TILEAl
WATER POLLUTION CONTROL RESEARCH SERIES • 12040 ENC 12/71
Color Removal From
Kraft Pulping Effluent
By Lime Addition
ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development and demonstration
activities in the Environmental Protection Agency, through
inhouse research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control -Research
Reports should "be directed to the Chief, Publications Branch
("Water), Research Information Division, R&M, Environmental
Protection Agency, "Washington, B.C. 20^60.
-------
COLOR REMOVAL FROM KRAFT
PULPING EFFLUENT BY LIME ADDITION
Interstate Paper Corporation
Riceboro, Georgia 31323
for the
Environmental Protection Agency
Program #12040 ENC
Grant #WPRD 183-01-68
December 1, 1971
-------
EPA-Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency nor -does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
For sale by the Superintendent ol Documents, U.S. Government Printing Office, Washington, D.C. 20102 - Price $1.26
ii
-------
ABSTRACT
A prototype color removal system was designed, constructed
and operated as an integral part of a tertiary treatment system
for total process effluent from a kraft linerboard mill. The basic
system includes a lime precipitation process for the removal of
color combined with primary clarification folio-wed by natural
biochemical lake stabilization and mechanical aeration.
Operating results show that the color removal system can operate
successfully under -widely varying conditions to give a relatively
constant effluent color in the range of 125 ppm APHA color units
at treatment levels of 1000 (+_ 50) ppm of calcium hydroxide with
untreated effluent colors in the range of 1200 (+ 200) ppm. Treat-
ment at this level reduces lime cost to $53. 73 per million gallons
with lime at $15.35/ton (90% CaO). Performance is directly
related to cqntrol of lime feed. Equipment evaluation indicated
substantial savings in capital cost for future installations.
Recovery of calcium used was carried out under mill conditions
on a continuous basis following a statistically designed program.
Results and full size design factors are given.
Performance of natural biochemical stabilization following lime
treatment is shown graphically. Overall BOD^ reduction for the
tertiary treatment system is 98% with a final discharge average
BOD5 of 6 ppm.
This report was submitted in fulfillment of Program No. 12040
ENC, Grant No. WPRD 183-01-68 under the partial sponsorship
of the Water Quality; Office, Environmental Protection Agency.
-------
CONTENTS
Section Page
I Conclusions 1
II Introduction 5
III Objectives 7
IV Chronology 9
Mill Site 9
Stream Surveys and Cooperative 9
Research Studies
Construction Permit 12
Mill Description 12
Effluent Treatment Facilities 13
Estimated Cost and Federal 14
Participation
V Lime Treatment - Materials and Methods 17
A. System 17
Mill Operating Conditions 17
Process Effluent Flow 17
Chemical Flow 19
B. Operation and Results 21
Initial Operation 21
Operating Period - June 27, 1968 through 22
September 30, 1968
Operating Period - October 1, 1968 through 23
October 31, 1968
Operating Period - November 1, 1968 25
through November 30,
1968
Operating Period - December 1, 1968 27
through December 3.1,
1968
Operating Period - January 1, 1969 28
-------
January 31, 1969
Operating Period - February 1, 1969 30
through February 28,
1969
Operating Period - March 1, 1969 32
through March 31,
1969
Operating Period - April 1, 1969 34
through April 30, 1969
Operating Period - May 1, 1969 36
through May 31, 1969
Operating Periods - June 1, 1969 through 39
July 31, 1969
Operating Period - August 1, 1969 through 44
August 31, 1969
Operating Periods - September 1, 1969 47
through October 31,
1969
Operating Periods - November 1, 1969 50
through December 31,
1969
C. Summary 54
Equipment Evaluation 54
Lift Pumps 54
Chemical Feed System 54
Flash Mixing 55
Flocculation 55
Clarification 56
Natural Stabilization 56
Operating Results 58
Lime Treated Effluent Color 58
Lime Treated Effluent COD & BOD 58
Natural Stabilization of Lime Treated 61
Effluent
Lime Treatment Cost 63
VI Calcium Recovery - Materials and Methods 65
A. Carbonation Process 65
vi
-------
B. Pilot Plant
65
Design 65
Layout and Orientation °°
Startup and Shakedown 7^
Test Procedures 71
Calcium 72
Carbon Dioxide for Sodium 72
Alkalinity
77
Percent Carbon Dioxide in '
Gas Feed
Densator Slurry Underflow 72
Specific Gravity
Calcium Content
Filter Leaf Test 72
73
Slurry Carbonate Purity
C. Methodology Used in Computer Program 73
•70
Program '
Data Analysis System 7°
Study Results 76
Mathematical Models of System 86
Reaction Rate Model 86
D. Carbonate Slurry 91
E. Instrumentation Evaluation 94
F. Summarization of Carbonation Process 94
VII Acknowledgments 97
VIII References 99
IX Appendix 101
vii
-------
FIGURES
No. Page
1 Vicinity Map 10
2 Effluent Treatment Flow Diagram 18
3 Treated Effluent Color Vs Lime
Concentration 59
4 Treated Effluent Color Vs COD and BODg 60
5 Stabilization Lake Color and BOD Profile 62
6 Percent Color Reduction and Cost Vs
Lime Concentration 64
7 Carbonator 67
8 Densator 67
9 Pilot Plant Flow Diagram 68
