conference room
Kitchen
Sanitary facilities, and septic tank and tile field
Sleeping quarters
Possible Future Construction
Pre-coat vacuum filter
Recirculate rock filter and/or Densator unit effluent to rock
filter influent line
Rock filter cover
Water storage tank
Proctor No. 1 Pumping Station
Pump suction downstream from confluence of 2 mine drainage dis-
charges originating on north side of U .S. Route 255
Point of withdrawal from stream Concrete sump set in
bed of stream, shape
stream channel as
needed
Sump capacity 1 )000 gallons
Type of pump housing structure Below grade, concrete,
acid resistant, locate
top above 1200
elevation
257
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Pump
Type Centrifugal, constant
speed with by-pass for
varying delivery
Capacity 350 gpm
Construction Acid resistant
Suction Flooded
Discharge Control building pipe
gallery and holding
lagoon
Piping PVC
Miscellaneous
Flow indicator; automatic pump stop; dry well suinp and pump; dry
well electric unit heater and ventilation; electrical controls
(duplicate on-off control in control building)
Proctor No. 2 Pumping Station
Pump suction from Proctor No. 2 and Tyler Run
Point of withdrawal from streams Concrete sumps set in
bed of both streams,
shape stream channels
as needed
Surnp capacity - each 1,000 gallons
Type of pump housing structure Below grade, concrete,
acid resistant, locate
top above 1205 elevation
Pumps
Number 2 (1 for Proctor No. 2
and 1 for Tyler Run)
Type Centrifugal, constant
speed with by-pass for
varying delivery
Capacity - each 350 gpin
Construction Acid resistant
Suction Flooded
Discharge Control building pipe
gallery and holding
lagoon
258
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Piping PVC
Miscellaneous
Flow indicators; automatic pump stops; dry well sump and pump;
dry well electric unit heater and ventilation; electrical con-
trols (duplicate on-off controls in control building)
Bennett Branch Pumping Station
Pump suction from Bennett Branch
Point of withdrawal from stream Portion of Bennett
Branch diverted from
main channel to con-
crete sump located by
shoreline, shape
stream channel as
needed
Sunip capacity 1,000 gallons
Type of structure Below grade, concrete,
acid resistant, locate
top above Bennett
Branch flood level
Pump
Type Centrifugal, constant
speed with by-pass
for varying delivery
Capacity 350 gpm
Construction Acid resistant
Suction Flooded
Discharge Control building pipe
gallery and holding
lagoon
Piping vc
Miscellaneous
Flow indicator; automatic pump stop; dry well. sump and pump;
dry well electric unit heater and ventilation; electrical con-
trols (duplicate on-off controls in control building)
259
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Holding Lagoon
Capacity 258,000 gallons
Dimensions
Diameter 112 feet
Center depth (liquid level to bottom) 5.6 feet
Freeboard 2 feet
Side slopes 4:1
Construction Earth and beritonite
mixed, and placed in
compacted layers
Inlet 1-leader to receive
mine drainage from
piping gallery and
discharge tangentially
at periphery; mechanical
liner at discharge point
Outlet Located at bottom
center of lagoon
protective screen
Piping PVC
Miscellaneous
Lagoon by-pass and overflow; piping arranged to flush outlet
protective screen; rate-of-flow control on discharge line;
influent and discharge to be metered (indicatorrecorder--
totalizer); continuous pH, temperature, iron and turbidity
recording; drain to Tyler Run; effluent piping to flash mixer,
Densator unit influent su!np, and biological treatment units
Flash Mixer
Dimensions (L x W x SWD) 12 x 3 x 6.5 feet
Volume 1,750 gallons
Location of inlets Bottom of unit
Location of outlets Top of unit
Construction Steel
2 6 0
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Mixers 2; portable so that
position within unit
can be varied
Piping PVC
Miscellaneous
Unit to be constructed so that detention time can be varied by
inserting or removing 2 baffles located 4 feet from each end;
provision for overflow from each compartment; bottom drain;
unit to receive mine drainage from holding lagoon or directly
from pipe gallery; effluent piping to oxidation unit and
Densator unit influent sump
Oxidation Unit
Dimensions (L x W x SWD) 15 x 10 x 10 feet
Volume 11)200 gallons
Location inlets Bottom of unit
Location of outlets Top of unit
Construction Steel, acid resistant
Turbine and surface aerators
Oxygen transfer capabilities 1 - 30 lbs. 0 2 /hr.
Blowers
Number 2
1 at 250 cfm
1 at 500 cfni
Air headers PVC
Piping PVC
Miscellaneous
Unit to be constructed so that detention time can be varied by
inserting or removing baffle located at mid-point; influent
and piping to allow for use of each compartment; air to be sup-
plied by diffusors, turbine aerators, and surface aerators;
unit to receive mine drainage from pipe gallery, holding lagoon
and flash mixer; effluent piping to Densator unit influent sump
and settling lagoon; flow indicator on air lines; bottom drain;
continuous pH and DO recording in unit (location of probes to
be varied)
z 1
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Rock Filter
Dimensions (Diam. SRD) 35 x 5 feet
3
Volume 4,800 ft.
Media
Type Silica rock
Size 1.5-2 inches
Piping PVC
Rotary distributor Acid resistant
Construction Concrete, acid resis-
tant
Miscellaneous
Unit to receive mine drainage from pipe gallery and holding
lagoon; effluent piping to Densator unit influent sump; make
provision for future recirculation of rock filter effluent
to influent line
Densator Unit
Dimensions
Overall (Diain x SWD) 25 x 14 feet
Primary reaction zone (Diam. x SWD) 12 2.25 feet
Secondary reaction zone (Diam. x SWD) 12 x 8.50 feet
Clarification zone
Surface area 378 ft. 2
SWD 10.5 feet
Volumes
Primary reaction zone 1,900 gallons
Secondary reaction zone 7,200 gallons
Clarification zone 29,400 gallons
Construction Steel, acid resistant
tnfluent pumps and sump
Sump
Capacity 1,000 gallons
Construction Concrete, acid resis-
tant
Overflow Tyler Run
262
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Pumps
Number 2 (1 on stand-by)
Type Centrifugal, manual
variable delivery
Capacity (each) 35-450 gpm
Suction Flooded
Construction Acid resistant
Sludge pumps
Number 2 (1 on stand-by)
Type Positive displacement,
manual variable
delivery
Capacity (each) 10-125 gpm
Suction Directly from bottom
of unit; pumps located
at outside bottom of
unit
Construction Acid resistant
Discharge Vacuum filter, tank
truck, sludge drying
lagoon, oxidation unit
influent, recycle into
Densator unit or
influent sump, settling
lagoon
Piping PVC
Miscellaneous
Unit to receive mine drainage from pipe gallery, holding laggon,
rock filter and oxidation unit; effluent piping to plant water
sunip and Tyler Run, limestone neutralization unit, and flash
mixer; flow indicators on influent and sludge lines; sample
lines into Densator unit; bottom drain; scraping mechanism
modified for thickening; continuous pH recording in Densator
unit; make provision for future recirculation of overflow to
rock filter influent line
263
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Settling Lagoon
Construction Earth placed in corn-
pactecl layers, mechanical
liner on bottom of floccu-
lation and settling corn-
partrnent
Freeboard 2 feet
Side slopes 2:1
Flocculation and settling compartment
First section
Volume 156,000 gallons
Dimensions (L x W x SWD) 86 x 52 x 8 feet
Bottom slope 3% toward influent
end, 6% toward
sludge withdrawal
line
Inlet Mechanical liner at
discharge point
Second section
Volume 63,000 gallons
Dimensions (L x W X SWD) 30 x 52 x 8 feet
Bottom slope 6% toward influent
end, 6% toward
sludge withdrawal
line
Polishing compartment
Volume 204,000 gallons
Dimensions (L x W x SWD) 70 x 70 x 8 feet
Sludge pump
Type Positive displacement,
manual variable
delivery
Capacity (each) 10-150 gpm
Construction Standard
Suction Flooded
Discharge Vacuum filter, tank
truck, sludge drying
lagoon, oxidation unit
influent
Type of structure Concrete, acid resis-
tant on outside
264
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Piping PVC
Miscellaneous
Sump and pump, electric unit heater, ventilation and electrical
controls (duplicate on-off controls in control building) in pump
structure; bottom drain to Tyler Run; discharge to be metered
(indicator-recorder-totalizer); locate top above 1205 elevation;
unit to receive mine drainage from oxidation unit, limestone
neutralizer, and flash mixer; effluent piping to plant water
sump and Tyler Run; flow indicators on sludge discharge lines
Drum Sludge Vacuum Filter
Filter
Area 125 ft. 2
Dimensions (Diam. x L) 6 x 7 feet
Construction Acid resistant
Type of drum discharge Roll
Coal storage and feed
Mesh -60 to +80
Storage hopper
Capacity 1,500 lbs.
Miscellaneous
Hand loading from ground floor with 100 pound bags of pre-
ground coal fines; level indicators; dust collector, vi-
brators; storage for 50-100 pound bags in control building,
additional storage in garage
Feeder
Feed rate
Maximum 500 lbs/hr
Minimum 20 lbs/hr
Type Dry volumetric feed
Control Manual variable
delivery
Slurry mix tank and pumps
Tank
Capacity 100 gallons
Mixer rapid
Pumps
Number 2 (1 on standyby)
265
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Type Positive displacement,
manual variable de-
livery
Capacity (each) 2-20 gpm
Suction Flooded
Construction Acid resistant
Discharge Conditioning tank
Conditioning tank
Volume 200 gallons
Construction Steel, acid resistant
Mixer Slow
Discharge Filter vat
Miscellaneous
Conditioning tank to receive underf low from Densator unit and
settling lagoon, polyelectrolyte slurry pump, and coal slurry
pump
Piping PVC
Miscellaneous
Filter wash and slurry water from plant water sump, or well; cake
conveyor to dump truck for hauling to local strip pits or refuse
piles; filtrate discharge to Tyler Run; float in filter vat to con-
trol start-stop position for sludge pump feeding conditioning
tank, and coal slurry pump; float in slurry mix tank to control
start-stop position of feeder and slurry water feed; flow indi-
cators on slurry water feed and pump discharge; make provision
for future pre-coat operation of filter
Limestone Neutralization Unit
Reactor
Dimensions (Diam. x L) S x 20 feet
Degrees from horizontal 8
Limestone
Size 0.25 inch
Volume ioo ft. 3
Construction Steel, acid resistant
Miscellaneous
Limestone added to upper end; mechanical rotation at 10 rpm for
2 minutes every 1/4 hour; reactor to receive mine drainage from
Densator unit; fixed overflow pipe at lower end; provision for
removal of spent limestone
266
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Reactor sump
Dimensions (L x W x SWD) 30 x 10 x 15 feet
Volume 11,200 gallons
Construction Steel
Miscellaneous
Bottom drain; effluent piping to settling lagoon
Storage hopper
Capacity i,ooo lbs.
Miscellaneous
Loaded by bucket on hi-lift vehicle; level indicators
Feeder
Feed rate
Maximum 80 lbs/hr
Minimum 2 lbs/hr
Type Dry volumetric feed
Control Manual variable de-
livery
Piping PVC
Miscellaneous
Storage hopper, feeder and upper end of reactor to be enclosed
Hydrated Lime Storage and Feed
Type of hydrated lime High calcium and dolo-
mitic to be fed
separately
Storage bin
Capacity 25 tons
Miscellaneous
Tank truck delivery (17-20 tons); blower hose; screw discharge
to lime hopper; level indicators, dust collector; circulating
fan; vibrators
Lime Feeder
Feed rate
Maximum 500 lbs/hr
Minimum 20 lbs/hr
Type Dry volumetric feed
Control Manual-automatic
variable delivery
267
-------
Table I (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Miscel laneous
Level indicators and vibrators on lime hopper; level indicators
to control operation of storage bin feed gate; feeder discharge
to be paced by pH measurement in oxidation and Densator units
Lime slurry
Mix tank
Capacity 60 gallons
Construction Steel
Miscellaneous
Float control tied in with slurry water feed; lime feeder,
and lime slurry pumpl flash mixer
Pump
Type Positive displacement,
manual variable drive
Capacity 15 gpm
Construction Caustic and acid
resistant
Piping Flexible hose, arrange
for flushing with raw
mine drainage
Discharge Flash mixer, Densator
unit primary reaction
zone and influent sump
Miscellaneous
Slurry water from treated mine drainage sump, or well; flow
indicator on slurry water and lime slurry lines
Quicklime Storage and Feed
Type of quicklime High calcium and dolo-
mitic pebble, to be
fed separately
Storage bin
Capacity 25 tons
Miscellaneous
Tank truck delivery (17-20); blower hose; screw discharge to
lime hopper; level indicators; dust collectors; circulating
fan; vibrators
268
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Lime feeder
Feed rate
Maximum 350 lbs/hr
Minimum 20 lbs/hr
Type Dry volumetric feed
by scres
Control Manual-automatic
variable delivery
Miscellaneous
Feeder discharge to be paced by pH measurement in oxidation
and Densator units
Lime slaker
Capacity 300 lbs/hr
Mix ratio 4 water:), lime,
by weight
Grit removal Manual
Pump
Type Positive displacement,
manual variable drive
Capacity 15 gpm
Construction Caustic and acid
resistant
Piping Flexible hose, arrange
for flushing with raw
mine drainage
Discharge Flash mixer, Densator
unit primary reaction
zone and influent sump
Miscellaneous
Dust collector and vapor condenser; mixers; slurry water from
treated mine drainage sunip, or well; flow indicator on slurry
water and lime slurry lines; float control in slaker tied in
with slurry water feed, quicklime feeder, and quicklime slurry
pump
Sodium Hydroxide Storage and Feed
Storage for 20-50 gallon drums incontrol building, additional
storage in garage
269
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Pump
Type Positive displacement,
manual variable drive
Suction Sodium hydroxide drum
Capacity 0.2 to 1.0 gpm of 50%
NaOH , specific gravity
of 1.54
Construction Caustic Resistant
Piping PVC
Discharge Flash mixer, Densator
unit primary reaction
zone and influent sump
Sodium Carbonate Storage and Feed
Storage hopper
Capacity 1,500 lbs.
Miscellaneous
Hand loading from ground floow with 100 pound bags; level
indicators; dust collector, vibrators, storage for 50-100
pounds bags in control building, additional storage in
garage
Feeder
Feed rate
Maximum 50 lbs/hr
Minimum 20 lbs/hr
Type Dry volumetric feed
Control Manual variable delivery
Sodium carbonate slurry
Mix tank
Capacity 60 gallons
Construction Steel
Miscellaneous
Float control tied in with slurry water feed, sodium carbonate
feeder, and sodium carbonate slurry pump; flash mixer
Pump
Type Centrifugal, manual
variable drive
Capacity 15 gpm
Construction Caustic and acid resistant
27Q
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIM1NAj y
BASIS OF DESIGN
Piping PVC
Discharge Flash mixer, Densator
unit primary reaction
zone and influent sump
Miscellaneous
Slurry water from treated mine drainage sump, or well; flow
indicator on slurry water and sodium carbonate slurry lines
Polyelectrolyte Storage and Feed
Storage for 6-100 pound bags in control building
Polyelectrolyte slurry
Mix tank and flash mixer
Volume 100 gallons
Construction Steel
Slurry pump
Type Positive displacement,
manual variable drive
Suction Flooded
Capacity 0.06 to 2 gprn of 3%
polyelectrolyte slurry
Construction Standard
Piping PVC
Discharge Mixing tee
Dilution water
Volume 5 gpm
Miscellaneous
Supply from plant water pump, or well; discharge to settling
lagoon, Densator unit primary reaction zone and influent sump,
and sludge conditioning tank
Plant Water Pumps and Sump
Sump
Capacity 1,000 gallons
Constructjo Concrete
Miscellaneous
Sump to receive discharge from settling lagoon and Densator unit;
overflow to Tyler Run; continuous pH, temperature, iron and
turbidity recording; rateof-flow control on holding lagoon
discharge line to close at pre-set pH level
271
-------
Table 1 (continued). UNITS AND FACILITIES FOR
COLLECTING AND TREATING MINE
DRAINAGE DISCHARGES - PRELIMINARY
BASIS OF DESIGN
Pumps
Number 2 (1 on stand..byr)
Type Centrifugal, Constant
speed with by-pass
Suction Flooded
Capacity 70 gpm
Construction Standard
Piping Cast iron
Discharge Control building plant
water piping
Sludge Drying Lagoon
Construction Earth placed in com-
pacted layers, mech
anical lines at in-
lets, drainage pro-
vided by 1 foot of
sand and 3 feet of
culm underlying lagoon
Freeboard 2 feet
Side slopes 2:1
Compartnient s
Number 3
Dimensions (L x W x SWD) 50 x 25 x 3 feet
Volume/compartment 34,000 gallons
Supernatant overflow Tyler Run
Piping PVC
Miscellaneous
Locate top of sand above 1205 elevation; provide access road
around lagoon; cover 1 compartment (wooden trusses and plastic
covering - opening on 4 sides); lagoon to receive sludge from
settling lagoon and Densator unit, and filtrate from vacuum
filter; drainage channel to Tyler Run for underflow
272
-------
Table 2. SPECIFICATIONS FOR SURFACE AERATORSa
(Lightnin Model LAR-60)
impeller Diameter 59 (4-blade - 316 S.S.)
