EPA-600/2-76-285
December 1976
RECOVERY OF LIME AND MAGNESIUM
IN POTABLE WATER TREATMENT
by
C. G. Thompson and G. A. Mooney
Black, Crow and Eisdness, Inc.
Montgomery, Alabama 36104
Grant No. S803194-01-4
Project Officer
Gary S. Logsdon
Water Supply Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental
Research Laboratory, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the
U.S. Environmental ProtectionAgency, nor does mention of trade
names or commercial products constitute endorsement or recommendation
for use.
ii
-------
FOREWORD
The Environmental Protection Agency was created
because of increasing public and government concern about
the dangers of pollution to the health and welfare of the
American people. Noxious air, foul water and a spoiled1 land
are tragic testimony to the deterioration of our natural
environment. The complexity of that environment and the
interplay between its components require a concentrated
and integrated attack on the problem.
Research and development is that necessary first
step in problem solution and it involves defining the
problem, measuring its impact, and searching for solutions.
The Municipal Environmental Research Laboratory develops
new and improved technology and systems for the prevention,
treatment and management of waste water and solid and hazardous
waste pollutant discharges from municipal and community sources,
for the preservation and treatment of public drinking water
supplies and to minimize the adverse economic, social health
and aesthetic effects of pollution. This publication is one
of the products of that research, a most vital communications
link between the researcher and the user community.
The research reported herein was directed to the
reduction of sludge discharges from water plants practicing
lime softening and coagulation. Sludge reduction is accom-
plished by recovering the required water treatment chemicals 7
carbon dioxide, magnesium carbonate, and lime. The study
reports that chemical recovery cost effectively eliminates
an average of 85 percent of the sludge discharged from these
types of water treatment plants.
Francis T. Mayo, Director
Municipal Environmental
Research Laboratory
-------
ABSTRACT
A hard, turbid surface water was successfully
treated using the magnesium carbonate process in a 2 MGD
pilot plant at the treatment works of Water District Mo. 1
of Johnson County, Kansas, for one year during 1975 and 1976.
During this study, froth flotation was used to separate river
sediments from calcium carbonate formed in the treatment
process.
Both bench scale and pilot plant flotation tests
have shown that sludges formed by softening turbid waters
can be processed to yield a relatively pure calcium carbonate
suitable for lime recovery. This significant breakthrough
is important for existing lime softening plants having a
serious waste disposal problem, as to date lime sludge from
surface water treatment has not been useable for lime reclama-
tion.
Process variables affecting both magnesium
carbonate recovery and calcium carbonate beneficiation were
studied in this work. Magnesium carbonate was successfully
produced on a continuous pilot scale from recycled magnesium
bicarbonate liquor. The cost per ton to produce relatively
pure MgO was estimated to range from $147 to $185. As in
previous studies, water treatment efficiency offered by the
recycled magnesium coagulation was found to equal or exceed
conventional lime softening.
Process economics were favorable. A comparison
of capital and operating costs for magnesium carbonate
treatment, sludge flotation and lime recovery with present
operating costs, including waste disposal, indicated that
annual costs would be considerably lower with the new tech-
nology. When the facilities are operating at full design
capacity, greater than $260,000 per year in savings is
projected.
This report was submitted in fulfillment of
Project Number S803196-01-4 by the Water District #1 of
Johnson County, Kansas, under the partial sponsorship of
the Environmental Protection Agency. Work was completed as
of July, 1976.
iv
-------
CONTENTS
Page
Foreword iii
Abstract iv
List of Figures vi
List of Tables viii
Acknowledgment ix
I Introduction 1
II Conclusions 4
III Recommendations 6
IV Project Description 7
V Results and Discussion 29
VI References 79
Appendices
A. Weekly Summaries - Pilot Plant Operating
Data 82
B. Weekly Summaries - Pilot Plant Solids
Balance 88
C. Flotation Summaries 94
D. Project Photographs 114
Glossary 117
v
-------
LIST OF FIGURES
Number Page
1 Float CaCC>3 vs. CaO Content After Calcining 17
2 Pilot Plant Facilities 20
3 Pilot Flotation Flow Diagram 2 3
4 Daily Operating Record 25
5 Carbonator Feed Solids vs. Magnesium In The
Resulting Recycle 2 8
6 Monthly Average Turbidity and Magnesium Dosage . . • - 30
7 Effect of Sludge Solids In Underflow on
Magnesium Loss 33
8 Magnesium Remaining In Sludge Cake at Indicated
Carbonation pH 35
9 Summary of Thickening Characteristics of Carbon-
ated Sludge 37
10 Summary of Thickening Characteristics of Uncarbon-
ated Sludge 38
11 Particle Size Analysis Comparison 44
12 Comparison of Recovery Efficiency At Indicated
Collector Level and Type 47
13 Effect of pH on CaC03 Flotation 48
14 Effect of % Solids In Pulp On Flotation
Performance 51
15 Typical Flotation Circuit Water and Solids
Balance 52
16 Comparison of Calcium Carbonate Fractions in
Indicated Flotation Streams 58
17 Thickening Characteristics of Flotation
Tailings 59
18 Flotation Costs as Related to Tonnage Processed ... 63
vi
-------
LIST OF FIGURES (Cont'd)
Number Page
19 Thickening Characteristics of MgCO^-SI^O 74
20 MgO Production Facilities 76
vii
-------
LIST OF TABLES
Number Page
1 Chemical Analysis of Johnson County's Raw
Water 7
2 Water Plants Practicing Lime Recovery 9
3 1975 Average Chemical Dosages and Current
Costs - Johnson County, Kansas 21
4 Silica Removal Using Recycled Magnesium 32
5 Summary of Flotation Results, February and
March, 1975 40
6 Particle Size Analysis Summary 42
7 Dresinate Comparison 46
8 Effect of Added Magnesium on Flotation 49
9 Flotation Results on Low Grade Conventional
Plant Sludge 56
10 Flotation Economics at 50 tons/day Calcium
Oxide 61
11 Effect of Additional Cleaning Stages on Product
Grade 64
12 Summary of Capital Cost - Johnson County,
Kansas 66
13 Lime Recovery Costs 67
14 Chemical Cost Savings 68
15 Economic Comparison of Present and Proposed
Treatment 69
16 Summary of Magnesium Production Studies 73
17 Capital Cost for 3 Tons/Day of MgO Production. . . 77
18 Summary of Annual Cost for MgO Production 77
viii
-------
ACKNOWLEDGMENT
Acknowledgment is extended to the entire Board of the Water
District #1 of Johnson County, Kansas. Their cooperative
enthusiasm was gratifying from early project planning to
completion of this report. Special recognition for extra-
ordinary assistance throughout the project is extended to
Mr- R. L. Chandler, General Manager, Mr. Ralph Wyss, Director
of Operations, Mr. Bennett Kwan, Superintendent of Production,
Mr. William Schuetz, Assistant Superintendent of Production,
Mr. William Bush, Assistant Superintendent of Production,
and Mr. James Rumsey, Laboratory Supervisor.
Mr. H. A. Dawson of the Galigher Company, Salt Lake City,
Utah provided considerable assistance during the flotation
studies. His contributions were sincerely appreciated.
Dr. Gary Logsdon, Project Officer, is extended acknowledgment
for his guidance and counsel during the latter project phases.
ix
-------
i
-------
SECTION I
INTRODUCTION
In 1946 the American Water Works Association
appointed a committee to study and report on the disposal
of wastes from water purification and softening plants.
That committee published six reports, the first in 1947
and the last in 1953.
The 1953 report of the committee noted that more
than 96% of 1/530 reporting plants discharged sludge to
streams or lakes without treatment and only 3% discharged
sludge to drying beds. The report concluded with the pre-
diction that there would be an increasing trend toward the
enactment of state and federal laws to prevent water pol-
lution.
In 1969 a new and larger committee of AWWA held
a two-day meeting in New York City to present and discuss
a full and in-depth study of this problem which has assumed
serious proportions. In the words of the committee, "The
water utility industry must now take action to solve its
problem of waste disposal."
The 1973 Inventory of Municipal Water Facilities
(not yet published), identifies some 7,539 municipal water
treatment plants producing a total of 44,217 MGD, that use
treatment processes which require waste discharge. In 1967,
the Census of Manufacturers (2) reported a total of 6,087
industrial water treatment plants, producing 12,178 MGD
that produce similar waste discharges.
Data collected and Supplement B of the Draft
Development Effluent Guidelines for the Water Supply Indus-
try (3) indicate a total sludge production from these plants
of approximately 3 million tons/year on a dry solids basis.
Of this total, approximately 1.5 million tons are produced
from water softening processes. PL 92-500 identifies
Water Supply as an industry, and as such, water plants must
treat waste discharges to meet established guidelines for
Best Practical Control Technology set for 1977, and Best
Available Control Technology set for 1983. Considerable
interest therefore, has been given to alternative means of
sludge treatment.
Efforts to date have been predominantly aimed
at sludge dewatering and hauling to landfills, with the
available technology requiring considerable economic and
1
-------
energy expenditure. In many cases, landfilling dewatered sludge
merely translates the problem from waterborne pollution control
to solid waste management.
The efforts of the study reported herein were
directed to the reduction of water plant sludge problems
by recovering the very chemicals used to treat the water.
Utilizing new water treatment technology to treat 2-5 MGD
of water, studies were conducted to recover lime and mag-
nesium compounds from the sludge produced. The study took
place at the Tom C. Hansen Water Treatment Plant of the
Water District No. 1 of Johnson County, Kansas, located near
the Kansas River.
Study Description
Although primary study emphasis was directed to the
recovery and purification of calcium carbonate from the sludge
produced, considerable attention was given to evaluating the
water quality obtained. Comparisons were made between the
settled water produced by the pilot and the conventional
plant. As the pilot plant settled water was mixed with the
total plant production for pH stabilization and filtration,
comparisons beyond the sedimentation process were impossible.
Surface and well water are utilized by the Johnson
County plant although only Kansas River water was treated by
the pilot facility. Raw surface water turbidities ranged as
high as 1,000 JTU, with presedimentation reducing the turbidity
in the water to be treated to a maximum of 150 JTU. The hard-
ness was as high as 390 mg/1 as CaCC>3 during a portion of the
study.
Bench and pilot scale froth flotation studies were
performed evaluating such process variables as: calcium
carbonate particle size, pulp pH, temperature, residence
time, silica, and collectors used. An attempt was made to
optimize the process from both economic and performance
considerations. The pilot plant operation and flotation
studies were conducted over a 12-month period.
Magnesium carbonate trihydrate was produced from
the recycled magnesium bicarbonate liquor in pilot scale.
Bench scale unit process studies were conducted for all
aspects of facility design for the flotation and magnesium
recovery processes.
2
-------
Objectives of the Study
The program for this study was planned with the
following objectives as its goals:
1) Compare the effectiveness of conventional
coagulation with recycled magnesium on a highly
turbid surface water.
2) Evaluate the feasibility of recovering
magnesium compounds from highly turbid sur-
face waters.
3) Determine if froth flotation is a feasible
method of beneficiation prior to recalcination.
4) Develop design information for all unit opera-
tions and processes involved.
5) Provide an economic analysis of each unit process
or operation, and for the overall process.
6) Provide an overall process evaluation as compared
with the existing process, including treated
water quality and cost of sludge disposal.
3
-------
SECTION II
CONCLUSIONS
Treatment of a hard, turbid river water with the
use of magnesium carbonate as a recycled coagulant produced
results that equalled or exceeded those obtained with con-
ventional treatment. Results obtained compared favorably
with those from the Melbourne and Montgomery studies. Al-
though the hydraulic loading rate on the water treatment unit
under study was less than in the conventional plant, no diffi-
culties would be expected with full-scale implementation of
the process. The high pH of coagulation would eliminate the
need for pre-chlorination, which should reduce the level of
chlorinated hydrocarbons in the treated water.
Froth flotation, using fatty acid collectors, is an
effective, economical process for the beneficiation of cal-
cium carbonate from water softening sludges. Studies conducted
in excess of one year duration found that a minimum of 92%
calcium carbonate purity could be obtained treating sludges as
low as 65% calcium carbonate. Product recovery is anticipated
to range from 85-90%. Optimum flotation conditions were
found to be pH 9.0-9.5, Dresinate collector feed (TX-60W) 5-7
lb/ton, and little or no silica.
Sludge recycle is necessary to increase calcium carbon-
ate particle size for successful flotation separation. Generally,
calcium carbonate with a D]_q less than 2y could not be floated.
Froth flotation has been found to be successful even where
low dosages of polymers and sodium aluminate have been fed. Sludge
carbonation for magnesium removal is necessary prior to flotation.
Froth flotation for calcium carbonate beneficiation will
allow lime recovery from all softening sludges. Implementation
will be dictated by the economics of lime recovery, initially
limited to the larger water plants. Lime recovery will drastically
reduce the objectionable water plant sludge for ultimate disposal.
Recovery of magnesium carbonate trihydrate from
the saturated magnesium bicarbonate recycle liquor was
successful although the product recovered contained calcium
carbonate, silica, and, to a lesser extent, iron. Laboratory
studies indicated that the calcium carbonate and silica can
be easily removed by satisfactory clarification. The cost
per ton to produce relatively pure MgO was estimated to range
from $147 to $185.
4
-------
Implementation of magnesium recycle, flotation,
and lime recovery was found to be economically attractive
for Johnson County. Recovery of magnesium compounds for
sale should be delayed pending results of a market survey
to be completed in the fall of 1976.
5
-------
SECTION III
RE C OMMENDATIONS
The results of studies reported herein demonstrate
that incorporation of this technology offers advantages in
water treatment efficiency, economics, resource and energy
conservation, and reduction in sludge disposal requirements.
It is therefore recommended that magnesium recycle, froth
flotation, and lime recalcination be utilized by the Johnson
County water treatment plant. It is further recommended that
magnesium carbonate recovery for sale be delayed pending the
completion of the marketing and economic feasibility study now
underway sponsored by the City of Dayton, Ohio.
Froth flotation should be attractive for all coag-
ulation-lime softening plants of sufficient size to consider
lime recovery. Process economics must include sludge treatment
and disposal for valid alternative comparisons. The size
requirement is dependent upon the local conditions? however,
generally in excess of 25 tons per day of CaO should be
recoverable for the economics to be favorable. Additional
studies should be conducted to determine if froth flotation
can be successfully applied where conventional coagulants are
employed in the coagulation-softening process.
Studies should be conducted to determine if froth
flotation can be used to beneficiate lime-sewage sludges.
Successful separation of calcium carbonate from calcium
phosphate would have significant economic benefits for sewage
plants practicing lime recovery. Proper combination of
temperature and silica level may result in successful flotation
separation.
Additional research is recommended to study the
effectiveness of the magnesium process for the removal of
certain heavy metals, asbestiforms (particularly chrysotile),
and organic precursors.
6
-------
SECTION IV
PROJECT DESCRIPTION
Water District No. 1 of Johnson County, Kansas
The Water District No. 1 of Johnson County serves
some 14 municipalities in suburban Southwest Kansas City,
Kansas.
Raw water is obtained from both shallow wells and
the Kansas River, which has similar characteristics to the
Missouri River. A partial chemical analysis of these raw
waters is shown in Table 1:
TABLE 1. CHEMICAL ANALYSIS OF JOHNSON COUNTY'S RAW WATER
Constituent £
pH
Temperature (°C)
Turbidity JTU
Ca (mg/1 as CaC03)
Mg (mg/1 as CaCOO
Total Alkalinity
(mg/1 as CaCO-)
Total Dissolved
Solids (mg/1)
Si09 (mg/1)
CI * (mg/1)
1975 Annual Averages
r Kansas River
Settled River
Well
8.3
7.9
7.2
17
18
16
100
30
8
205
200
357
59
59
67
188
173
301
443
449
514
12.7
13.3
27.9
84
89
65
7
-------
Present treatment includes presedimentation, lime-
soda ash softening of the blended well and settled water,
use of polymers and alum for clarification, single stage
carbonation, and high rate mixed media filtration. The
present capacity of the well field is 9 MGD; however, the
production is declining. Treatment plant capacity is de-
signed for a maximum of 60 MGD with an average annual produc-
tion of 22.5 MGD.
In 1975, maximum daily values for river water
hardness and turbidity were 390 and 1,000 JTUr respectively.
Maximum daily turbidity of the settled raw water during 1975
was 150 JTU. Presedimentation greatly reduces the turbidity
loading to the water treatment plant? however, the suspended
solids remaining are generally more difficult to coagulate.
Treatment of this turbid, hard water produces in excess of
2,500 lbs of dry solids/MG of water produced. The predominantly
calcium carbonate sludge, in excess of 11,000 tons/year
dry solids, is presently dewatered in lagoons and hauled
to a landfill at a cost exceeding $80,000 per year. As
suitable landfill locations become more difficult to obtain,
these costs are projected to increase dramatically.
Lime Recovery
Several water treatment plants, as shown in Table
2, have successfully produced high quality quicklime from
their softening sludge. In every case, these plants treat
water very low in suspended solids; thus, the sludge pro-
duced is relatively pure calcium carbonate. In recalcining
water plant sludge, the calcium carbonate is converted to
CaO asj
Heat 1700-2'J00°F
CaC03 ^ CaO + C02
Sludge contaminants not volatized on calcination would in-
crease in concentration as a result of the molecular weight
difference between calcium carbonate and calcium oxide.
Slaking problems have been reported when the calcium oxide
purity drops below 80-85%. The lime impurities would also
increase with recycle and reuse.
8
-------
TABLE 2. WATER PLANTS PRACTICING LIME RECOVERY
Date Production3 Type of
Installed Tons/Day Calciner
City CaO
Miami, Florida
Salina, Kansas
Lansing, Michigan
Dayton, Ohio
Ann Arbor, Michigan
St. Paul, Minnesota
1948
90
1954
-
1954
16
1960
92
1968
8
1968
25
Rotary kiln
Flash calcination
Fluidized reactor
Rotary kiln
Fluidized reactor
Fluidized reactor
a average actual daily production
Recalcining operations at Dayton, Ohio, and Miami,
Florida, have been notably successful. Dayton, Ohio, had sold
more than $1,000,000 in excess lime by 1973.
When recalcining water plant sludge is compared with
purchasing commercial quick lime, energy and resources are
conserved. Typical operations required for commercial lime
production and distribution require:
1) Mining of limestone or marble from quarry
2) Transporting to kiln
3) Crushing and grinding before burning
4) Calcination in kiln
5) Transporting in bulk shipment, frequently
hundreds of miles
Recalcining of the water plant sludge would eliminate 1/2,3 and
5 of above. In addition, hauling of dewatered water plant sludge
to ultimate land disposal would be eliminated. Energy require-
ments for the recalcination operation are similar for both
commercial and on-site recovery operations. A large percentage
of the energy used is lost either in radiant heat, or
in the stack gases. Thus, the 35% or less moisture remaining
in the dewatered water plant sludge does not add significantly
9
-------
to the energy requirements. Plants practicing magnesium recovery
will utilize waste heat from recalcining, thus providing additional
energy conservation.
Lime kilns designed prior to recent energy conserva-
tion efforts generally did not incorporate features that would
reduce fuel requirements. As recently as the early 1970"s, fuel
costs as low as ?6.00/ton of lime calcined were common. Dependent
upon fuel source, this cost may today be as high as $20.00/ton.
Lime kilns designed and constructed today incorporate a number
of devices to conserve energy. Thus, regardless of whether the
application is for recalcining water plant sludge or for a
commercial limestone operation, energy requirements are lower
than for the older installations.
Commercial lime operations reduce natural limestone
reserves. Quarry operations leave unsightly terrain which
one day may require restoration, further increasing energy
requirements and lime costs. Lime production from oyster shells
is coming under increasing attack by environmentalists as a
result of the dredging operations required.
In the upgrading of calcium carbonate, or benefi-
ciation, two parameters are of primary concern: the grade
and recovery of final product. These parameters are termed
simply grade and recovery and are generally expressed in
percentage figures. The most commonly used scheme for
beneficiation of calcium carbonate sludges, prior to recal-
cination, utilizes two-stage centrifugation. Centrifugation
uses the specific gravity difference between the calcium
carbonate and contaminant to make the separation. Rejection
of magnesium hydroxide has generally been of primary concern?
however, one recently completed investigation was directed
to the separation of silt from calcium carbonate. (4) The
primary disadvantage of this method is that considerable
calcium carbonate is wasted with the contaminants in order
to achieve an acceptable grade. Two-stage centrifugation
has been employed on lime-treated sewage sludge to separate
calcium carbonate from calcium phosphate. (5) Again, problems
with recovery of calcium carbonate have been reported. Air
classification following calcination has also been inves-
tigated for lime-sewage sludges with limited success reported.(5)
The product losses using these classification
methods are necessary to prevent contaminant build-up
on recycle. Product grade is generally not sufficient to allow
reuse on a closed cycle system.