10 Pilot Plant Orientation Photograph 69
11 Function Block Diagram Carbonation
Pilot Plant 74
12 Reaction Rates, CO2 Transferred, % 88
13 Efficiency, Total Calcium Loss,% 89
14 Reaction Rate Vs Carbonation Depth and
Densator Recycle Rate 90
15 System Efficiency Vs Gas Feed
Temperature and Carbonation pH 92
16 EIMCO Filter Leaf Test Densator Sludge 93
vlii
-------
TABLES
No. Page
1 BOD Reduction Vs Color Reduction 26
2 BOD Reduction Vs Color Reduction 28
*
3 COD & BOD Reduction Vs Color Reduction 30
4 COD & BOD Reduction Vs Color Reduction 32
5 COD & BOD Reduction Vs Color Reduction 34
6 COD & BOD Reduction Vs Color Reduction 36
7 COD & BOD Reduction Vs Color Reduction 38
8 COD & BCD Reduction Vs Color Reduction 41
9 Inline Mixer Evaluation Treatment Data 42
10 COD & BOD Reduction Vs Color Reduction 47
11 Effluent Distribution 49
12 BOD Reduction Vs Color Reduction 50
13 Effluent Distribution 51
14 BOD Reduction Vs Color Reduction 52
i
15 Clarifier Calcium Balance 53
16 Average Mill Effluent Discharge Comparison 61
17 Carbonation Experimental Program
Parameters 75
18 List of Other Variables Measured 77
19 Carbonation Pilot Plant Operating Conditions 78
ix
-------
TABLES
No. Page
20 Results of Statistically Designed Carbonation
Experiment 79
21 Carbonation Pilot Plant Operating Conditions
Duplicate Runs 80
22 Results of Statistically Designed Carbonation
Experiment Duplicate Runs 81
23 Means and Standard Deviations of All
Variables 83
24 List of Variables Eliminated Due to
Correlation 84
25 Means and Standard Deviations of Selected
Variables 85
26 Simple Correlation Coefficients 87
-------
SECTION I
CONCLUSION
A color removal system in a tertiary treatment process for a kraft
pulp mill can operate successfully under widely varying conditions to
give a relatively constant effluent color in the range of 125 ppm APHA
color units at treatment levels of 1, 000 ppm (+ 50 ppm) of calcium
hydroxide with untreated effluent colors in the range of 1, 200 ppm
(jf 200 ppm). Treatment at this level offers advantages in cost
savings and in obtaining sludges -which are more suitable for dewater-
ing. Lime costs are $53. 73 per million gallons with lime at $15. 35/
ton (90% CaO). Performance is directly related to control of lime
feed.
Operating experience and studies of the process have shown that flash
mixing of lime slurry with the mill effluent prior to flocculation has
no effect on the reduction of color. This is particularly so as floccu-
lation has to be at stirring rates sufficient to keep carbonate mud and
other solids in the mill effluent in suspension. These other solids
including wood fiber aid clarification. Because of this, clarifiers for
this service can be conservatively designed at rise rates of 0. 5
gallons per minute per square foot of clear rise area. Flocculation
time can be reduced to thirty-five minutes. The smaller equipment
requirements offer substantial savings in future installations. How-
ever, cost should not be a factor in the selection of untreated effluent
pumps. As calcium precipitation and stabilization following clarifi-
cation of treated effluent is not part of the color removal process at
Interstate, 140 acre lake surface area is required for six mgd for
natural absorption of carbon dioxide from the atmosphere before
natural biochemical treatment is effective.
Additional advantages of the color removal process are that it
preconditions the effluent for biochemical treatment allowing rapid
degradation. Foaming is eliminated and phosphorus concentration
in the effluent is reduced thereby preventing possible eutrophication.
High alkalinity in the process provides sterilization so treated
domestic sewage does not require chlorination when handled through
the system.
Because lime precipitation for color removal in tertiary treatment
of kraft pulp mill effluent is a new development, the facilities at
Interstate Paper Corporation's mill have been sponsored by the
Environmental Protection Agency as a demonstration plant. In
-------
keeping with this intent, operating information has been made available
to interested parties and to others in the pulp and paper industry.
Unpublished interim reports have been presented at an Auburn
University engineering short cpurse and at a regional meeting of the
National Council for Air and Stream Improvement. The plant has
been open for inspection. The number of visitors to date number 238
including paper mill representatives from across the United States
and representatives from Canada, Finland, Sweden, and France. With
the interest shown in the operating experiences at Interstate there will
be further improvements and acceptance of the process.
The impact of the treated effluents from the mill on the environment
is particularly noteworthy. The stabilization lake has become a
natural preserve for egrets, ducks, and geese. There are also otters,
alligators, and deer which feed in the area. The lake has been stocked
with bream and bass by the Georgia State Game and Fish Commission.
It is,to the author's knowledge,the first time that an effluent treatment
facility is also being used as a sport fishing lake.