Impeller Speed 19-56 RPM (variable)
Impeller Motor Drive 7-1/2 H.P. - 1750/583 RPM
Oxygen Transfer Rate 30 lbs 02 @ STP/hr
4.0 lbs 0 2 /hr/H.P.
Pumping Capacity 20,000 GPM @ max. speed
Impeller Elevation Adjustable 6 1 1 in 1 increments
from normal position
aThe specification oxygen transfer rate and liquid pumping capacity
are valid for the impeller position as shown in Figure 13 where
the top of the impeller blade is at the liquid level and with the
indicated tank geometry. If the impeiler immersion is increased,
the pumping rate would increase but the oxygen transfer rate would
decrease. If the impeller is raised, both the pumping capacity and
the oxygen transfer rate would decrease. Since each oxidation tank
has a capacity of approximately 10,000 gallons, the nominal pumping
rate of the aerators in normal position produced a complete tank
turn-over every one-half minute.
273
-------
Table 3. POWER CONSUMPTION BY VARIOUS UNIT OPERATIONS
Designation
Power
rating,
hp
Current
rating
amperes
Voltage
Measured
power conswlption
KWHr/Hr
05 a
(cents/hr.)
high
low
Pumps:
average
average
high
Low
Water Collection
Proctor No. 1 15 19.0 440 3.98 9.67 0.75 6.8 16.4 1.3
Proctor No. 2 25 31.0 440 6.59 17.68 0.94 11.2 30.1 1.6
Tyler Run 25 31.0 440 6.68 9.79 1.58 11.4 16.6 2.7
Bennetts Branch 30 36.6 440 6.09 17.60 0.12 10.4 29.9 0.2
Process
Vacuseal Pump 10 13.5 440
Densator Feed 10 16.5 440 1.16 2.93 0.11 2.0 5.0 0.2
Lime Slurry Circulation 10 13.5 440 3.60 8.73 0.17 6.1 14.8 0.3
Precoat Feed 2 4.2 440 0.29 0.33 0.25 0.5 0.6 0.4 .
Filter Feed 2 3.6 440 0.20 0.60 0.01 0.3 1.0 0.02
Process Water (2 motors)b 7.5 9.8 440 2.46 5.23 0.13 4.2 8.9 0.2
Compressor - Housewater 0.75 1.55 440 0.10* 0.2
Compressor - Autovalve 1.0 13.8 115 1.00* 1.7
Sample Pump No. 1 0.75 9.0 115 0.70* 0.1
Sample Pump No. 2 0.75 9.0 115 0.70* 0.1
Drain Sump 0.33 1.5 440 0.50* 0.1
Sludge Sump Drain 0.33 5.4 115 0.40 0.1
Chlorinator -- 1.2 115 0.01* 0.01
Unit Operations:
Densator
Mixers 1.5 3.3 440 0.38 0.57 0.10 0.6 1.0 0.2
Sludge Scraper 0.5 1.4 440 0.16 0.33 0.03 0.3 0.6 0.05
Recirculation 1.5 3.3 440 0.34 2.40 0.02 0.6 4.1 0.03
Total 3.5 8.0 0.88 3.30 0.15 1.5 5.6 0.3
-------
Table 3 (continued). POWER CONSUMPTION BY VARIOUS UNIT OPERATIONS
Power
rating,
Designation hp
Current Measured
rating
amperes Voltage average
power consumption
KWHr/Hr
05 a
(cents/hr.)
average
high low
high
low
Vacuum Filter
Vacuum Pump 30 36.5 440 10.70 17.40 1.88 18.2 29.6 3.2
Agitator 1.5 2.4 440 0.22 0.40 0.07 0.4 0.7 0.1
Knife Retractor 1 1.8 440
Drum Drive 1 2.8 440 0.22 0.40 0.10 0.4 0.7 0.2
Filtrate Pump 3 4.6 440 1.07 1.40 0.32 1.8 2.4 0.5
Sludge Conveyor 2 3.6 440 0.19 0.42 0.08 0.3 0.7 0.1
Conditioning Tank Mixer 0.33 4.0 115
Total 38.83 55.7 17.07 27.70 3.29 29.0 47.1 5.6
Chemical Oxidation
.J Surface Aerator No. 1 7.5 10.5 440 1.82 11.11 0.42 3.1 18.9 0.7
Surface Aerator No. 2 7.5 10.5 440 1.36 4.40 0.13 2.3 7.5 0.2
Reac t or s
Limestone Rotary Reactor 20.0 25.6 440 7.49 20.20 0.10 12.7 34.3 0.2
Limestone Reactor Feeder -- 5.0 220 0.8* 1.4
Flash Mixer Agitator No. 1 0.33 5.4 115 0.4* 0.5
Flash Mixer Agitator No. 2 0.33 5.4 115 0.4* 0.5
Reagent Control
Hydrated Lime
Line Conveyor No. 1 1 2.5 440 0.05 0.07 0.04 0.09 0.1 0.07
Lime Conveyor No. 2 1 1.8 440 0.04 0.05 0.04 0.07 0.09 0.07
Mixer 0.5 1.1 440 0.17* 0.09
Helix 0.25 115 0.30* 0.5
Air Blower 5 7.1 440 0.05 0.07 0.04 0.09 0.1 0.07
Vibrator -- 2.0 115 0.16* 0.3
Recirculation Fan 1 1.82 440 0.41* 0.7
Total 8.75 16.3 1.23* 0.9
-------
Table 3 (continued). POWER CONSUMPTION BY VARIOUS UNIT OPERATIONS
aBased on 1.7 per KWHr. Actual rate schedule varies with demand and consumption.
bRatings are for the individual motor, consuinptions for the system.
*Estimated
Designation
Power
rating,
hp
Current
rating
Voltage
Measured
power consumption
KWI-Ir/Hr
Costa
(cents/hr.)
Pebble Lime
average
high
low
average
high
low
Vibrator Conveyor - 2
L7 5
115
0.13*
Lime Conveyor No. 3
1
1.8
440
0.03
0.03
0.02
Mixer
l. 5
2.3
440
0.51
0.71
0.03
0.05
0.03
Helix
0.25
115
0.30*
0.13
0. 9
1.2
0.2
Grit Removal
0.25
5.2
115
0.40*
0.5
Recirculation Fan
1
1.82
440
0.56*
0.7
Total
4.0
12.87
1.93*
1.0
33
Liquid Reagents
0.33
5.6
115
0.43*
0.7
Distribution Pump
Sludge Control
Transfer Pump
3
5.1
440
0.15
0.29
0.04
0.3
(Settling Lagoon)
0.07
Sludge Sump Mixer
0.5
5.5
220
.85*
1.4
Precoat Feeder
0.25
4.4
115
0.34*
Precoat Agitator
0.25
3.2
220
0.48*
0.6
Precoat Vibrator
--
2.0
115
0.15*
0.3
Flocculant Preparation
0.33
5.6
115
0.43*
Mixer
Distribution Pump
0.5
1.1
440
0.34*
0.7
.O
-------
Table 4. DI?4ENS IONS OF SETTLING LAGOON COMPARThIENTS
4
Top
Volume
Total
Dimensions
Area
Sludge
Storage
Total
Cujrulative
ft.
sq. ft.
Cu. ft.
gallons
cu. ft. gallons
gallons
Ini 1uent Section
52W x 65L
3,380
5,400
140,395
19,240
Middle SectIon
52W x SQL
2.600
5,000
37,403
16,500
Effluent Section
70W x 54L
3,780
23,516
Pond Depth - 5 ft.
minimum - 8 ft.
maximum (at
center
of
lagoon)
Average Depth approx. 6 ft.
143,925
123,429
175 ,927
143 ,925
267 ,354
4 3,281
-------
2
cc
00
C
, -1
- - I
0
a
C
0
0
4 -4
0
-c
0.
CU
)-1
t a O
0
1 - i
0
C
04
I
a )
: 1
00
-
-------
_ ;- η T
Figure 2.
Application of Bentonite Clay to Holding Lagoon
___ _________ as-
_ _ _ a
S
-
-
S .
0 . ,p
. A
__ _#j* $ _ e
Figure 3.
Filling Holding Lagoon
279
it :
r
It$
I
L.
-------
.
I
-.
1
J::a -
p
0
03
0
r-4
4J
C u
0)
cL
0
4J
..- C
3 - 1
a)
1
>4
a 3
at
I ;4 [ jI
4-
a)
3- 4
aD
1-4
ηx4
/
7 / - ,
-------
J.
F:
Cfi
rt
OQ
OQ
0
0
Ii
-------
-4
r1
a
4J
a)
U
a)
:3
r4
4 - 4
4-1
bO
0
I a
0
4 1
C .)
(U
a)
a)
1
0
4 1
U)
a)
E
- I
0
a)
$4
:3
be
-4
C l
k s
I , }
Lj
A
- 4c
-------
Pzj
I -
Pt
CD
F - 4
pJ.
S
U)
rt
0
S
S
S
n
0
P1
(n
c 0
L I) :
OQ
FTI
CD
rt
CD
ft
S
Pt
I-
CD
CD
r)
S
)-J.
U)
P.
, : - , ...- -_
i v
4tt#c
-------
__
- A
4 ,
E
- v .4
0
.r4
-I
p 4
-4
p4
-4
U)
r4
-4
00
a)
-4
bO
1 . 4
-------
Figure 9. Internal piping and unit operations
j k ;
. .
00
U i
-
ci
1-
> ---
r t
: ri-
*
;__--
(
. i
I
-
- -
r
-------
Figure 10. Sump designs
-
± ±t i
Of nu l ls LAGOON OFFUJGN1 STIUCTOOO
I
A-t- J,J-
A t rAti*i.
. 7
k άG PtWI& $!AfifU
k M. I I It
00
N
PWOCTOG M C I PIING ITOTION
ka-
-------
tigure ii. ettiing iagoon aesign
/
/
T !
±
00
L
-------
Figure 12. Sludge drying basin design
kM N
- -H )
pu
ii
IOJ U$0N OtTALI
j
-
1LcT I
288
-------
Figure 13. Oxidation tanks with surface aerator
I ?
I SO
330
1
PVC IPIPLUINT
PVC INFLUENT
SAFF LI 0
I
OUPIP AC S
SRATOI
ON P EL0CIR I. L rooupo.
-;7 OUSPACI
M
1 J
i
t V
pv IPFLUIU4T
CONCHII (NCAOEMLNT
289
PVC 05001
-------
Figure 14. Infilco densator design
EL. 1 .17
OVERFLOW
RMUIIG TOP OF BEAM
DOOR
Li -
12 12
CONC.
RETURN TO CONTROL BLDG. 12 I 1.6.. I 16 1R I i I 12
I t. EL. 129310 -r t -- -t-
I 80
46 . J2 1 ID. OF CONC. TANk 2311 -
1
41!2W1 1
-------
Figure 15. Surface biochemical oxidation reactor
FIN
LL. 1JJ.5U
EL.
to
TOP OF STONE
EL. 1213.00
EL. 1205.59
BOTT. OF
EL. 1206.34
ALUM. GRATING
BOTT. OF CHANNEL
EL. 1206.84
-------
Figure 16. Schematic of Limestone Reactor
20 Ft.
2MM
WEDGE WI R
SCREEN
RAW WATER
EFFLUENT SUMP
MANHOLE
VIBRATIONS FEEDER
1 In. OPENINGS SCREEN
2 In, OPEN
3 In.
S SCREEN
NINGS SCREEN
GROUND LEVEL
N
C )
N
DRAIN PLUG
-------
APPENDIX C
DETAILED PLANT DATA
Table 1. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME
PROCESS CONDITIONS (PROCESS 10)
Water Source - (- χ
Flash Mixer + Oxidation
Tank #2
+
C
+
Air
Settling
Effluent
to Stream
Lagoon
+
Sludge
H 2 0 +
Ca (OH) 2
Lime Slurry Sump
+
Dissolver
+
Lime Feeder
+
Conveyor
+
+ Storage Bin
Average
Run
Water
flow,
duration
treated
Test
Water source gpm
hrs
gal
No.
Tyler Run
182
47.5
520,912
10-70
247
67.5
1,001,060
9-70
Bennetts Branch
153
240
317
48.0
48.0
90.0
441,400
693,600
1,716,650
3-70
2-70
1-70
Proctor No. 1
Proctor No. 1 and
112
182
251
298
342
18.0
24.0
60.0
48.0
48.0
120,636
263,200
904,400
859,570
984,960
8-70
4-70
5 7O
6-70
7-70
Bennetts Branch
(290+52)
Process Flow:
293
-------
Table 2. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
RAW WATER ANALYSES
Test
Water
No.
Source
970
Tyler
Rur
1070
Tyler
Run
370
Bennetts
Branch
2-70
Bennetts
Branch
170
Bennetts
Branch
870
Proctor
No. 1
470
Proctor
No. 1
570
Proctor
No. 1
670
Proctor
No. 1
770
Proctor No. 1
Bernetts Branch
Temp.°F, Av. 60 57 57 55 51 52 52 51 55 53
Range 5665 5661 5560 5158 4657 5154 5053 4853 5260 5255
pH, Av. 3.96 3.47 4.08 4.05 3.96 3.50 3.32 3.29 3.29 3.28
Range 3.406.98 3.453.49 4.034.12 3.914.33 3.674.73 3.423.54 3.293.42 3.253.33 3.263.31
Acidity, Av. 89 90 74 68 101 446 511 513 599 384
mg/i Range 8993 8991 7080 5776 12139 418474 502519 451580 4501030 196492
Fe II, Av. 1.9 1.8 13 13 23 55 58 60 60 49
mg/i Range 13 12 1115 617 1.734 5456 5759 5863 5862 2666
Fe III, ing/1, Av. >0 >0* 6* 6* >0* >0* >0* 0* 0*
Al, mg/i, Av. 9* 9* 26* 26* 17* 80* 84* 84* 82* 73*
Ca, mg/i, Av. 59* 59* 36* 36* 36* 96* 96* 96* 96* 80*
Mg, mg/i, Av. 64* 64* 4Q* 40* 40* 76* 76* 76* 76* 60*
S0 4 ,mg/i, Av. 180* 180* 179* 179* 179* 730* 730* 730* 730* 600*
Conductivity, Av. 925 932 560 554 772 1186* 1170* 1174* 1182* 1806
micromhos Range 856946 9001000 470630 470640 3201010 17601850
* Values established from monthly analyses.
-------
Table 3. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
970
1070
370
270
170
870
470
570
670
770
Water
Tyler
Tyler
Bennetts
Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor Nol
Source
Run
Run
Branch
Branch
Branch
No. 1
No. 1
No. 1
No. 1
Bennetts
Br.
Influent Fe II
lbs. 15.5 9.1 48.6 70.0 332.7 55.3 126.8 453.8 451.0 435.4
lbs./hr 0.235 0.191 1.018 1.448 3.697 3.072 5.282 7.563 9.396 9.072
Fe II Reacted, lbs 7.3 4.8 39.9 41.8 164.7 51.9 120.0 422.0 427.0 402.6
Minimum Oxidation Rate
mg/I/hr 0.9 1.1 10.8 7.2 11.5 51.9 54.5 55.9 59.6 49.0
1
t Theoretical Reagent Required
Cr1 lbs 585.8 303.3 214.7 310.1 1139.9 353.7 884.3 3050.2 3385.0 2486.6
lbs/hr 8.6 6.5 4.5 6.5 12.7 19.7 36.8 50.8 70.5 51.8
lbs/l,000 gal 0.56 0.59 0.49 0.45 0.66 2.93 3.36 3.37 3.94 2.52
Reagent Utilized
lbs 563 340 182 170 1350 680 986 2820 3020 3100
lbs/hr 8.3 7.2 3.8 3.5 15.0 37.8 41.1 47.0 62.9 64.6
lbs/l,000 gal 0.56 0.65 0.41 0.23 0.79 5.64 3.75 3.12 3.51 3.15
Alkali Consumption Utilized/Theoretical
96.1 110.4 84.8 54.8 118.4 192.3 111.5 92.5 89.2 124.7
Theor. + Eff 1. Alk .