10
-------
In 1957, Black and Eidsness (6) showed that the
gelatinous, highly basic magnesium hydroxide could be rapidly
and selectively dissolved from calcium carbonate by carbonating
the sludge with kiln stack gas containing about 20% CC^.
This method has been successfully used at Dayton since 1958.
The recovery of both the calcium and magnesium
values present in the Dayton sludge was described by Black,
Shuey and Fleming in 1971 (7).
In the same year, Thompson, Singley and Black (8,9)
found that magnesium carbonate recovered from such sludges
may be used as a recycled coagulant for the treatment of most
types of waters. The new technology may be summarized by the
statement that all types of waters and many types of waste
waters (hard or soft, clear, turbid or colored) may be treated
by three chemicals normally present in such waters: calcium,
magnesium and carbon dioxide. In so doing, the problems of
sludge disposal are eliminated or minimized and chemical treat-
ment costs significantly reduced.
The New Magnesium Process
A CAPSULE REPORT, "Magnesium Carbonate - a recycled
coagulant for water treatment, " (10) has been prepared and
published by the Technology Transfer Division of the U. S.
Environmental Protection Agency (EPA) which describes in con-
siderable detail the application of the new technology for the
treatment of various types of waters. The process has been
previously studied by three different projects supported by EPA.
(11,12,13)
The first was a laboratory study carried out at
Gainesville, Florida, as EPA Project 12120 ESU, which was
completed in May, 1971.
The second was a Demonstration Project, 12120 HMZ,
conducted at Montgomery, Alabama, and Melbourne, Florida. At
Montgomery, a pilot plant was constructed and studies made of
the treatment of a very soft, turbid river water. The pilot study
was followed by application of the process in the 20-MGD municipal
water treatment plant, with one-half of the plant operated
with alum as the coagulant and the other with a magnesium salt
as the coagulant. The work was then extended to Melbourne,
Florida, where a second plant-scale study was made of the
treatment of a highly colored, low turbidity lake water. As
at Montgomery, one-half of the plant was operated with alum
as the coagulant and the other half with a magnesium salt as
the coagulant.
11
-------
The third project, EPA Demonstration Project 12130
HRA, was carried out at Gainesville, Florida, and was a lab-
oratory and pilot plant study of the use of the new technology
for the treatment of municipal and industrial wastes.
This, the fourth project of the series, differs in
a number of respects from the earlier studies.
First, it has been carried out in a plant of major
size treating one of the most turbid river waters in this
country. The fact that the plant possesses a small "stand-by"
2 MGD Permutit Precipitator treatment plant complete with
chemical feeders has provided the opportunity to treat a
2 MGD flow of the same water being treated in the large plant
but with recycled magnesium bicarbonate as the coagulant.
It was, therefore, only necessary to provide facilities for
sludge storage, sludge thickening, sludge carbonation, mag-
nesium recycling and recovery, and calcium carbonate benefi-
ciation by froth flotation.
Second, it was possible for the first time to study
froth flotation not only on a laboratory scale, but also in a
continuous pilot plant operation.
Third, since this is the only water studied to date
containing substantial magnesium, it has been possible to study
magnesium recovery on an expanded scale. Such studies have been
carried out previously at Dayton, Ohio, but with a clear
wellwater relatively free from possible contaminants.
Froth Flotation
For a number of years, the mining industry has used
froth flotation to separate valuable minerals from an ore slurry.
It differs from dissolved air flotation, primarily in that the
desired slurry components are selectively floated.
There are several different schools of thought
regarding the basic mechanisms involved in froth flotation.
Readers are referred to three highly regarded sources (14,15,16)
for in-depth discussions. The present generally accepted theory
appears to be that certain chemicals attract the desired minerals
to an air-water interface provided by a bubble. This chemical,
called a collector/ attaches by its polar end to the mineral
and its nonpolar end to a bubble interface. The bubble pro-
vides the needed air-water interface and physically carries the
mineral particle from the flotation cell. The process is a
highly sensitive system involving both physical and chemical
forces.
12
-------
In the flotation of calcium carbonate fatty acid
collectors or their soaps are used. They are anionic materials
having 18 carbons in the chain and are typically a mixture
of oleic, linoleic, or resin acids, usually more effective in
their soap form.
Frothers are chemicals added to promote abundant
bubbles by lowering the surface tension of the water. They
are used in many flotation circuits but are rarely needed when
fatty acid soaps are used since these collectors are highly
surface active and produce ample frothing.
Modifiers are used to activate or depress the action
of collectors and generally provide greater selectivity.
They range from sodium carbonate for pH control to a soluble
silicate to depress or enhance calcium carbonate collection.
Since surface chemistry plays a major role in
flotation technology, many investigators have utilized measure-
ment of the zeta potential of the calcium carbonate particles
in an attempt to increase process efficiency. However, as in
water coagulation, the point of maximum efficiency is not
always at the point of zero particle charge. It has been
found, however, that in general, best collector efficiency
is obtained if the charge is adjusted near zero. Zeta
potential is, of course, affected both by change in the
solution pH as well as by multivalent counter ions, present
or added.
Froth flotation was patented in 190 6. The first full-
scale application in this country was in 1911, when it was
employed at the Basin Reduction Company plant for the bene-
ficiation of a copper ore. The first flotation circuits
were mainly used for the separation of minerals naturally
flotable, and required only frothers.
In 1926 froth flotation was first used to recover
limestone from iron ore tailings using fatty acid collectors.
In 1934, at Conshohocken, Pennsylvania, limestone was purified
by flotation for the manufacture of Portland cement. (17) As
late as 1960, only one plant in this country was reported to
be using flotation for the benefication of calcite. This plant,
operated by the Universal Atlas Cement Company, began operation
in 1952, and presently floats about 1,000 tons of calcite per
day. (18) In 1963, the Permanente Cement Company began
floating about 2,000 tons of calcite per day. (19) It is
reported that only one operator is needed to operate the
flotation circuit and downtime the first year was less than
9%. In 1964, the Georgia Marble Company placed into operation
a plant at Sylacauga, Alabama, floating 1,000 tons of calcite
per day. (20)
13
-------
Mined Versus Precipitated Calcium Carbonate
Frequently, terms used to denote or describe cal-
cium carbonate are used incorrectly. Review of geologic
terminology should assist in providing a better understanding
of these commonly used terms.
"Rock" can be defined as a naturally formed mass
of mineral matter. A mineral occurs naturally and is
generally crystalline and inorganic in nature. Limestone is
the predominant rock form containing calcium carbonate? how-
ever, marble is a rock consisting of crystalline metamorphosed
limestone. Another rock, chalk, consists of finely grained
pulverized calcium carbonate.
Calcium carbonate is found in nature in two min-
eral forms: calcite and aragonite. Calcite is the predom-
inant mineral form of calcium carbonate. Aragonite is
less common than calcite and much less stable. It will convert
to calcite at elevated temperatures or in solution. It is
generally formed at lower temperatures, near surface deposits.
A considerable degree of calcium substitution is
possible. Calcite may have in excess of 10% magnesium sub-
situted as well as manganese, iron, and zinc. Generally, the
degree of substitution increases with an increase in the rate
of precipitation and concentration of foreign ions in solution.
In addition to the substituted impurities, considerable impur-
ities exist cemented between calcite crystals and layers of
rock.
Calcite mined for cement or lime manufacture is
most often obtained from the rock, limestone. The rock is
crushed, generally to a size range of 100-200 ji, prior to
flotation. Typical feed grade ranges from 80-85% calcium
carbonate and only slight upgrading is required. Often
flotation is preceded by wet cyclone desliming to remove
much of the silt and fine materials.
Calcium carbonate found in water plant sludges
is formed by the reaction of CO2 and calcium bicarbonate
with lime in very dilute solutions. Sheen and Lammers (21)
state that the sludge produced by selective lime softening
at the Wright Aeronautical Plant has an average particle
size of 5-7 p. Black (22) found that for a particle size
distribution of sludge obtained in softening the water of
Miami, Florida, where no magnesium is removed, 9 9% were
finer than 24 p, 72% were finer than 11.5 ji, and 30% finer
than 4.9 ju,
14
-------
A number of investigators have shown that sludge
recirculation increases the kinetics of calcium carbonate
precipitation and have suggested an increase in particle
size. One study evaluated the effects of sludge recir-
culation on filtrability and thickening characteristics,
indirect measures of particle size. With increased recir-
culation, the sludges thickened and dewatered much more
readily. (23)
It is apparent, therefore, that the industrial
flotation experience is not necessarily applicable to
floating calcium carbonate from softening plant sludges,
where the surface area is far greater due to the very small
particle size. There is also the possibility, when calcium
carbonate is precipitated in the presence of silt turbidity,
that encapsulation of clay particles might take place, making
separation by flotation impossible.
Of the several factors which control the flotation
of calcium carbonate, particle size appears to be the most
important. This is dramatically brought out in the studies
to be reported . A method for the control of particle
size is also discussed.
Calcite Flotation Theory
As stated previously, the pH of the sludge to be
floated greatly affects the efficiency of froth flotation of
calcium carbonate. Somasundoran (24) reports that calcite
in an aqueous suspension with a pH between 9.0-S.5 usually
has a zeta potential of zero. Most investigators have re-
ported that fatty acid soaps are the preferred collector
for calcite and that efficiency usually increases as the
length of the collector chain increases. Maximum recovery
is usually obtained in the pH range of 9.0-10.0.
As has been pointed out earlier, silica or a ;
soluble silicate is often used to depress calcium carbonate
collection. As the temperature of flotation increases, this
depressant action of silica increases. The presence or add-
ition of polyvalent cations is reported to increase the
depressant action of the silica (25). Klassen (14) reported
that Si02= is less of a depressant than H2Si02? thus higher
pH flotation tends to reduce the effect of the silica.
It has been reported that additives which increase
mineral hydration depress flotation, while additives which
dpcrease hydration increase flotation.
15
-------
At extremely high pH values, hydroxide ions compete
with the collector, thus depressing recovery. The use of
soda ash as a pH modifier has been reported to be more
effective than sodium hydroxide.
To summarize the effect of particle size on the
flotation of calcium carbonate, the smaller the particle
size, the larger the surface area and more collector will
be required. The particle size of freshly precipitated
calcium carbonate is so small as to make flotation difficult.
Anything which will bring about a significant increase in
the particle size will greatly improve flotation efficiency.
As discussed, sludge recycle for crystal seeding is an
effective means of controlling particle size. Flocculation
during flotation has also been used with success.
Several investigators have shown that carbon dioxide
flotation of calcite increases collector efficiencies and
thus reduces collector requirements. (26,27)
Higher pulp solids have been found effective in
floating fine particles whereas lower pulp solids are more
effective when large particles are present. There appears
to be considerable controversy over optimum pulp density.
Froth Flotation with the Magnesium Process
The liquid phase of a fully carbonated sludge
contains a concentration of magnesium ion from a low of a
few hundred mg/1 to as high as 3,800 mg/1.
The sludge is dewatered, removing as much of the
liquid phase as possible. The sludge cake is reslurried,
conditioned by the addition of flotation reagents and the
slurry piped to the first series of cells referred to as the
rougher. The float from the rougher, called rougher concen-
trate, is pumped to a second series of cells called the
cleaner cells. The fraction of the pulp not floated is termed
the tails.
Approximately 1.4 times more CaO is produced in
water softening than is fed? therefore, 100% recovery in
beneficiation and recalcination is not required for the process
to be self-sustaining. The grade of CaO must generally be
greater than 85% for proper slaking and feed. Therefore, from
Figure 1, a 91% calcium carbonate grade feed to recalcining is
required, assuming the impurity fraction is not reduced.
16
-------
100-
o
m
o
*
LEVEL OF ACCEPTANCE
% CaC03
FIGURE 1. FLOAT CaC03 vs. CaO CONTENT AFTER CALCINING
(Assumes no further beneficiation and contaminants
nonvolatile).
17
-------
Recovery and grade, therefore, are the two impor-
tant factors in determining flotation feasibility. The
brief review of flotation theory would indicate the following
operating parameters to effect recovery and grade:
Collector type
Collector level
PH
Alkali used to adjust pH
Particle size
Grade of feed pulp
Residence time in stages
Pulp density
Pulp temperature
Depressants
Foreign ions in pulp
Number of cleaners used
Description of Pilot Plant Facilities
The study emphasis was directed to calcium carbonate
beneficiation and magnesium recovery. Considering only these
aspects, the purpose of the water treatment units was to
generate sufficient solids for separation and production
recovery studies. However, water treatment was not neglected
from study. Attention was given to monitoring the removal
efficiency of suspended solids, silica, and magnesium.
The water treatment units were operated continuously
throughout the study. Except during initial start-up, these
units were operated and maintained by plant personnel. The
sludge studies were conducted on an intermittent basis.
Approximately one week out of every month was directed to
field studies of the separation process. The sludges studied
over the one-year period were representative of a wide range
in raw water quality. A considerable portion of the separation
studies was conducted in the laboratory using bench scale
equipment.
Water Treatment Phase
As previously stated, the Tom C. Hansen Memorial
Water Treatment Plant has a maximum capacity of 60 MGD with
a present average production 22.5 MGD. The original plant,
constructed in 1956, utilized a Permutit Precipitator with
a design capacity of 10 MGD. This portion of the plant was
used in recent years only to meet peak production requirements.
The study utilized one half of this facility along with the
rapid mix and chemical feeders. Only minimal conversion was
required. The settled water was mixed with the water from
18
-------
the other plant treatment units prior to stabilization and
filtration. Therefore, following clarification, no study
comparisons were made. Figure 2 illustrates the water
treatment units discussed.
Table 3 summarizes the average chemical dosages
fed in 1975, along with current unit costs. These data rep-
resent annual averages for treating well and river water.
The treatment facilities allow raw Kansas River water, after
presedimentation, to flow by gravity to the plant where river
and well water can be combined in a desired ratio prior to
treatment. During the study period, only settled river water
was treated in the Permutit basin.
Prior to the study, settled river water turbidities
often exceeded several hundred units. Careful presedimentation
coagulant control has considerably reduced turbidities; thus
eliminating the need for blending river and well water to
reduce turbidity to desired levels during the study.
In the pilot operation, recovered magnesium bi-
carbonate was recycled to the rapid mix, and sufficient lime
slurry fed to maintain a pH of 11.0-11.3. A number of investi-
gators have shown that virus and bacterial disinfection occurs
at this pH (28,29). The water then flowed approximately 30 ft
through a flume into a 25-ft x 140-ft rectangular upflow pre-
cipitating basin utilizing internal flocculation and baffling.
Precipitated calcium carbonate, magnesium hydroxide, and clay
turbidity settled into V-shaped sludge collection troughs
running the length of the basin.
Basin sludge was normally pumped to the carbon-
ation cells at 8-10 GPM with a solids content of 5-10%.
Carbonation to pH 7.0-7.2 was accomplished in four 10 ft^
cells (Galigher #24 flotation cells ) with either pure or
18% C02, selectively dissolving the magnesium hydroxide into
soluble magnesium bicarbonate. A 20-ft-diameter thickener
was used for solids-liquid separation. Magnesium bicarbonate
liquor flowed by gravity to the 1500-gal. storage tank prior
to recycle to the rapid mix at controlled rates. A rotameter
was used to measure recycle flows. The thickened sludge at
15-30% solids was normally pumped to the existing lagoon.
Sludge was collected by gravity using drilled pipe
at the bottom of the V-shaped troughs. The thick sludge
tended to plug the collection piping. A sophisticated back
flushing system operating on a time cycle was used to reduce
Product of the Galigher Company, Salt Lake City, Utah
19
-------
LIME
RIVER
WATER
TO
STABILIZATION
AND FILTRATION
RAPIO
MIX
10 MGD PERMUTIT BASIN
30 x 40' STEEL BUILDING
-^1 CARBONATION
PRODUCT
TAILS
FLOTATION
SLUDGE
THICKENEI
V 20' 0 ,
VACUUM
FILTER
Q Q^(hc°3i
storage
HEAT
EXCHANGE!
OVERFLOW
OVERFLOW
TO WASTE
FIGURE 2. PILOT PLANT FACILITIES.
/
-------
TABLE 3. 1975 AVERAGE CHEMICAL DOSAGES AND CURRENT COSTS
JOHNSON COUNTY, KANSAS
Chemical
Dosage (mg/1)
Cost ($/Ton)
Chlorine
Pre 5.4
Post 3,1
Alum
Presedimentation 0.2
Plant 1.3
Lime 135.0
Soda ash 13.8
Sodium Aluminate 2.0
Polymer
Presedimentation 0.9
Plant 0.7
Carbon Dioxide 20.4
Carbon 1.5
Potassium Permanganate
Presedimentation 0.4
Poly Phosphate 0.4
Fluoride 1.2
200
200
126
126
38
77
412
898
898
38
420
1234
528
294
21
-------
plugging problems. The system back flushed for a period of
seconds, and the flow valved to waste to remove the clear
water remaining in the pipe. When the timer was carefully
set, little, if any, sludge was wasted; however, if out of
sequence by only a few seconds, hundreds of gallons of sludge
were wasted from the system. This loss of sludge represented
magnesium lost from the system.
The level of the sludge blanket was maintained
to not exceed a pre-determined maximum height. Sludge was
wasted when this level was exceeded.
Flotation Circuit
During flotation studies a 3-ft-diameter by 1-ft-
wide vacuum filter dewatered the thickened sludge, with the
filtrate returned to the magnesium bicarbonate storage tank.
The resulting filter cake, varying from 40-60% solids, was
reslurried in two 1.5-ft^ cells followed by two similar cells
where flotation reagents were added. Diaphragm solution
metering pumps were utilized to feed reagents: frcm fcur 55-gal.
chemical mixing tanks.
Flotation was performed in three stages, one rougher
stage and two cleaner stages, with each stage consisting of
four 1.5-ft3 cells. Air was supplied at 6.7 cfm per cell (or
80 cfm to the flotation cells) by a centrifugal blower. A
process flow diagram of the flotation circuit is shown in
Figure 3. The conditioned feed flowed by gravity to the
rougher cells. Rougher concentrate overflows the cells as
froth is collected by launders and pumped to feed the first
cleaner. The rougher tails, containing clay turbidity and
calcium carbonate not collected, are wasted. The first cleaner
concentrate is collected by launders and pumped to feed the
second cleaner. The second cleaner concentrate is collected
by launders and, in this case, is the final flotation product.
The final product is pumped to the 10-ft-diameter thickener
for storage. All cleaner tails flow counter current, where
remaining calcium carbonate may be collected. The rougher
tails were wasted to the sewer. Following each flotation
study, the product could be dewatered for further evaluation
using the vacuum filter.
The flotation circuit was capable of processing
from 150-200 lb/hr of feed solids on a dry basis.
22
-------
ROUGHER FEED
WATER
3-4 GPM REAGENTS
CAKE FROM
VACUUM
FILTER
i
RESl URR>
i
1st CLEANER
CONCENTRATE
CONMTIO
S :CON
CLE iNER
K)
OJ
PRODUCT TO
THICKENER
-3
» , t U t , t {
1RST
CLE)
NER
ROl
Jj HER
*
TAILS
ROUGHER CONCENTRATE
O
FIGURE 3. PILOT FLOTATION FLOW DIAGRAM.
-------
Magnesium Recovery
During studies of magnesium carbonate recovery, the
magnesium bicarbonate liquor was valved to flow through a
hot water heat exchange unit where the liquor was warmed to
30-45°C. The heated liquor, piped to the flotation cells,
was air stripped of the carbon dioxide. Aeration at the proper
temperature range precipitated magnesium carbonate trihydrate.
Recycle of the treated liquor was employed for particle seeding.
Precipitated magnesium carbonate trihydrate was thickened and
vacuum filtered for further testing and analyses.
DESCRIPTION OF PILOT PLANT STUDIES
Water Treatment Phase
Magnesium sulfate was fed initially until sufficient
sludge had accumulated to allow recycle. Once sufficient sludge
inventory was achieved, treatment parameters were held relatively
constant. Water district plant personnel monitored the treat-
ment process, along with basin influent and effluent water
quality parameters. A typical daily operating log indicating
the parameters monitored is given as Figure 4. A solids balance
around treatment units was calculated from daily average oper-
ating data. This solids balance provided a means of sludge
inventory and flow control to the various units.
Seasonal raw water quality variations and temperature
fluctuations were encountered in the course of the study.
Generally, no other water treatment process variations were
attempted.
Studies were conducted to determine the effects
of sludge temperature, carbon dioxide purity, retention time,
and final pH on the rate and efficiency of magnesium hydroxide
solubilization in sludge carbonation. Thickening tests were
performed on basin sludge before and after normal carbonation,
and vacuum filter leaf tests were conducted on the thickened
sludge.