The extensive combination primary, secondary and tertiary treatment
plant at Interstate Paper Corporation has successfully demonstrated
that the pulp and paper industry can be productive and protect a
healthy environment.
In designing a carbonation system for recovery of calcium carbonate
and the stabilization of tie highly alkaline kraft pulp mill effluent
following color removal, the major factor is the formation of the
calcium carbonate precipitate in the carbonation zone. With careful
attention to this part of the physical and chemical reaction process,
maximum calcium recovery can be obtained without the use of
coagulant aids at high clarification flow rates. There are several
standard designs of water treatment equipment on the market by
different companies which can be used with some modification.
The following general specifications for process equipment are
conservative at a 10 million gallons per day rating.
Carbonation Zone
Counter current gas and water flow
Carbonation depth - 10. 5 ft
Residence time - 16-20 min
-------
Recycle to Carbonator Feed
Volume - 25% of total flow
Solids - 135-275 lbs/1000 gal
Clarification Zone
Rise rate - 2. 50 gal/min/sq ft
Residence time - 60 min
Sludge zone - 20% of total volume
Blow down - 0. 33% of total feed
Sludge concentration - 2.25 Ibs/gal
If a combined carbonation and clarification unit is selected, the
approximate size of the carbonation zone would be 40 ft in diameter
10. 5 ft deep in a 72 ft diameter clarifier with a 15 ft water depth.
For this size clarifier a maximum rake torque rating of 100, 000 ft
pounds will be more than adequate.
The residence time for the carbonation zone is based only on the
physical requirements of the gas diffuser.
With an average mill lime kiln stack gas temperature of 180° in a
range of 160-200° F, the average calcium loss will be 1. 025 Ibs
CaCC>3 per 1,000 gallons of waste treated (123 pptn). By with -
drawing the stack gas before it enters the venturi scrubber and
removing the particulate matter in a separate cyclone, gas temper-
ature could be raised to 325° F. This would decrease the calcium
loss to approximately 0. 43 Ibs CaCO3/l, 000 gal or 50 ppm. How-
ever, there will be added cost for blowers, piping and diffusers.
Polyelectrolites could possibly reduce calcium loss. In this case
operating cost would be? increased. The economics of the individual
mills will dictate which plan to follow.
-------
SECTION II
INTRODUCTION
Color of kraft pulp mill effluents has been considered objectionable
in receiving streams mainly from an esthetic standpoint - particularly
in recreational areas. In downstream sources of municipal and
industrial water supplies, treatment cost may be increased. Color
could interfere with biological activity by retarding transmission of
sun light into the stream and indicate undesirable dissolved organic
chemicals which could possibly be harmful to commercial and sport
fishing.
It was this last factor which was the greatest concern of the Georgia
State Water Quality Control Board as the effluent from a kraft pulp
mill in the Riceboro vicinity would enter a virgin coastal area, well
known as a highly productive breeding ground of commercial shrimp,
oysters, and crabs, as well as for its excellent sport fishing. To
assure protection for the aquatic life, chemical treatment to remove
color was made a requirement in the construction permit for the
Interstate Paper Corporation kraft linerboard mill at Riceboro,
Georgia.
The color removal process using slaked lime which was developed,
designed and constructed to meet the Georgia State requirements
had industry wide application provided it was economically feasible.
Because of this the Federal Water Quality Administration, now
EPA-WQO, entered into a research study with Interstate Paper
Corporation in June of 1968.
This report is a summarization of the operation of the color removal
process for a two year period. It goes beyond the scope of the
Federal Water Quality Administration study to give operating results
of natural biochemical stabilization following the color removal
process for the second year of operation. A third part of the report
covers a joint pilot plant scale research project undertaken by the
Interstate Paper Corporation and the Continental Can Company, Inc.
under their respective FWQA grants to investigate the recovery of
calcium from lime treated kraft pulp mill effluent under mill
operating conditions.
In the color removal process being used at the Interstate mill,
calcium hydroxide in a slurry of constant concentration is mixed
with the total process effluent in direct proportion to flow. The
-------
mixture is retained in a flocculator for thirty-five minutes and then
clarified in a center feed clarifier. Color of the combined effluents
average 1160 APHA units with a range of 260 to 2300 units. The
colored substances (mainly lignin by-products) are precipitated as
calcium salts and are removed from the system in the underflow of
the clarifier together with fiber and other settleable solids and
pumped to a holding pond. The decolorized effluent overflowing the
clarifier is saturated with calcium hydroxide at a pH of 12. 2. On
entering an oxidation lake, following the treatment process, the
treated effluent absorbs carbon dioxide from the atmosphere and the
calcium is precipitated as carbonate, •while the pH drops to an
acceptable range for biochemical stabilization to occur in the lower
lake areas. Following stabilization, the effluent is aerated by a
single floating aerator on discharge under controlled conditions to
the receiving stream.
-------
SECTION III
OBJECTIVES
The objectives of the project were to develop, install, and demonstrate
a new chemical process for removing color from kraft pulp mill
e f fluen t.
The color removal process was designed as one phase of a multi-step
effluent treatment system which includes primary clarification and
secondary biochemical stabilization along with post reaeration.