Utilized
107 103 67 156 71 55 82 103 111 85
Retention Time (minutes)
Flash Mixer 6.0 8.1 9.7 6.2 4.7 13.2 8.1 5.9 5.0 4.3
Oxidation Tanks 40.2 54.5 64.8 41.4 31.3 88.6 54.5 39.6 33.3 29.0
1. Based on 93.98% Ca(OH) 2
-------
Table 3 (continued). TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED
LIME - PROCESS AND REAGENT REQUIREMENTS
Test
No.
970
1070
370
270
170
870
470
570
670
770
Water
Tyler
Tyler
Bennetts
Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
No. 1
Source
Run
Run
Branch
Branch
Branch
No.
1
No.
1
No.
1
No. 1
Bennetts
Br.
Theoretical Oxygen Requirement
lbs 2.2 1.3 7.0 10.0 47.7 7.9 18.2 65.0 64.6 62.4
lbs/hr 0.034 0.027 0.146 0.207 0.530 0.440 0.757 1.083 1.346 1.299
lbs/l,000 gal 0.0022 0.0025 0.0159 0.0144 0.0278 0.0655 0.0691 0.0719 0.0752 0.0634
ft 3 (std) 25.2 14.4 78.3 111.5 535.3 88.9 203.9 729.3 725.1 699.9
ft 3 /hr (std) 0.38 0.30 1.64 2.32 5.95 4.94 8.50 12.16 15.11 14.58
Oxygen Utilized as Supplied by Aerators
Maximum
lbs 50.8 15.0 5.7 6.0 23.6 7.4 17.2 60.5 61.2 57.7
lbs/hr 0.770 0.315 0.120 0.125 0.262 0.414 0.716 1.008 1.274 1.201
lbs/l,000 gal 0.0507 0.0288 0.1291 0.0087 0.0137 0.0613 0.0653 0.0669 0.0712 0.0586
Minimum
lbs nil nil nil nil nil nil nil nil nil nil
lbs/hr nil nil nil nil nil nil nil nil nil nil
Oxygen Transfer Rate Based on Retention Time (1
ppm 0 2 /water 4.1 3.2 1.7 0.7 0.9 10.9 7.2 5.3 4.7 3.4
ppm/hr 0 2 /water 6.2 3.5 1.6 1.0 1.7 7.4 7.9 8.0 8.5 7.0
Oxygen Transferred as Required by Iron
4.4 8.6 121.7 165.6 202.2 106.3 105.7 107.4 105.7 108.2
Oxygen Transferred as Aerator Rating
2.6 1.1 0.4 0.4 0.9 1.4 2.4 3.4 4.2 4.0
Sludge Production
al/l,000 gal treated 12.5 15 20 20 35 145 95 112 120
solids content,mg/l 5200 4340 1400 2400 2770 2350 3460 2880 3390
lbs dry solids/l,000
gal treated 0.54 0.54 0.23 0.40 0.81 2.84 2.74 2.70 3.39
-------
Table 4. TREAThENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
OXIDATION TANK AND SETTLING LAGOON
(PROCESS NO. 10)
3.3
2.14.5
ANALYSES
-.4
Test No. 970
1070
370
270
170
870
470
570
670
770
Water Tyler
Source Run
Tyler
Run
Bennetts
Branch
Bennetts
Branch
Bennetts
Branch
Proctor
No. 1
Proctor
No. 1
Proctor
No. 1
Proctor
No. 1
Proctor
Proctor
No.
No.
1 &
2
pH, Av. 6.87
Range 6.607.36
6.70
6.607.10
6.22
5.806.68
6.15
4.506.43
6.24
5576.61
6.48
6.446.50
6.43
5.986.72
6.57
6.146.87
6.19
5.296.75
6.46
6.256.55
Acidity, Av. -
Range
-
+36
+19 +49
+9
+5 +13
--14
20 +55
-
+43
+41 +45
+23
10 +67
+6
15 +25
Alkalinity, Av. -6
Range 15 +3
12
15 6
-
-29
30 27
-24
40 12
FeII,Av. 1
mg/l Range
1
2
13
3
24
4
17
3
16
328
4
26
4
27
4
26
Fe Total, ΐy. --
mg/l Range --
--
--
--
--
--
--
6
--
--
--
--
-
--
--
--
Suspended Solids, Av. --
mg/l Range
--
-
3
23
5
46
12
11-12
--
--
10
5
3.-7
8
2
l3
Temperature, Av. 64
Range 6268
61
6063
--
56
5360
50
4753
-
--
--
-
--
Conductivity, ΐy. 759
Range 668836
948
9001000
580
560594
519
430594
822
504950
1771
16801880
1488
1370-1700
1898
17361990
1958
19002000
1767
15901920
Effluent Rate, ΐy. 285
gpm Range 142410
182
158250
182
180280
282
255288
36].
-
214
200214
287
338
310338
380
380385
Turbidity, ΐy. 1
Range 17
5
47
2
23
5
222
3
210
2
23
3
24
2
23
3
24
2
112
Oxidation
Tank Analysis
pH, ΐy.
Range
6.71
6.297.77
6.56
6.137.17
6.507.98
6.256.50
5.738.53
6.717.04
6.947.29
6.167.88
6.74
5.807.54
6.74
6.587.03
Dissolved
mg/l
Oxygen, ΐy. 6.1
Range 4.5-7.8
--
--
--
--
--
--
--
--
-
--
--
Suspended
Ay.
Range
Solids, ml/1
28
48
3660
97
65124
341
331351
324
297351
339
327351
406
227505
65
6465
-------
Table 5. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
PROCESS CONDITIONS (PROCESS NO. 11)
Process Flow:
Oxidation Tank #l>,Thjck
Water Source.- - F1ash Mixer_ ._- 4 oxidation Tank #2
I l
Air Effluent Sludge
to Stream
Lime Slurry Sump
) DissLver
Lime Leder
I
Conveyor
p
Ca (OH) r Storage Bin
Average
Run
Water
Flow,
Duration
Treated
Test
Water Source gpm
hrs
gal
No.
Bennetts Branch
103
45.5
281,440
7171
Proctor
No.
1
96
100
26.8
35.0
154,700
210,000
7071
3871
Proctor
No.
2
87
89
97
100
106
134
178
22.8
19.5
5.3
19.8
24.0
7.0
15.5
119,470
104,980
30,860
119,140
153,310
56,370
165,830
6671
5671
6871
6771
6971
5771
6571
H 2 0
298
-------
Table 6. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
RAW WATER ANALYSES
Test No.
71 71
3971
7071 6671 6771 - -
Water
Bennett s
Proctor
Proctor -
Source
Branch
No. 1
No. 1
Temp.
°P,
2W.
56
50
50
Range
3.40
3.03.90
273
381028
73
3824 4
2.69
2. 383 . 45
417
40542 3
69
6970
567 1
Proctor
No. 2
52
515 3
2.94
2.832.86
1305
250
pH, At?.
Range
to Acidlty, ΐy.
O mg/i Range
Fe II, Ar.
mg/i Range
Fe III, mg/i, Ar.
Al, mg/I, Ar.
Ca, mg/i, ΐY.
Mg, mg/i, Ar.
SO4 mg/i, Ar.
Conductivity, Mr.
rU cronthos Range
3.21
3.013.40
467
45 1482
74
657 1
Proctor
No. 2
6 1
5568
3.03
2.793.37
1343
13241362
304
Proctor
No. 2
61
6064
3.50
1516
293
Proctor
No. 2
59
5762
2.92
2.833.02
13 80*
290*
6871
Proctor
No. 2
63
636 4
2.95
2. 922. 98
13 80*
290*
14
>Q*
16
62*
41
36*
96*
96*
19
76*
36
402
529
761
360
740*
1800
23*
191*
143
60
1796
3300
16003740
>0*
192*
201*
97*
1822
39QQ*
6971
Proctor
No. 2
53
5254
2.84
2.792.89
1399
1249 1531
341
338344
>0*
81.5
201*
70. 5
1921
4964
388056 80
>0*
192*
201*
97*
1918 *
3800
5 771
Proctor
No. 2
53
52 5 3
2.83
2.922.83
13 75 *
300*
191*
201*
97*
1985*
3773
37503790
>0*
89
201*
97*
1918*
5500
>0*
192*
201*
97*
19 18*
3725
36703780
* Values established from monthly an 1yses.
-------
Table 7. TREATMENT OF COAL MINE DRAINAGE IVITH HYDRATED LIME -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
717 1
3871
7071
5671
6571
6671
6771
6871
6971
5771
Water
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Source
Branch
No. 1
No. 1
No. 2
No. 2
No. 2
No. 2
Proctor
No. 2
Proctor
No. 2
Proctor
No. 2
Influent Fe I
lbs
lbs/hr
I
Fe II Reacted
lbs 168.9
Minimum Oxidation Rate
mg/I/hr 71.9
171.2 121.9 95.3 217.1 419.8 280.9 287.4 74.6 434.2 140.8
3.763 3.483 3.555 11.136 27.081 12.322 14.514 14.078 18.090 20.119
106.3
60.7
Theoretical Reagent Required 1 1 1
lbs 504 576
Lbs/hr 11.1 12.7
lbs/l,000 gal 1.79 2.0
93.9
72.8
475
17.7
3.07
217.1
247.8
1
900
46.1
8.57
418.4
303.2
2
1699
109.0
10.19
280.0
280.8
2
1373
60.2
11.49
286.4
288.1
2
1246
62.9
10.46
74.4
289.5
2
328
60.9
10.47
431.6
337.5
1
1410
58.8
9.20
140.4
298.7
1
510
72.9
9.05
Reagent Utilized
lbs 588 108.6
lbs/hr 12.9 31.0
lbs/1.000 gal 2.09 5.17
751
28.0
4.85
916
47.0
3.73
2852
184.0
17.20
3890
170.6
32.56
2714
137.1
22.78
727
137.2
23.56
2301
95.9
15.00
764
109.1
13.55
Alkali Consumption titilized/ Theoretical
116.6 18.9
158.1
101.8
168.9
283.3
217.8
221.6
163.2
149.8
Retention Time (minutes)
Flash Mixer 14.4 14.8
Oxidation Tanks 192 199
Theoretical Oxygen Requirement
lbs 24.5 17.5
lbs/hr 0.539 0.499
lbs/l ,000 gal 0.0871 0.0833
ft (ctd) 275.3 196.0
ft /hr (std) 6.05 5.60
15.4
206
13.6
0.509
0.0879
153.2
5.72
16.6
223
31.1
1.595
0.2962
349.1
17,90
8.3
112
60.1
3.879
0.3624
674.8
43.34
17.0
228
40.2
1.765
0.3365
451.6
14.8
199
41.2
2.079
0.3458
461.9
15.3
204
10.7
2.017
0.3467
119.9
14.0
187
62.2
2.591
0.4057
698.0
11.1
148
20.2
2.882
0.3583
226.4
a
a
-------
Table 7 (continued). TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED
LIME - PROCESS AN!) REAGENT REQUIREMENTS
Oxygen Utilized as Supplied by erators
Maximum
lbs 2 .2
lbs/hr 0.533
lbs/1,000 gal 0.0860
Minim jiu
The
lbs/hr
Oxygen Transfer Rate Based
ppm 0 2 /water
ppm/hr 0 2 /water
Oxygen Transferred as Required by Iron
$ 101.3
Oxygen Transferred as Aerator Rating
0.9 0.7
153.5 137.3 140.8 151.2 105.6
40.4 41.4 41.4 48.5 42.8
100.3 100.3 100.3 100.6 100.3
2.9 3.5 3.4 4.3 4.8
1. Based on 93.98% Ca (ON ) 2
2. Bulk hydrated lime h d become carbonated during storage. Assume
3. T o hour settling tine.
Test
No.
7171
3871
7071
5671
6571
6671
6771
6971
69-71
57iT
Water
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor Proctor
j
Source
Branch
No. 1
No. 1
No. 2
No. 2
o. 2
No. 2
No. 2
No. 2
No.2
1.4
0. 031
on Retention Time
33.0 28.9
10.3 8.7
15.2 13.5 31.1 59.9 40.1 41.0 10.7 61.8 20.1
0.435 0.502 1.595 3.866 1.759 2.072 2.010 2.576 2.872
0.0738 0.0873 0.2962 0.3612 0.3366 0.3441 0.3467 0.4031 0.3566
nil 0.9 22.7 47.6 31 3 32.0 8.4 49.4 15.5
nil 0.035 1.162 3.072 1.377 1.61 1.589 2.058 2.215
35.7
10.4
114.7
133.1
35.8
81.0
43.4
Sludge Production
gal/l,000 gal treated 3
solids eontent,. ing/l
lbs dry solids/1,000
gal treated
101.4 100.0 100.3
0.8 2.7 6.4
1.67
4.90
47
85
100
242
(260)
(260) (260) C260)
4250
6900
4960
5780
5150
6540 6580 5760
4.14
11.70
11.20
14.15 14.25 12.50 15.50
90 (260
26500 rs) 5760
12.50
81.43% Ca °j 2 equiv.
-------
Table 8. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED LIME -
OXIDATION TANK ANALYSES (PROCESS NO. 11)
Test No.
7171
3871
7071
5571
6571
6671
6771
8871
8971
57_711
Water
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Source
Branch
No. 1
No. 1
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
No. 2
tH, Av.
8.58
8. 147
8.32
7.9 4
7.22
9.11
8.80
8.25
8.18
7.05
Range
7.159.43
7.73-9.30
6.789.30
7.288.38
6.008.05
7.0510.05
8.5149.09
8.208.30
7.038.65
Acidity or Alkalinity*
mg/i, Av. +18 33 +26 17 +8 +21 _11 12 6
Range 23 +49 . . .345 +60 +15 +149 7 24 +1 +74 19 2 .. 14Q +15
IronTI,Av. 1 4 1 1 1 1 1 1 2 1
mg/i Range 19 - - - 12 - -_ 114
Oxidation-Reduction Potential
milliv, Av. 162 353 166 190 275 176 169 200 178 t)
Range 110215 3021400 130215 162200 250295 130250 138200 136210
Suspended Solids, Av. 200 588 496 13 140 1700 1710 1860
mi/i Range 188212 328848
Thickener Effluent Analyses
pH, Av. 7.57 7.57 9.69 9.77 9.75 7.60
Range 7.557.62 7.547.67 9.659.75 9.709.79 7.537.74
Negative value is Alkalinity.
1 Average values (mg/i)
Sulfate 1542
Calcium 568
Magnesium - 48
-------
Table 9. TREATMENT OF COAL MINE DRAINAGE WITH CALCINED LIME -
PROCESS CONDITIONS (PROCESS NO. 10)
Average
Run
Water
Flow
Duration
Treated
Test
Water Source gpm
hrs
gal
No.