Flotation Studies
Open circuit bench and continuous pilot flotation
studies were periodically conducted on the carbonated,
thickened and vacuum filtered sludge. The objectives of the
flotation studies were to: determine the degree of benefi-
ciation achievable and the amount of product recoverable
using the flotation process? determine chemical dosages,
process control parameters, and pretreatment requirements,
if any; and develop design data to optimize the flotation
process.
24
-------
Softening
Basin
Sludge
Date
Rapid
Carbonator
Chemicals
Cartoonator
Thickener
Basin Influent
Chemical8 Applied
Nix
Basin
effluent
Influent
Applied
Effluent
Under1
low
Plow
ItJ°l
CaH*lNgHb
Turb.
Lime
847A
Mg(HCO,) ?
P«
pH
Caul MgH
Total
Turb.
PH
Dry vt.
C02
Mr
PH
Total
Alk
* Dry
Tiae
NGD
"¦q/1
as CaCOr-"
FTU
•Q/1
«J/1
Lnvel
OTtt
oa/lCaCO
FTU
FTU
Solids
cfm
cf»
lag/1)
wt.solids
0100
2.0
J
170
0.21
4' 7"
9
11.6
12
0
7.1
0200
2.0
170
0.21
4'9"
9
11.4
11
4
140
12
7.4
12
0
7.1
0300
2.0
170
0.21
S'2"
10
11.4
12
0
6.9
0400
2.0
170
0.21
4'6"
10
11.4
11
3
148
4
6.6
0.8
12
0
7.1
0500
2.0
170
0.21
4 * 1"
10
11.4
12
0
6.9
0600
2.0
170
0.21
3*8"
10
11.4
11
3
128
16
7.5
1.3
12
0
7,0
0700
2.0
132
140
44
170
170
0.21
4*0"
8
10.5
12
0
6.9
0800
2.0
170
0.21
4'2"
9
11.3
11
3
130
20
7.5
1.7
12
0
6.9
9000
8.6
0900
2.0
170
0.21
4'2"
9
11.3
12
0
6.9
1000
2.0
170
0.21
4'2"
9
11.5
11
2
116
24
6.4
1.6
12
0
7,0
1100
2.0
170
0.21
4* 3"
9
11.4
12
0
6.9
1200
2.0
170
0.21
4' 4*
9
10.5
11
3
118
24
7.6
1.5
12
0
6.9
11000
1300
2.0
170
0.21
4' 4"
9
11.3
10
0
6.8
1400
2.0
170
0.21
4'4"
9
11.3
11
2
112
28
6.2
1.4
10
0
7.2
1500
2.0
170
170
40
45
170
0.21
4'4"
9
11.1
10
0
7.2
1600
2.0
170-
0.21
4'4"
9
11. 3
11
2
102
28
6.3
1.3
10
0
7.3
8400
1700
2.0
170
0.21
4*4"
9
11.2
10
0
7.2
1800
2.0
IBS'
0.21
4'5"
9
11.2
11
2
96
32
4.7
1.2
10
0
7.5
1900
2.0
185
0.21
4'5"
9
11.0
10
0
7.3
2000
2.0
185
0
4' 6"
10
11.3
11
2
116
12
4.5
2.8
5.0%
10
0
7.3
9600
2.9
10.6
2100
2.0
185
0.21
4*4"
10
11.3
10
0
7.3
2200
2.0
185
0.21
4*0"
10
11.5
11
2
118
34
6.2
1.8
10
0
7.5
2300
2.0
164
202
38
45
185
0.21
3*8"
10
11.1
10
0
7.4
2400
2.0
185
0.21
3'6"
10
11.3
11
1
120
26
7.5
1.7
10
0
7,4
7400
a - c«
lciuiB
Hardness
b
- Hag
rvesiua t
&rdne<
s
AVERAGE
162
173
41
07
174
0.20
9.3
11.2
11
2
120
22
6.5
1.4
5.0*
11
0
7.2
9080
2.9
9.6
Figure 4. Daily Operating
Record
-------
In the normal operation of the Permutit unit,
a relatively long sludge retention time (approximately
20 days) was maintained in the settling basin and thickener.
This storage capacity provides a sludge of relatively constant
character over short study periods. A wide range in sludge
calcium carbonate purity was encountered, resulting from fluc-
tuation in raw water turbidity. Ranges in particle size
distributions, residual magnesium contents, water temperatures,
and levels of impurities and interferences were also experienced
in the course of the project.
Process control variations studied included
collector type and dosage, pH, pulp density, pulp temp-
erature, number of cleaning stages, residence time in
stages, modifying and dispersing agents and dosages, and
artificial as well as background impurities and interfer-
ences .
In order to further evaluate product purity,
5-10 gram samples of the flotation product were dewatered,
dried, and ignited at 1900 °F for four hours using a lab-
oratory muffle furnace. Pure analytical grade calcium car-
bonate was also calcined under the same conditions for
comparison.
Magnesium Recovery
Magnesium was recovered as magnesium carbonate
trihydrate on several occasions during the study period.
The recovered product was analyzed for type and amount of
impurities.
PROJECT LIMITATIONS
While the existing Permutit precipitator saved
considerable construction expense and time, several problems
resulted from its use. The unit was designed to treat a
total of 20-MGD, 10 million in each side. The flow metering
and control device was, therefore, sized to handle a much
higher rate than the normal 2-MGD treated during the study.
The flow control, at 2-MGD, varied depending on the head
available from the presedimentation tanks. This problem
was partially solved by increasing the water treatment flow
to 4-5-MGD. In that range, more accurate metering and control
was found.
As discussed previously, the sludge withdrawal
system did not provide positive solids removal from the
basin. As the water and solids balance is extremely
important from an experimental standpoint, the desludging
system proved to be a major deficiency. Accurate inventory
26
-------
of solids was impossible to maintain. Also, sludge wasting
during backwashing of the collection system occurred fre-
quently. These problems would not be associated with a normal
clarification system.
The deficiencies in the sludge withdrawal system
caused low solids in the sludge underflow such that the
average recycle alkalinity was only 6,000 mg/1. In previous
studies with a positive sludge collection system, carbonated
alkalinities of 15-16,000 mg/1 could be easily obtained.
Figure 5 represents a relationship between carbonator feed
sludge solids and magnesium level obtained after carbonation
assuming 13.8% Mg (OH)2 in the sludge.
As discussed previously, no attempt was made to
monitor the process after clarification. It would have been
desirable to include two-stage stabilization and filtration
for comparative purposes. Previous projects have shown,
however, that successful stabilization and filtration are
readily achievable assuming that the water has been properly
treated in prior units. Present single-stage stabilization
caused the effluent from the demonstration plant to have a
much higher hardness than effluent from the conventional
process. This added hardness results from the dissolved
calcium, at the elevated pH, not precipitating during carbon-
ation. The inability to properly stabilize the pilot plant
settled water caused a slight increase in the water hardness
from the plant. Laboratory stabilization studies were used
to evaluate water quality following two-stage stabilization.
The turbidity in the pre-settled water was much
lower than had been experienced in previous years. As
discussed, the lower turbidity resulted from more careful
coagulant control in the presedimentation process. Blending
of well and river water was not found to be necessary to
control water quality. In fact, the suspended solids were
below desired levels during most of the study.
27
-------
20.0
_ 15.0
CO
O
O
co
o
10.0
5.0
1 f i i > f i i I i i i i i I I
0 5.0 10.0 15.0
% SOLIDS IN FEED
FIGURE 5. CARBONATOR FEED SOLIDS vs. MAGNESIUM
THE RESULTING RECYCLE.
28
-------
I
SECTION V
RESULTS AND DISCUSSION
Operation of the water treatment units began the
week of January 2, 19 75. Carbonation of basin sludge began
the following week. Flotation studies did not begin until the
month of February.
Weekly summaries of the operating records are included
as item A in the appendix. The inability to accurately
measure and control raw water flow caused erroneous data re-
porting in the early study phases. Increased flow, beginning
in August, allowed more accurate flow measurement and subsequent
chemical dosage calculations. The chemical feed data reported
in Appendix A was unavoidably subjected to considerable error
and can only be used cautiously. Poor correlation between
soluble magnesium determination using EDTA titration and atomic
absorption methods was found. Titration technique and testing
procedures were carefully checked and found to be satisfactory.
Residual magnesium levels reported in the appendix often contra-
dict previous study findings (8,9). Analytical interferences
are suspected; thus caution is advised when attempting to
correlate residual magnesium with coagulation pH. Of the more
than 30 raw waters studied to date, the Kansas River water
presented the only problem achieving a reasonable magnesium
determination using EDTA. Additional studies using the various
inhibitors available could result in more accurate EDTA titration
for residual magnesium.
Figure 6 summarizes the water treatment data on a
monthly basis. Excellent turbidity removal was obtained through-
out the study period. As practiced in previous studies, acid
turbidities were used to measure true residual turbidity not
removed in coagulation and clarification. Conventional turbidity
analysis was also performed and reported. The difference
between the two turbidity values is caused primarily by suspended
calcium carbonate, which is readily removed by filtration and
does not significantly shorten the length of the filter cycle.
Acid turbidity is highly correlated with the filtered water
turbidity; thus, is an excellent indicator of coagulation-
sedimentation efficiency.
Inventory of solids in the various operating units
was maintained by mass balance calculations. These balances
proved to be of assistance primarily from an operational stand-
point. The solids balance summary sheets are shown as Item B
29
-------
IPRECIPITATEI
SETTLED RIVER
EFFLUENT I ACID TURBIDITYl
FIGURE 6.MONTHLY AVERAGE TURBIDITY AND MAGNESIUM DOSAGE.
-------
in the appendix. The balances shown were calculated from raw
water quality and recycle data. The contributors to production
of calcium carbonate, suspended solids, and magnesium hydroxide
were identified. The turbidity to suspended solids ratio was
found to be approximately 1.
The literature documents the effectiveness of magnesium
hydroxide for removing dissolved silica. Silica removal is impor-
tant, particularly for industrial boiler and cooling tower uses.
A study was conducted between October and January to evaluate
the effectiveness of the recycled magnesium on silica removal.
Silica was measured on the settled river water, Permutit basin
effluent, and on three occasions on conventional treatment
effluent. Table 4 compares silica levels in the settled river
and effluent from the Permutit basin. Silica removal averaged
72% over the two-month study period. The conventional treat-
ment provided an average of 28% removal of silica.
Silica concentrations in the recycle liquor increased
to 1,650 mg/1 before an equilibrium was reached. Silica removal
efficiency in water treatment was apparently not reduced as a
result of the high level of silica in the recycled magnesium
bicarbonate liquor.
Although silica removal with magnesium hydroxide has
been used successfully many times, the ability to recover
the magnesium, recycle the coagulant for hundreds of cycles,
and effectively remove raw water silica is an important finding.
Considerable expense is required for silica removal
for industrial users, thus simultaneous removal of silica with
water treatment may be of major interest.
The excessive magnesium loss in the thickener blow-
down required initial make-up feed of magnesium sulfate. This
magnesium source, expressed as calcium carbonate, costs in
excess of $.30/lb of magnesium. Figure 7 illustrates the effect
of thickener underflow solids, which are controlled by pumping
rate, and magnesium loss. Based on the assumptions shown, an
additional make-up cost of $60/day would be required if 15% solids
were pumped, as opposed to 40%. During the study period an
average of 25 mg/1 of magnesium as calcium carbonate was pre-
cipitated from the raw water; thus, a magnesium loss of 416 lb/
day could be accepted without need for make-up. Make-up magnesium
was not required when thickener underflow pumping rates were
carefully controlled.
31
-------
TABLE 4. SILICA REMOVAL USING RECYCLED MAGNESIUM
Silica (mg/1 Silicon)
Permutit
Date Settled River Effluent % Removal
10-5-75
30.7
6.7
78
10-12-75
16.7
8.4
50
10-19-75
24.7
12.8
48
11-9-75
47.8
11.7
76
11-16-75
43.7
21.7
50
11-30-75
76.0
12.0
84
12-14-75
78.0
38.0
51
Note: All samples are 24 hr composited. Samples were
treated with acid prior to Silicon determination
by atomic absorption analysis.
32
-------
NOTE: ASSUME THICKENER FEED SOLIDS
OF 10% AND SLUDGE PARTICLE
SPECIFIC GRAVITY OF 2.65
% SOLIDS IN UNDERFLOW
FIGURE 7. EFFECT OF SLUDGE SOLIDS IN UNDERFLOW
ON MAGNESIUM LOSS.
33
-------
Carbonation Studies
A series of experiments were conducted to determine
the effect of carbonation pH on undissolved magnesium using
pure and 20% carbon dioxide. In these studies, the carbon
dioxide and air rotameters were adjusted, 30 min allowed to reach
equilibrium, and a sludge sample collected from the carbonator.
The slurry was filtered, the resulting cake dried to 103°C,
and a representative 1-gram sample acidified with IN HC1.
After dilution to one liter with distilled water, magnesium
was determined by atomic absorption analysis. The results
were expressed as milligrams of magnesium per gram dry cake.
It is recognized that this procedure measured
magnesium present in the clay turbidity which would not be
recovered by carbonation. On three occasions, sufficient
raw water was filtered to obtain enough dry solids to analyze
as described above. An average of 10 mg/gram magnesium was
found in the suspended solids and was treated as a blank to
be subtracted based on the relative percentage of silt present.
As an example, assume a 1 gram sample taken from the carbonator
consisted of 0.7 gram calcium carbonate and 0.3 gram of raw
water contaminant. A blank of 0.3 x 10 mg/1 or 3.0 must be
subtracted from the magnesium value obtained to determine the
true magnesium remaining after carbonation.
Figure 8 illustrates typical relationships found.
These results would indicate a pH of 7.1-7.2 as the desirable
operating range when using pure carbon dioxide.
The pH of the slurry during carbonation is in dynamic
equilibrium with the system conditions. A sample collected for
pH determination slowly reaches equilibrium with atmospheric
conditions. As the partial pressure of carbon dioxide is con-
siderably lower, the pH will slowly rise, reaching 8.0 or higher,
dependent upon the alkalinity present. This pH increase is exper-
ienced in actual operation as the pH of the thickener underflow
sludge typically is 7.5-8.0. The pH values shown in Figure 8
closely represent the process control pH, as the samples were
quickly analyzed. The process control pH, when using 20% car-
bon dioxide, would be expected to be slightly higher than for
pure carbon dioxide.
Carbonating more concentrated sludges to produce
magnesium levels near equilibrium, could result in precipi-
tation and loss of soluble magnesium as a result of the pH
increase. A covered thickener, using a carbon dioxide blanket,
would prevent the carbonated slurry from contact with the atmos-
phere, Capital costs are added and may not provide advantages
34
/
-------
11.0-
10.0-
9.0-
8.0-
7.0-
o>
5 6.01
CO
u>
Ui E
5 5.0i
B
CD
S
4.0-
3.0-
2.0
1.0'
6.0
—r~
6.5
T
7.0
—I-
7.5
8.0
T=20 C
Initial Mgs 26.3 mg/gram as Mg
Mg determined by atomic absorption
X-PURE C02
0-18% C02
T
8.5
~T
9.0
9.5
PH
FIGURE 8. MAGNESIUM REMAINING IN SLUDGE CAKE AT INDICATED CARBONATION pH
JOHNSON COUNTY, KANSAS
-------
when carbonation is below saturation levels. Problems with
reprecipitation during thickening were not experienced with
magnesium levels of 8r000-10,000 mg/1 as calcium carbonate.
Using 20% carbon dioxide, foaming must be controlled
by water sprays. The water sprays diluted the magnesium
bicarbonate liquor by approximately 10% in this study. The
problems with controlling carbonation sludge density and sub-
sequent recovered magnesium concentrations required increased
carbonation-recycle flow rates to maintain the desired
coagulant level. Further dilution by required carbonation
water sprays would have further taxed the recycle system's
hydraulic capacity. Twenty percent carbon dioxide was not
used routinely for this reason. However, a number of studies
were conducted which demonstrated that either gas concentration
would effectively solubilize the magnesium at the operating flow
rates (see Figure 8).
Carbonation temperatures as high as 28°C were
common during summer months. The chemical reaction is slightly
exothermic? however, less than 1°C rise was experienced when
10 g/1 or less magnesium was solubilized. When carbonating
below saturation values, problems with precipitation of mag-
nesium carbonate were not experienced using either pure or
20% carbon dioxide although residual magnesium values in-
creased slightly. During July, the average residual magnesium
was 3.2 mg/gram as opposed to 1.6 mg/gram in December, an in-
significant difference. Precipitation could be expected to
cause problems at elevated carbonation temperature if operation
near saturation levels was required. Sludge chilling would be
required for successful operation under these conditions.
Solids Handling Studies
Bench scale thickening studies were conducted through-
out the year on both carbonated and uncarbonated sludge. The
results were analyzed using the Talmadge and Fitch method (30)
and all results obtained are plotted in Figures 9 and 10. The
upper and lower test results were plotted with the shaded area
between encompassing all other tests. The wide range of initial
sludge concentrations and the different sludge particle sizes
encountered during the study explain the divergence of sludge
thickening characteristics. Thickening characteristics would
be expected to improve as the sludge particle size increased.
Carbonated sludge was found to thicken more readily
although feed solids concentration was extremely important.
Results obtained were similar to those reported in previous
studies. A loading rate of 40-50 ft^/ton/day should produce
an underflow jf greater than 30% solids.
36
-------
130*
120-
110H
100^
< 90
80
70
60-
50-
I . 40
H <
30-
20-
10-
Cqs 4.6 %
NOTE: C0= INITIAL SOLIDS CONCENTRATION
Cn=l2.1 %
10
15
—J—
20
—r~
25
—r~
30
—r—
35
—r~
40
—r~
45
—i—
50
UNDERFLOW SOLIDS %
FIGURE 9. SUMMARY OF THICKENING CHARACTERISTICS OF
CARBONATED SLUDGE.
37
-------
160-1
= 4.2%
140
120
>
<
o
V.
o 100
I
UJ
OH
2
uj O
^2
Oq
X<
K O
-I
<
Ui
cc
<
80-
60-
40'
20-
NOTE:C0 = initial solids concentration
C0= 13.7%
—I™
10
15
—I-
20
25
30
35
UNDERFLOW SOLIDS %
FIGURE 10. SUMMARY OF THICKENING CHARACTERISTICS OF
UNCARBONATED SLUDGE.
38
-------
Vacuum filtration studies were conducted on both
a batch and continuous scale. Leaf filter batch tests were
found to closely correlate with results obtained with the pilot
vacuum filter. Filter rates ranged from 10-25 lb/ft^/hr.
Filtration rates increased with an increase in calcium car-
bonate particle size and grade,
FLOTATION STUDIES
Initial flotation work began the week of February 10,
19 75. Typically, one week out of each month was directed to on-
site flotation work. During the study period, 18 sludge samples
were collected for flotation evaluation. A summary of all flo-
tation results is shown as Item C in the appendix.
Pilot vs. Bench Scale
Bench scale flotation is used to evaluate flotation
parameters on a batch basis just as jar tests are used to eval-
uate water treatment processes. Generally, several hundred grams
of dewatered sludge are floated in a 3-liter cell. All process
variations can be studied, requiring only a fraction of the time
and materials that a similar pilot scale study would require.
Impellor speed, air flow rates, and pulp temperature are readily
controlled. Typically, 10-15 sets of operating conditions
could be studied each day.
Pilot scale testing generally required 3-4 hours
for each series of process conditions. Typically, 100-150
lb/hr dry solids were dewatered and reslurried for flotation.
Considerable effort was necessary to 'balance' the process.
Samples were collected every 30 minutes and composited over
the duration of the run. Generally, only two sets of flotation
operating conditions could be studied each day.
Obvious advantages of the bench scale studies en-
couraged greater use. It was important initially to demon-
strate that the bench scale accurately reflected results ob-
tained in the pilot scale. When treating similar sludges
under the same process conditions, the concentrate grades and
recoveries were found to be very nearly the same. This proved
as expected based on experiences* reported in the mining industry.
Table 5 summarizes flotation data collected during
the months of February and March. Comparing pilot versus bench
scale results, note the close agreement between the final grade
obtained in the number 2 cleaner, which is the product. This
comparison can be made for all pilot and bench scale data, sum-
marized as Item C in the appendix.
* Communication with the Galigher Company
39
-------
TABLE 5. SUMMARY OF FLOTATION RESULTS
February and March
Bench Scale
Sample
«
Feed
% Solids pH
Chemicals #1 Ton
Silicate Soda Ash Dresinate
Feed
% CaCOj - by Acid Base
Tails
RoCon #1C1. #2C1.