Application of the process was for a new 400 ton kraft linerboard mill
to meet the following limitations stipulated by the Georgia State Water
Quality Control Board:
Effluent discharge 10 mgd
Biochemical oxygen demand 800 Ibs/day
Suspended solids 10 ppm
Color (APHA cobalt units) 30 ppm
At the time this effluent treatment system went into service, March
15, 1968, there was no known proven color removal process in
operation for kraft pulp mill effluents in the United States. In the
public interest, it -was the intent to demonstrate that the color removal
process would:
a) Remove undesirable and possible harmful color matter from
kraft mill effluent.
b) Preserve the esthetic qualities of the stream, thus enhancing
its value for recreational purposes while at the same time
permitting use of the stream for vital industrial purposes.
c) Be a technically and economically feasible color removal
process applicable throughout the kraft pulp industry.
d) Provide an early test of the efficiency of the process in actual
use.
e) Be adaptable to treatment systems of other mills in operation.
The objectives of the carbonation pilot plant study were to obtain the
design criteria for the most efficient system at minimum cost for the
recovery of calcium used in the process and the conditioning of the
-------
effluent from the color removal process for immediate biochemical
stabilization, either by the natural or by the accelerated processes.
8
-------
SECTION IV
CHRONOLOGY
Mill Site
Interstate Container Corporation is a producer of corrugated boxes
in the northeastern states. It was with the intent to supply its own
requirements for linerboard that Interstate Paper Corporation was
formed and the mill at Riceboro, Georgia, was built and put into
operation March 15, 1968.
The mill is located on a 1, 900 acre site in Liberty County, sixteen
miles inland from the Atlantic Coast and borders Riceboro Creek
and the North Newport River. It is thirty-two miles south of
Savannah. See vicinity map Figure 1 page 10.
The North Newport River flows through marsh lands to the east and
empties into St. Catherine's Sound. No other industrial waste enters
the coastal waters between the Savannah River to the north and the
Altamaha River to the south, a distance of fifty miles.
The marine biologists and conservationists refer to these marsh
areas and coastal waters as the most productive breeding grounds of
commerical shrimp, oysters, and crabs in the world. There is
good sport fishing from Half Moon Landing on the North Newport
River, 15.1 stream miles below the mill site into the sound and in
the off-shore waters.
Stream Surveys and Cooperative Research Studies
As a major part of the mill site evaluation, stream surveys were
conducted to establish existing chemical, biological and physical
conditions and to estimate stream potential for effluent assimilation.
Six foot tides and complex estuarine conditions made this a difficult
assignment. Two surveys can be summarized in a few words.
1. For all practical purposes, Riceboro Creek is a dead end of the
estuary. Discharge of 10 mgd mill waste would, at times,
occupy the full cross section of the stream, requiring fifteen to
twenty days to reach the Sound.
2. Sea water does not extend far upstream from Half Moon
Landing; and for this reason, tides could not be relied upon
-------
FIGURE 1
o
INTERSTATE PAPER CORP
MILL SITE .*'
-------
to introduce fresh sea water to supply dissolved oxygen and
dilute mill effluent. Oxygen replenishment of the stream is
mainly through surface contact with the atmosphere.
3. The river water dissolved oxygen (DO) of 4. 8 to 5. 5 ppm and
the biochemical oxygen demand (BOD) of 0. 5 to 1. 0 ppm
indicated no significant source of pollution other than general
marsh drainage.
4. To maintain a minimum DO of 2 ppm, three to four miles below
the mill and a minimum of 4 ppm DO above the Half Moon Land-
ing would limit the effluent discharge to between 500 and 1, 000
Ibs of BOD per day. It would also be required that the waste
have a DO concentration of 6 ppm.
5. Treated mill waste could be discharged into the North Newport
River without serious effects on the water below Half Moon
Landing or damage to fishing near the ocean. Above Half Moon
Landing there is no extensive sport fishing and no competitive
uses of the river water. Below this point there is extensive
sport fishing and some oyster harvesting.
The complexity of estuarine conditions required extensive study
beyond the preliminary surveys. For this reason and to record any
changes in the receiving water, cooperative investigative and
research studies of the North Newport and St. Catherine's Sound
were established, one by the Georgia State Water Quality Control
Board and the University of Georgia Marine Institute at Sapelo Island,
and one by the Georgia State Water Control Board and the U. S.
Geological Survey. Both of these studies were for four years; two
of these years were prior to mill discharge for comparison with the
next two years after mill startup. In the first program, -water
samples have been collected and analyzed, fish and marine organisms
have been collected, counted and identified for population density
and bioasseys have been performed to determine tolerance levels of
aquatic life to unbleached kraft mill waste. The first part of the
biological studies was presented at the TAPPI Air & Water
Conference, Jacksonville, Florida, in April, 1969^ ^ '^ '.
The second cooperative program is mainly a hydrological study of
the estuary tidal movement and fresh water entry. Seven stations
have been established on the North Newport River and in
St. Catherine's Sound, five to record stage and two to record flow
direction and velocity. Two of these stations automatically monitor
DO, pH, turbidity, temperature and conductivity. Wind direction
and velocity are also recorded. Water quality and hydrological data
11
-------
can be obtained from the U. S. Geological Survey office in Atlanta,
Georgia.