Tyler Run 101 23.5 143,220 16-71
199 37.0 443,770 14-71
Bennetts Branch 109 45.3 297,500 8-71
197 62.5 741,490 771
289 60.0 1,040,560 6-71
Proctor No. 1 107 21.5 138,030 9-71
196 45.3 534,960 10-71
Proctor No. 2 103 40.3 250,590 11-71
195 29.2 342,900 1371
199 41.0 490,210 12-71
200 50.0 601,330 15-71
Proctor No. 1 and 325 29.3 571,560 17-71
Proctor No. 2 (260+65)
Process Flow:
Water Source 4F1ash Mixer ) Settling Lagoon
Q
D air effluent sludge
to stream
Lime Slurry Sump
I
slaker
Lime Feeder
Conveyor
CaO *stoLge
303
-------
Table 10. TREATMENT OF COAL MINE DRAINAGE WITh CALCINE!) LIME -
RAW WATER ANALYSES
Test No. 16-71 14-71 8-71 77 ! 671 9-71 10-71 11-71 13-71 1271 15-71 17-71
Water Tyler Tyler Bennetts Bennetts Bennetts Proctor Proctor Proctor Proctor Proctor Proctor Proctor i
Source Run Branch Branch Branch No. 1 No. 1 No. 2 No. 2 No. 2 No. 2 Proctor 2
Temp. o ΐy. 37 41 34 34 46 45 52 45 36
Range 35-42 40-43 4S47 4349 51-52 4054 35_37
pU , ΐy. 3.82 4.02 3.94 3.04 3.02 3.87 3.47 3.02 3.52 3.95 2.92 2.81
Range 3.79-3.89 3.78-4.23 3.51-4.37 2.40-3.92 2.99-3.05 2.99-5.13 3.36-3.80 2.38-3.79 3.51-3.54 3.11-4.82 2.72-3.05 2.68-3.0
Acidity mg/i, Av. 74 73 161 140 281 477 501 1736 1603 1580 1390 1518
Range - 70-79 114-208 106-176 202-430 475-479 493-507 1652-1848 1443-1762 1478-1696 - -
Iron lI,mg/l, Av. 1 1 32 26 32 83 84 419 82 263 239 320
Range - - 29-35 25-27 2635 81-85 8086 413421 8083 147-402 212-261 -
Iron I II , mg/i ΐy. 3 3
-------
Table 11. TREATMENT OF COAL MINE DRAINAGE WITH CALCINED LIME -
PROCESS AND REAGENT REQUIREMENTS
1.2 5.7 79.4 859.0
0.51 0.100 I,46 2.58,3
S. II 0.4
lb.. 1.2 5.7 70.0 142.2
M.niac Oxidation
kite, me/I/hr. 1.0 1.0 20,0 23.0
Theoret Ical Rrajett
Required
lb .. 58.7
lb ../hr. 2.20
lbn./I ,000 gal. 0.36
R.igent Utilinel
lb.. 40
lb../lu. 8.70
lb ,./1000 Oil. 0.2$
277. 95.0 374.9 074.8 261.9 1 1r4.4 1406.3 15224
4.626 4.445 8.240 :1.599 0.955 26.193 23.923 52.030
857.0 233.3 505.6 1431.0 320.? 1305.4 2110.9 2677.2 3772.5 401.l 4225,9
4.26 5.15 8.09 23.81 14.8 20.02 52.5$ 91.69 02.0! Ot.42 144.2
0.56 0.70 0.60 1.30 2.32 2.44 0.45 .0! 7.70 0,77 .39
Alkali C on u o.ption
Th.or. Ical 776 843.0 163.1 148.2
Alkali Consumption
Their. Effl. bIb .
Ut ! l i ,.J
iet.ntion Ii .. (.18.)
P1 1.6 Miner 04.7
0 ,ld.tiot link. 186
The9r.t it o! Onygen
Requi r.nnt
lb.. 0.2
16./hr. 0.007
16./1,000 aol. 0.0054
ft. ° ).td.) 1.0
?t./hr. (old.) 0.08
Ooypn Utilized II
Supplied by Aerator,
Mac l it
lbs. 13.4
lb../hr. 0.50$
16./1.000 gal. 0.0084
010 1.0$
lb.. nil
lbs/it ,. nil
Oxygen lnn.f.r Rate
$ 10.4 on R.t.nt ion
Ti ..
ppm 0 3 /w.ter 30.6
ppm/hr. 0,/Oiler 81.2
Ooyg.n To.tt.fefr$d a.
Aenitor lit itg, I 0.9
Sludge Produttlon
$ lI.fl,000 108.
treated 113)
mIld, content,
ni/I 1000
Sb.. dry ,oiIds/
1,000 .l. tr.ut.d 0.55
I0.ed 08 09.08 CoO
5 T ,to hour 1.168(01 tim.
8220 500 2160 300 670 10390 9t2 5060
20.33 23.26 47.68 91.0! 229.69 251.1 175.54 234.13
1.17 3.62 4.04 14.77 19.55 21.01 14.58 12.00
Inficent In II
lb..
lbs/hr.
loot No.
lb.!
14.71
0-?!
771
671
9-7!
14-71
Il ? !
13.71
l2 1
151
l7-1
Outer
Tyler
Tyler
Bennetts Oroonttn bennetts Proctor
Proctor
Proctor
Proctor
Proctor
Peoctol
lroctor
1 5 2
Soorce
Ron
Ron
llrooch
Oranch
Branch
No, 1
110. 1
140. 2
To.
No. 2
No.
231.6 95.7 370.4 070.4 229.9 1069.1 1196.2 1517.7
27,0 03,2 83.8 416.5 00.4 258,1 13n,3 310,3
227 300 780
6.14 0.39 12.00
0,51 8.20 1.01
85,3 106,1 150.2 875.0 258.5 :2.2 21.0 lO !.
52 113 17 . 07 41 r
7,0 83.6 7.5 5.8 13.4 7.6 14.4 2.6 .4 4 4.6
100 102 101 60.6 106 101 883 102 100 88 61.1
0.5 11.4 22.0 41.6 13.7 53.7 120.5 57.5 153.5 171.4 288.1
0.014 0.200 0.367 0.693 0.617 1.100 3,094 1.303 0,743 3,427 7.456
0.0061 0.0303 0.2066 0.0400 0.0892 0.1004 0.5400 0.1494 0.513! 0.2050 0.3816
5.9 827.7 247.3 466.7 153.7 602.7 1406.3 421.1 1722.3 1922.9 2448.0
0.86 2.81 4.82 7.70 7.85 13.246 34.724 14.45 42.01 38.45 03.60
40.7 82.0 20.4 87.1 83.7 04,0 124.6 50.9 190.1 213.7 252.2
1.800 0.263 0.340 0.474 0,744 1.946 3.077 2.083 4.758 4,273 0.520
0.0987 0.0403 0.2758 0.0164 0.0982 0.1570 0.4972 0.1780 0.3900 0.3554 0.4411
87.2 nil nil oil 8.4 35.4 803.76 50.! 153.0 51.0 106.0
0.466 ,iI nil nIl 0.067 0.777 2.562 1.027 3.732 3.26 7.012
80.3 84.6 5,7 3.8 43.1 31.6 192.0 35.0 79.7 30.4 54.0
81.0 4$ 3.4 3.3 13.9 10.0 59,7 20.8, 47,9 42.? 53.0
63.9 100.8, 8,3,7 70.? 40.2 96.5
8.8 0.4 0.6 0.8 1.2 3.1 5.! 3.4 7.9 7.1 4.4
(13) 35 25 23 - 200 286 (2425 1250) 1250) 5250)
5000 4720 3240 4260 .- 2000 5459 9450 5920 60130 6000
0.05 1.3$ 0.61 3.50 10.90 09.10 02.31 12.00 12.50
oxygen Transferred in
0.q ,cir*d by Iron, 1 9.2 8.3 90.8 107.9. 116.2 00.6
305
-------
Table 12. TREATMENT OF COAL MINE DRAINAGE WITH CALCINED LIME -
OXIDATION TANK ANALYSES (PROCESS NO. 10)
Test
No.
1671
1471
871
771
671
971
1071
1171
1371
1271
1571
1771
Water
Tyler
Tyler
Bennetts Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Source
Run
Run
Branch
Branch
Branch
No.
1
No. 1
No. 2
No. 2
No. 2
No. 2
No. 1 & 2
pH, Av. 9.05 9.34 6.85 7.03 7.53 6.95 7.81 8.14 8.96 8.79 8.50 7.70
Range 6.7411.69 8.3111.01 5.878.25 6.057.74 6.418.88 6.018.04 5.999.40 7.458.93 7.90-9.52 7.229.52 4.4410.39 4.219.48
Acidity or
Alkalinity,mg/1 *
Av. +14 _7 55 +3 +2 16 6 13
Range 6 +20 15 1 176 +22 15 +28 63 +178 72 4-18 36 +11
FeII,Av. 1 1 5 2 11 1 1 1
mg/i Range 113 010 029 0 -1
Suspended Solids
rg/1, ΐy. 165 81 98 430 1903 2289 1480 0
Range 9699
Dissolved Oxygen
mg/i, Av. 11.1 10.9 0.4 0.4 0.3 7.3 9.1 10.3 8.4
Range 8.012.0 9.211.4 0.20.3 0.215.0 7.411.2 4.711.9
OxidationReduction
Potential ,milliv
Av. 215 198 361 371 414 327 295 213 208 254 271
Range 170260 110285 260405 280440 345443 200475 240330 212214 180214 175525 170580
* Negative value is alkalinity.
-------
Table 13. TREATMENT OF COAL MINE DRAINAGE WITH CALCINE I) LIME -
SETTLING LAGOON EFFLUENT ANALYSES (PROCESS NO. 10)
Test
No.
16 71
1471
871
771
671
971
1071
1171
1371
1271
1571
1771
Water
Tyler
Ty ler
Bennetts
Bennetts Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
Source
Run
Run
Branch
Branch
Branch
No.
1
No. 1
No. 2
No. 2
No. 2
No. 2
1 6 2
pit, A ,. 8.36 9.44 7.54 7.34 2.32 8.06 8.01 8.23 9.40 9.14 7.18
Range 6.509.67 8.9010.61 7.008.93 6.679.67 6.1811.38 5.959.37 6.759.51 7.459.51 9.209.62 8.759.53 4.23-9.42
Acidity, ΐy. -1-17 +32 +13 +29 +10 33 9
nigh Range 24464 1464 1+26 +31-64 +2+20 15-SO 33+25
FeIl,Ay. 1 3 4 13 1 0 1 1
aghl Range 14 26 - - - 02 - - -
1 PeTota1 ,mg/1,Av. 1 - S 1 1 -
Suspended Solids
nigh, ΐy. 6 10 9 6 6 3 23
Range 57 113 944
Tentp.°F,Av. 35 41 32 32 32 35 38 40 43 39 41 41
Range 3441 3847 3237 3739 3841 4147 3641 4044 3445
Conductivity, As. 1180 686 1115 1014 1290 1684
Range 8401760 560800 11001140 9401120 11601760 11402020
Effluent Rate
gpni ΐy. 105 256 127 380 106 273 125 249 297 266 370
Range 105206 83218 375405 100110 210273 67163 248250 243297 203370 145445
SO 4 , ag/i, Av. 275 296 698
Calcius, ag/i, ΐy. 94 94 246
-------
Table 14. Treatment of Coal Mine Drainage with
Sodium Hydroxide Process Conditions (Process No. 10)
Average
flow,
Water source gpm
Run
duration
hrs
Water
treated
gal
Test
no.
Tyler Run 91
48.0
263,230
1-71
Bennetts Branch 87
30.0
157,280
2-71
199
48.0
573,900
3-71
292
24.0
421,380
4-71
Proctor No. 1 88
36.0
190,570
5-71
204
21.5
263,160
18-71
255
24.0
368,250
19-71
Proctor No. 2 87
13.5
70,880
20-71
195
25.3
296,110
21-71
227
27.0
367,760
22-71
Water Flow:
Source -+ Flash Mixer
Oxidation Tanks
Settling
Lagoon
+
l 2
+
+
Reservoir
+
0
.
Air
Effluent
to Strean
Sludge
.+
NaOH (drums)
50% solution
308
-------
Table 15. TREATMENT OF COAL MINE DRAINAGE WITH SODIUM HYDROXIDE -
RAW WATER ANALYSES (PROCESS NO. 10)
Test No.
171
271
37 1
471
5-71
Water
Source
Tyler
Run
Bennetts
Branch
Bennetts
Branch
Bennetts
Branch
Proctor
No. t
Proctor
1971
Proctor
2071
Proctor
2171 22-71
Proctor Proctor
Temp., At.
°F Range
34
35
3435
34
34
45
4446
1
46
4547
No. 1
46
4448
No. 2
41
No. 2 No. 2
51 46
pH, At.
Range
3.97
3.474.43
4.19
4.024.47
4.49
4.404.55
4.38
4.154.58
3.50
3.123.79
3.50
3.47 3. 53
3.49
3.453.51
3.00
2.943.08
2.84 2.83
2.832.87 2.802.66
Acidity, At.
mg/i Range
123
114140
138
117152
129
62172
98
87109
591
516660
376
302450
403
386419
1142
12501634
1616 1533*
Fe II, Ay.
mgIi Range
1
28*
2630
32
2936
24
2029
78
7284
73
7274
39
331
307 316*
Fe III, mg/i, At.
3 *
?0*
6*
>0*
?0
32*
56*
53* 43*
Al, mg/i, At.
16*
18*
19*
20*
80*
7 4*
74*
207*
212* 218*
Ca, mg/i, An.
59*
36*
36*
36*
96*
96*
96*
201*
201* 201*
Mg, mg/i, At.
64*
40*
40*
140*
75*
76*
76*
97*
97* 97*
SO , At.
mgh Range
962
8801000
966
9601000
1091
10401160
1012
9401120
1885
16002400
753*
753*
2800
18003200
2577 2540*
22003700
Conductivity, ΐy.
micromhos
374*
592
592*
592*
1260*
1195
1195*
2580*
2580* 2580*
(4
0
0
*
Values established from monthly analyses.
-------
Table 16. TREATMENT OF COAL MINE DRAINAGE WITh SODIUM HYDROXIDE -
OXIDATION TANK ANALYSES (PROCESS NO. 10)
Test j 171
27 1
371
471
571
1871
1971
2071 2171
2271
Water Tyler
Source Run
Bennetts
Branch
Bennetts
Branch
Bennetts
Branch
Proctor
No. 1
Proctor
No. 1
Proctor
No. 1
Proctor Proctor
Mo. 2 No. 2
Proctor
No. 2
pH, Av. 6.29
Range 5.158.70
6.12
5.636.72
6.67
5.609.00
6.93
6.557.32
6.74
6.117.40
9.03
7.3310.00
8.01
7.529.10
9.05 7.42
8.2510.00 4.338.60
59l
5.188.05
Acidity or Alkalinity
mg/i, Av. 418
Range
+29
43
58 21
+9
35 72
11 58 118 26
75 23
Iron II, Av. -_
mg/i Range -
-
--
5
1-11
1
1
11
1-27
1
Suspended Solids, Av. 44
mi/i Range 3948
86
91
8595
106
259
393
1530 1210
Dissolved Oxygen
ppm, Av.
Range
0. 14
0.30.4
0.4
0.30.4
0.4
0.30.4
0.2
0.20.3
10.0
9.810.2
10.3
10.110.6
8.8 5.5
8.210.4 4.8.5.8
5.6
4.98.3
Oxidation-Reduction
Potential, milliv.
Av. 5140
Range 525581
1416
380450
1432
400480
414
400420
536
527544
262
21 14 309
218 128
180272 40180
329
2114560
0
Thegative value is Alkalinity.
-------
Table 17. TREAThIENT OF COAL MINE DRAINAGE WITH SODIUN HYDROXIDE -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
171
271
371
471
571
1871
1971
2071
2171
2271
Water
Tyler
Bennetts
Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Proctor
2
Source
Run
Branch
Branch
Branch
No. 1
No. 1
No. 1
No.
No.
Influent Fe II
lbs. 2.2 36.9 167.8 82.14 123.7 160.2 220.5 188.1 813.2 1012.2
lbs./hr. 0.046 1.219 3.187 3.508 3. 435 7 1453 9.189 13.933 32.205 37.490
Fe II Reacted
lbs. 2.2 36.9 123.9 61.8 117.7 160.2 220.5 187.5 776.2 1009.2
Minimum Oxidation Rate
mg/i/hr. 1.00 28.1 25.9 17.6 74.0 73.0 71.8 316.7 314.3 328.8
Theoretical Reagent Required
lbs. 216.0 l 4 4 .8 493.9 275.5 751.14 660.2 990.1 5140.1 3192.6 3761.5
lbs./hr. l4 5Q 4.83 10.29 11. 48 20.87 30.71 41.26 40.0 126.19 139.31
lbs./1,000 gal. 0.82 0.92 0.86 0.65 3.94 2.51 2.69 7.52 10.78 10.23
Reagent Utilized
lbs. 178.6 95.7 350.9 3414.5 618.9 11514.8 1295.1 880.4 4561.7 59146.2
lbs ./hr. 3.72 3.19 3.71 14.35 17.19 53.71 53.96 65.22 180.31 220.23
lbs./1,000 gal. 0.68 0.61 0.61 0.83 3.25 4.39 3.52 12.42 15.41 16.17
Alkali Consumption Utilized/Theoretical
82.7 66.1 71.0 125.0 82.4 1714.9 130.8 163.0 142.9 168.1
Theor. + Effi. Alk .
Utilized
96 2141 97 52 118 714 614 73
Retention Time (minutes)
Flash Mixer 16.3 17.0 7.14 5.1 16.8 7.3 5.8 17.0 7.6 6.6
Oxidation Tanks 218 228 100 68 226 97 78 228 102 87
-------
Table 17 (continued). TREAThIENT OF COAL MINE DRAINAGE WITh SODIUM
HYDROXIDE - PROCESS AND REAGENT REQUIREMENTS
Test
No.