2/10-2
2/10-4
2/24-1
2/25-1
2/25-2
2/27-2
3/17-1
O
10
10
10
10
10
9.1
9.0
9.3
9.6
2.6
2
1
2a
2
2
4
89.7
61.3
96
89.7
62.9
97
86.3
89.6
86.3
69.0
92.2
86.3
97.8
75.0
60.0
88.8
91.8
83.3
95.1 95.7
98.3
a _
Mixture of Dresinate a:id Sodium Alkyl Sulfate (1:1)
Pilot Scale
2/11/75
2/13/75
2/14/75
2/25/75
3/19/75
8.7
8.95
8.90
0.67
5
4.8
6.7
5.5
1.25
3.3
2.7
b - Emulsion of 1% Dresinate and 0.05* Sodium
Alkyl Sulfate
1.53
87.1
82
—
93
95
8
87.2
85.1
90.8
92.3
95.1
lib
94.8
68.3
94.8
98.5
—
7.04
87.3
80.8
89.3
90.8
98.8
5.5
89.7
87.5
90.2
93.0
98. 3
-------
While the data for the chemical feed and grade of
the tails shown in Table 5 do not compare favorably, the cause
of this difference will be discussed in following sections.
Product grade was found to be the determining factor
for successful flotation. Recovery and chemical efficiency
were secondary considerations as improvement could be made by
process optimization. All data collected support the con-
clusion that successful bench scale results indicate success
for the pilot flotation studies.
The pilot scale studies were necessary in order to
produce sufficient material to study thickening, clarification,
and filtering of the concentrates and tails. In the latter pro-
ject stages, this was the primary function of pilot scale studies.
Effect of Calcium Carbonate Particle Size on Flotation
Early flotation study results indicated little dif-
ficulty in producing high concentrate grades with modest re-
covery. The studies conducted in February and March generally
found concentrate grades of 97-98% possible as shown in Table 5.
Recovery was usually found to be less than 70%; however, initial
concern was producing an acceptable grade.
In April studies, the results indicated that there
was no collector selectivity. The concentrates often were
lower in calcium carbonate than in the tails. The froth
physical character had drastically changed. Initial efforts
were directed to various chemical feed combinations. Review
of the data indicated that the level of residual magnesium had
increased from approximately 3.0 mg/1 to approximately 6.0 mg/1.
Efforts were then directed to reduce the residual magnesium
by a number of means. The filter cake was washed, the carbona-
tion was more carefully controlled, the carbonation temperature
was artificially reduced, the slurry was washed and repulped,
the magnesium level was reduced using ion exchange resin, plus
other treatments. All efforts found only marginal success in
lowering residual magnesium and offered no improvement in
flotation results.
Beginning in June, the particle size of the sludge
was determined using a Coulter Counter* along with microscopic
examination. There are a number of descriptive denominators
for particle size; however, D^q, which is the size exceeded by
90% of the particles, appears to be unusually important. Other
descriptors include and D90 which are defined in a
similar manner. The uniformity coefficient Cu (Dgo/D10)
describes how uniform the particles are in size. Results of
all particle size analyses are shown in Table 6 along with
feed and product grades found in flotation studies.
*a product of Coulter Electronics Co., Inc., Hialeah, FL
41
-------
TABLE 6. PARTICLE SIZE ANALYSTS SUMMARY
JOHNSON COUNTY, KANSAS
Particle size in Microns (u)
Sample
Max. Min.
d65
Carbonated Sludge
(6/9)
Carbonated Sludge
(6/12)
Carbonated Sludge
(8/28)
Lagoon Sludgea
Accellator Sludgea
(9/28)
Accellator Sludge3
(10/30)
Carbonated Sludged
(10/30)
Carbonated Sludge3
(11/17)
Carbonated Sludge a
(12/15)
60
20
16
25
40
40
65
25
33
33
0.8
0.8
0.8
2.9
1.6
5.6
4.2
6.1
18.2
8.2
1.0 16.2
1.0 14.5
3.7
5.4
1.5 15.5 14.4
2.5 18.0 16.8
17.0
7.8
15. 3
13.H
'10
Uniformity
Coefficient
1.5
1.2
1.8
4.4
5.3
6.7
3.8
5.6
4.9
3.33
3.22
3.0
3.27
3.17
2.54
2.08
2.73
2.85
l- Sludge recirculation practiced
-------
A sample of lagoon sludge, primarily resulting from
conventional treatment, was collected for bench scale studies.
Following carbonation to remove residual magnesium, excellent
selectivity and recovery were found, A product grade of 97%
with recovery as high as 91% was found. It was noted that the
particle size distribution was considerably larger than previous
sludges that were not successfully floated. The magnesium
residual after carbonation was 7.9 mg/gram, thus, refuting the
argument that high residual magnesium caused flotation inter-
ference.
It was reasoned that an aging process had resulted
in the crystal growth and possibly altered the form of the
residual magnesium, eliminating the interfering action.
A sample of sludge directly from the Accelator
was collected on 8/28. This sludge was carbonated and de-
watered prior to bench scale flotation study. Results in-
dicated that the calcium carbonate was readily floatable.
Again, particle size data indicated this sludge to be sub-
stantially larger than the carbonated sludge produced in
the pilot plant.
Early in September, the rapid mixer motor on the
Permutit system failed. Inspection found the mixer to be
heavily coated with calcium carbonate indicating that sludge
recycle had not been practiced. Inspection of the recycle
pump found it out of service. Sludge recycle has been used
to prevent encrustation on mixing equipment and as discussed
in Section 1, particle size is significantly influenced by
sludge recycle.
The recycle pump was repaired, the basin cleaned,
and the system allowed to reach equilibrium. Sludge samples
collected had a D^q of 3.8/1, considerably higher than pre-
vious samples having a Diq of 1.5 p. In flotation studies,
once again, collector selectivity was found and the calcium
carbonate was easily separated from the sludge.
A brief study was made to distinguish between
calcium carbonate and raw water contaminant in the size
analysis. A sample of sludge collected on June 12 was
separated into two representative samples. One sample was
treated with acid to dissolve all of the calcium carbonate
present. This sample was then neutralized with sodium
hydroxide to pH 7.0. Both samples were analyzed for particle
size distribution with the results shown in Figure 11. These
results indicate that, in the upper size range, the sw water
contaminants contribute only to a minor extent. For comparative
purposes, particle size analysis for a sample collected after
sludge recirculation was reinstated (10/30) is included.
43
-------
4^
c
u
M
U
L
A
T
I
V
E
P
E
R
C
E
N
T
P
A
S
S
I
N
G
100-
6-12-75 ACIDIFIED
D10 =1.06
Cy-2.02
D10-1.15
Cu=3.22
SILT
CLAY
6-12-75
10-30-75
D10 = 3.7
500.0
100.0
t—r
50.0
10.0
SIZE IN MICRONS
FIGURE 11. PARTICLE SIZE ANALYSIS COMPARISON
i
-------
Effect of Collectors, Modifiers/ and Pulp pH
Chemical feed variations were studied in all flota-
tion experiments; however, the work with the sludge collected
on December 15 was the most extensive.
Three types of Dresinate* collector were studied:
81, TX-60W, and DS-60W. A general comparison of properties
and costs is included in Table 7. Figure 12 illustrates
the product recovery effectiveness of the collectors at
various feed rates at the conditions stated. It was found
that product grade was relatively unaffected by either level
or type of collector studied. Economics and performance
would dictate that TX-60W be selected as the logical choice
of collectors.
The level of collector added greatly affects
recovery, but has little effect, if any, on grade. From
a practical standpoint, 4-5 lb collector/ton dry solids to
be floated is the maximum feed rate which can be handled.
Increased collector causes severe foaming in the float.
Staged feeding of the collector to the rougher was found
helpful.
The adjusted pH of the conditioned pulp was found
to drastically affect product recovery, but demonstrated only
a minimal effect on grade. As illustrated in Figure 13, an
optimum pH range of 9.0-9.5 was found, generally in agreement
with the reported literature. As discussed in Section I, this
is the pH range of minimum zeta potential for the calcium car-
bonate as well as the pH necessary to insure that the Dresinate
be in the soap form.
The effect of sodium silicate addition to condi-
tioning was somewhat difficult to evaluate. As discussed
previously, silica removal was accomplished in the water
treatment process. Some of the silica was solubilized on
magnesium recovery. The silica level in the recycle stream,
however, stabilized, thus not requiring 'blow-down1 of
recycled magnesium bicarbonate. The silica present in the
recycle reached 1,650 mg/1 as silica during most of the
period of flotation study. Moisture remaining in the filter
cake would add approximately 2 lb silica/ton dry solids.
It was, therefore, impossible to evaluate the effect of
*
Products of Hercules, Inc., Wilmington, Delaware
45
-------
TABLE 7. DRESINATE COMPARISON
Dresinate 81
Dresinate TX-60W
Dresinate DS-60W
Description
87% solids, modified aqueous
solution of sodium soaps of
dark rosin and pale processed
or modified rosins
60% solids, aqueous solution
of a fully saponified sodium
soap of tall oil
60% solids, aqueous
solution of a sodium
soap of a specially
processed tall oil
fraction.
Typical Use
Mainly as an emulsifier in-
gredient in solvent cleaners
Soap emulsifier in manu-
facture of road surfacing
asphalt anionic emulsions
(RS and MS types)
Low cost, anionic
flotation reagent
Free Alkali
Content, % dry
basis NaOH
0.5 Max
2.0 Max
Viscosity (Stor-
mer), poises at
60 °C
5.7
5
30
Color
Brown
Dark Brown
Dark Brown
pH of a 25%
aqueous solution
-
-
11
Availability
Carload, or drum
Bulk only
All
Cost $/per 100 lb
26.15 (Carload)
10.45 (Bulk)
5.00 (Bulk)
FOB Savannah, Ga.
-------
100
90
80-
70*
60
50
GRADE
/
/
/
cc 40*
30
20'
10-
81
TX-60W
DS-60W
05 LB/T0N,SILICA
pH- 9.0
1 2 3 4 5 6 7 8
LEVEL OF COLLECTOR (lb/TON)
10
—r-
11
12
FIGURE 12.COMPARISON OF RECOVERY EFFICIENCY AT
INDICATED COLLECTOR LEVEL AND TYPE.
47
-------
100
90
80
70
60
50
40
30
20
10
0
Grade
Recovery
Temperature - 22°C
/ton
SiO-
Collector-2-8*/ton? ores TX 60W
Bench Scale
-i 1 i 1 1 i v i
6.5 7 7.5 8 8.5 9 9.5 10
pH UNITS
13. EFFECT OF pH ON CaC03 FLOTATION.
-------
zero silica; however, it is concluded that added silica does
not increase, and probably decreases, calcium carbonate col-
lection. Silica is an effective dispersant which can be used
to disperse the clay particles. Unfortunately, silica has
been shown to also disperse the calcium carbonate; thus, the
overall effect is often lower collection efficiency.
A series of bench scale experiments was performed
to investigate the effect of added magnesium bicarbonate on
the flotation process. The tests were performed on the pilot
plant sludge sample collected December 15, 1975, having a
feed grade of 86.3%. This sludge contained 3.0 mg/gram of
residual magnesium, which is quite low in comparison with the
other sludges studied.
A magnesium bicarbonate solution was used to add
increasing levels of magnesium to the pulp prior to condition-
ing. Results of these tests are shown in Table 8.
TABLE 8. EFFECT OF ADDED MAGNESIUM ON FLOTATION
Added
Magnesium
(mg/gram dry
solids)
Total
Magnesium
(mg/gram dry
solids)
Results
(%CaC03)
Concentrate Tails Recovery
o
3.0
98.4
36.1
92
3.2
6.2
97.6
57.6
81
CO
•
00
11.8
97.6
63.5
76
14.4
17.4
100.0
67.7
67
19.1
22.1
97.7
76.8
52
Note: Collector dosage - 5 lb/ton TX-60VJ
Silica - 0.5 lb/ton
pH - 9.0
These data indicate that the added magnesium reacts
with the collector, much like soap and water hardness, reducing
the collection efficiency. The added magnesium does not inter-
fere with collector selectivity, however. The form of the re-
sidual magnesium is obviously important in assessing its effect
on the flotation process. Soluble magnesium, as evaluated in
this study, would be expected to have the greatest effect on
reducing collector efficiency.
49
-------
Figure 14 illustrates the effect of pulp density
on product recovery and grade. Again, results agree with
the published data.
Temperature was-* not found to be effective in
increasing flotation performance. Perhaps residual silica
effects may have masked the effect of increased temperatures.
As discussed in Section I, silica becomes more effective in
reducing collection efficiency at elevated temperatures.
The use of carbon dioxide replacing air was
found to lower the pulp pH below acceptable levels for flo-
tation; thus, efficiency was reduced. Possibly another col-
lector, effective in the acid form, would have produced more
encouraging results.
The sludge samples collected from lagoons and the
conventional treatment plant contained activated carbon. An
increase in collector requirements could be anticipated,
although no definitive data were developed. Johnson County
has changed the feed point of carbon so that it is added
after softening and coagulation; thus will not be present in
future sludge.
Size Requirements of Flotation Circuit
Various design parameters have been used to size
flotation circuits, including: hydraulic residence time,
solids removal rate (lbs/ft^/min), solids residence time,
and hydraulic overflow rates (gal./ft2/min.). During the
pilot plant operation several water and solids balances
were performed around each flotation unit. Countercurrent
feed makes the balances and subsequent sizing of the various
units difficult. Figure 15 illustrates a typical balance
found during the pilot studies. The balance was prepared
as follows:
1) Flow measurements were made on the rougher tails,
rougher and cleaner feeds, as well as floats from
the two cleaners.
2) Measurement of the feed and float from the second
cleaner allowed calculation of underflow feed to
the first cleaner.
3) A similar procedure was used on the first cleaner
to calculate underflow to the rougher.
50
*
-------
90
80
70
60
50
40
30
20
10
0
PRODUCT GRADE
RECOVERY
FEED GRADE- 86.3 % CaC03
PULP pH - 9.0
TEMPERATURE-20°C
fill
5 10 15 20
% SOLIDS IN PULP
14. EFFECT OF % SOLIDS IN PULP ON FLOTATION PERFORMANCE.
-------
TAILS
FEED
4.1 GPM - 5.7 % SOLIDS
ROUGHER
CLEANER
TAILS
1 GPM-10.3%
SOLIDS
f
FIRST CLEANER CONCENTRATE
3 GPM-2.9% SOLIDS
FIRST
CLEANER
ROUGHER CONCENTRATE
3.0GPM- 4.9% SOLIDS
CLEANER
TAILS
1 GPM-3.7 %
SOLIDS
SECOND
CLEANER
T
PRODUCT
2 GPM-2.5% SOLIDS
Cn
fo
NOTE' DILUTION WATER ADDED TO ROUGHER
CONCENTRATE AND BY LAUNDER SPRAYS.
FIGURE 15. TYPICAL FLOTATION CIRCUIT WATER AND SOLIDS BALANCE.
-------
4) Solids determinations were made on all streams.
Solids were identified as either calcium car-
bonate or 'other'.
Using the above procedure does not produce an exact
balance. Overflow losses and solids accumulation in the junc-
tion boxes are causes of imbalance. Foaming of the condition-
ing pulp occasionally causes an overflow to the second cleaner,
resulting in contamination of the product. Fortunately,
foaming could be avoided by proper operator attention.
Bench and pilot scale study results indicate that
hydraulic residence time is the most important design criterion.
Pulp density determines solid removal rate for a given resi-
dence. Pulp densities above 15% were found to cause excess
foaming and solids settling in the launders and junction boxes
resulting in overflow from the units. These problems may be
a result of the small-scale operation and might not be found
in a full-scale application.
The pilot studies found that most of the flotation
was accomplished in the first of the two cells for each stage.
This was overcome somewhat by stage feeding the collector to
the second series of rougher cells. While the change in col-
lector feed was helpful, the froth was only lightly loaded
in the second cells indicating that the system was not loaded
to capacity. The variation in solids in the float was
progressively apparent comparing the rougher with the second
cleaner. Each stage had volume of 44 gal. Typical feed flow
of 3-5 GPM would result in a hydraulic residence time of
approximately 9-15 minutes, not considering underflow to the
rougher. Underflow was found to be usually 1 GPM or less.
Maximum solids loading with successful flotation was approx-
imately 218 Ibs/hr of dry solids for the 2 number 15 Galigher
flotation cells used in each stage.
Recovery is established by performance in the rougher
cells. Calcium carbonate^lost in the rougher tails represents
material that will not be recovered? thus design of the rougher
circuit is of primary importance.
Variations were made in hydraulic and solids handling
to determine system responses. The analysis of the data in-
dicates that the flotation circuit is relatively ir~ansitive
to these changes. At increased solids loading and flows, more
careful attention is required to balance the system. Due to
the relatively inexpensive capital costs of the flotation
system and possible scale-up problems, a design basis of 10
minutes hydraulic residence time per stage, feeding a maximum
of 15% solids is recommended. The design residence time
should be the same for each added cleaner stage. The actual
53
-------
residence time will be slightly greater for the two cleaners
and slightly lower for the rougher depending upon the amount
of dilution water added.
Grade and Recovery
Throughout the study period the feed grade ranged
from 65-95% CaCC>3. The concentrate grade ranged from 92-
100% CaCC>3 when sludge recycle was employed. As discussed,
concentrate grade appeared relatively insensitive to most
system parameters. A concentrate grade of 92% was possible
when the collector was selective. Excellent agreement between
pilot and bench scale results was demonstrated for product
grade. Slightly higher collector feed was required to main-
tain good recovery in the pilot system indicating an increased
collector efficiency for the bench scale system. Incremental
collector feed in the pilot system was obviously not as
efficient, but was required for froth control.
Recovery was affected by a number of variables
previously discussed, including; pulp pH, level and type
of collector, residence time and silicate feed. Recovery
was calculated using two methods, a weight balance and a
mathematical equation.
In bench scale tests there was excellent agreement
between methods for calculation. All tails and concentrate
produced were dewatered and analyzed for calcium carbonate.
A check comparison was also made for the feed samples. The
formula for recovery is as follows:
_ _ lOOc(f-t)
"f(c-t)
WHERE: R = Recovery in %
c = Concentrate grade
f = Feed grade
t = Tail grade
In using this formula, the flotation circuit must
be considered as a single system.
Recoveries found in bench scale tests under optimum
conditions were consistently 90-92%. In the pilot scale,
recoveries of 75-30% were typical with poor agreement between
the formula and gravimetric methods of recovery determination.
The difference between methods was partially explained by
54
-------
solids lost from the circuit or accumulation within junction
boxes. It has been shown that recovery is sensitive to oper-
ating conditions which are more difficult to control with
the pilot system. It was impossible to feed all of the col-
lector to the conditioning stage because of foaming. Although
more collector per ton of feed solids was reported for the
pilot scale, the loss in efficiency made comparison difficult.
A full-scale installation should provide more stable
operating conditions allowing more efficient collector usage.
The results obtained would indicate that under proper oper-
ating conditions, a minimum of 85% recovery should be possible.
Effect of Low Grade Feed
Most of the successful flotation tests were performed
on sludge samples having a calcium carbonate feed grade in
excess of 35%. Although the-sludge will seldom be of lower
quality, flotation of low grade sludge was undertaken for
evaluation under extreme conditions. Unfortunately, sludge
recycle failed concurrently with the seasonal rise in raw
water turbidity. Pilot plant operation was discontinued in
January, 1976; however, samples of the conventional plant
sludge were collected on April 21. River water turbidity
was 1,400 FTU, and settled feed to the treatment plant was 160
FTU• These conditions represent extremes not exceeded in
1975. Three mg/1 of sodium aluninate and 0.75 mg/1 of
cationic polymer were fed as coagulants along with 95 mg/1
of lime.
The calcium carbonate grade in the sludge collected
for study was 65%. The sludge was carbonated to dissolve
magnesium present, pressure filtered to 65% solids and re-
slurried with tap water for flotation study. Residual mag-
nesium after treatment was 7 mg/gram dry cake, subtracting
the contribution from the clay.
Flotation treatment and results are shown in Table
9. A product grade of 92% was obtained with a Dresinate
TX-60 W feed of 7 lb/ton at a flotation pH of 9.0. Added
silica appeared to reduce recovery without significantly
affecting product grade.
It is impossible to evaluate the effects of the
added coagulants, polymer and sodium aluminate. It is
probable that improved results may be found in their absence.
A considerable amount of activated carbon was present in the
sludge. Although the effect on flotation has not been demon-
strated, it is proposed that increased collector was required
because of the activated carbon.
55
-------
TABLE 9. FLOTATION RESULTS ON LOW GRADE
CONVENTIONAL PLANT SLUDGE
Number
Feed
Chemicals Addec
(lb/ton)
% CaCO,
%
Solids
PH
Silicate
Soda
Ash
Dresinate
Feed
Tails
RoCon #1 CI. #2 CI.
#3 CI.
#4 CI.