Construction Permit
Based on the preliminary surveys, primary clarification, and natural
biochemical stabilization followed by mechanical aeration were
submitted for effluent treatment in the application for construction
permit from the Georgia State Water Control Board. In meetings and
discussions following, the Board took the position that effluent treat-
ment as proposed would possibly not be adequate protection for the
estuarine shrimp and fish breeding grounds. This was their greatest
concern. As an added precaution, two suggestions were made; one,
provide a minimum of 200 to 1 dilution at Half Moon Landing by
diverting flow from the Medway River on the north; or, two, provide
chemical treatment to give a residual waste color of 30 ppm APHA
units.
On evaluating the two proposals, Interstate selected chemical treat-
ment and the Board set the following limitations:
Waste discharge not to exceed 10 mgd
Effluent BOD not to exceed 800 Ibs/day
Suspended solids not to exceed 10 ppm
Color not to exceed 30 ppm APHA color units
The intent of the limitation on effluent color was to define the degree
of chemical treatment for the removal of tannic acid and lignin
compounds which are not readily biodegradeable.
The color standard is particularly noteworthy as there was no back-
ground information on the effectiveness of color removal, and it is
the first time that it has been used as a measure of chemical treat-
ment required by a regulatory agency. In the past, effluent color
has been considered a detriment to steam photosynethesis, or to the
stream's esthetic value.
Mill Description
The mill was designed and built by The Rust Engineering Company
of Pittsburgh, Pennsylvania, for a rated capacity of 400 tons of
unbleached kraft linerboard per day. A detailed description of equip-
ment has been published in several trade journals Wl*^ Besides
the modern design, an important feature is the steps taken to reduce
12
-------
fresh water usage and losses. Process water requirements,
exclusive of cooling water, were 12, 500 gallons per ton compared to
25, 000 gallons per ton average for similar production in the industry.
Biochemical oxygen demand to the waste treatment plant was
anticipated to be in the order of 25 Ibs per ton compared to the 35 Ibs
per ton average.
At the end of the operating period March 15, 1970, present production
capacity has increased to 563 tons per day without additions to the
facilities. Process water has decreased to 9, 800 gallons per ton with
29 Ibs BOD per ton to the effluent treatment plant.
Effluent Treatment Facilities
The color removal process was developed and designed from bench
test data without the benefit of pilot plant studies. Time was a major
factor. Subsequent research at the Paper Institute in Appleton,
Wisconsin under a FWQA grant ( ' and at Syracuse University, New-
York, funded by the National Council for Stream and Air Improvement
of the Pulp and Paper Industry has developed the mechanics of the
lime reactions with kraft effluent color bodies. This information will
lead to further refinements of the process and the reduction of cost.
Under the original concept for effluent treatment as first proposed,
flow was by gravity to a remotely placed primary clarifier discharg-
ing to a natural stabilization lake. To provide for color removal, the
primary clarifier was relocated to -within the mill yard. A lift station,
lime handling system, flash mixing and flocculator were added ahead
of the clarifier. The design was based on a maximum waste flow of
10 mgd and a BOD& loading of 14, 000 Ibs per day. Description of the
process and equipment as installed has been given in an interim
progress report to the TAPPI Air and Water Conference in Jackson-
ville, Florida, April 1969 ^7^ and to the AlChE 63rd Annual Meeting
in Chicago, Illinois 1970 (8>.
In review, the equipment as installed and in the order of flow includes:
1. Pumping station - consisting of a sump with a fixed bar screen
to which all process wastes flow by gravity, three 3, 500 gpm
vertical lift pumps and recirculation piping for level control.
2. A lime feed and flocculation system including a 210 ton lime
silo, screw conveyors, 65 ton/day slaker, 750 gallon slurry
tank with agitator, inline mixer, 45 ft diameter, 33 ft high floe
mix tank with agitator, piping and instrumentation to control the
13
-------
lime feed in proportion to waste flow.
3. Centerfeed primary clarifier - 200 ft in diameter and 15 ft water
depth providing for twelve hours retention at 5 mgd average flow
and six hours retention at maximum flow of 10 mgd.
4. Sludge lagoon with decant pumps - 167 acre feet capacity.
5. Process effluent holding lagoon with barge mounted pump and
return piping to process sump; 48 million gallon capacity.
6. Natural stabilization lake - 650 acres with an average depth of
4.4 feet providing 180 days retention at 5 mgd average flow not
including rainfall or losses due to evaporation.
7. Post mechanical aeration - including a 75 HP mechanical
floating aerator and a concrete lined retention basin forming
integral part of the discharge structure.
8. Discharge structure - consisting of a fixed bar screen, motor-
ized gate valve, flow nozzle and recorder, motorized flow
control valve, continuous sampling device and 0. 5 mile of 36"
pipe for submerged discharge into Riceboro Creek.
Estimated Cost and Federal Participation
Total capital investment for effluent treatment at the Riceboro mill
is in the order of $2, 500, 000 representing 10% of the total mill cost.