171
271
371
471
571
1871
1971
2071
2171
2271
Water
Tyler
Bennetts
Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Ppoctor
Proctor
Source
Run
Branch
Branch
Branch
No. 1
No. 1
No. 1
No. 2
No. 2
No. 2
Theoretical Oxygen Requirement
lbs. 0.3 5.3 214.0 11.8 17.7 23.0 31.6 26.9 116.5 11414.9
lbs./hr. 0.007 0.175 0. 1157 0.502 0.492 1.068 1.316 1.996 4.613 5.370
lbs./1,000 gal. 0.0011 0.0337 0.01418 0.0280 0.0929 0.08714 0.0859 0.3795 0.39314 0.39140
ft. 3 (std.) 3.8 59.3 269.7 132.14 198.8 257.6 3514.5 302.14 1307.3 1627.2
ft. 3 fhr. (std.) 0.08 1.96 5.13 5.63 5.52 11.98 114.77 22.140 51.77 60.27
Oxygen Utilized as Supplied by Aerators
Maximum
lbs. 0.3 5.5 19.5 10.0 17.1 146.0 63.1 32.0 1214.7 161.7
lbs./hr. 0.007 0.183 0.1408 0.1426 0.14714 2.1140 2.627 2.373 14.940 5.990
lbs./l,000 gal. 0.0011 0.0350 0.3140 0.0237 0.0897 0.17148 0.17114 0.4515 0.4211 0.4397
Minimum
lbs. nil nil nil nil nil 1414.1 30.3 25.3 100.4 131.l
lbs./hr. nil nil nil nil nil 2.053 1.264 1.875 3.975 14.855
Oxygen Transfer Rate Based on Retention Time
ppm 0 2 /water 0.5 16.0 6.7 3.3 140.7 8.6 28.8 207.1 86.0 76.14
ppm/hr. 02/water 0.15 14.2 14.0 2.9 10.8 5.3 20.6 54.5 50.6 52.7
Oxygen Transferred as Required by Iron
5 100.0 95.6 112.0 117.8 103.8 149.9 50.1 84,1 93.14 89.6
Oxygen Transferred as Aerator Rating
0.01 0.3 0.7 0.7 0.8 3.6 14.0 8.2 10.0
Sludge Production
gal./l,000 gala
treated 30 35 (35) (35) (1140) (140) (200) (200)
solids content,
mg/i 11465 21460 2500 3020 1850 2800 7650 6050
lbs. dry solids!
1,000 gal.
treated 0.367 0.717 0.760 0.882 2.16 3.28 12.75 10.10
aTWO hour settling time.
-------
Table 18. TREAThIENT OF COAL MINE DRAiNAGE WITH SODIUM HYDROXIDE -
SETrLING LAGOON EFFLUENT ANALYSES (PROCESS NO. 10)
met No 1-71 2-71 3-71 471 571 18-71 1971 20 -71 tl _ 72 22 _ 71
_________ __________
Water Tyler Bennetts Bennetts Bennett s Proctor Proctor Proctor Proctor Proctor Proctor
Source Run Branch Branch Branch No. 1 No. 1 No. 1 No. 2 No. 2 Mo. 2
p H, ΐy. 6.76 8.75 6.75 6.84 6.74 8.37 8.54 8.92 6.31 5.48
Range 6.327.60 6.9810.46 6.287.15 6.597.00 6.406.92 6.899.16 8.009.23 6.47..g33 5.757.59 5.ie5.86
Acidity, Av. +26 +41 +33 +16 +15
ag/i Range 8 +45 +28 +62 +11 +54 +7 +22
Alkalinity, Av. - 83 -70
mg/i Range - - - -92 - -6
(# 4 lronII,Av. 0 0 6 5 3 0 i c
mg/i Range - - 211 46 26 0 1
IronTota i,Av. 1 1 7 1 1 0
mg/i
Suspended Solids, Av. 4 10 5 7 22 5 29
mg/i Range 15 9il 341
Temp., ΐy. 32 32 32 32 33
°F Range - - 3235
Conductivity, Av. 966 1013 10 146 1016 1428
micromhos Range 940-1040 9801040 10001080 10001040 10001920
Effluent Rate, Av. 70 123 232 351 92 302 329 1 08 128 60
gprn Range 42107 95123 185270 351380 40100 285302 300380 97320 40180 5861
Tubidity, ΐy. 2 10 3 8
jtu Range 23 713 26 215
5014, mg/i, I Sv. 212 234
-------
Table 19.
TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE -
PROCESS CONDITIONS (PROCESS NO. 10)
Source- - tF1ash Mixer )Oxidation Tanks Cettling
I 1C2
0
I
Reagent Sump
I
M 2 0 Dis solver
at
Feeder
I
Sodium Carbonate
Average
Run
Water
Flow,
Duration
Treated
Test
Water Source gpm
hrs
gal
No.
Tyler Run
276
24.0
398,150
2571
Bennetts Branch
100
1 14 1 +
196
288
22.8
14.3
25.0
24.0
137,869
123,800
294,880
415,000
3071
2771
2971
2671
Proctor No. 1
106
169
222
23.8
24.3
45.5
152,370
246,731
607,300
3271
3171
2871
Proctor No. 2
105
187
21.8
12.8
137,860
144,200
3371
3471
Water Flow:
Effluent
to Stream
Lagoon
S 1 d g e
314
-------
Table 20. TREATMENT OF COAL dINE DRAINAGE WITH SODIUM CARBONATE -
RAW WATER ANALYSES (PROCESS NO. 10)
Test No. 2571 2671 2971 2771 3071 2871 3171 3271 3 471 3371
Water Tyler Bennetts Bennetts Bennetts Bennetts Proctor Proctor Proctor Proctor Proctor
Source Run Branch Branch Branch Branch No. 1 No. 1 No. 1 No. 2 No. 2
Temp., Av. 38 36 41 41 4]. 46 145 45 50 50
°F Range 26 141 3438 4350
pH, Av. 3.76 4.09 4.88 4.26 4.73 3.68 3.55 3.50 2.83 2.92
Range 3.703.80 3.794.21 4.485.72 4.024.62 4.6944.60 3.504.03 3.513.60 3.483.53 2.782.87 2.912.93
Acidity, Av. 71 84 62 130 72 417 332 567 1243 1398
mg/i Range 6972 8286 5269 86173 5186 397440 3283148 2301434 11741311
Fe II, Av. 1 8 14 28 15 61 57 52 118 241
mg/i Range 710 1414.5 2035 1218 5067 5658 4559
Fe II I, mg/i, Av. 3* 9* 3* >0* 2* 10* 20* 20* 155* 126
Al, mg/i, Av. 114* 18* 18* 18* 19* 65* 63* 62* 245* 240*
Ca, mg/i, Av. 59* .35* 36* 36* 36* 96* 96* 96* 201* 201*
Mg, mg/i, Av. 614* L 4Q* 140* 140* 140* 76* 76* 76* 97* 97*
SO 4 , mg/i, Av. 131 1141 130* 160* 1114 520* 550 1163 1628 1942*
Range 4861840
Conductivity, Av. 320* 604 532 600 148 4 300* 1926 1906 270* 2312
micromhos Range 560680 500580 560640 480500 19201940 18401980 18002550
*ValueS established from monthly analyses.
-------
Table 21.
TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE -
PROCESS AND REAGENT REQUIREMENTS
Test
c.
25-71
2671
29-71
2771
3071
2671
31-71
Water
Tyler
Bennetts
Bennetts
Bennetts
Bennetts
Proctor
31471
33-71
Fource
Run
Branch
Branch
Branch
Branch
No. 1
Proctor
No.
Proctor
Proctor
Proctor
Influent Fe II
J.bs./hr,
33
0.138
3L
1.1441
35.1
1.403
29.3
2.018
17.3
0.716
Fe II Reacted
lbs.
3.3
31.1
28.5
28.2
10.9
inii um Oxidation Rate
mg/i/ho.
1.0
9.0
11.6
27.14
9.5
Theoretical Reager.t Required
lbs.
ire. /hr.
lbs.!1,000 gal.
Reagent Utilized
lbs.
lbs. /hr,
lbs.il,000 gal.
308.4 115.7 65.5 141.7 275.5
6.777 4.821 2.759 11.0143 12.6614
2149.8 308.1
101.1 12.85
0.63 0.714
562 963
22.142 140.13
1.141 2.32
Alkali C nsum tjon Utilized/Theoretical
225.0 312.6
161.6
6.1+7
0.55
375
15.0
1.27
232.1
113.7 614.3 139.1 264.6
55.2 50.6 115.6 230.0
142.2
9.9;
1.15
3014
21.26
2.146
213.8
87.6
3.85
0.64
145
6.36
1.05
165.5
Alkali Consumption Theor
+ Effl. Alk.
Utilized
ketenticri Time (minutes)
Flash 5 ixer
OxIdation Tanks
303.3
59.8
2238.1
149.19
3.69
60014
131.96
0, 89
268.3
50
108 49
723.9
29.79
2.93
2011
82.75
8.15
277.8
47
79 70 75
763.5
32.08
5.01
1117
46.93
7.33
1146.3
83
15814.1 1703.3
123.76 78.13
10.99 12.36
4648 45142
363.13 208.35
32.23 32.95
293.4 266.7
41 41
5.14 5.1
72 69
7.6 10.3 114.8 6.7
103 138 199 89
8.8 14.0 7.9 114.1
118 187 106 189
-------
Table 21 (continued). TREAThENT OF COAL MINE DRAINAGE WITH SODIUM
CARBONATE - PROCESS AND REAGENT REQUIREMENTS
Test
No.
257].
2671
297].
2771
3071
2871
3171
3271
3471
3371
Water
Tyler
Bennetts
Bennetts
Bennetts
Bennetts
Proctor
Proctor
Proctor
Proctor
Proctor
Source
Run
Branch
Branch
Branch
Branch
No. 1
No. 1
No. 1
No. 2
No. 2
Theoretical Oxygen Requirements
lbs. 0.5 5.0 5.0 14.2 2.5 144.2 16.6 9.4 20.3 39.5
lbs.Ihr. 0.020 0.206 0.201 0.289 0.109 0.971 0.691 0.395 1.582 1.814
lbs./1,000 gal. 0.0012 0.0120 0.0169 0.0339 0.0181 0.0728 0.0673 0.0617 0.1408 0.2865
ft. 3 (std.) 5.3 55.6 56.4 47.0 27.8 495.7 186.0 105.3 227.8 442.8
ft. 3 Ihr. (std.) 0.222 2.35 2.26 3.24 1.22 10.89 7.75 4.43 17.75 20.35
Oxygen Utilized as Supplied by Aerators
Maximum
lbs. 20.7 25.2 14.6 10.3 6.6 65.7 21.7 11.6 33.6 39.4
lbs./hr. 0.861 1.051 0.585 0.711 0.284 1.444 0.907 0.488 2.619 1.811
lbs./l,000 gal. 0.0520 0.0607 0.0495 0.0832 0.0471 0.1082 0.0879 0.0761 0.2330 0.2858
Minimum
lbs. nil nil nil nil nil 11.6 nil nil 11.1 27.8
lbs./hr. nil nil nil nil nil 0.255 nil nil 0.861 1.276
Oxygen Transfer Rate Based on Retention Time
ppm 0 2 /water 7.4 8.14 10.3 22.8 18.9 19.3 21.0 28.7 49.5 108.4
ppm/hr. 0 2 /water 6.2 7.3 6.0 9.9 5.7 13.0 10.7 9.2 28.0 34.4
Oxygen Transferred as Required by Iron
2.3 19.6 34.4 40.6 38. 14 67.2 62.7 80.9 60.4 100.1
Oxygen Transferred as Aerator Rating
1.14 1.8 1.0 1.2 0.5 2.14 1.5 0.8 4.14 3.0
Sludge Production
gal./l,000 gal. treated 3 10 (10) (10) (10) 50 30 30 80 110
solids content, mg/i 10600 50700 4400 6000 3700 4400 7000 5940 10000 21100
lbs. dry soiids/
1,000 gal. treated 0.250 0.476 0.367 0.500 0.309 1.84 1.75 1.48 6.67 19.35
Two hour settling time
-------
Table 22. TREATMENT OF COAL MINE DRAINAGE WITh SODIUM CARBONATE -
OXIDATION TANK ANALYSES (PROCESS NO. 10)
Test Jo. 2571 2671
2971
2771
3071
2871 3171
3271
3471 3371
Water Tyler Bennetts
Bennetts
Bennetts
Bennetts
Proctor Proctor
Proctor
Proctor Proctor
Source Run Branch
Branch
Branch
Branch
No. 1 No. 1
No. 1
No. 2 No. 2
oH, A. 7.96 7714
7.70
7.95
7.44
7.55 7.49
7.80
7.44 7.31
Range 7.2410.00 7.408.08
6.827.03
7.568.21
7.128.00
7.248.01 7.007.85
7.568.08
7.317.73 6.907.74
Acidity or Alkalinity
g/l, Av. 81 43
2
50
0
135 137
39 432
Range 212 40 33 58
12 +8
33 66
26 +26
120 1147 136
138
293 570
Fe II, Av. 6
3
1
8
1 3.
-
9
mgJl Range --
12 +8
118
00
Fe III, mg/i, Av. 1 1
1 0
1
Suspended Solids, Av. 31 57
44
60
37
220 210
178
800 2318
mg/i Range - - -
--
156-20
Dissolved Oxygen, Av. 6.1 6.0
4.4
6.0
4.3
t43 2.7
1.9
2.8 1.2
ppm Range 5.96.4 4.86.2
1.45.2
5.86.3
4.05.0
3.46.0 2.14.0
1.82.1
2.62.9 1.01.8
Oxidation-Reduct ion
Potential, milliv.
Av. 332 284
383
313
357
198 85
154
100 220
Range 240370 260360
370420
270340
340370
75300 8090
100215
Negative value is Alkalinity.
-------
Table 23.
TREAThENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE -
SETTLING LAGOON EFFLUENT ANALYSES (PROCESS NO. 10)
Water
Source
Tyler
Run
Bennetts Bennetts
Branch Branch
3071
Bennetts Bennetts
Brsnch
2871 3171
Proctor Proctor
3271 3471
Proctor Proctor
3371
Proctor
-
ph, Av.
6.56
7.79 7.26
Branch
No. 1 No. 1
No.1 No.2
No. 2
Range
7.62 9 .78
7.5-8.14 6.947.78
7.64 7.52
7.66-8.15 6.95-8.10
7.71 7.53
7.45-6.10 7.288.02
7.51 7.55
1.29
Alkalinity, Av.
101
45 5 1
7.25-7.64 7.17-8.29
6.95- 5.11
Range
34 195
58 24 100 1
17
92
-35
137 1 05
124 262
-148
+1
1SS 120 120 76
136 110 397
-170
Fe I I, mg/i, A s.
1 1
1 1
1
-120
Ye Total, mg/i, Av.
si
2
2
Suspended Solids, Av.
3
5 3
6
2
mg/i Range
-
27
512
7 7
410
Temp. , At.
°F Range
37
3440
35 39
3436 3642
34 37
3336 3539
41 Lz2
3346 3646
Conductivity, Av.
800
856 945
955
micromhos Range
740-940
740980 760i320
816
940980 760880
1090 1233
7601420
-
Effluent Rate, At.
3 149
368 205
-
_
gpm Range
348395
320395 165205
120
180340
267 220
170267
207 166
122
Turbidity, A s.
2
5 8
5
- 150195
jtu Range
15
27 610
5
56
6 9
7 20
18
2
16
338
46
4 348
127
1
1
26
43
4 246
2
3
4
44
4345
-------
Table 24. TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE
PROCESS CONDITIONS (PROCESS NO. 11)
Average
flow,
Water source gpm
Run Water
duration treated Test
hrs gal No.
Proctor No. 2 202
7.0 84,880 35-71
253
22.3 338,610 37-71
256
14.8 227,445 36-71
Water Flow:
Source + - Flash Mixer
Oxidation Tanks , . Thickener
-f
l 2
,
Settling Lagoon
4
+
Reagent 5u p
Effluent Sludge
to Stream
H 2 0 Dissolver
+
Feeder
1
Sodium Carbonate
320
-------
Table 25. TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE
RAW WATER ANALYSES
Test No.
36-71
37-71
35-71
Water
Proctor
Proctor
Proctor
Source
No. 2
No. 2
No. 2
Temp., °F, Av.
50
50
50
pH, Av.