Recovery
1
15
9.0
0
6
2.5
65.0
86.6
2
15
9.0
1
6
2.5
65.0
85.2
3
10
9.0
0
4
5.0
65.0
35.3
82.7
86.2
77
4
15
9.0
0
5
5.0
65.0
44.8
75.0
78.6
72
5
10
9.0
2
5
5.0
65.0
37.3
81.3
78
6a
15
9.0
0
5
5.0
65.0
42.8
79.1
74
7
10
9.0
1
7
7.0
65.0
36.7
71.6 81.9 85.4
91.2
73
8
10
9.0
2
7
7.0
65.0
38.5
83.9
75
9
10
9.0
4
6
7.0
65.0
40.8
87.1
70
10
10
9.0
0
6
7.0
65.0
44. 3
87.5
92.0
61
11
10
9.0
2
6
7.0
65.0
55.5
91.4
92.2
37
12
10
9.0
4
6
7.0
65.0
50.1
90.6
92.0
50
a
Conventional plant s
.udge was not
pretreated by carbonation
-------
Bench scale tests were not closed cycled; thus
tails from cleaner stages were not recycled to the processing
stage. This method of testing reduces recovery with increasing
cleaner stages as the data demonstrated. Earlier studies
with higher grade feed sludges exhibited poor correlation
between feed and product grade.
Froth flotation provided acceptable beneficiation
to a sludge far lower in calcium carbonate content than would
be expected at Johnson County. Product recovery, in full-
scale operation, could be expected to exceed the 70-75% found
in the lab study.
Flotation Tails
Initial observation of the recovery efficiencies
indicated would appear confusing considering the relatively
high percentage values of calcium carbonate in the tails.
Review of a typical example will provide insight as to the
various relationships making up recovery.
Consider a feed assay of 95%, concentrate assay
of 98% and tails assay of 50% CaC03. Low recovery would
be concluded when observing the relative high grade of the
tails. The recovery equation predicts an 85% recovery. A
graphical analysis of these values is shown as Figure 16.
The histogram illustrates that while the calcium carbonate
percentage in the tails is high, the relatively low 'other'
content in the feed explains the high recovery found.
A silt-calcium carbonate sludge exhibits thicken-
ing and dewatering characteristics that are increasingly
favorable as the calcium carbonate percentage increases.
Thus, having tails with 30-40% calcium carbonate allows
the sludge to be readily thickened and dewatered.
Flotation tails will contain 20-30% of the total
sludge produced at a treatment flow of less than 0.5% of the
water flow rate.
Thickening curves for several bench scale studies
are presented in Figure 17. Leaf filter tests indicate that
3-5 lb/ft^/hr could be expected dewatering with a vacuum
filter. Clarification studies indicate that a suspended
solids content of 70 mg/1 could be expected on the thickener
overflow.
Grade and Recovery Sensitivity Analysis
The initial flotation studies were designed in hope
that regression equations could be obtained that would quantify
recovery and grade as a function of the various operating parameters.
57
-------
~ OTHER
7ZX Ca C03
.159
85g
01g
• 71 g
14g
14g
FEED CONCENTRATE TAILS
FIGURE 16.COMPARISON OF CALCIUM CARBONATE FRACTIONS
IN INDICATED FLOTATION STREAMS.
58
-------
MOTE: 0o= INITIAL SOLIDS COWCENTEATfiQtt
UNDERFLOW SOLIDS %
FIGURE 17 . THICKENING CHARACTERISTICS Or FL0TA7IC
TAILINGS
59
-------
These equations would allow prediction of flotation efficiencies
as well as provide a means of optimizing the flotation process.
This has not been the case although the data obtained will pro-
vide a substantial basis for such equations that will, hopefully,
be developed on a project soon to be initiated.
SUMMARY OF FLOTATION STUDY RESULTS
The results do allow a number of qualitative con-
clusions, which may provide additional insight into the flo-
tation process. The grade was found to be relatively unaffected
by most of the operating parameters except in the case of the
extremely low quality feed. Feed grade appeared to have only
a slight effect on the grade of the product. Again, the one
series of tests with low grade sludge were performed on an
atypical sludge in which coagulants, carbon, etc., were present.
Feed particle size has an obvious influence on grade. Of the
various particle size descriptors, D^q would appear the most
significant in determining the lower limit for successful
flotation.
Recovery was affected by a number of system param-
eters, including collector concentration and type, silica feed,
pulp pK, and pulp density. Feed grade did not appear to have
a marked effect on recovery. Feed particle size below a min-
imum size prevents collector selectivity; thus, recovery.
Sufficient data was collected to select optimum conditions
for each of these parameters; however, the interrelationship
between parameters has not been fully demonstrated.
Flotation Economics
The cost of flotation is dependent upon a number of
specific local factors. A budgetary cost estimate is made,
however, for purposes of relative comparison. The following
equipment is included in the capital cost estimate: flotation
machines, blowers recycle pumps, samplers, reagent feeders,
housing and the necessary mechanical and electrical service.
In Johnson County's case, the existing lagoons will be used
to clarify and dewater the tails, thus adding no additional
costs. Polymer costs to treat this waste stream are con-
sidered insignificant.
The flotation mechanical equipment is of modular
construction, thus tending to be easily expandable. Flotation
will be required in Johnson County only two-thirds of the year,
reflecting reduced labor and power costs. Historical water
quality data indicate that a sludge grade acceptable for
recalcination without the need for beneficiation can be ex-
pected at least four months of the year. Table 10 summarizes
the flotation costs assuming a design and operating capacity
60
-------
TABLE 10. FLOTATION ECONOMICS AT 50 TONS/DAY CALCIUM OXIDE
CAPITAL COST
$250,000 (7% for 20 Years)
LABOR
12 Hours/Day @ $6/Hour
POWER
175 HP @ $0•02/KWH
2/3 of the Year
CHEMICALS
5 #/Ton of Dresinate
2 #/Ton of Soda Ash
TOTAL
$/YEAR $/TON OF CaO
23,258 1.42
17,345 1.06
15,232 0.78
16,753 1.02
$ 67,004 $ 4.28
61
-------
of 50 tons/day. All costs are expressed in terms of
the final product, calcium oxide.
For comparative purposes, costs for 80 and 160
tons/day plants are also made and included in Figure 18.
Unit labor and capital costs are greatly reduced with the
large production facilities.
The system estimated includes three cleaner stages.
This additional cost may be justified during high water tur-
bidity. The extra stage may be bypassed to reduce power costs
when not required. Table 11 illustrates the need for additional
cleaner stages for the typical and low feed grade sludge
conditions.
Chemical costs will vary, but the effect will be
minor as they are only a minor portion of the total cost.
Additional Process Economic Considerations
A budgetary cost analysis to include the various
unit operations required to implement the process at Johnson
County follows. The units are designed for maximum plant pro-
duction of 50 MGD and include:
1) Sludge Carbonation - the cost estimate includes
sludge piping, carbonation cells, piping and
transport of kiln exhaust gas, and pumping to
and from the cells.
Design Flow 150 GPM
Retention Time 60 min.
Alkalinity 8,000 mg/1 as CaC03
Solids 10%
2) Carbonated Sludge Thickener - The thickener is de-
signed based on thickening studies reported earlier
Design Loading 50 ft^/ton/day
Surface 4,500 ft^
Diameter 75 ft
3) Vacuum Filter - The cost estimate for the vacuum
filter does not include housing, as this has been
provided in the flotation circuit. A mechanical
conveyor has been included to transfer the filter
cake to the flotation conditioning cells.
Design Loading 15 lb/ft^/hr
Surface Area 500 ft
62
-------
6-
1*
I I l I » i I I I I I I 1 I I 1—
20 40 60 80 100 120 140 160
TONS OF CaO PER DAY
FIGURE 18. FLOTATION COSTS AS RELATED TO TONNAGE PROCESSED.
-------
TABLE 11. EFFECT OF ADDITIONAL CLEANING STAGES ON PRODUCT GRADE
Minimuma Typical^
Stage Grade Grade
Feed 65.1 88.3
Rougher 71.6 92.8
First Cleaner 81.9 96.0
Second Cleaner 85.4 96.9
Third Cleaner 91.2 97.4
Fourth Cleaner 92.2
aSludge collected 4/30/76. pH to 9.0f 7.0 lb/ton of Dresinate
TX-60 Collector.
^Sludge collected 10/8/75. pH to 9.0, 2.8 lb/ton of Dresinate
TX-60 Collector.
64
-------
4) Product Centrifuge - The centrifuges are designed
for feed solids of 8-10%, almost pure calcium
carbonate/ at a total flow rate of 250 GPM. Two
centrifuges are planned.
5) Tails Dewatering - This will be accomplished in
the existing sludge lagoons. Overflow supernatant
will be recycled back to the plant influent. La-
goon cleaning, hauling and disposal costs should
be reduced by 80%.
6) Rotary Lime Kiln, Feed, and Storage Facilities -
Actual kiln size will depend upon future plant
expansion considerations, however, a 50 ton/day
kiln is assumed. The necessary facilities for
coal storage, pulverizing, and feed are included
as are the necessary air pollution control devices.
Table 12 summarizes the capital costs for all necessary
facilities including flotation. The total capital cost for fa-
cility implementation is $3,727,600 including engineering and con-
tingencies. All capital and operating costs are expenses against
lime production for economic comparison. Table 13 summarizes
these.
The unit costs were compared for average and full-
scale facility operation. As the average water production in-
creases, the unit costs will decrease substantially.
Energy requirements of the recalcining process require
careful consideration as to fuel source. The cost differential
for the various fuel sources is significant. Natural gas in the
Kansas City area is, at best, only seasonally available, thus
another fuel source is required. We used $32/ton as the cost
for coal. Most commercial lime producers are currently using
coal in winter months when gas supplies are interrupted. The
storage, pulverizing, and feed facilities are easily operated
and relatively uncomplicated. Coal as a secondary, if not
primary fuel is proposed. Using coal as fuel, the total cost
per ton of lime ranges from $42.88 to $68.77 dependent upon
lime production.
Implementation of the magnesium recycle, flotation,
and lime recovery will reduce costs in a number of ways. First,
sludge disposal costs will be reduced by 80%, or $64,000/year.
This cost savings can be expected to increase yearly. Table
14 summarizes chemical costs which would be greatly reduced or
eliminated.
t
65
-------
TABLE 12~ SUMMARY OF CAPITAL COST
JOHNSON COUNTY, KANSAS
Item Cost
1}
Sludge Carbonator Cells
50,000
2)
Carbonated Sludge Thickener
175,000
3)
Vacuum Filters
215,000
4)
Product Centrifuges
200,000
5)
Rotary Lime Kiln
11600j 000
6)
Piping (15%)
336 ? 000
7)
Electrical and Instrumentation
448,000
(20%)
Sub-Total
$3,024-000
Engineering and Contingencies 453,600
(15%)
Sub-Total $3,477,600
9) Flotation Circuit3 250,000
TOTAL $3,727,600
aTotal cost including engineering and contingencies has been
calculated prior6
66
-------
TABLE 13. LIME RECOVERY COSTS
50 tons/day CaO
(Includes all additional plant facilities)
Annual Water Production
Cost 30 MGD 60 MGD
($/yr) ($/ton CaO)
CAPITAL COSTS:
3,727,600 @ 7% for 30 yrs 297,582 32.61 16.30
LABOR COSTS; 91,500 10.03 5.01
4 Operators @ $12,000/yr
3 Ass!t Operators @ $9,000/yr
1 Maintenance @ $12,000/yr
0.25 Management @ $18,000/yr
REPAIRS & MAINTENANCE 2.00 1.00
FUEL:
Natural Gas ($l/million BTU) 9.00 9.00
Fuel Oil ($2.20/million BTU) 19.80 19.80
Coal ($1.25/million BTU) 11.25 11.25
POWER:
Electricity 108,307 11.86 8o30
CHEMICALS:
Flotation Reagents 8, 376 - 16, 753 1.02 1.02
TOTALa 627,525 68.77 42.88
to
782,560
a - Assume use of coal
67
-------
TABLE 14. CHEMICAL COST SAVINGS
Chemical
Average Annual
Dosage
(mg/1)
Tons/yr
(Assume 30 MGD
Production)
$/yr
Carbon Dioxide
20.4
930
35,354
Chlorine
5 . 4a
246
49,255
Alum
1.5
68
8,620
Sodium Aluminate
2.0)
91
37,580
Polymer
1-6
73
65,528
$196,336
a Prechlorination only
(NOTE: Based on 1975 chemical usage)
It is doubtful that these chemicals could be totally
eliminated; therefore, an 80% reduction is assumed or $157,068.
Approximately 20% of the lime recalcined will be in
excess of water treatment requirements; thus, available for
sale. Implementation of the magnesium process will require a
lime feed of 165 mg/1 as opposed to the 135 mg/1 presently fed.
Table 15 compares the present method of water treatment
with the proposed process for directly affected cost elements.
Implementation of lime recovery at the initial average production
of 30 MGD produces an annual additional cost of $105,542; however,
as the facilities are fully utilized, a cost savings of $261,408
results. Additional chemical cost savings may be realized in
that a reduced soda ash feed is required when additional mag-
nesium carbonate hardness is removed, as would be the case with
the process studied. An average water production of 36 MGD
would be required for the process to be economically favorable
considering only the data shown in Table 15.
Design of facilities to operate more nearly at capacity
on start-up would reduce or eliminate the additional annual costs.
Water plant sludge could be stored to reduce peak requirements
and facility sizing. The economic data were developed for com-
parative purposes and do not necessarily represent the least cost
facility design.
68
-------
TABLE 15. ECONOMIC COMPARISON OF PRESENT AND PROPOSED TREATMENT
Comparable Cost
Elements
Annual Cost ($/yr)
30 MGD Production
60 MGD Production
Present
Proposed
Present
Proposed
Lime
Additional Chemical
Costs3
Sludge I-Iauling Costs'
Excess Lime Sales*5
237,041
157,068
80,000
- 0 -
627,526
- 0 -
16,000
<63,875>
474,002
314,136
160,000
- 0 -
782,560
- 0 -
32,000
<127,750)
TOTAL
$474,109
$579,651
$948,218
$686,810
NET ANNUAL SAVINGS
<105
,542>
261,408
- See Table 14, assumes 80% reduction
- Assumes sales at $35/ton, 20% excess
-------
The sale of excess magnesium, which may add to the
cost savings, will be discussed in the next section,
Magnesium Recovery
As noted previously in Table 1, the magnesium content
of the raw water averaged 62 mg/1 considering the average ratio
of well to river water. Dependent upon coagulation pH, up to
52 mg/1 as CaCOo would be precipitated under normal operation.
Assuming a 10% loss in the sludge, approximately 390 lb/MG of
magnesium, expressed as calcium carbonate, can be recovered=
Magnesium can be recovered from the saturated recycle
stream in a number of forms. In 194 6, the Department of the
Interior reported a process for magnesium carbonate recovery
by using air to strip out the carbon dioxide causing precipi-
tation. (31) Controlling the temperature between 35-45°C pro-
duced magnesium carbonate trihydrate (MgCO^ ' 3 H^O) having
a solubility of approximately 2.2 g/1. Boiling the liquor
precipitates basic magnesium carbonate (4 MgCC>3 • Mg(CH) 4
4 HjO) which has a solubility of approximately 0.1 g/1. Either
of the compounds can be converted to various grades of MgO
by heating to 550°C or higher.
In a 1972 market survey, Innes (32) reported that the
United States consumption of basic carbonate had stabilized
at an annual level of 8,000 tons. Most of this consumption
is in the drug, rubber and plastics industries. The market
for magnesium oxide is greatly expanded representing a con-
siderable market for a significantly lower grade product.
Preliminary results of a market survey in progress
indicate a considerable market for light burned magnesium oxide.
(33) By controlling the temperature of calcining to approxi-
mately 550°C the magnesium oxide produced is quite reactive.
Uses and typical market prices for the product include:
Use Cost Approximate Market
($/ton) (tons/yr)
Electrical Insulation
Refractories
Electrical Steel
Rubber
Pharmaceutical
800 10,000
200 Large
900 1,000-2,500
74 0 Unknown
14 60a Unknown
This survey reports magnesium supplies to be 1 tight7 with con-
cern over a possibility of shortages developing.
Production of the trihydrate allows only partial re-
covery of the magnesium because of its relatively high solubility«
Recycle of this process stream can be accomplished so that it is
70
-------
not wasted from the system. Recycle to the raw water will re-
quire lime for reprecipitation, although it may be useful as a
coagulant for those raw waters with only moderate magnesium
levels. One flow scheme proposed utilizes dilution water prior
to carbonation to control solids. In this case, the water from
the magnesium production process could be used for this purpose,
reducing lime requirements.
Magnesium Carbonate Production Pilot Studies
In the operation of the magnesium production portion
of the pilot plant, the recycle stream was valved through hot
water heat exchanger to the flotation cells. The recycle rota-
meter was used to measure flows of the liquor<, The temperature
was controlled by the water heater thermostat.
The flotation cells, after thorough cleaning, were
used to strip carbon dioxide from solution, precipitating
the magnesium carbonate trihydrate. The cells used had a
volume of 135 gal. providing a residence time of 22-45 minutes
dependent upon flow rate.
Recycle of precipitated material was maintained at
a flow rate of 1 GPM and was provided to assist precipitation
by seeding the reactions.
Pilot scale studies were performed in June, July,
August, and November. Influent alkalinities ranged from
3,400-6,800 mg/1. As previously discussed, the low alka-
linities resulted from the mechanical sludge collection
deficiency in the Permutit basin. Higher sludge solids in
the carbonator feed would have produced respectively higher
alkalinities; unfortunately, increased solids feed was a
mechanical impossibility.
After aeration, the alkalinity ranged from 2,800-
3,000 mg/1 as CaC03. During the November study, approximately
54 lb/hr of the product were precipitated. Vacuum filtration
produced a 40% solids cake. The filtration characteristics
were excellent producing in excess of a 1 in. thick, easily
discharged filter cake.
All products were lower in purity than expected.
The primary impurity was calcium carbonate which was carried
over in the thickener supernatant. Either a sand or diato-
maceous earth filter prior to the heat exchanger would have
eliminated or greatly reduced this problem.
The silica level in the recycle stream was reflected
in the product. Filtration would have little effect in reducing
values found. However, greatly reduced levels would be expected
when production, not recycle was employed. The precipitated
product was slightly brown in color, resulting from the organic
71
-------
color in the solution- The color can be removed by contacting
the magnesium bicarbonate liquor with activated carbon as re-
ported in previous studies. Table 16 summarizes the results of
the four pilot scale production studies.
Laboratory Studies
Samples of the recovered magnesium carbonate were
treated in the laboratory to determine how product purity
might be improved as well as to develop additional unit pro-
cess design data.
A sample of magnesium carbonate produced in the
pilot plant was redissolved using pure carbon dioxide pro-
ducing a solution with a magnesium concentration of approxi-
mately 10,000 mg/1 as CaCO^. This solution was used in all
subsequent studies.
Prolonged settling insured that only a clear solu-
tion was used in the precipitation studies. A laboratory scale
precipitator reactor was fed the clear magnesium solution con-
trolling such variables as temperature, air flow and mixing
conditions, and the presence of preformed magnesium carbonate
crystals. The magnesium carbonate produced was evaluated for
purity and the precipitate used for bench scale thickening and
dewatering studies.
Sufficient settling of the magnesium bicarbonate
liquor prior to precipitation of the magnesium dramatically
improved the quality of the product. A comparison of im-
purities contained in the product produced in laboratory and
pilot studies follows:
Contaminant Pilot Laboratory
(mg/q)
Silica 34.0 0.5
Calcium 7 3.0 3.4
Iron 0.9 1.7
The improved quality of the laboratory produced
magnesium carbonate supports the conclusion that filtration
of the magnesium bicarbonate liquor will dramatically improve
product purity. Figure 19 illustrates thickening characteristics
of the magnesium carbonate trihydrate. Leaf vacuum filter
tests indicated filter loading rates in excess of 100 lb/ft^/hr
are possible producing a 40% solids cake.
72
-------
TABLE 16. SUMMARY OF MAGNESIUM PRODUCTION STUDIES
Alkalinity
(mg/1 as CaCO^)
Product
Analysis (mq/q)
Date Flow
Rate (GPM)
Temperature (°C)
In
Out
MSL
Ca
Si Fe
6/11
3
40
3922
2407
170
-
-
7/12
6
40
109
95
-
8/28
3
39
3400
3175
54
-
11/20-21
5
41
6700
3200
141
60
34 0.88
- denotes that analysis
not performed
-------
160°
140-
120-
>
<
Q
**«s.
Z
o 100-
CM
i
HI
-------
A summary of design criteria developed from the labo-
ratory studies is as follows:
1) The temperature for maximum precipitation of
magnesium carbonate trihydrate was 45°C. Above
or below this temperature a slight increase in
solubility resulted.
2) An air flow rate of 4 ft^/gallon of liquid was
required to optimize tankage and blowers.
3) Heat exchanger surfaces must be designed to re-
duce scaling from the saturated magnesium solution.