Construction cost of the color removal plant was estimated at $454,100
including the plans, specifications and construction supervision. Site
cost and preparations were estimated at $27, 500, giving a total
capital cost of $481, 600. Operations were estimated at $269, 000 per
year for chemicals, power, and labor. Other costs including adminis-
tration, contingencies, and post construction studies and reports were
estimated at $133, 100, bringing the first year cost to $883, 700.
Assuming a straight line ten year depreciation and based on 140, 000
tons per year, the total cost of the lime color removal system would
be $3. 19 per ton of production for the first year.
Since the lime color removal system was unique and of industrywide
application, it was believed that the Riceboro mill system could well
serve as a demonstration plant. Accordingly, a formal application
for a Research and Development Grant was submitted to the Federal
Water Pollution Control Administration on August 11, 1967. On
14
-------
June 13, 1968 a grant of $466, 895 was offered to and accepted by
Interstate Paper.
15
-------
SECTION V
LIME TREATMENT - MATERIALS AND METHODS
A. Systems
Mill Operating Conditions
The production of paper is a dynamic performance of many men
controlling machines and using large quantities of raw materials
The chemical and physical properties of effluent from this
process are ever changing as the position of a bouncing ball.
Variations in effluent flow at the Riceboro mill are as much as
100 percent from the average because of machine wire and
clothing changes and plugging of the effluent sump bar screen.
The latter causes short duration surges. Intermittent flows of
green liquor dregs and lime mud from the recausticizing area
and stock from stock chest cleanouts and floor washups in the
pulp mill and machine areas can cause inaccurate flow measure-
ments. Effluent chemical properties of the combined effluent
stream are far from constant because of the variation in volume
from all sources and because of pullovers from the evaporators
following boilouts and washups in the chemical recovery area.
All process effluents flow by gravity to a sump and lift station
as well as rain run-off from paved and curbed areas around
chemical and liquor storage areas. Sewage septic tanks also
drain to the process sewers. Uncontaminated cooling water is
diverted to Riceboro Creek. As a precaution against spills or
storage tank ruptures, there is a motorized gate operated by a
conductivity probe at the outfall of the cooling water. Should
contamination occur, this gate closes and an alarm is sounded
so that the water c'an be diverted to a holding pond.
Figure 2 page 18 is a flow diagram of the tertiary treatment
system including all changes to date. There may be further
modifications as experience dictates.
Process Effluent Flow
As a quick rundown of the treatment system, the process
effluent stream flows through a fixed bar screen at the entrance
to the sump. Collected trash is removed manually as needed.
Flow is then through the lift pumps and under pressure through
17
-------
(TV
LIME STORAGE TANK
VARIABLE SPEED
SCREW CONVEYOR
EMERGENCY
DIVERSION BOX
IGATES
STEAM
1
4.9 MGD
• 8200* BOD/DAY
FLOAT
MOUNTED
AERATOR
Ixn
FTU
(10 MGD MAX)
MGD
-800# BOD/DAY
(10 MOD MAX)
3 MOD
14000B BOD/DAY
BIOCHEMICAL TREATMENT LAKE
650 ACRES - 900 MG
(180 DAYS RETENTION)
^MOTORIZED SLUICE GATE
{SAMPLER!
VLC
EFFLUENT LIFT PUMPS
3500 GPM 50' TDH
HOLDING LAGOON 48 MG
RICEBORO CREEK
FIGURE 2
EFFLUENT TREATMENT FLOW DIAGRAM
-------
an eccentric orifice and mixing tee into the floe tank. Slaked
lime slurry is fed into the effluent stream at the mixing tee.
Calcium hydroxide in the slurry combines with effluent color
constituents and forms a brown precipitate in the floe tank.
The flow is then by gravity through the clarifier and on through
the remaining part of the system.
In the clarifier the settled floe, inert solids and fiber are moved
to the center by a mechanical rake and withdrawn by two Moyno
pumps discharging to one of the holding lagoons. Foam and scum
are removed from the surface of the clarifier by a skimmer
attached to the sludge rakes and are likewise discharged to the
same lagoon. Supernatant from the sludge is pumped to the
adjacent process waste holding lagoon. The decolorized and
clarified effluent overflows the clarifier through the peripheral
collection launderer and pipe line to the stabilization lake or it
is diverted to the second holding lagoon if the color is not accept-
able. Untreated effluent is also diverted to this lagoon when
there is an equipment outage or a major mill upset. Diverted
flows are pumped back to the sump for retreatment. Normally,
by-pass lines are not shown on a simplified flow diagram. With
treatment for color removal, this is an important feature •when
a color standard is to be met. It will be noted on referring to
Figure 2 that flows can be diverted from any point in the system.
The decolorized effluent entering the stabilization lake at a pH of
12. 2 has in solution 30 to 72% of the calcium hydroxide added to
the system depending on the treatment level. The calcium is
ultimately precipitated as calcium carbonate by carbon dioxide
absorbed from the atmosphere. Calcium precipitation occurs in
the first 100 acre section of the lake. The effluent is then suit-
able for biochemical stabilization which occurs in the adjacent
areas of the lake. .' Before controlled discharge to Riceboro
Creek, mechanical aeration brings the dissolved oxygen up to
saturation.