Range
3.45
3.20-3.92
3.50
3.43-3.69
3.57
3.83-3.95
Acidity, Av.
mg/i Range
1167
1104-1259
1121
---
1 3 86a
Fe II, Mr.
mg/i Range
149
143-154
159
252 a
Fe III, mg/i,
Av.
111 a
101 a
21 a
Al, mg/i, Av.
233 a
223 a
240 a
Ca, mg/i, Av.
201 a
201 a
201 a
Mg, mg/i, Av.
97 a
97 a
97 a
SOt 1 , mg/i, Av.
1542
1748
1938
Conductivity,
micromhos
Av.
2700
2740
2700
aValues established from monthly anaiyses.
321
-------
Table 26. TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
3671
3771
3571
Water
Proctor
Proctor
Proctor
Source
No. 2
No. 2
No. 2
Influent Fe II
lbs. 261.6 448.0 178.3
lbs./hr. 19.090 20.132 25.476
Fe II Reacted
lbs. 229.6 448.0 178.3
Minimum Oxidation Rate
mg/l/hr. 121.0 158.5 261.8
Theoretical Reagent Required
lbs. 2345.8 3357.1 1039.7
lbs./hr. 158.50 150.54 1148.53
lbs./1,0 0 0gal. 10.31 9.91 12.25
Reagent Utilized
lbs. 5230 9787 2330
lbs./hr. 353.38 438.88 332.86
lbs./l,000 gal. 22.99 28.90 27.45
Alkali Consumption Utilized/Theoretical
222.6 291.5 224.1
Retention Time (minutes)
Flash Mixer 5.8 5.9 7 3
Oxidation Tanks 77.6 78.5 98.3
Theoretical Oxygen Requirement
lbs. 40.3 64.2 25.5
lbs./hr. 2.735 2.8814 3.649
lbs./1,000 gal. 0.1772 0.1900 0.3004
ft. 3 (std.) 452.6 720.1 286.7
ft. 3 /hr. (std.) 30.69 32.36 40.96
Oxygen Utilized as Supplied by Aerators
Maximum
lbs. 333 64.2 27.2
lbs.fhr. 2.255 2.884 3.892
lbs./l,000gal. 0.1464 0.1900 0.3204
Minimum
lbs. 14.0 35.5 20.0
lbs./hr. 0.951 1.595 2.862
Oxygen Transfer Rate Based on Retention Tine
ppm 0 2 /water 22.8 29.8 63.1
ppm/hr. O 2 /water 17.6 22.8 38.5
Oxygen Transferred as Required by Iron
121.3 100.0 93.8
Oxygen Transferred as Aerator Rating
% 3.8 6.5
Sludge Production
gal./l,000 gal. treated (110) (110) (110)
solids content, mg/l 7000 7500 5650
lbs. dry solids/
1,000 gal. treated 6.43 6.87 5.10
1 Two hour settling time.
322
-------
Table 27. TREATMENT OF COAL MINE DRAINAGE WITH SODIUM CARBONATE
OXIDATION TANK ANALYSES (PROCESS NO. ii)
Test No. 36-71 37-71
35-71
Water - Proctor Proctor
Procotor
Source No. 2 No. 2
No. 2
ph, Av. 6.83 - 7.39
7.53
Range 5.58-8.08 6.038.89
Acidity or Aikalinitya
7.05-7.95
mg/i, Av. -36 -914
-406
Range -634-+293 ---
---
Fe II, Av. 33
mg/i Range 19-56 ---
Fe III, mg/i, Av. 1 -- -
1
Suspended Solids,
mg/i, Av. 770 824
610
Dissolved Oxygen, Av. 0.2 -
2.3
ppm Range -_ ---
2..02.7
Oxidation-Reduction
Potential, miiiiv.
Av. 206 242
176
Range 170-260 220-280
168-194
aNegative value is Alkalinity.
Table 28. TREATMENT OF COAL MINE DRAINAGE WITH
SODIUM CARBONATE
THICKENER EFFLUENT ANALYSES (PROCESS
NO. 11)
Test No. 36-71 37-71
35-71
Water Proctor Proctor
Proctor
Source No. 2 No. 2
No. 2
pH, Av. 7.60 7.71
7.56
Range 7.56-7.63 7.50-7.81
7.48-7.60
Temp., Av. 46 47
44
°F Range 45-46 46-49
42-46
323
-------
Table 29. TREATMENT OF COAL 1INE DRAINAGE WITH HYDRATED DOLOMITE -
PROCESS CONDITIONS
Average
Flow,
Run
Duration
Water
Treated
Test
Water Source gpm
hrs
gal
No.
Process No. 11
Proctor No. 2 83 13.5 67,900 5971
105 11.3 71,700 5871
106 21.3 135,740 6071
Raw Water9 1Flash Mixer\ Thickener Settling Lagoon
Lime Slu ry Sump\ Feed Sludge Effluent
I \ Sump to Stream
Water - Solution Tank \
I Oxidation Tanks
Lime Feeder 1 and #2
I
Hydrated Dolomite
324
-------
Table 30. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED DOLOMITE
RAW WATER ANALYSES
Test No.
59-71
58-71
60-71
Water
Proctor
Proctor
Proctor
Source
No. 2
No. 2
No. 2
Temp. °F, Av.
Range
53
53-54
53
52-54
52
52-53
pH, Av.
Range
2.87
2.83-2.90
2.84
2.83-2.84
2.91
2.90-2.91
Acidity, Av.
mg/i Range
1352
1331-1364
1365 a
1413
Fe II, mg/i, Av.
266
305 a
305
Fe III, mg/i, Av.
0
>0
>0
Al, mg/i, Av.
191 a
191 a
191 a
Ca, mg/i, Av.
158
149 a
149
Mg, mg/i, Av.
62
61 a
61
SO , mg/i, Av.
1773
1937 a
1793
Conductivity, Av.
micromhos
3777
3750-3790
3780
3760-3790
3802
3800-3810
aValues established from monthly analyses.
325
-------
Table 3].. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED DOLOMITE -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
5971
5871
6071
Water
Proctor
Proctor
Proctor
Source
No. 2
No. 2
No. 2
Influent Fe II
lbs. 1149.2 177.4 321.3
ibs./hr. 11.049 15.765 15.119
Fe II Reacted
lbs. 148.9 177.4 321.3
Minimum Oxidation Rate
mg/i/hr . 263.1 296.7 283.6
*
Theoretical Reagent Required
lbs. 5140.6 576.2 1129.1
lbs./hr. 40.0 51.0 53.0
ibs./1,000 gal. 7.96 8.05 8.32
Reagent Utilized
lbs. 5014 8140 1200
lbs./hr. 37.3 74.3 167.14
lbs./1,000 gal. 7.42 11.72 26.26
Alkali Consumption Utilized/Theoretical
93.2 1145.8 106.3
Alkali Consumption
Theor. + Effl. Alk .
Utilized
110 70 32
Retention Time (minutes)
Flash Mixer 17.8 l4.1 14.0
Oxidation Tanks 239 189 187
Theoretical Oxygen Requirement
lbs. 21.1+ 25.4 1+6.0
lbs./hr. 1.583 2.258 2.166
lbs./l,000 gal. 0.3151 0.3542 0.3389
ft. 3 (std.) 239.9 285.1 516.5
ft. 3 /hr. (std.) 17.76 25.34 24.31
Oxygen Utilized as Supplied by Aerators
Maximum
lbs. 21.3 25.4 46.0
lbs./hr. 1.580 2.258 2.166
lbs./1, 0 00gal. 0.3137 0.3542 0.3389
Minimum
lbs. 15.5 19.6 314.9
lbs./hr. 1.151 1.746 1.640
Oxygen Transfer Rate Based on Retention Tine
ppm 02/water 151.4 135.1 127.2
ppm/hr. 02/ water 38.0 42.9 40.8
Oxygen Transferred as Required by Iron
100.2 100.0 100.0
Oxygen Transferred as Aerator Rating
% 2.6 3.8 3.6
Sludge Production (18
gal./l,000 gal. treated 110 hours) - 334
solids content, mg/i 13200 5800
lbs. dry solids/
l 0O0 gal. treated 12.1 11.4 16.2
*
Based on titre value of 81.03%.
326
-------
Table 32. TREATMENT OF COAL MINE DRAINAGE WITH HYDRATED DOLOMITE
OXIDATION TANK ANALYSES (PROCESS NO. 11)
Test No. 58-71
60-71
Water Proctor
Proctor
source No. 2
No. 2
p11, Av. 7.51
8.03
Range 7.01-8.20
7.31-8.95
Acidity of Alkalinitya, Av. -28
-26
mg/i Range ---
-84-+36
Iron II, Av. .1
1
mg/i Range ---
Oxidation-Reduction Potential
miliiv., Av. 202
229
Range 192-214
220-242
Suspended Solids, Av. 1370
1940
mi/i Range ---
1870-1990
aNegative value is Alkalinity.
Table 33. TREATMENT OF COAL MINE DRAINAGE WITH
HYDRATED DOLOMITE
SETTLING LAGOON EFFLUENT ANALYSES
Test No. 59-71 58-71
60-71
Water Proctor Proctor
Proctor.
source No. 2 No, 2
No. 2
pH, Av. 7.48 7.52
7.54
Range 7.20-7.58 7.51-7.53
7.50-7.57
Alkalinity, Av. -36 ---
---
mg/i Range ---
---
Iron II, Av. 1
mg/i
Calcium, Av. 580 ---
---
mg/i
Magnesium, Av. 53 ---
mg/i
327
-------
Table 34. TREATMENT OF COAL MINE DRAINAGE WITH PEBBLED DOLOMITE
RAW WATER ANALYSES (PROCESS NO. 11)
Test No.
61-71
62-71
63-71
64-71
Water
source
Tyler
Run
Bennetts
Branch
Proctor
No. 1
Proctor
No. 2
Temp. °F, Av.
Range
52
48-62
61
60-65
55
52-60
53
53-56
p1-I, Av.
Range
3.60
3.51-4.00
3.54
3.20-3.81
2.81
2.80-2.82
2.39
2.38-2.39
Acidity, Av.
mg/i Range
102
6-188
119
45-170
438
429-446
1395
1336-1474
Fe II, Av.
mg/i Range
4
1-16
18
13-22
69
310
308-311
Fe III, Av.
mg/i
>0
>0
>0
>0
Al, mg/i, Av.
12 a
26 a
86 a
192 a
Ca, mg/i, Av.
Range
40
31-58
34
31-37
96
153
Mg, mg/i, Av.
Range
17
12-25
11.5
11-12
76 a
---
65
---
SOS ,, mg/i, Av.
Range
169
133-192
198
191-204
726 a
---
1869
1855-1883
Conductivity, ΐy.
micromhos Range
542
500-600
631
590-690
1540
600-1800
4275
900-5400
a
Values established from monthly analyses.
328
-------
Table 35.
TREATMENT OF COAL MINE DRAINAGE WITH PEBBLED DOLOMITE -
PROCESS AND REAGENT REQUIREMENTS
Influent Fe II
lbs.
lbs. /hr.
Fe II Reacted
lbs.
Theoretical Reagent
lbs.
lbs ./hr.
lbs./l,000 gal.
Reagent Utilized
lbs.
lbs. /hr.
lbs./l,000 gal.
13.3 25.3
0.122 0.559
24.1 211.0
1.071 13.187
23.14 208.3
65.9 302.8
Test
No.
6171
6271
6371
61471
Water
Tyler
Bennetts
Proctor
Proctor
Source
Run
Branch
No. 1
No. 2
Minimum Oxidation Rate
rn /1/hr. 3.0
10.0 23.9
Required 1
206.5
1.89
0.51
16.8
101.6
2.214
0.60
207 189
1.90 14.17
0.51 1.11
93.5
4.16
2.20
565
25.11
13 .27
Alkali Consumption Utilized/Theoretical
100.2 186.0
577.2
36.08
7.00
552
314.50
6.69
Retention Time (minutes)
Flash Mixer 214.3
Oxidation Tanks 326
6014.3 95.6
Theoretical Oxygen
lbs.
lbs ./hr.
lbs,/l,000 gal.
ft. (std.)
ft. 3 /hr. (std.)
Requirement
1.9
0.018
0.00147
21.14
0.20
23.9
320
3.6
0.080
0.0211
140.7
0.90
Oxygen Utilized as Supplied by Aerators
Maximum
lbs. 1.4 3 14
lbs./hr. 0.013 0.076
lbs./].,000 gal. 0.0035 0.0200
Minimum
lbs. nil nil
lbs./hr. nil nil
47.9
6141
3.5
0.153
0.0822
38.7
1.72
3.4
0.1149
0.0799
0.023
0.001
Time
102.6
9.6
102 .7
0.2
Oxygen Transfer Rate
ppm 02/water
ppm/hr. O 2 /water
17.14
234
30.2
1.889
0.3661
339.2
21.199
29.9
1.865
0. 3624
23.2
1.448
170.8
143.8
101.3
3.1
Based on
2.3
0.43
Retention
12 8
2.14
Oxygen Transferred as Required by Iron
138.5 105.3
Oxygen Transferred as Aerator Rating
_______________________ 0.02 0.1
Sludge Production
gal./l,000 gal.
treated
solids content,
mg/l
lbs. dry solids/
1,000 gal.
treated
10
10
50
170
8300
14200(10
hrs 000
5380
0.69
1.18
1 Based on calcium content of 93.24% of theoretical.
3.75
10.140
329
-------
Table 36. TREATMENT OF COAL MINE DRAINAGE WITU PEBBLED DOLOMITE
OXIDATION TANK ANALYSES (PROCESS NO. 11)
Test No. 61-71
62-71 63-71
64-71
Water Tyler
Bennetts Proctor
Proctor
source Run
Branch No. 1
No. 2
pH, Av. 7.44
7.63 7.89
7.87
Range 5.50-8.25
7.22-8.60 7.20- 9.34
7.108.26
Acidity of
Alkalinitya mg/i, Av. -10
+3 +21
35
Range -61-+48
-17-+21 +4-+38
-149-+24
Iron II, ΐy. 1
1 2
4
mg/i Range ---
--- 1-2
127
Oxidation-Reduction
Potential Milliv., Av. 293
242 197
238
Range 265-450
232-254 162-213
236-240
Suspended Solids,
mg/i, Av. 83
142 ---
1245
Range 4-150
102-196 ---
1240-1250
Settling Lagoon
Effluent Analysis
pH, Av. 9.10
9.48 9.56
9.64
Range 7.53-9.78
8.73-9.80 9.28-9.71
9.60-9.70
aNegative value is Alkalinity.
330
-------
Table 37. TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC LIMESTONE -
PROCESS CONDITIONS
Average
Run
Water
Flow,
Duration
Treated
Test
Water Source gpm
hrs
gal
No.
Process No. 20
Tyler Run 89 13.0 69,930 5071
Process No. 23a
Tyler Run 136 24.0 196,880 5171
Process No. 23
Bennetts Branc1 45 17.5 47,650 52-71
139 16.5 137,790 5371
Process No. 214
Proctor No. 1 100 45.0 270,2140 5471
78 30.5 11414420 5571
331
-------
Table 38. TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC LIMESTONE -
RAW WATER ANALYSES
Test
No.
5071
5171
5271
Water
Tyler
Tyler
Bennetts
5371
Bennetts
5471
5571
Source
Run
Run
Branch
Proctor
Temp. °F, Av. 51 57 54 53 50 53
Range 4955 4764 5356 5056 4756 4958
pH, Av. 3.69 3.87 4.61 5.02 3.80 3.68
Range 3.453.80 3.813.90 3.875.00 4.855.13 3.505.13 3.613.71
Acidity, Av. 65 86 43 28 260 354
mg/i Range 6069 2661 2235 224300 293414
Fe II, Av. 1.5 1 6 5 43 65
mg/i Range 12 58 1954 5672
Al, mg/i, Av. 12* 16* 16* 18* 80* 80*
Ca, mg/i, Av. 22* 23 16 16 48 68
Range 2123 4749 6273
Mg, mg/i, Av. 9* 9 6 7 20 22
Range 67 1920 1925
S0 4 ,mg/1, Av. 138* 134 85 69 412 591
Range 135140 133135 5981 399424 588624
Conductivity, ΐy. 475 383 270 252 1150 1496
micromhos Range 400600 350400 210330 240270 10601240 14001580
* Values established from monthly analyses.
-------
Table 39. TREANENT OF COAL MINE DRAINAGE WITh DOLOMITIC LIMESTONE -
PROCESS AND REAGENT REQUIREMENTS
Test
No.