Economics of Magnesium Recovery
Figure 20 illustrates the units required to produce
MgO from the recycled magnesium bicarbonate liquor. Heat ex-
change, to warm the liquor to 45°C, is provided by scrubbing
the multiple hearth furnace exhaust gases. The CC>2 partial
pressure of the gases will reduce precipitation problems.
The capital cost for three tons per day MgO pro-
duction is shown in Table 17. Annual costs are summarized
in Table 18. The cost per ton of MgO produced is estimated
at $155.
The lime kiln or multiple hearth furnace can possibly
be used for both lime recalcining and MgO production. The mag-
nesium carbonate thickener can be slightly increased in size
to allow storage of material for a four-day period. Lime
recalcining would operate four days and store calcium car-
bonate one day when the kiln is used to produce MgO. Sufficient
storage is available in the calcium carbonate thickener to
allow the proposed operation. The extent of MgO contamination
can be evaluated in pilot studies.
Using the furnace or kiln for dual purposes would
allow a capital cost savings of approximately $400/000 along
with an estimated labor savings of 10,000 per year. These
savings would reduce the cost of producing MgO to approximately
$117 per ton. Additional lime feed is required to precipitate
all of the magnesium. The added cost for this additional lime
is for fuel costs only as the lime recovery facilities will
adequately handle the additional capacity. Assuming an addi-
tional feed of 50 mg/1 of CaO and a fuel cost of $11.25 per
ton of CaO, the added cost per MG treated is $2.34. At a MgO
production rate of 0.08 tons per MG of water treated, the re-
sult is an additional cost of $30 per ton of MgO. Therefore,
a cost range of $147 to $185 per ton of MgO can be expected.
75
-------
cn
MAGNESIUM ^
Bl CARBONATE
PUMP
FILTER
nL
fflHTk
1
1 • 1
1 • 1
11 • 1
J
1
BLOWER
AERATION CELLS
SEPARATOR
SLURRY RECYCLE
OVERFLOW
TO SLUDGE
DILUTION
l~TJ Zt
SLURRY PUMPS
i
SCRUBBER
CYCLONE
onTARY DRYER
VACUUM
&
T\'
f
FILTRATE
PRODUCT
STORAGE
* ~ ~
~
~
~
~_
n_
~D
if
=5 ^-^Tl
ifSB
multiple
HEARTH
FURNACE
FIGURE 20. MgO PRODUCTION FACILITIES
-------
TABLE 17. CAPITAL COST FOR 3 TONS/DAY
OF MgO PRODUCTION
Item Cost ($)
Aeration Cells (60 minute retention) 50,000
Thickener (30 ft diameter) 60,000
Vacuum filter (9 ft^) 25,000
Rotary Drier 115,000
Multiple Hearth Furnace (includes
(scrubber, storage, etc.) 333,000
Piping (15%) 87,500
Electrical (20%) 116,600
Sub-total 787,100
Engineering & Contingencies (20%) 157,420
TOTAL $944,520
TABLE 18. SUMMARY OF ANNUAL COST
FOR MgO PRODUCTION
Element Annual Cost ($/yr)
Capital Cost Recovery (30 years 75,372
0 7%)
Maintenance (3% of Capital Cost) 28,335
Fuel (6 million BTU/ton MgO) 14,454
Electricity (3C/KWH) 23,224
Labor (4160 hrs/yr 3 $7/hr) 29,120
TOTAL ANNUAL COST $17 0,5 05
COST PER TON MgO - $155
77
-------
Magnesium recovery can be added without affecting
the existing processes. It is, therefore, recommended that
recovery be delayed until Dayton has conducted the proposed
market survey and economic analysis. It is very unlikely
that magnesium recovery will be more attractive in Johnson
County than Dayton, as Dayton is in close proximity to a
considerable magnesium market and Dayton's product can be
produced in a purer form at a lower cost. This results pri-
marily from Dayton's clear groundwater source containing
little or no objectionable product contaminants and larger
production capability.
Recycle of magnesium as a coagulant, as opposed
to precipitation of all magnesium present, will result in
reduced lime requirements. As the market for the product
develops and experience with recovery indicates economic
justification, recovery facilities can be added.
78
-------
SECTION VI
REFERENCES
1. Environmental Protection Agency, "1973 Inventory of Muni-
cipal Water Facilities", Washington, D. C., (Not yet
published).
2. Bureau of the Census, "Water Use in Manufacturing, 1967
Census of Manufacturers", U.S. Department of Commerce,
U.S. Government Printing Office, Washington, D. C.
3. Southern Research Institute, "Draft Document Guidelines
for the Water Supply Industry", March, 1975.
4. Burris, Michael A., Cousens, Kenneth W., and Mair, David
M. , "Softening and Coagulation Sludge Disposal Studies
for a Surface Water Supply", Presented at the 36th Annual
Conference of the Ohio Section of the American Water Works
Association, November, 1974.
5. Young, K. W., Matsch, L. C., and Wilcox, E. A., "Sludge
Considerations of Oxygen Activated Sludge", Presented at
the University of Texas Water Resources Symposia, November,
1972.
6. Black, A. P., and Eidsness, F. A., "Carbonation of Water
Softening Plant Sludge", Jour. AWWA Vol. 49, Oct., 1957,
p. 1343.
7. Black, A. P., Shuey, Bruce, and Fleming, Paul, "Recovery
of Calcium and Magnesium Values from Lime-soda Softening
Sludges", Jour. AWWA Vol. 63, Oct., 1971, p. 616.
8. Thompson, C. G., Singley, J. E., and Black, A. P., "Mag-
nesium Carbonate - A Recycled Coagulant for Water Treatment,
Part 1 - Laboratory Studies", Jour. AWWA Vol. 63, Oct.,
1971, p. 11.
9 . , "Part 11 - Practical Applications and
Economic Evaluation", Jour. AWWA Vol. 63, Nov., 1971, p. 93.
10. Environmental Protection Agency, "Magnesium Carbonate, A
Recycled Coagulant for Water Treatment", Technology Transfer
Capsule Report.
11. Gainesville, City of, Florida, "Magnesium Carbonate, A Re-
cycled Coagulant for Water Treatment", EPA Office of Research
and Development, June, 1971 (12120 ESW 06/71),
79
-------
12. Black, A. P., DuBose, A. T., and Vogh, R. P., "Physical-
Chemical Treatment of Municipal Wastes by Recycled Mag-
nesium Carbonate", EPA Office of Research and Development,
June, 1974, (EPA-660.2-74-055).
13. Black, A. P., and Thompson, C. G. , "Plant Scale Studies
of the Magnesium Carbonate Water Treatment Process", EPA
National Environmental Research Center, Office of Research
& Development, May, 1975, (EPA-660/2-75-006).
14. Klassen, V. I., and Mokrousov, V. A., "An Introduction to
the Theory of Flotation", Butterworths, 1963.
15. Gaudin, A. M., "Flotation", McGraw-Hill Book Co., Inc.,
1957
16. Sutherland, K. L., and Wark, I. W., "Principles of Flo-
tation", Australian Institute of Mining and Metallurgy,
1955.
17. Whelan, P. E., "Froth Flotation: A Half Century Review",
The Industrial Chemist, Vol. 32, 1956, po. 315-318, 409-
411, 489-491.
18. Boucher, L. J., "Cement Rock Beneficiation at the Universal
Atlas Cement Co., Northamoton, Pa.", Mining Engineering,
Vol. 5, pp. 289-293.
19. Kleiber, J. C., "Floating Limestone at Permanente", Mining
Engineers, Vol. 16, No. 3, March, 1964, pp. 39-44.
20. Trauffer, W. E., "New Calcium Carbonate Products Plant,
Last Word in Automation, Efficiency", Pit & Quarry, Vol.
58, No. 4, 1965, pp. 86-90, 97, 98.
21 Sheen, R. T., and Lammers, H. B., "Recovery of Calcium Car-
bonate or Lime from Water Softening Sludges", Jour. AWWA,
Vol. 36, Nov., 1944, p. 819.
22. Black, A. P., "Disposal of Softening Plant Wastes. Lime and
Lime-soda Sludge Disposal", Jour. AWWA, Vol. 41, Sept., 1949,
p. 819.
23. Glenn, R. W., Judkins, J. R., and Morgan, J. M., "Filtrability
of Water-Treatment-Plant Sludge", Jour. AWWA, Vol. 65, June,
1973, pp. 414-417.
24. Somasundaran, P. and Agar, G. E., "The Zero Point of Charge
of Calcite", Jour. Colloid & Interface Science, Vol. 24,
1967, pp. 433-440.
80
-------
25. Fuerstenau, M. C., Guierrez, G., and Elgillani, D.,
"Influence of Sodium Silicate in Nonmetallic Flotation
Systems", Society of Mining Engineers, AIME, No. 241(3),
Sept,, 1968, pp. 319-323.
26. Biswas, A. K., "Role of Carbon Dioxide in Flotation of
Carbonate Minerals', Indian Journal of Technology, Vol.
5, No. 6. June, 19 , pp. 187-189.
27. Kumar, S., Mohan, N., and Biswas A. K., "Fundamental
Studies on the Role of Carbon Dioxide in a Calcite Flo-
tation System", Society of Mining Engineers, AIME,
Vol. 250(3), Sept., 1971, pp. 182-186.
28. Riehl, M. L., H. H. Weiser and R. T. Rheins. Effect
of Lime Treated Water Upon Survival of Bacteria.
J.AWWA. 44:466 (May 1952) .
29. Chaudhure, Maley and R. S. Engelbrecht. Removal of
Viruses From Water By Chemical Coagulation And
Flocculation. J.AWWA. 62:563 (September 1970).
30. Talmadge, W. P., and E, B, Fitch, "Determining Thickener
Unit Areas", Ind. Engr. Chem., Vol. 47 (1955), p. 38.
31. Doerner, H. A., Holbrook, W, F., and Fortner, 0. W.,
"The Bicarbonate Process for the Production of Magnesium
Oxide", U. S. Bureau of Mines, Fed. Paper 684, 1946.
32. Innes, George L,, "The U. S. Market For Basic Magnesium
Carbonate", Report to the City of Dayton, Jan. 4, 19 72.
33. Rice, Owen, "A Report on the City of Dayton", August 2, 1976.
81
-------
WEEKLY SUMMARY
PILOT PLANT OPERATING DATA
JOHNSON COUNT*, KANSAS
February *> March
NNk
Ending
Plow
IMCZ>)
Basin influent
Chanicals Added
Rapid
Mix
pH
Basin
Effluent
Carbona tor
Thickener
Underflow
Ibtal
Alk.
—taq/1
Ca 1*3
Hard.[Hard.
as CaCO^--
rurb.
FTU
LiM
mg/1
PolyMr
¦9/1
Recycle
¦9/1
pH
¦C
Hard.
mg/1-
*9
Hard.
CaCO^
Total
Turb.
FTU
ACld
Turb.
mi
Inf.
pH
Dry Wt.
Solida
CO j
c£b
Sludge
9P»
Kffluent
gptn
* Dry
Wt Solidi
pH
Total
Alk.
2/7/75
2
200
228
50
48
207
0.20
64
11.29
11.2
127
35
5.92
1.25
11.23
9.65
7.2
7719
2.6
15.6
2/14/75
2
219
251
59
20
204
0.12
28
11.26
11.1
144
23
5.8
1.19
11.17
5.44
7.2
8020
1.0
21.15
2/21/75
2
214
246
55
24
188
0.10
16
11.35
11.2
173
19
3.53
0.51
11.16
6.2
4.43
6.15
7.3
6919
2.2
12.11
2/28/75
2
189
209
53
43
163
0.20
31
11.19
11.1
136
29
4.4
1.21
11.15
7.9
8.0
9.4
7.
7632
S. 52
12.54
3/7/75
2
206
234
69
30
166
0
29
11.22
11.1
146
36
4.0
1.18
11.12
6.3
6.25
8.16
7.3
6943
7.22
8.5
3/14/75
2
206
229
68
17
158
0
31
11.39
11.2
147
36
4.7
1.5
11.3
6.7
7.69
8.5
7.2
R086
5.06
10.S
3/21/75
2
204
234
66
19
129
0
34
11.37
11.3
135
46
3.6
0.8
11.3
7.14
7.0
8.0
7.2
7 305
3.07
12.8
3/28/75
2
205
232
69
25
123
0.02
20
11.38
11.3
119
43
5.6
0.8
11.3
7.3
5.4
7.7
7.2
5754
4.54
15.6
APPENDIX A
-------
WEEKLY SUM4ARY
PILOT PLANT OPERATING DATA
JOHNSON COUNTY, KANSAS
April 6 May
CO
LO
week
Ending
Plow
(NGD)
Basin
influi
nt
Chmicals i
kdded
Rapid
MIX
pH
Basin
Effluent
Carbonator
Thickener
Underflow
TOtAl
AUc.
--nq/1
Ce |Mg
Hard.l Hard.
as CaCOa—
Ttirb.
FTU
Line
¦9/1
Polymer
wq/l
Recycle
og/2
PH
Ca Kg
Kerd.liierd.
nq/l-CaCO,
Total
Turb.
FTU
Acid
Turb.
FTU
Inf.
pH
Dry Wt.
Solids
co2
cfm
Sludge
9J*>
Effluent
gpm
% Dry
WL Solids
PH
Total
Alk.
4/4/75
2.2
191
204
58
47
133
0.05
10
11.23
11.3
92
43
9.4
3.3
11.1
9.4
2.6
4.9
6.99
3621
2.4
10.7
4AV7S
3.0
193
200
63
40
172
0.04
4
11.11
10.9
80
40
14.4
3.1
10.77
9.0
1.03
5.6
6.95
2110
3.1
12.3
(847A)
0.71
(607)
4/18/75
2.0
180
181
57
43
161
0.17
17
11.17
U.l
75
37
12.3
1.55
10.83
8.7
3.51
7.54
7.09
3429
1.0
13.7
(847A)
0.93
(607)
4/25/75
2.0
175
177
50
56
168
0.22
17
11.25
11.2
82
29
23
2.3
10.87
6.6
4.6
7.8
7.15
3756
2.4
10.9
(847A)
1.66
(607)
5/2/75
2.0
170
179
57
48
169
1.50
13
11.13
11.1
90
25
18.7
1.66
10.83
6.2
5.3
6.1
7.06
4638
3.3
10.7
(607)
5/9/75
2.0
177
1%
58
29
196
0.45
ie
11.12
11.1
123
22
19.6
2.22
10.79
6.8
5.2
7.1
7.19
5125
2.9
12.0
(847A)
5/16/ 75
2.0
143
158
65
24
201
0.45
26
11.12
11.1
126
15
12.5
1.42
10.75
5.7
7.2
7.3
7.2
7010
3.5
9.7
(847A)
3/23/ 75
2.0
152
172
70
19
215
0.29
44
11.00
11.0
130
18
15.8
1.00
10.59
5.0
9.0
7.8
7.13
7400
3.3
9.3
(B47A)
5/30/ 75
2.0
144
159
56
54
221
0
34
11.05
11.0
1
118
24
12.9
1.46
10.65
4.2
7.4
7.9
7.11
6460
3.5
9.0
APPENDIX A
-------
MEEKLY StJ>#4ARY
PILOT PLANT OPERATING DATA
JOHNSON COUNTY, KANSAS
June t July
NMk
Flow
Basin
lAflua
At
Ci
«calcala Added
Rapid
Basin
Effluent
Carbonator
Thickener
Underflow
Ending
(HGD)
Total
Ca
Hq
rurb.
Liae
Polymer
Recycle
Mia
pH
Ca
H9
Total
Acid
Inf.
Dry Wt.
co2
Sludge
Effluent
gpn
% Dry
Wt Solid#
AXk.
--mq/1
Hard.1 Hard,
as CaCOi—
FTU
og/1
mq/l
ag/1
PH
Hard.
nxj/l-
Hard.
CaCOj
Turb.
FTU
Turb.
pro
pu
Solids
cfn
91®
pH
Total
A1X.
6/6/75
2.0
137
145
42
74
183
0
22
11.01
11.0
108
16
16.0
2.93
10.84
5.6
7.2
8.1
7.2
5660
3.4
11.2
6/13/75
2.0
137
137
45
69
227
2.7
(607)
24
10.99
11.1
113
17
14.4
1.83
10.84
5.7
8.3
8.4
7.1
5670
3.1
11.6
6/20/75
2.0
145
150
48
29
306
0.34
(607)
19
10.96
10.9
115
14
16.2
1.79
10.79
4.7
7.8
8.5
7.1
5040
3.3
9.6
6/27/75
2.0
133
143
43
65
250
0.2
(607)
18
10.99
10.9
146
13
15.7
2.9
10.76
5.4
7.4
7.9
7.2
5270
3.8
8. 1
7/4/75
2.0
126
137
42
62
210
0
18
11.1
11.0
164
14
15
4.3
10.8
4.7
6.0
7.0
7.2
4600
3.0
8.2
7/11/75
2.0
133
151
43
38
176
0
19
10.9
11.0
136
16
18
2.9
10.6
5.4
6.8
7.7
7.2
6200
4.3
6. J
7/18/75
2.0
137
169
52
33
204
0
15
10.9
10.8
134
17
22
3.0
10.5
4.3
5.2
7.5
7. 2
4750
4.1
7.1
7/25/75
2.0
153
183
45
23
243
0
5
10.9
10.9
139
14
23
1.9
10.5
4.0
2.9
2.9
7.0
3860
1.9
6.3
APPENDIX A
-------
WEEKLY SUMMARY
PILOT PLANT OPERATING DATA
JOHNSON COUNTY, KANSAS
August & September
Meek
Biding
Flow
(MOD)
Baein influent
Chemicals Added
Rapid
Mix
PH
Be Bin
Effluent
Carbonator
Thi
Und
ckener
erflow
ratal 1 ca
Alk. 1 Hard.
*9
Hard.
CO,--
rurb.
FTU
Lino
ng/1
Folyaer
¦9/1
Recycle
mg/1
£>H
ca
Hard.
rag/1-
Mg
Total
Acid
Turb.
FTU
Inf.
PH
Dry Wt.
Solids
co2
cfn
Sludge
gjm
Effluent
gpo
% Dry
Wt. Solid*
Hard.
CaCO^
Turb.
FTU
PH
Total
Alk.
8/4/75
2.0
134
165
46
33
272
0
19
10.9
10.9
135
10
20
3.3
10.5
4.0
2
7.5
7.0
5222
3.6
6.0
6/11/7?
2.0
119
139
56
28
140
0
16
11.0
10.9
171
13
13
2.2
10.5
3.5
3
7.7
6.8
4138
3.5
5.5
0/10/7J
2.0
136
158
60
16
216
0
20
10.9
10.7
137
14
10
1.6
10.4
3.4
3
7.1
7.0
3950
2.4
9.3
8/25/75
2.0
147
169
63
30
290
0
25
11.0
10.9
151
14
14
1.9
10.6
3.1
4
7.9
7.1
3683
1.0
8.6
9/1/71
2.0
144
167
49
24
336
0
17
11.0
10.8
139
14
23
2.3
10.6
2.7
4
6.8
7.1
3961
0.7
16. 3
9/8/75
2.0
148
199
64
36
321
0
1J
11.0
10.9
165
17
24
2.7
10.6
3.2
4
7.9
7.0
3866
3.4
6.0
9/15/75
3.6
134
173
48
49
231
0
13
11.2
11.1
155
16
17
2.6
10.9
3.8
6
8.8
7.0
3829
2.1
15.1
4/22/75
4.0
174
217
49
37
186
0
17
11.2
11.0
135
24
29
4.3
10,8
4.8
13
11.0
6.7
4332
2.1
13.7
9/29/75
2.9
155
>91
61
40
195
0
20
11.2
11.1
135
26
24
3.8
10.9
5.1
8.6
7.1
5084
0.5
18.5
APPENDIX A
-------
WEEKLY SUM4ARY
PILOT PLANT OPERATING DATA
JOHNSON COUNTY, KANSAS
October & November
Weak
Plow
Basin
Influent
Ch«lcals Added
Rapid
Basin
Effluent
Carbonator
Thickener
Underflow
Biding
(MOD)
Total
Ca
*9
rurb.
Lime
Polymer
Recycle
Mix
P«
Ca
Mg
Total
Acid
Inf.
Dry Wt.
CO,
Sludge
Effluent
% Dry
A Ik. i B&rd.lttard.
—nw/1 aa CaCQ-^-
PTU
®g/i
mi/1
¦9/1
P«
Hard.
mg/1-
Hard.
CaCO-,
Turb.
PTU
Turb.