Chemical Flow
In the order of chemical flow, calcined lime (CaO), is with-
drawn from the silo and fed to the slaker by a variable screw
conveyor controlled by an air bubbler type specific gravity
measurement device on the discharge side of the slaker. Water
flow to the slaker is controlled by a d/p cell and level controller
in the slaked lime slurry tank. Slaking temperature is
19
-------
maintained at 200° F. by a direct steam, water heater controlled
by a controller and thermocouple placed in the slaking compart-
ment of the slaker. Under normal operating conditions the lime
slurry to the slurry tank is held between 1. 05 and 1. 06 grams
calcium hydroxide per liter at 190° F.
Water is added to the slurry tank to trim out the lime concentra-
tion at 1. 04 grams per liter at 170° F. Slurry concentration at
this point in the system is also controlled by a similar gravity
measurement device.
Slurry is withdrawn from the slurry tank by a 120 gpm constant
discharge pump at 180 ft head. On the discharge side of this
pump, the lime slurry flow is split by two control valves. One
flow is through a magnetic flow meter to the mixing tee ahead of
the floe mix tank. The other flow is recirculated to the slurry
tank and to a timer operated sampler. These two flows are
controlled by a controller and the eccentric orifice in the process
waste line on the discharge side of the lift station. Lime slurry
flow is in direct proportion to •waste flow. All variables are
recorded. Should the lime slurry demand fall below 20 gpm,
the slurry automatically goes into full recycle, the slurry sampler
and flow integrator trip out and the pipe line to the floe tank
is flushed with water. When the demand increases to above
20 gpm, the flushout system reverses itself.
Instrument settings for automatic control are made from charts
of calcium hydroxide slurry concentrations from 1. 03 to 1. 07
grams per liter at temperatures of 170 and 190° F; and tables of
effluent and lime slurry flow for treatment levels of 1, 000 to
2, 600 ppm calcium hydroxide. With interruption of automatic
control for maintenance or repair, the system is put on manual
control based on average waste flows and charts of lime screw
output.
The system is monitored by weighing grab samples of lime slurry
from the slaker and slurry tank for comparison with instrument
readings. At twenty-four hour intervals, operating charts are
collected, a lime slurry feed composite sample is checked for
calcium hydroxide concentration,and integrator readings of
lime slurry and process effluent flows are recorded.
The transition from a concept of a process to a full scale
operating plant can be a rugged but challenging experience.
20
-------
This is particularly so when the system is a prototype design
which mast go on stream with a new mill startup. The Riceboro
mill treating plant was no exception. The following is a
commentary on the components of the system.
B. Operations and Results
Initial Operation
As the process -was a prototype design, it was necessary to
determine the limitations and adequacy of the equipment as it
has been installed. To do this, range of lime feed, color of
untreated and treated process waste, and waste flow were
recorded during the first 90 days of operation.
There are two continuous timer-operated sampling stations on
the process waste flow set to collect fifty gallons composite
samples for each 24 hour period. The one for the total waste flow
to the lime treating process is of a splitter box flow design to
minimize plugging from suspended solids. The one for the treated
waste overflow from the clarifier is of the solenoid valve design.
A third continuous timer operated sampling station was added at
•the lime slurry feed tank. It is a splitter box design similar to
that in use on the untreated waste flow line. This sample station
is used to monitor the calcium hydroxide concentration of the
lime feed to the floe tank.
On a routine basis, samples were collected from the two
process waste flow sampling stations every twenty-four hours
and tested for pH, alkalinity, and color. Twenty-four hour
composite sample,s of lime slurry were analyzed for calcium
hydroxide concentration. Waste flow and lime feed were
recorded for the corresponding period of sampling.
On every eight hour shift and as often as indicated by mill
operating conditions, grab samples of treated and untreated
waste were tested for color to check level of treatment.
On as needed basis or when time permitted the following tests
were performed:
1. Jar coagulation test to check level of treatment.
2. Available CaO of calcined lime.
21
-------
3. BODg of untreated and treated waste.
4. Dichromate oxygen consumed (DOC).
Operating Period - June 21, 1968 through September 30, 1968
Untreated waste color averaged 1080 APHA platinum cobalt units
with a high and low of 8, 000 and 200. The average was slightly
above the indicated normal operating range of 600 to 1, 000 ppm.
Treated waste color averaged 95 color units adjusted to pH 7.6
with a low and high of 450 and 40 units. Percent reduction was
fairly constant in the range of 90%, irrespective of the untreated
waste color.
Average color reductions per 100 ppm of calcium hydroxide
feed for the ranges of color and lime feed rates are shown on
page 102. Lime usage as 90% CaO averaged 37. 51 tons per
operating day.
Process waste flow to the color removal system averaged 5. 15
mg for a total of 453. 32 mg for the period. The process waste
was 53. 3% of the combined process and cooling water flow.
During the period, 23. 7 mg of treated waste was diverted to
the holding pond and 42. 1 mg was returned from the holding
ponds for treatment giving a net reduction in retained waste of
18.3 mg.