5071
5171
5271
5371
5471
5571
Water
Tyler
Tyler
Bennett s
Bennetts
Proctor
Proctor
Source
Run
Run
Branch
Branch
No. 1
No. 1
Influent Fe II
lbs. 95.9 77.2
lbs.fhr. 2.132 2.530
Fe II Reacted
lbs. 93.7 98.4
Theoretical Reagent Required 1
lbs. 36.6 129.9 16.48 31.07 578.2 411.0
lbs./hr. 2.82 5.41 0.94 1.88 12.85 13.48
lbs./1,000 gal. 0.52 0.66 0.35 0.23 2.14 2.85
(#1
...i Reagent Utilized
CaCO 3 MgCO3
lbs. 500 100 75.0 140 290
lbs./hr. 38.46 5.71 4.55 3.11 9.51
lbs./1,000 gal. 7.15 2.10 0.54 0.52 2.01
CaCO 3
lbs. 111.0 501 1251
lbs./hr. 4.63 11.13 41.02
lbs./1,000 gal. 2 0.56 1.85 8.68
Total CaCO 3 MgCO 3 Equivalent
lbs. 102.0 601 1440
lbs.fhr. 4.25 13.35 47.2
lbs.fl,000 gal. 0.52 2.22 9.99
Alkali Consumption Utilized/Theoretical
1366.1 78.5 606.8 24l. & 103.9 350 k
Alkali Consumption
Theor. + Eff 1. Alk .
Utilized
146 106 30
Retention Time (minutes)
Flash Mixer 16.6 10.9 32.9 10.7 14.8 19.0
Oxidation Tanks 256 200
-------
Table 39 (continued). TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC
LIMESTONE - PROCESS AND REAGENT REQUIREMENTS
Test
No.
5071
5171
5271
5371
5471
5571
Water
Tyler
Tyler
Bennetts
Bennetts
Proctor
Proctor
Source
Run
Run
Branch
Branch
No. 1
No. 1
Theoretical Oxygen Requirements
lbs. 13.7 11.1
lbs./hr. 0.305 0.362
lbs./l,000 gal. 0.0507 0.0770
ft. 3 (std.) 1S .2 124.0
ft, 3 fhr. (std,) 3.427 4.07
Oxygen Utilised as Supplied by Aerators
Maximum
lbs. 13.4 10.9
lbs./hr. 0.298 0.357
lbs./l,300 gal. 0.0496 0.0756
Sludge Production
gal./l,000 gal. treated 4 4 14 (20) (20)
solids content, mg/I 91000 34000 54000 70000 70000
lbs. dry solids/l,000 gal.
treated 3.014 1.77 5.00 11.70 10.8
Assumed to be 95.28% CaCO 3 MgCO 3 .
2 Based on limestone supplement only.
-------
Table 40. TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC LIMESTONE -
LIMESTONE REACTOR EFFLUENT ANALYSES
C u
pH, Av.
Range
Av.
mg/i Range
2 25
1 5
Put oxidation tanks
Test No.
5071
5171
5271
5371
514_711
5571
Water
Tyler
Tyler
Bennetts
Bennetts
Proctor
Proctor
Source
Run
Run
Branch
Branch
No. 1
No. 1
After
Limestone
Addition
Settling
Lagoon
Influent
1 2
7.3
5.7
7.2
6.2
7.9
7.28.5
5.s6.0
6.57.7
6.l_6.L4
7.18.0
1 1 1
Alkalinity, Av.
39,
13*
5
9*
10
15
mg/i
Range
16118
9*_17*
2l_8*
017
17_l7*
Suspended Solids, mg/i
600
216
36t4
136 1300
7.3 6.8 6. 146
7.27.148 5.027.0 4.997.30
*
Acidity
Could not maintain Fe levels even with CaCO 3 addition. Needed longer reaction time and oxidation.
in circuit.
48
2966
15 91*
17_l55* 5l*_189*
-------
Table 41. TREATMENT OF COAL MINE DRAINAGE WITH DOLOMITIC
LIMESTONE AND PULVERIZED LIMESTONE OXIDATION
TANK ANALYSIS (PROCESS NO. 24)
Test No.
54-71
55-71
Water
source
Proctor
No. I
Proctor
No. 1
pH Av.
Range
7.76
7.50-8.05
7.21
6.50-7.70
Acidity or Aikalinitya
mg/i, Av.
Range
-22
-36-17
-1
- 17-+4i
Iron II, Av.
mg/i Range
1
---
4
1-20
Iron III, mg/i, Av.
1
1
Suspended Solids, Av.
mi/i Range
571
550-592
775
761-788
aNegative value is Alkalinity.
336
-------
Table 42. TREATMENT OF COAL MINE DRAINAGE %VITU DOLOMITIC LIMESTONE -
SETTLING LAGOON EFFLUENT ANALYSES
Test No.
5071
5171
5271
5371
5471
5571
Water
Source
Tyler
Run
Tyler
Run
Bennetts
Branch
Bennetts
Branch
Proctor
No. 1
Proctor
No. 1
pH, Av.
Range
7.29
7.297.30
7.25
6.907.50
7.47
7.427.5].
7.47
7.427.52
7.50
7.007.60
7.54
7.287.63
i
Alkalinity, Av.
mg/i Range
-
--
-13
18 8
--
-
28
-52 +1
13
--
Iron IT, Av.
mg/i Range
--
--
-
-
-
-
-
-
2
05
--
--
Iron, Total, Av. ,
mg/i
3
1
Suspended Solids.,
mg/i
Av.
Range
-
24
-
-
-
11
1011
9
-
Calcium, Av. , mgi].
54
41
55
76
Magnesium, Av. , mg/i
13
7
11
14
-
-------
APPENDIX D
PROCEDURES FOR THE ANALYSIS OF COAL MINE DRAINAGE WATERS
The methods employed to analyze coal mine drainage waters for its
various components and properties vary widely depending upon the
purposes and history of the particular laboratory. Since some
Constituents in these waters are arbitrarily defined, minor proced-
ural variation can change the results signIficantly. The acidity
determination is especially subject to these discrepancies. Most
commonly, the base of the procedures employed has been Standard
Methods for the Examination of Water and Waste Water, American
Public Health Association. 86 These analytical methods are widely
used and commonly accepted procedures which have been adapted to
this application. During the studies they were consistent with
procedures employed by the Pennsylvania Department of Environmental
Resources, formerly the Department of Health. 87 The U.S. Environ-
mental Protective Agency--through its former agency, the Federal
Water Quality Administration- - issued a manual of analytical proced-
ures 88 which has recently been updated 89 but it does not specify
procedures for coal mine drainage for some components. The details
of the procedures herein described are available in report form. 9 °
SAMPLE COLLECTION
Samples were collected) transported and stored by recognized proced-
ures to be representative of the materials being collected and to
minimize change prior to analysis. Water samples containing
hydrolyzable salts are especially subject to change.
Two types of special stream samples were collected. One was used to
determine ferrous iron content and the other to determine total iron,
aluminum, calcium, and magnesium content. Plastic bottles with a
capacity of 100 milliliters are employed for the collection of 25 ml
samples by pipette. For the sample to determine ferrous iron con-
tent, ten milliliters of a phosphoric-sulfuric acid mixture is added
to the container prior to going into the field. This acid mixture
is prepared as follows: mix 150 ml of sulfuric acid (1.84 sp. gr.)
and 150 ml of phosphoric acid (1.75 sp. gr.) with enough deionized-
distilled water to measure 1000 ml. This particular reagent is
employed to prevent or minimize the oxidation of the ferrous iron
until the sample can be analyzed.
Prior to collecting the sample to be used in determining the total
iron, aluminum, calcium and magnesium content, two milliliters of con-
centrated HC1 is added to the container to prevent the hydrolysis of
either ferric iron or aluminum sulfate.
338
-------
DETERMINATION OF pH
Determination of pH is carried out as soon as possible after a sample
has been returned to the laboratory. Change which occurs in a
sample as it stands prior to measurement could be related to modifi-
cation in carbon dioxide content, iron II oxidation, or hydrolysis
of metals salts. Analysis of the sample is made electrometrically
using a pH meter with a glass electrode and a reference potential
provided by a saturated calomel electrode. In continuous plant
measurements, a coating of the electrodes with a gypsum-iron
hydroxide film can be a serious problem. Cleaning of the electrodes
with 1:2 hydrochloric acid has been adequate to keep them responsive
(86d); the frequency of cleaning could vary with the systems and
electrode locations.
DETERMINATION OF CONDUCTIVITY
Conductivity has significance in regard to the kind and number of
ionic species present. It may also be employed to check other
analyses. Such procedures have been detailed (86g) including a table
of conductance factors for ions commonly found in waters. This
parameter is especially useful for indicating changes in ionic con-
centration. In this work the value was measured with an Industrial
Instruments, Inc., Conductivity Bridge, Model RC16B2.
An interesting relationship, based upon monthly stream analyses, was
observed between conductivity and sulfate concentration. These data
are presented in Figure 1. On this plot, a line is given based upon
Equation 10, developed by least squares fit as a fifth degree
polynomial expression. This relationship can be helpful for control
and other relative purposes due to the simplicity of conductivity
determinations. The average error is 6.5 percent. This relationship
applies for all of the Hollywood waters when conductivity is determined
on a fresh sample.
The conductivity of mine drainage tends to increase upon standing
due to iron II oxidation and oxide precipitation with subsequent
increase in hydrogen ion concentration and its higher conductivity
response. After standing about three months, a sample evaluated by
this relationship yields sulfate values averaging about 18 percent
high.
SOL+ (mg/l) = -4.4293552 + O.43640856Y + 2.752911 x lO 6 Y 2
+ 1.8950735 x 10 7 Y 3 - 6.3826551 x l0 1 Y
+ 5.9328758 x 10 15 Y 5 (10)
where Y = conductivity as micromhos at 20°C
339
-------
Figure 1.
to
I-
z
- I
1 )
z
z
C
4
I-
z
M a
L i
z
0
U
Ma
I -
4
Correlation between conductivity and sulfate concentration
in mine drainage waters
S
r rr r
3600 4000 4400 4800 5200 5600 6000
CONDUCTIVITY N MICRO-MHO UNITS
340
-------
DETERMINATION OF ACIDITY-ALKALINITY
Coal mine drainage is a mineralized or brackish water containing ions
subject to complex asssociations and equilibria. These components
make the acidity-alkalinity determination arbitrary. Existing
procedures are not fully adequate, but should relate to the environ
mental conditions experienced or be otherwise defined.
The acidity (86a) of a water is defined as the capacity of that water
to donate protons. This includes the non-ionized portions of weakly
ionizing acids such as carbonic acid, acid salts, and hydrolyzable
salts, as ferrous, ferric, and aluminum sulfate. When the sample
has a low pH value, mineral acids contribute to the acidity. The
term, acidity, as applied in these mine drainage studies expresses
in stoichiometric equivalence the amount of alkali necessary to
develop a stable condition within the water having a pH of 8.3
(phenolphthalein end point). Possible complex equilibria conditions,
which will vary with pH, Eh and temperature, appear to function as a
buffer, yielding an environment inconsistent with a stable condi-
j 0 tt 9 1 According to Pyne and Yeates 92 , acidity values may be high
due to reactions with magnesium, aluniinate, and manganous ions.
These potential difficulties were previously cited by Lovell 61 .
The alkalinity (86b) of a water is the capacity of that water to
accept protons. Alkalinity is usually imparted by the bicarbonate,
carbonate, and hydroxide components. In many natural waters, the
alkalinity (and acidity) values are influenced by an equilibrium
between these constituents.
In addition to the arbitrary basis of acidity-alkalinity evaluation,
an additional uncertainty in the values from treated water samples
has been related to sample treatment prior to analysis. There is
evidence that samples may contain fine, insoluble, alkaline matter
which will react during the analysis. Thus, an acidity-alkalinity
value can vary depending upon the prior settling or filtration history
of the sample.
The procedure for alkalinity-acidity employed is presumed to give
results which are net value between acidity and alkalinity. For
samples whose pH is below 4.5, the acidity is determined by adding
hydrogen peroxide, heating to boiling to oxidize the iron, and
followed with titration of the hot solution to a phenolphthalein end
point with a standard base. When the sample has an initial pH above
4.5, the sample is first treated with a known volume of a standard
acid solution and, following the steps just listed, back titrated
with a standard base. The resulting value will express a net acidity
or alkalinity.
341
-------
DETERMINATION OF SOLUBLE SULFATES
Soluble sulfate content is determined gravimetrically by precipita-
tion of barium sulfate (86f).
DETERMINATION OF SUSPENDED SOLIDS
Suspended solids values are established on a weight per unit volume
basis developed by membrane filtration 86 . In samples containing
large amounts of suspended solids, as thickener or settling lagoon
underflow, quantitative filter paper is employed rather than filter
membranes. The solids are normally dried at 105°C, although this
temperature is increased to 180°C if concern exists regarding the
presence of chemically-bound water in the crystals.
DETERMINATION OF DISSOLVED SOLIDS
To determine dissolved solids content, a sample is evaporated from a
platinum vessel in a sand bath and finally dried at 105 or 180°C.
DETERMINATION OF TURBIDITY
Water turbidity is expressed in terms of arbitrary units rather than
in terms of actual suspended material as weight per unit volume. It
is evaluated by some light scattering procedure 8 b. The actual
suspended material is the more useful concept in mine drainage waters
treatment and control although turbidity measurement is more rapid,
convenient and may be adapted for Continuous evaluation. A correla-
tion between turbidity and suspended solids as well as the range of
limitations of a given turbidity procedure must be established for
each system considered.
A continuous].y...reading device functioning along these principles, and
called Surface Scatter Turbidimeter (Hach Chemical Company) was
employed satisfactorily in attaining platit data for this report.
DETERMINATION OF SPECIFIC GRAVITY
Mine Drainage Waters
The specific gravity of samples of mine drainage waters was determined
at 20 C utilizing a hydrometer. The particular instrument employed,
A.H. Thomas Urinometer (Squibb), catalog No. 3512, covered the range
0.0000 to 1.0600 and was readable to ±0.0001 units.
342
-------
Sludge Samples
A standard SO ml pycnolneter, Gay-Lussac, ASTM, glass (Fisher
Scientific Company, catalog No. 3-2400), was employed to determine
the specific gravity of sludge samples. Total suspended solids are
determined first. Dried sludge samples may be evaluated by this
procedure utilizing boiled, deionized-distilled water as the dis-
placement liquid. This liquid is probably less than satisfactory
for this purpose due to solubility of some sludge components.
Ethanol or acetone would likely be a more appropriate displacement
liquid. A more convenient procedure applicable to fluid sludges
simply established the weight of a given volume of sludge (50 or
100 ml), taking care to expel any trapped air bubbles.
DETERMINATION OF IRON
Ferrous and ferric iron content are determined by dichromate titra-
tion using diphenylamine sulfonate as an indicator. Total iron is
determined by prior iron reduction with stannous chloride. When the
iron content is less than five ppm, this procedure is insensitive
and, for such samples, it is replaced by a colorimetric procedure
(86c), utilizing the absorption of the iron - 1,10phenanthroline
complex at 508 millimicrons.
DETERMINATION OF ALUMINUM
Aluminum content is determined gravimetrically by difference involving
separation of the R 2 0 3 group and then subtracting the iron concentra-
tion expressed as the oxide.
DETERMINATION OF CALCIUM
Calcium content is determined gravimetrically by precipitation of
calcium oxalate and its subsequent ignition to the oxide.
DETERMINATION OF MAGNESIUM
Magnesium content is usually determined on the filtrate from the
calcium determination gravimetrically by precipitation of the magnesium
ammonium phosphate (86a).
DETERMINATION OF METALS AND SULFATES IN SLUDGES
To determine metal and sulfate content in sludge, the dried sludge
was solubilized with diluted hydrochloric acid and if necessary, any
343
-------
insolubles were fused. The cations and sulfates were determined by
the procedures cited for water samples. Specialized procedures
utilizing atomic absorption relied on lithium metaborate fusion
procedures 93 . S ectrochemical procedures employed were detailed by
ONeil and Suhr 9
A special series of analytical procedures were developed by Bodkin
for these samples and this project. They involved a series of
determinations from a single sample breakdown. The aluminum, iron,
calcium, and magnesium were determined by chelometric titrations
and back titrations with EDTA and CDTA. Magnesium and iron are
obtained by difference t in a series of titrations in which the
calcium and aluminum are determined directly. Sodium is determined
by atomic absorption. The details of these procedures will be
published elsewhere 90 .
DETERMINATION OF MANGANESE
The standard absorption rQcedure utilizing potassium periodate as
an oxidant was employed 61 to determine manganese content.