FTV
PH
Solids
cfm
9(XB
PH
Total
Alk.
gym
Wt Solids
10/6/75
2.6
153
194
52
18
163
0
16
11.2
11.1
128
29
12
1.6
11.0
6.7
7
5.5
7.0
5662
0.4
28.2
10/13/75
3.6
147
179
73
22
190
0
17
11.1
11.0
118
38
21
2.6
10.9
6.3
8
12.2
7.1
5339
3.1
18.6
10/20/75"
1.3
239
11.0
1.3
10/27/75
3.9
192
235
71
11
227
0
21
11.1
11.0
124
49
7
0.8
10.8
8. 3
8
11
7.0
6439
1.9
27.3
11/3/75
3.9
214
256
70
12
206
0
24
10.9
11.0
125
50
11
1.1
10.8
9.0
6
12. 3
7.1
5509
1.0
32.0
11/10/75
3.8
204
238
72
21
234
0
28
11.0
11.0
136
38
8.7
I .0
10.8
7.7
10
13.2
7.1
6525
1.8
28.9
11/17/75
3.2
221
272
GO
12
189
0
23
10.9
11.0
135
59
6.1
1.3
10.8
8.9
6
10.1
7.1
5128
1.2
20.0
11/24/75
4.S
223
264
70
12
228
0
14
11.1
11.1
133
53
9.8
1.6
10.9
10.4
7
6.3
7.1
6929
0.6
22.6
Pilot Plant Off
APPENDIX A
-------
MEEKLY SUMMARY
PILOT PLAOT OPERATING DATA
JOHNSON COUNTY, KANSAS
December & January
Week
Ending
Plow
-------
WEEKLY SUMMARY
pir/rr pijint soi.tds balance
Johnson County, Kansas
APRIL & MAY
Week
Ending
Softening Basin Solids Production
Produced
Total Solids
Carbonator Inf.
Total Solids
Thickener Eff.
Total Solids
Turb.
Alk.
FUw Water
Recycled M
CaCO*
Mg(OH)t
mq/1
lb/day
lb/day
ag/1
lb/day
lb/day
4/4/75
42
866
116
3R01
15
?60
151
10
355
113
866
4585
265
5717
4.9
9.4
5284
2.4
10.7
3197
4/11/75
19
976
11ft
5RR1
22
521
303
5.4
248
72
976
6123
375
7480
5.6
9.0
5998
3.1
12.3
3358
4/10/75
43
715
105
3507
20
336
195
16
643
170
715
4093
365
5056
7.4
8.9
7919
3.4
13.6
5400
4/25/75
56
941
100
3341
21
357
207
15
510
148
941
3711
355
5146
7.8
6.6
6200
3.4
15.3
5401
5/2/75
50
827
95
31R3
31
524
304
12
395
115
827
3612
419
4857
6.1
6.3
4S57
3.4
10.7
4228
5/9/75
29
480
102
3398
3b
593
344
17
572
166
486
3970
510
4966
7.1
5.0
5593
2.7
11.7
3706
5/16/75
25
412
68
2011
50
634
484
24
aoi
232
412
3083
716
4211
7.3
5.4
4725
3.5
9.8
4070
5/23/7 5
19
319
76
2588
5*
915
531
39
1305
378
319
3829
909
5057
7.8
5.0
4742
3.3
9.3
3711
5/30/75
54
097
69
2317
32
537
311
31
1036
300
896
3352
611
4859
8.1
4.1
3955
!
i"*i i
9.0
3777
APPENDIX B
-------
WEEKLY SUMMARY
PILOT PLANT SOLIDS BALANCE
Johnson County, Kansas
FEBRUARY « MARCH
1
t
week
Softeninq Basin Solids Production
Produced
Total Solids
w«r bona tor inf.
Total Solids
Thickener L'ff.
Total Solids
Turb.
A1X.
Raw war. <»
Recycled Maqnesiua
Turb.
CaCOi
Mg(0H)->
Total
Sludge
% Dry
lb/day
gjm
% Dry
lb/day
Ending
mq/1
lb/day
og/1 lb/day
CaCOi
Mq (OH),
CttCO-»
Mq (OH) ¦}
9F®
Solidq
Solidi
mg/1 lb/day
lb/day
»g/l lb/day
lb/day
2/28/7 5
42
695
117 3903
26 428
248
31 1030
298
695
4933
54 7
6175
9.2
7.6
8367
4.5
11.9
5464
1/7/75
3f>
498
131 4356
33 546
317
28 882
256
498
5238
573
6308
8.2
6.3
6160
7.2
8.5
6847
3/14/75
17
286
131 4356
32 536
311
28 929
270
286
5284
5131
6151
8.5
6.7
6842
4.5
11.0
5898
3/21/75
16
310
129 4303
20 33B
196
32 107/
312
31U
5 380
508
6199
8.0
7.1
6778
2.2
6.3
2883
3/20/75
25
417
130 4332
26 431
250
18 615
178
41/
4947
428
5792
7.7
7.3
4.1
14.0
5639
APPENDIX B
-------
WEEKLY SUMMARY
PILOT PLANT SOLIDS' BALANCE
Johnson County, Kansas
JUNE & JULY
Week
finding
Softoninq Basin Solids Production
Produced
Total Solids
Carbonator inf.
Total Solids
Thickener Eff.
Total Solidtt
Turb.
Alk.
Raw water
Recycled Magne&iua
Turb.
OtC03|Mg(OH)2
lb/day —
TotaJ
Sludge
gpa
% Dry
Solids
Lb/day
gpo
% Dry
Sol Ids
Lb/day
ng/1
lb/day
mg/1
lb/day
CaCO,
mq (oh) t
CaCO,
Mq (Oil) i
mg/1
lb/day
lb/day
mg/1
>i
!
lb/day
6/6/75
74
1229
68
2264
26
483
263
20
672
195
1229
2916
458
4623
8.0
5.6
5514
3.4
11.3
4530
6/13/75
69
1153
62
2054
29
489
284
27
743
216
1153
2797
299
4450
8.4
5.7
5703
3.1
11.6
4467
6/20/75
29
484
70
2335
34
569
330
17
SRI
16ft
404
2916
499
3899
6.5
4.7
4040
3.3
9.6
J7fl2
6/27/75
64
1075
58
1768
29
4R8
283
17
572
166
1075
2340
449
3863
7.9
5.0
4715
3.8
8.1
3054
7/4/75
62
1041
51
1701
27
455
265
17
553
160
1041
2254
425
3720
7.8
4.7
4390
3.0
8.2
2943
7/11/75
38
633
58
1935
28
467
212
17
576
167
633
2511
4 39
3503
7.7
5.4
4959
4.3
6.4
3339
7/18/75
33
550
62
2-68
35
573
334
14
481
139
550
2549
473
1
3572: 7.5
4.3
3832
4.1
7.1
3476
7/25/75*
24
384
384
2.9
4.0
1.9
6.3
• Solids blanket lost on 7/22/75; MySO^ 25 lb/hour fed until July 30, 1975.
APPENDIX B
-------
WEEKLY SUMMARY
PILOT PLANT SOLIDS BALANCE
Johnson County, Kansas
AUGUST 6 SEPTEMBER
Week
Sc
fttnina Basin
solids Production
Produced
Total Solids
Carbonator Inf.
Total Solids
Thickener Eff.
Total Solids
Turb.
AIX
Raw Water
Recycled Magnesium
Turb.
CaCO3|Hg(0N)2
Total
Sludge
% Dry
lb/day
gpo
% Dry
lb/day
Ending
ng/1
lb/day
tog/1
lb/day
CaCO,
Mq (OH) •>
CaCO,
Kq(OU)-,
gpo
Solids
Solids
tog/1 lb/day
lb/day
mg/1
lb/day
lb/day
a/4/75
33
548
59
1973
36
602
351
18
591
171
548
2564
522
3630
7.5
4.0
3600
3.6
6.5
2771
8/11/75
26
462
44
1482
43
706
412
16
519
ISO
462
2001
562
3026
7.7
3.8
3508
3.5
5.6
2354
8/18/75
16
274
61
2051
45
746
435
19
634
184
274
2684
641
3600
7.1
3.4
2890
2.6
9.3
2225
8/25/75
30
508
72
2412
49
809
472
2 J
762
221
50ft
3174
67ft
4359
7.9
3.1
2968
1.0
8.6
958
Vl/75
24
405
67
2302
35
573
334
16
546
159
40b
3850
493
3748
7.4
2.7
2411
0.8
13.0
750
9/8/75
36
593
73
2426
46
761
444
13
434
126
593
2859
569
1022
7.9
3.4
3226
2 .9
6.1
2101
9/15/75
49
620
59
1959
46
533
311
22
734
21 3
820
2692
524
2036
8.8
3.9
4093
2.1
16.7
3711
9/22/75
37
620
99
3307
25
417
243
34
1139
330
620
4446
573
>639
11.0
4.9
6493
2.1
16.2
3505
9/29/75
41
393
66
1639
44
422
246
42
801
232
393
2440
478
3311
5.1
5.1
4192
0.3
16.0
S22
APPENDIX B
-------
WEEKLY SUMMARY
PILOT PLANT SOLIDS BAIANCE
Johnson County, Kansas
OCTOBER S NOVEMBER
Keek
Ending
Softenlnq Basin
Solids Production
Produced
Total Solids
2arbonator Inf.
Total Solids
Thick«n«r Eff.
Total Solids
TUTb.
AIX.
Raw wate
r
Recycled Maqneoium
Turb.
CaC03|Mg(OH)2|Total
—lb/day
Sludge
gpo
% Dry
solid*
Lb/day
9I»
% Dry
Solid;
Lb/day
mg/1
lb/day
®g/i
lb/day
caco-»
Mq (OH) •?
CaCO-i
Kq(0M)->
og/l lb/day
lb/day
og/1
lb/day
lb/day
10/6/75
1ft
176
76
1162
22
206
120
40
366
112
176
1149
177
1411
4.6
6.6
3640
0.2
28.2
778
10/13/75
23
276
73
1443
41
490
286
28
867
251
276
2602
537
3416
8.7
7.0
7340
2.5
21.7
9570
10/20/75*
10/27/75
12
138
lie
2807
22
257
150
43
1034
300
138
3841
450
4347
9.1
8.4
9230
1.5
26.4
4361
11/3/75
12
284
139
6349
18
406
237
42
1983
575
284
8332
812
9428
12.3
9.1
13636
1.0
30.1
3444
11/10/75
21
677
129
8296
34
1067
622
49
3161
916
677
11457
1539
13672
13.2
7.8
12261
1.6
28.7
5544
11/17/75
12
323
146
7877
19
511
298
34
1685
547
323
9762
84S
1093G
10.1
8.8
10211
0.6
26.4
2152
11/24/75
12
441
14B
11150
19
700
406
30
2019
585
365
11711
932
13006
6.0
9.4
7002
0.^
22.8
1452
* Pusp off all of this week
APPENDIX B
-------
WEEKLY SUMMARY
PILOT PLANT SOLIDS BALANCE
Johnson County, Kansas
DECEMBER & JANUARY
VO
u>
Week
Ending
Softanina Basin
Solids Production
Produced
Total Solids
2arbonator Inf.
Total Solids
Thickener EEC.
Total Solids
Turb.
Alk.
Raw Watai
Recycled Magnesium
Turb.
CaCOj|Mg(OH)2
lb/day
Total
Sludge
gpa
% Dry
Solid*
lb/day
gj*B
% Dry
Solids
Lb/day
ny/i
lb/Uay
09/A
lb/day
caco?
Mq(OH),
CaCOi
M«7 (OH)-,
ng/1 lb/day
lh/day
«g/l
lb/day
lh/day
12/1/75
25
822
125
8859
12
394
230
27
1943
564
822
10802
794
12418
11.5
9.0
12413
4.1
18. 7
9227
12/8/75
30
995
85
5292
18
482
281
20
1383
401
995
6675
682
8352
9.9
4.3
5162
5.1
9.9
5788
12/15/75
22
362
112
3746
27
448
261
25
820
238
362
4566
499
5427
9.6
7.3
8321
4.6
11.6
6611
12/22/75
31
517
99
3307
21
348
203
?0
572
192
517
3970
395
5025
7.0
6.2
9082
1.2
21.9
1954
12/29/7S
24
403
101
3355
15
249
145
9
291
84
403
3646
229
4278
7.1
8.7
7487
3.9
8.5
3915
1/5/76
16
317
113
3876
13
218
127
'
20
677
196
317
4552
323
5192
10.3
9.7
12038
2.6
18.6
APPENDIX B
-------
PILOT SCALE FLOTATION SUM4ARY
(Page 1 of 2)
Run
REAGENT FEEDS (lb/ton)
pH
RESULTS
COMMENTS
*
Collector
Na2C03
SiO-
% CaCO,
% Recovery
£
Cone.
Feed
Tails
X
1.53(a)
1.25
0.33
8.7
95.0
87.1
82.0
42.8%
2/11/75
2
3.6 (a)
12.6
0.6
9.0
94.8
87.5
96.5
13.0
2/12/75
3
8.1 (a)
31.2
2.8
8.95
95.1
87.2
85.1
22.9
2/13/75
4
7.6 (a)
3.8 (b)
40.0
2.4
98.5
94.8
68.3
91.2
2/14/75, #1 CI
5
7.0 (a)
3.3
3.4
9.0
98.8
87.3
80.8
40.9
2/25/75
6
5.2 (a)
2.7
2.2
8.8
72.8
87.3
74.6
2/26/75
7
5.5 (a)
2.7
2.7
8.9
98.3
89.7
89.5
2.5
3/19/75
8
3.4 (a)
2.2
1.7
9.2
96.7
89.5
90.4
3/20/75
9
3.6 (a)
2.7
1.8
9.15
93.8
96.3
87.3
3/20/75
10
15.8 (a)
4.1
2.0
9.0
71.4
78.0
62.0
4/15/75
11
6.6 (a)
1.6
1.7
9.0
69.8
82.2
83.6
4/15/75
12
2.3 (a)
1.8
1.4
8.75
72.2
81.9
82.4
4/16/75
13
14
2.3 (a)
3.3 (a)
33.7
(KOH)
26.2
(KOH)
1.4
1.4
10.5
10.5
76.1
71.5
81.4
81.5
•H CO
en «
CD Q0
4/16/76
4/16/76
APPENDIX C
-------
PILOT SCALE FLOTATION SUMMARY
(Page 2 of 2)
Run
REAGENT FEEDS (lb/ton)
PH
RESULTS
COMMENTS
*
Collector
Na2C03
' sio2
% CaCO,
* Recovery
Cone.
Feed
Tail 8
17
4.2 (a)
12.5
0
8.5
82.9
71.0
61.6
51.5%
8/27/75
18
4.6 (a)
16.3
0
8.6
81.7
71.0
67.2
30.2
8/27/75
19
2.2 (a)
6.6
2.5
8.9
80.5
82.7
82.5
9/22/75
20
2.5 (a)
8.3
1.0
9.0
80.5
81.5
80.9
9/23/75
21
2.5 (a)
0
1.0
8.1
80.8
81.5
81.3
9/23/75
22
2.5 (a)
0
1.0
8.3
81.0
81.5
81.6
9/23/75
23
7.8 (a)
11.8
0
9.15
95
80.6
56.8
73.4
12/16/75
24
6.1 (a)
11.6
0
9.2
94.2
80.9
58.5
73.1
12/16/75
25
11.6 (a)
14.6
0
9.22
94.2
79.0
53.7
74.5
12/16/75
26
9 (a)
10.8
1.6
9.5
98.7
94.1
76.8
82.9
1/12/76
4.8% Solids
27
4.3 (a)
7.7
(a) I
(b)
1.7
resinate 81
Sodium Alky
9.3
1 Sulfate
99.4
95.7
1
i
i
77.5
86.3
1/13/76
6.7% Solids
APPENDIX C
-------
BENCH SCALE FLOTATION SUMMARY
Peed Grade - 86.3% CaC03
Date Collected - 2/24-27/75
Run
#
REAGENT FEEDS
(lb/ton)
PH
RESULTS
COMMENTS
Collector
Na23
sxo2
% CaC03
% Recovery
Cone.
Tails
2/24
1
Kb)
l
1
9.0
89.6
First Clnr.
2/24
2
2(b)
l
1
9.0
88.9
Rougher
2/24
3
2(b)
1
2
9.2
91.4
77.5
Rougher
2/24
4
2(b)
l
3
9.0
90.2
76.8
Rougher
2/25
1
2(g)
1
2
9.0
97.2
69.0
69.1
Second Clnr.
2/25
2
2(b)
l
2
9.0
97.8
Second Clnr.
2/27
1
2(b)
1
2
9.4
98.3
52.5
2/27
2
2(b)
0
2
9.3
98.2
60.0
2/27
3
2(b)
4
2
9.65
98.9
60.0
(b
Drcsinate
81
»
Mixture Sodi
lsi Aklyl Sulfc
te and Dresii
ate
APPENDIX C
-------
Peed Grade - 79.0% CaC03 BENCH SCALE FLOTATION SUMMARY
Magnesium - 2.9 ng/g
Date Collected - 2/28/75
Run
*
REAGENT FEEDS
(lb/ton)
pH
RESULTS
COMMENTS
Collector
Na2C03
si°2
% CaC03
% Recovery
Cone.
Tails
l
2(b)
1
2
9.1
78.5
67,5
5% pulp
First Clnr
2
2(b)
1
2
9.2
88.3
72.5
10% pulp
Second Clnr
3
2(b)
1
2
9.1
96.8
72.5
15% pulp
Second Clnr.
(b) t
iresinate 8J
APPENDIX C
-------
BENCH SCALE FLOTATION SUMMARY
Feed Grade - 91.8% CaC03
Date Collected - 3/17-18/75
Run
REAGENT FEEDS
{lb/ton)
RESULTS
COMMENTS
*
Collector
Na2C03
Si02
pH
% CaC03
% Recovery
Cone.
Tails
3/17 1
0.5(b)
1
2
9.45
97.4
89.7
28.9
Second Clnr.
3/17 2
4(b)
1
2
9.45
98.3
83.3
60.7
M «l
3/17 3
2(b)
1
0.5
—
97.5
84.0
61.4
«t II
3/17 4
2(b)
1
4
—
96.8
85.6
58.4
*1 II
3/18 1
0.5(g)
1
2
—
95.4
90.5
First Clnr.
3/18 2
4(g)
1
2
—
96.0
28.8
97.4
Second Clnr.
(b)
(g) Mj
Dresinate 8
xture Sodium
I
ilkyl Sulfate
and DresJ
nate (1:1)
APPENDIX C
-------
Feed Grade - 87.0% CaCO-
Magnesium - 4.7 mg/g
BENCH SCALE FLOTATION SUMMARY
Date Collected - 4/14-16/75
Rui>
i
Collector
REAGENT FEEDS (lb/ton)
Na2C03
SiO.
PH
RESULTS
% CaCO.
Cone. Tails
% Recovery
COMMENTS
4/14 1
4/14 2
4/14 3
4/15 1
4/16 1
2.8(g)
5.7(g)
5(g)
4 (b)
4 Cb)
2
2
2
(f)
9.20
9.35
9.33
11.0
11.0
76.4
91.6
76.3
87.7
79.8
79.8
68.1
79.3
87.8
75.4
85
No Selectivity
No Selectivity
No Selectivity
No Selectivity
(b) Dresi
-------
Peed Grade - 78.4% CaCO-
BENCH SCALE FLOTATION SUMMARY
Date Collected - 4/18/75
Run
#
Collector
REAGENT FEEDS (lb/ton)
Na2C°3
sla
pH
RESULTS
tcsot:
COMMENTS
Cone
Tails
% Recovery
3(b)
3(b)
3(b)
3(b)
3(b)
3(b)
3(b)
NOT I
(f)
(f>
(f)
(f)
(f)
(f)
(f)
(b) Dres
(f) KOH
w& ;
inate 81
added
High pH f
magnesium
otation was
as Hg(OH)2 -
inoperative at this tii
10.8
10.9
9.1
9.0
11.0
11.0
9.0
78.2
77.9
73.4
79.9
77.8
78.0
74.2
81.2
78.4
79.4
77.5
80.0
78.5
76.7
e raluated in at attenpt
: lecycle of si
*dge was ;
to.precipiti
uspected of
ite
being
No Selectivity
No Selectivity
No Selectivity
No Selectivity
No Selectivity
No Selectivity
No Selectivity
APPENDIX C
-------
Feed Grade - 86.7% CaC03
Magnesium - 5.3 mg/g
BENCH SCALE FLOTATION SUMMARY
Date Collected - 5/13-14/75
Run
#
REAGENT FEEDS (lb/ton)
Collector
Na2C°3
SiO,
PH
RESULTS
% CaCO.
Cone. Tails
% Recovery
COMMENTS
O
5/13 1
5/13 2
5/13 3
5/13 4
5/14 1
5/14 6
8(d), 2(b)
2(b)
9(d)
2(d)
1(b)
2(d)
1(b)
6(d)
(Total of 40 T^sts)
0
0
0(f)
0
8.5
8.5
11.1
11.1
9.0
9.0
85.9
85.1
83.1
84.8
86.3
84.5
86.2
86.2
85.1
85.8
84.6
84.7
NOTE:
(b) Dresifcate 81
(d) Palomyn
(f) KOH usee
to adjust pH
oleic acii
Forty (40) bencq scale flotation tests wer^ perforaejd
coi
Pa]
add
cbinations of JPalomyn, a relatively pure
omyn showed rjo additional selectivity w}th poor
ed frother wajs of no benefit i
using var
arid Ores
rothing.
Aj i
ous
nate
y
No Selectivity
No Selectivity
No Selectivity
No Selectivity
No Selectivity
No Selectivity
APPENDIX C
-------
Feed Grade - 86.5% CaC03
Magnesium - 5.8 ng/g
BENCH SCALE FLOTATION SUMMARY
Date Collected - 6/10/75
Particle Size in Microns
Cu
Max.
Min.
D65
d60
D10
20
0.8
5.6
5.0
1.5
3.33
Run
*
REAGENT FEEDS
(lb/ton)
pH
RESULTS
COMMENTS
Collector
Na2C03
Si02
% CaCO^
fc Recovery
Cone.
Tails
1
3(a)
2
2
8.7
84.5
Tests 1 through
5, #1 Cleaner
only
2
2(a)
1
0
8.6
85.4
—
—
—
1(d)
3
3(a)
2
l
8.7
86.3
—
—
Washed
4
2(a)
2
l
9.0
86.0
Acidified to
pH 6.5, decant
5
2(a)
3
l
9.0
90.2
—
—
Acidified to
pH 6.5, vacuum
filtered
(a) Dresinate
81
(d) Paxnolyn
APPENDIX C
-------
BENCH SCALE FLOTATION SUMMARY
Feed Grade - 11.1% CaC03
Magnesium - 6.2 ng/g Date Collected - 7/3/75
Run
REAGENT FEEDS
(lb/ton)
pH
RESULTS
COMMENTS
.#
Collector
Na2C°3
si02
% CaC03
% Recovery
Cone.
Tails
1
4(a)
2
2
9
69.1
Acidified
decant
2
3(a)
2
2
9
69.1
—
—
Scrub, decant
twice
(a) Dresina
te 81
!
APPENDIX C
-------
BENCH SCALE FLOTATION SUMMARY
Feed Grade - 71.8% CaCO^
Magnesium - 5.0 rag/g Date Collected - 8/27/75
Run
I
REAGENT FEEDS
-------
BENCH SCALE FLOTATION SUMMARY
Peed Grade - 82.7% CaC03
Magnesium - 5.0 mg/g Date Collected - 9/22/75
Run
#
REAGENT FEEDS
(lb/ton)
PH
RESULTS
COMMENTS
Collector
Na2C03
Si02
% CaC03
% Recovery
Cone.
Tails
1
3(a)
4
2
9
83.0
80.7
87.3
Ro only tests
1-5
2
3(a)
4
2
9
84.5
79.3
66.8
Scrub/ pres-
sure filter
3
3(a)
4
2
9
82.9
81.0
94.6
Temp. 36°C
4
3(a)
7
2
9
84.4
78.6
72.1
Acidify, pres-
sure filter
5
4(a)
w/HCl
2
7.5
81.9
—
—
6
3(a)
4
2
9
87.7
79.9
38.1
Scrub, pres-
sure filtef
7
3(a)
4
2
9
88.8
79.5
36.9
n n
8
3(a)
—
2
7.5
90.0
77.3
46.3
ft H
9
3(a)
4
0
9
88.6
80.1
32.8
(a) Dresinat
e 81
1
APPENDIX C
-------
Peed Grade - 88.3% CaCO-
BENCH SCALE FLOTATION SUMMARY
Date Collected - 10/8/75
Particle Size in Microns
Cu
Max.
Min.
°65
°60 !dio
40
1.5
15.5
14.4 4.4 13.27
Run
*
REAGENT FEEDS
(lb/ton)
pH
RESULTS
COMMENTS
Collector
Na2C03
Si02
% CaC03
% Recovery
Cone.
Tails
1
1.9(b)
3.3
1.0
9.0
96.8
73.3
70.0
2
2.8(b)
3.3
1.0
9.0
97.2
70.4
73.5
3
3.8(b)
3.3
1.0
9.0
96.6
61.3
83.7
4
2.8(b)
3.3
2.8
9.0
96.9
70.4
74.1
5
2.3(b)
3.3
0.0
9.0
96.7
65.6
79.9
6
2.8(b)
3.3
1.0
9.0
96.1
64.0
82.4
+1.0 lb/ton
7
2.8(b)
0.0
1.0
7.0
96.9
86.7
17.2
pH adjusted
v/HCl (d)
8
2.8(b)
0.0
1.0
8.0
96.6
74.7
67.9
9
0.5(b)
3.3
1.0
9.0
88.3
0.0
10
1.0(b)
3.3
1.0
9.0
97.1
85.9
23.6
11
4.7(b)
3.3
1.0
9.0
96.7
57.1
86.3
12
5.7(b)
3.3
1.0
9.0
95.8
50.0
90.7
13
2.8(b)
3.3
0.0
9.0
97.0
68.4
76.4
14
2.8(b)
3.3
0.5
9.0
96.7
67.3
78.2
15
2.8(b)
3.3
3.7
9.0
96.8
65.3
80.0
16
2.8(b)
3.3
4.5
9.0
97.2
67.4
77.2
17
2.8(b)
4.6
1.0
9.5
97.1
63.1
81.5
18
2.8(b)
5.1
1.0
10.0
97.6
72.5
69.6
19
2.8(b)
3.3
1.0
9.0
#3-97.4
67.4
76.8
#2-96.9
63.0
83.1
#1-96.0
59.4
85.8
Ro-92-8
54.5
92.7
(b) Dr«
sinate TX60W;
(d) Palomyn
APPENDIX C
-------
BENCH SCALE FLOTATION SUMMARY
Feed Grade - 91.9% CaC03
Magnesium - 6.3 ag/g Date Collected - 10/8/75
Run
i
REAGENT FEEDS
(lb/ton)
PH
RESULTS
COMMENTS
Collector
Na2C03
SX02
% CaC03
% Recovery
Cone.
Tails
i
3(a)
4.0
0.2
9.0
92.0
91.6
75.1
RO only tests
1-11
2
3(d)
4.0
0.0
9.0
93.3
89.0
68,5
3
3(a)
4.0
0.2
9.0
92.9
92.1
75.2
Scrub, pres-
sure filtered
92.7% CaCO-j
4
3(a)
1.9 Ca(OH)
0.0
9.0
92.5
90.8
65.1
5
3(a)
1.9 Ca (OH) _
o.o
9.0
92.7
90,2
68.6
6
3(a), 3(c)
0.3 Ca(OH)2
0.5
9.0
92.7
90.7
60.5
7
3
-------
. „ BENCH SCALE FLOTATION SUMMARY
Feed Grade - 94.6% CaC03
Magnesium - 6.3 mg/g Date Collected _ 10/30/75
Particle Size in Microns
Cu
Max.
Min.
D65
d60
D10
25
1.6
8.2
7.8
3.75
2.08
O
oo
Run
REAGENT FEEDS
(lb/ton)
RESULTS
COMMENTS
«
Collector
Na2C03
Si02
pH
% CaC03
% Recovery
Cone.
Tails
1
7(b)
2.3(d)
6
1.5
9.1
96.8
89.2
72.7
One Cleaner
2
4(b)
3(e)
2.3(d)
4
1.0
9.0
96.3
92.0
61.6
n m
3
5(b)
1.2(d)
4
1.5
9.4
95.9
90.1
78.7
Scrub Pres-
sure filter
4
4(b)
0.9(d)
1.5
1.5
9.15
95.2
90.0
94.5
w
(b) Dresina
;e TX60W; (d)
Pamolyn; (e)
Sodium Alk
yl Sulfate
APPENDIX C
-------
Feed Grade - 96.1% CaC03 BENCH SCALE FLOTATION SUMMARY
Magnesium - 15.2 mg/g DATE COLLECTED - 10/30/75
Particle Size in Microns
Cu
2.54
Max.
Min.
Dfif>
DfiO
Dm
65
2.85
18.2
17
6.7
Run
#
Collector
REAGENT FEEDS (lb/ton)
Na2C03
sla
PH
RESULTS
tot:
Cone.
Tails
% Recovery
COMMENTS
4(b)
1.4(d)
4(b)
4(b)
3(b)
1.7(d)
1.0
2.5
2.5
3.5
9.03
9.3
9.35
9.35
97.8
97.4
97.7
98.1
(b) Dresiriate TX60W
(d) Sodium Alkyl Sulfate
78.4
89.9
85.9
84.0
94.1
83.8
87.9
87.6
Carbonated, pres
sure filtered
96.4% CaC03;
11 mg/g Mg;
#1 CI only
Scrub, pres-
sure filtered
APPENDIX C
-------
Feed Grade - 92.3% CaC03 BENCH SCALE DOTATION SUMMARY
Magnesim - 6.0 mg/g Date Collected - 11/17/75
Particle Size in Microns
Cu
Max.
Kin.
DfiS
Dfio
Din
33
1
14.5
13.8
4.85
2.85
Run
#
REAGENT FEEDS
(lb/ton)
pH
RESULTS
COMMENTS
Collector
Na2C°3
Si62
% CaC03
% Recovery
Cone.
Tails
1
3(b)
3.5
0.0
9.0
96.0
93.1
35.2
Peed 94.1,
tests 1, 2
2
5(b)
3.5
0.0
9.0
97.3
87.9
68.2
3
6(b)
3.5
0.5
9.0
95.7
82.7
76.7
Peed 92.3
4
6(a)
3.5
0.5
9.0
95.1
90.6
38.9
5
2.4(b)
6
0.7
9.3
93.0
49.8
Peed 90.4
6
2.4(b)
4
0.7
9.6
93.2
35.9
7
5(b)
12
1.0
9.4
93.7
60.0
3 Clnrs
tests 7-13
8
5(b)
10
1.0
9.5
94.7
65.1
9
5(b)
10
0.0
9.5
94.9
43.6
10
5(b)
10
0.0
9.7
95.1
68.6
11
5(b)
23
1.0
9.7
94.0
86.0
12
5(b)
23
0.0
9.6
93.8
85.6
13
5(b)
12
1.0
9.3
94.2
82.6
(a) Dresin
ate 81; (b) D
'esinate TX60
1
APPENDIX C
-------
„ „ „ , BENCH SCALE FLOTATION SUMMARY
Peed Grade - 86.3% CaCO-j
Magnesiua - 3.0 mg/tf Date Collected - 12/15/75
Particle Size in Microns
Cu
2.85
Max.
Min.
Dfin
DlO
33
1
14.5
13.8
4.85
Run
REAGENT FEEDS
(lb/ton)
PH
RESULTS
COMMENTS
#
Collector
Na2C°3
sio2
% CaC03
% Recovery
Cone.
Tails
1
4(b)
3.0
0.5
9.1
98.2
44.2
88.7
2
4(a)
3.0
0.5
9.1
97.5
38.6
91.5
3
4(c)
3.0
0.5
9.1
98.6
64.5
73.0
4
1(b)
3.5
0.5
9.0
98.9
67.1
69.2
5
3(b)
3.5
0.5
9.0
97.7
46.6
88.0
6
5(b)
3.5
0.5
9.0
97.6
42.0
90.1
7
8(b)
3.5
0.5
9.0
97.2
36.2
92.5
8
5(b)
3.5
2.0
9.0
98.5
46.3
87.5
9
5(b)
3.5
0.0
9.0
97.3
38.6
91.6
10
5(c)
3.5
0.5
9.0
98.9
53.8
82.6
11
5(a)
3.5
0.5
9.0
97.9
38.8
91.2
12
3(c)
3.5
0.5
9.0
99.1
76.7
49.2
13
3(a)
3.5
0.5
9.0
97.9
44.7
88.7
14
3(b)
3.5
0.5
9.0
93.5
77.4
60.0
Temp. - 35°C
15
5(b)
3.5
0.5
9.0
94.1
83.7
27.3
Tenp. - 35#C
16
5(b)
3.5
0.5
9.0 - 6.6
83.4
85.4
—
Float with C02
17
5(b)
3.5
0.5
9.0
98.3
53.5
83.4
5% solids pulp
18
5(b)
3.5
0.5
9.0
98.4
36.1
91.9
10% solids pulp
19
5(b)
3.5
0.5
9.0
97.8
34.4
92.8
15% solids pulp
20
5(b)
3.5
0.5
9.0
95.6
36.1
93.5
20% solids pulp
21
5(b)
17.0
0.5
9.0
97.6
57.6
81.1
3.2 mg/g Mg
22
5(b)
40.0
0.5
9.0
97.6
63.5
75.6
8.8 mg/g Mg
23
5(b)
80.0
0.5
9.0
100.0
67.7
66.7
14.4 OKj/g Mg
24
5(b)
145.0
0.5
9.0
97.7
76.8
51.5
19.1 tag/g Mg
25
7(c)
3.5
0.5
9.0
98.1
54.7
62.8
26
12(c)
3.5
0.5
9.0
97.6
46.9
87.9
27
5(b)
3.5
0.0
9.1
28
5(b)
3.5
3.0
9.1
29
5(b)
3.5
0.0
9.1
Temp.- 27°C
30
5(b)
3.5
3.0
9.1
Tenp. - 27°C
31
5(b)
3.5
0.0
9.1
Temp. - 35®C
32
5(b)
3,5
3.0
9.1
Tenp. - 35*C
(a) Dresinate 81; Cb) Dresinate TX60M; (c) Dresinate DS60M
APPENDIX C
-------
Feed Grade - 94.9% CaCO.
BENCH SCALE FLOTATION SUMMARY
Date Collected - 12/17/75
Run
#
Collector
REAGENT FEEDS (lb/ton)
Na2c°3
SiO.
pH
RESULTS
% CaCO.
Cone.
Tails
% Recovery
COMMENTS
to
5.2(b)
2.6(b)
4(b)
3.8
1.3
1.9
0.5
0.0
0.5
9.4
9.1
9.1
98.3
98.8
99.7
71.7
80.7
67.2
90. 3
81.7
86.8
(b) Dresinate TX60W
Carbonated
pressure
filtered
93.7% CaC03
APPENDIX C
-------
Peed Grade - 65.0% CaCO-
Magnesium - 6.5 rag/g
BENCH SCALE FLOTATION SUMMARY
Run
REAGENT FEEDS
(lb/ton)
RESULTS
COMMENTS
«
Collector
Na2C03
Si02
pH
% CaC03
% Recovery
Cone.
Tails
l
2.5
6
0
9.0
86.6
Third Clnr
2
2.5
6
1
9.0
85.2
—
n n
3
5.0
4
0
9.0
86.2
35.3
77
» n
4
5.0
5
0
9.0
78.6
44.8
72
•• n
5
5.0
5
2
9.0
81.3
37.3
78
- a
6
5.0
5
0
9.0
79.1
42.8
74
Second Clnr
7
7.0
7
1
9.0
85.4
36.7
73
Third Clnr
8
7.0
6
2
9.0
83.9
38.5
75
Second Clnr
9
7.0
6
4
9.0
87.1
40.8
70
Second Clnr >
10
7.0
6
0
9.0
92.0
44.3
61
Fourth Clnr '
i
11
7.0
6
2
9.0
92.2
55.5
37
1
Fourth Clnr j
12
7
6
4
9.0
92.0
50.1
50
Fourth Clnr !
NOTE:
Carbonated pr<
All Runs
ssure filter*
)resinate 1
>d accelat
X-60W
.or sludge
_L
APPENDIX C
-------
APPENDIX D
CONVENTIONAL SLUDGE DEWATERING LAGOON
PILOT EQUIPMENT
INCLUDING PRECIPITATOR, THICKENER,
PRODUCT STORAGE, AND SLUDGE HANDLING BUILDING
114
-------
APPENDIX D
CARBONATION OF SOFTENING SLUDGE USING 18% CO-
i
VACUUM FILTRATION OF CARBONATED SLUDGE
115
-------
APPENDIX D
PILOT FLOTATION CIRCUIT
FROTH OVERFLOW, CALCIUM CARBONATE FLOTATION
116
-------
GLOSSARY OF FLOTATION TERMS
activator: A reagent which increases the flotation activity
and aids flotation of a mineral or minerals,
beneficiation: Improvement of the mineral purity of a pulp?
removal of impurities from a valuable mineral thereby
increasing its value and suitability for use.
calcite: The predominant mineral form, crystalline in structure,
of calcium carbonate.
cleaners: Sequential flotation steps, in addition to the rougher,
to further purify concentrates.
collector: A substance which when added to the mineral pulp
attaches to, coats, or is adsorbed by normally nonfloating
minerals making them capable of adhering to gas bubbles.
concentrates: Flotation products of improved purity containing
a minimum amount of impurities.
depressant: A reagent which lowers the flotation activity
and suppresses or prevents flotation of a mineral or
minerals.
fatty acid: An organic compound consisting of the carboxyl
radical - COOH, and an alkyl (non-cyclic hydrocarbon) or
aryl (cyclic hydrocarbon) radical, usually represented
by R.
flotation circuit: An arrangement or sequence of flotation steps
designed to optimize the separation of impurities from
valuable mineral.
froth: A formation on the water surface of gas bubbles, pulp
liquor, collected mineral and a certain amount of uncollected
but entrapped mineral, resulting from the rise to the surface
of dispersed bubbles.
froth flotation: A process for separating finely divided solids
from each other in a water suspension, resulting from the
adhesion of some species of solids to generated or introduced
gas bubbles and the simultaneous adhesion of other species
of solids to water.
frother: Generally organic substances which, when added to water,
lower the surface tension and cause the formation of foam
on the water surface upon introduction of gas bubbles.
117
-------
grade: The relative purity (normally in percent) of a flotation
product? the percentage of a flotation product which
consists of valuable mineral.
limestone: The predominant mineral rock containing calcium
carbonate,
modifier: An additional flotation agent added to modulate
the action of collectors and to obtain the most effective
control.
pulp: Finely divided solids in a water suspension.
recovery: The percentage of valuable mineral present initially
which is now available in purified form.
rougher: The initial flotation step in which the bulk of separation
is formed.
selectivity: Measure of the ability of a collector or modifier
to affect the floatability of only a specific mineral or
group of minerals.
tailings: Flotation products to be discarded containing the
majority of impurities and a minimum of valuable mineral.
i
118
-------
TECHNICAL REPORT DATA
(Pleas* nod hutructions on the reverse before completing]
1. REPORT NO. 2.
EPA-600/2-76-285
3. RECIPIENT'S ACCESSlOr+NO.
4. TITLE AND SUBTITLE
RECOVERY OF LIME AND MAGNESIUM IN POTABLE
WATER TREATMENT
5. REPORT OATE
December 1976 (Issuina Date)
6. PERFORMING ORGANISATION CODE
7. AUTHOR(S)
C. G. Thompson and G. A. Mooney
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Black, Crow & Eidsness, Inc.
807 South McDonough
Montgomery, Alabama 36104
10. PROGRAM ELEMENT NO.
1CC614
11. CONTRACT/GRANT NO.
Grant No. S803194-01-4
12. SPONSORING AGENCY NAME AND AOORESS
Municipal Environmental Research Laboratory—Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY COOE
EPA/600/14
15. SUPPLEMENTARY NOTES
This project was carried out under Demonstration Grant No. S803194-01-4
with the Water District No. 1 of Johnson County, Kansas.
16. ABSTRACT
A hard, turbid surface water was successfully treated using the
magnesium carbonate process in a 2 mgd pilot plant at the treatment work >
of Water District No. 1 of Johnson County, Kansas, for one year during
1975 and 1976. During this study, froth flotation was used to separate
river sediments from calcium carbonate formed in the treatment process.
Both bench-scale and pilot plant flotation tests have shown that sludges
formed by softening turbid waters can be processed to yield a relatively
pure calcium carbonate suitable for lime recovery. Prior to this work
lime sludge from surface water treatment had not been useable for lime
reclamation. Process variables affecting both magnesium carbonate
recovery and calcium carbonate beneficiation were studied in this work.
Magnesium carbonate was successfully produced on a continuous pilot seal
from recycled magnesium bicarbonate liquor. Process economics were
favorable. A comparison of capital and operating costs for magnesium
carbonate treatment, sludge flotation and lime recovery with present
operating costs, including waste disposal, indicated that annual costs
would be lower with the new technology.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENOED TERMS
c. cos AT i Field/Group
beneficiation magnesium oxides
calcium carbonates reclamation
calcium oxides water softening
coagulation water treatment
flotation
froth
magnesium carbonates
Kansas River
Johnson County, KN
Vlagnesium recovery
lime softening
life recovery
magnesium coagulant t<
sludge recovery reuse
13B
:ch.
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. N-
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
frT/?7
EPA Form 2220-1 (»-73) \
1 >U S. GOKMMEIT PRMIIN6 OlflCE: 1977-757-056/5571* Region No. 5*11
-------
|