344
-------
APPEI4DIX E
ESTIMP INC IRON OXIDATION RATES
Table 1. COMPUTER PROGRAM FOR ESTIMATING BIOCHEMICAL IRON OXIDATION
RATES
FORTRAN I V C LEVEL 19 MAIN DATE 70342 13110/Of PAGF 0001
Co ol I FORM9T IFS.0.lx,F5.l,1x.F5.l.lK.F5.l,Lx. 13,lx,F4.l,Ix, 13, Ff2 04
6002 2 FORMAl ll , 1fl 5TART ,lX, ,2E, CAPACITv DC TANK IN GAL . , Ff2 Of
CFA.0,4X,LST FEZ IN TA1 K ,F f. 1,13. DFM ,4x, F1 ,F3.1. FF2 06
C1X,f.PM ,4X,FF2 INPuT . S.l, PPI) FF2 07
C0 3 ORM4T I .llX,tRS Ff2 IN TANK .F9.6,49. L8S CE? INPUT ,FI? OR
CF8.6.IW, PER INTERVAL) FE? 09
4 FORMAT ( ,FlO.6,IX. LBS ,40,C5.i.iA. pPM ,49.FS. 6 .15,.L8S CNANGFFE2 10
C,41,F#.6,iX,LfS GAtN,4X.I3,IK,MfN ) Ff2 II
COOS c FORMAT t O,75K, flx RATE ,F4.I,jA .PPMJH R,2f.OR .,29.F9.6. Ff2 12
C I X, LNSIINTFRVAL) Ff2 129
r)I4FNSION AP(i).AFE1).AFLOWI1I,ARFI i),INII (L),flxyRayIl, FF2 13
r.INTPVL I II FF2 139
C READ IN PIF4 CONSTANTS Ff2 If
C C U is
( (7 TEAD I5,1,ERR5000) *C*PU),AFE II1),AFLOWUI, AFEIN(1 1,IMIN,I, FE? 16
C,D RYR AT I I I,INTRVL(1) FF2 165
C FIGURE IFS FF2 IN TANK INITIALLY AND LRS Ff2 PFR RIM CORING IN Ff2 17
C FE? IA
2008 ST TLfACAPUI * 8.3413 * AFEI(1) 1000000. Ff2 if
(0 79 AKEPLRS1R TLF FF2 195
lINT 1NTFVL(1I FF2 197
CCII FEINLRAFLrIWI1I * 8.3453 * AFEINI I) / I000000.SAINO FF2 20
C012 flXIOSSIOXVRAT )1)1IOO0000./4O.SIACAP(I).ACLCw(i).AIwT,.8, 53 FF2 204
C*AINT) Ff2 207
C Ff2 21
C WRITE 8118 CONSTANTS *80 INITIAL CONDITIONS Ff2 22
C FE? 23
COlD WRITE )6,2.EPR 6O0O) ACAPR I),AFfi )1),AFLOW(I),AFFIMI I, FF2 24
(014 WRITE I6,3,ERR7000 1 STRTLR,FRINLB FF7 25
CCII 75 WRITE )6,5,ERR9000 ) OXYRAT )I),OXLOSS FF2 26
CC16 WIN a FF2 27
COIl *DELFEQ. FEZ 273
(DIR IMIP4lT IMIM( 1 ) FF2 215
7019 INTRV .ENTRVL) I) FF2 277
C020 00 500 IINTRV,IMINIT,IN1RV Ff2 2R
DOTS AINTIASTRTL R,FF INLR$O RLOSS Ff2 29
002? ANEWLRKIN1LFIAFLOWI I)SAINDFIACAPI I)+*FI )i )SAIpfl).AINyLR) Ff2 30
C023 NIN M!N*INTRv Ff2 31
6024 ANWPPMAN FWLR I*CAP )1 )/ A.3459*1 00 0 00 0_ FF2 32
£025 OELFE2ANEWLR-STRILR FF2 33
C02 6 O FEFLO D ELFF2O RLOSS Ff2 335
CO?? WRITE 16.4,EPRR0001 ANEWL R.ANWPPM.DELFFZ,DFFFLO.IIN FF2 34
2028 STRTL R.ANFWL R CR2 35
C029 ADFLFEADfIFf+A 8 1 30E 1FE2) FF2 354
CODA IF IAOFICE-I.0L*NEWLF)) 500.500.300 FF2 35
CO Di WOO CONTINUE Ff2 36
C032 STOP 1000 FF2 37
COT 700 OXYR*TIIIOFEFLORL00000 0.S6 0. /AINTF )AC*P )II$AFLOWII).*INT),A 3453 FF2 373
2034 SIPTL8AKERLR FF2 374
0035 OXLDSSDFEFL O Ff2 379
0036 GO TO 750 FF2 376
0037 5600 STOP 9000 FF2 38
0038 6000 STOP 8000 FF2 39
0039 1000 STOP 7000 FF2 40
0040 5000 STOP 8000 FF2 41
0041 9000 smp 9000 FF2 A?
0042 END
345
-------
Table 2. EXAMPLE OF THE ESTIMATION OF BIOCHEMICAL IRON (I I)
OXIDATION RATES
TO STA RT CAPACITY OP TANK IN GAL 9931, 1ST P 12 IN TANK 42.0 PPM PLOW 24.0 5PM P 12 iNPUT 351.0 PPM
LBS Ff2 IN TANK 3.480839 LBS PE2 INPUT 0.391103 PER INTEKYAL
1.786587 L aS 45,7 PPM 0.305748 ItS CHANGE
3.309748 LBS CAIN
tIE lATE 0.0 PPMIHB ( lB 0.0
B MIN
3,494489 LBS
42,0 PPM
0.003651 LBS
OX RATE
43.7
PPMflIB
OR
.303749
LB5 1N1fByA
3.498096 LBS
42.1 PPM
CHANGE
0.003607 Las
0,309399 LBS GAIN
B PIN
3.4 1A60 LBS
42.1 PPM
0.003564 LBS
0.309355 LBS GAIN
10 PIN
3.495181 LBS
42.2 PPM
0.003521 LBS
0 .309312 LBS GAiN
15 BIN
3.493660 LBS
42.2 PPM
CHANGE
0.003479 LAS
0.309269 LBS GAIN
20 HIM
3.502091 LBS
42.3 PPM
0.003438 LBS
0.309227 IRS GAIN
29 MIM
3.505495 LBS
42.3 PPM
0.003397 LBS CHANGE
0.3C9186 LBS GAIN
30 MIN
3.508891 LBS
42.3 PPM
0.001354 LBS
0. 309145 LBS GAIN
35 PIN
3.512SA7 LBS
42.4 PPM
0.003316 LBS
0.309104 LBS GAIN
40 MIMI
3.515444 LBS
42.4 PPM
CHANGE
0.003277
0.309064 IRS GAIN
45 PSIN
9.9(8682 LBS
42.5 PPM
CHANGE
0.003239 LBS CHANGE
0.309029 LBS GAIN
0.308986 LBS GAIN
SO PIN
53 PIN
3.48120) LBS
2.0 PPM
0.000431 LBS
OX RATE
44.2
PPP#HB
TB
. 308984
LBS/INtERVAL
9.481736 LBS
42,0 P M
0,000444
Q.309437 LBS GAIN
5 MIN
3. 4621 77 IRS
42.0 PPM
CHANGE
0,003441 LBS
0. 309432 LBS GAIN
10 PIN
3.4P2613 IRS
42.0 PPM
CHANGE
0.000436 LBS
0,309426 LBS GAIN
15 PIN
3. 483)43 155
42.0 PPM
0,0Qfl43Q
0. 309422 LBS GAIN
2) PIN
I.4M346B LBS
42.0 PPM
CHANGE
1.C0042S
0,309416 LBS GAIN
25 PIN
3.493349 LBS
42.0 PPM
CHANGE
0.000421 LBS CHANGE
0.309411 LBS GAIN
30 PIN
3.494303 LBS
42.0 PPM
0.000413
0, LBS GAIN
35 MfN
3.4B471.4 195
42.0 PPM
CHANGE
0.00 1NSO LBS CHANGE
0.309401 LBS GAIN
40 PIN
3.4B51I. LBS
42.1 PPM
0.003405 LBS CHANGE
0.309396 LBS GAIN
45 MI S
3.4B59l LBS
42.1 PPM
1.000401 LBS
0.309391 LBS GAIN
50 PIN
3.4 95 15 LBS
42.1 PPM
0,03034e LBS CHANGE
0.309386 LBS GAIN
55 PIN
3.486306 LBS
2.1 PPM
0.000391 LBS CHANGE
3.309381 LBS GAIN
60 BIN
3.486692 LBS
42.1 PPM
0.000986 LBS CHANGE
0.30937? LBS GAIN
69 PEN
3.457074 LBS
42.1 PPM
0.000381
0.309372 LBS GAIN
70 MEN
3.497455 LBS
42.1 PPM
CHANGE
0.000377 LBS CHANGI
0.309367 185 GAIN
75 BIN
3.487823 LPS
42.1 PPM
0.000373 LBS CHANGE
0.309362 LBS GAIN
60 BIN
3,489593 LBS
42.1 PPM
0.000368 LBS CHANGE
0.309359 LBS GAIN
85 PIN
3.448555 LBS
42.1 PPM
0,000363 165 CHANGE
0.309354 LBS GAEN
SC BIN
3.456914 LPS
62.1 PPM
0.000360 LBS
0.309949 LBS GAIN
95 PIN
3.489269 LBS
42.1 PPM
0.00)359 LAS CHANGE
0.309345 LBS GAIN
100 PIN
3. 489820 1.5
42.1
PPM
0.C05IBj LBS CHANGE
3.309340 LBS GAIN
103 PIN
3.489967 LBS
62.1
PPM
0.000347 LBS CHANGE
0.309337 LBS GA OPA
110 P iN
3.490310 LBS
3.490448 LBS
42.1
42.1
PPM
PPM
0.000942 LBS CHANGE
0.003339 LBS CHANGE
0.309333 LBS GAIN
0.309323 LBS GAIN
115 PIN
120 PIN
3.4 C9B3 LBS
42.1
PPM
0.300335 LBS CHANGE
0.309324 LBS GAIN
125 iIN
3. 491913 LBS
42.1 PPM
0. 000330 LBS CHANGE
0,309320 LBS GAIN
130 M3N
3.491639 LBS
42.1 PPM
0.000324 LBS
0.309316 LBS GAIN
135 MIN
3.491961 LBS
2,1 PPM
0.000322 LBS
0,3099 12 LBS GAIN
140 MEN
3.4 22B0 IRS
2.1 PPM
0.000310 LBS CHANGE
0.309308 LBS GAIN
145 BIN
.4°2595 LBS
42.1 PPP
0.000315 LBS CHANGE
0. 309304 LBS GA !9
150 BIN
3.492936 LBS
42,1 PPM
0.000311 LBS CHANGE
0.30910 0 LBS GAIN
155 BIN
3.493213 LMS
42.1 P99
0.000307 LAS
0.309297 LBS GAIN
160 MIN
3.493517 LBS
42.2 PPM
0.000304 LAS CHANGE
0.3C9293 LBS GAIN
169 PIN
3.4938 17 LBG
42.2 PPM
0.000300 LBS
0.309290 LBS GA1N
170 BIN
3.494114 LBS
62.2 PPM
0.000297 LBS.
0. 309284 LBS GAIN
109 PIN
3.494437 LBS
42,2 PPM
0.000293 LBS CHANGE
0.909242 LBS GAIN
I SO MEN
3.4946 97 LaS
42.2 PPM
0.000290 LBS
0.309278 LBS GAIN
lB S BIN
3.494983 LBS
42.2 PPM
0.0002BA LBS
0.309276 LBS GAIN
190 PIN
3.495289 LBS
42.2 PPM
0.000282 LBS
0.309272 LBS SAIN
193 PIr
3. 459944 LBS
42.2 PPM
0.000279 LBS
0.309269 LBS GAIN
203 MIN
3.495920 LPS
42.2 PPM
0.000276 LBS CHANGE
0. 309269 LBS GAIN
203 BIN
3.494093 LBS
42.2 PPM
0.000273 LBS CHANGE
0.309261 LBS GAIN
210 HEN
3.496062 LAS
42.2 PPM
0.00)249 LBS
0.309259 LBS GAIN
21 5 BIN
3.496829 LBS
42.2 PPM
3.000266 LBS CHANGE
0.309235 LBS GAIN
220 MIN
3.406891 LBS
42.2 PPM
0,000243 LBS
0,309232 LBS GAiN
225 PIN
3.N 7L50 LBS
42.2 PPM
CHANGE
0.Ooo2ss LBS
0.309249 LAS GAIN
230 PIN
3.497407 LBS
42.2 PPM
CHANGE
0.000237
0,309245 LAS GAIN
235 MIN
346
-------
Table 3.
EXAMPLE OF THE ESTIMATION OF AIR-IRON II OXIDATION RATESa
Water source
Fe 0 11
mg/i
FeFII
mg/i
Al
mg/l
pH
T°C
Rate (ing/l/hr)
Tyler Run
2
1
10
6
16
79
Bennetts Branch
15
7
30
6
16
172
Proctor No. 1
60
7
85
6
10
66
Proctor No. 2
400
7
200
6
10
180
Examples
5
0
1
5
10
150
5
5
5
50
50
50
50
125
125
125
125
450
450
450
450
0
0
0
7
7
7
7
7
7
7
7
7
7
7
7
1
1
50
1
1
1
50
1
1
1
125
1
1
1
200
5
7
7
5
5
7
7
5
5
7
7
5
5
7
7
20
20
20
10
20
20
20
10
20
20
20
10
20
20
20
150
156
48
10,500
10,500
10,920
360
16,140
16,200
16,800
420
31,560
31,620
32,820
900
C)
0
m
2
77
2
-4
2
-4
2
C)
0
.7
0
7
0
a
a
aSee Equation 6.
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
4. Title
AN APPRAISAL OF NEUTRALIZATION PROCESSES TO TREAT
COAL MINE DRAINAGE
7 Author(s)
Lovel], Harold L.
Supplementary Notes
Environmental Protection Agency
Report Number EPA670/273093, November 1973.
w
5. Report Date
6.
8. Performing Organization
Report No.
10. Project No.
11. Contract/Grant No.
14010 EFN
13. Type of Report and
Period Covered
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. bEPA TMENT OF THE INTERIOR
WASHINGTON, 0. C. 20240
1. Report No. 2. 3. Accession No.
9. Organization
Pennsylvania State University
College of Earth and Mineral Sciences, under Contract to
Pennsylvania Department of Environmental Resources
12.
15.
Sponaoring Organization Environmental Protection Agency
Pennsylvania Department of Environmental Resources
16, Abstract
Four different quality drainages were treated and detailed results tabulated. Appro-
priate unit operations, from water collection to sludge disposal were considered for
eight different reagents (calcium/magnesium carbonate, oxide and hydroxide; caustic
soda and soda ash). Necessary process variations were possible with a versatile
500,000 gpd facility.
Limestone has the least cost per neutraiization equivalent and may be used with
drainage containing up to 500 mg/i iron II. The resulting sludge is dense, rapid
settling and dewaterable. Lime can treat any drainage efficiently but may result
in excess consumption. Lime-produced sludges are voluminous and difficult to handle.
Unitized sludge recycle process, using lime, forms a dense, dewaterable sludge.
Iron II oxidation was accomplished by air in alkaline systems or with autotrophic
bacteria in acid system.
Thickeners are preferred to settling lagoons to separate sludge in larger plants.
Dewatered sludge requires least disposal volume. Drying basin dewatering is prag-
matic, but filtration or centrifugation are effective, but more costly. Settled
sludge may be discarded directly to abandoned deep or surface mines; while dewa-
tered sludge may be placed in landfill areas.
17a. Descriptors
*Coal Mine Acids, *Neutralization, Water Treatment, Operation and Maintenance,
*S],udge, Treatment Facilities, *Separatjofl Techniques, Bacteria, Dewatering,
Disposal, Capital Costs, Operating Costs, *Oxjdatjon Water Quality.
1 7b. Identifiers
Treatment Processes, Autotrophic Bacteria, Alkaline Reagents, HollywoOd (PA).
17c. CO WRR Field & Group
IS. Availability
19. S .curif 7 Class.
(Report)
20. S.curhy Clan.
(Pig.)
Abstractor Harold L. Lovell
21. No.of
Pages
22. Price
Institution Pennsylvania State University
WRs,! IV. JUNE 1 571)
Gpo t3.25t
------- |