From operating experience through this period, the following
limitations were observed:
1. Maximum feed of lime as Ca(OH) resulted in a
concentration of 2, 000 ppm at a waste flow of
6 MGD. Maximum feed was increased to give a
concentration of 2, 600 ppm by installing a parallel
lime feed line.
2. Above 6 MGD, waste flow was the governing factor
for lime concentration.
3. Below 4 MGD, lime feed was erratic. Density of lime
in slaker and slurry tank increased beyond the desired
level.
4. Rapid changes in waste flow gave a quick response from
22
-------
the dilution control valve between the lime slurry tank
and transfer pump suction. As a result dilution water
backed up into the slurry tank; the desnity of lime in the
slurry tank decreased and the level rose; water to the
slaker shut off and the lime screw continued to feed
lime at the rate of 0. 6 tons per hour.
5. Lime feed to the slaker fluctuated with the level of lime
in the silo, lime particle size and percent available
lime as CaO.
6. Lime feed to the inline mixer decreased with scaling in
the transfer line. As a result treatment was interrupted
to clean out the line.
7. The system had an excessive inherent lag.
Based on the above observations, the lime feed control system
was redesigned. The most important features were:
1. Dilution water was to be added to the slurry tank instead
of to the transfer pump suction.
2. A density controller and recorder was added to vary the
dilution water flow and record lime slurry density.
3. Lime feed to the inline mixer was to be controlled by an
air operated valve in direct ratio to waste flow.
4. Lime slurry was to be circulated back to the slurry tank
to provide for continuous sampling.
5. A magnetic' flow meter was to be installed to record
lime slurry feed.
Equipment was ordered and delivery was expected to be
completed by October 11, 1968. Installation was scheduled for
completion by November 8, 1968.
Operating Period October 1, 1968 through October 31, 1968
Modifications to Equipment
During the equipment startup, the floe agitator was modified in
23
-------
that one set of blades was removed when the floe tank discharge
was lowered to stream-line flow to the clarifier. Agitation was
reduced by forty-five percent and the tank volume was reduced
twenty-five percent. To compensate for the loss in agitation,
the agitator drive-speed was set at the maximum rpm.
When the floe tank was drained for the mill shutdown, October 13,
excessive buildup of solids was observed. The drain line was
closed by sludge and the tank had to be emptied through the inlet
line and by-pass to the holding pond. To reduce buildup of
solids, it will be necessary to replace the upper blades. This
will require the removal and machining of the agitator shaft to
provide a minimum of 48 inches spacing above the lower set of
blades. Work was scheduled for completion on November 22,
1968.
The clarifier had settled because of compaction of the bearing
soil. Most of this settling in the past had been at the center
column. A check of the launder levels on October 7 showed that
the north half was 1.3 to 1.4 inches lower than the south half
The launder ring was scheduled to be re-leveled by December 7,
1968.
Operating Results:
During the month of October, the mill was shut down from the
13th through the 20th because of a labor strike. For this reason
test data covers twenty-four (24) operating days.
The untreated process waste color averaged 1380 APHA platinum
cobalt units with a high of 5, 000 and a low of 400. The average
color is 28% higher than the preceding period reflecting the
higher losses during mill shutdown and startup. Treated process
waste color averaged 120 color units with a high of 300 and a low
of 50. Percent reduction in color was in the order of 90-92%.
Average color reduction per 100 ppm. of calcium hydroxide
concentration for the range of color and lime concentrations is
shown in the Appendix page 103. Lime usage as 90% CaO averaged
47. 32 T/D compared to 37. 51 T/D average for the preceding
period. The higher lime concentration figures are questionable
as they give a lirne usage higher than the end of the month
inventory withdrawals. It is expected that closer correlation will
be attained when the new lime feed metering system is completed.
24
-------
Average process waste flow for the period was 4. 74 mgd. This
is 49.4 percent of the total mill flow including uncontaminated
cooling water discharged to Riceboro Creek. Of the total
process waste, 22. 43 rng was diverted to the holding ponds and
14. 93 mg was returned from the holding ponds for re-treatment
giving an increase of 7. 5 mg retained.
Operating Period November 1, 1968 through November 30, 1968
Modifications to Equipment:
1. Lime Feed System
The modifications to the lime feed system were completed
except for the installation of one control valve and the
controller for the magnetic flow meter. Shipment was
delayed on both of these pieces of equipment.
2. Inline Mixer
Impeller of the inline mixer was cracked when the baffles were
removed during the start-up period to reduce pressure drop.
This impeller was replaced on November 21, 1968 when the
treatment plant was taken out of service to replace bearings
on the lime screw conveyor and to straighten the lime screw.
3. Floe Agitator
While the treatment plant was out of service for repairs, the
upper agitator blades were replaced. Work -was completed
one day ahead of schedule.
4. Clarifier Sludge (Lines
The sludge line discharge was extended 150 ft on November
15, 1968 to move ahead of the deposited sludge in the east
holding pond.
Operating Results:
Some of the typical mill operating problems occurred during
the month of November. As a result, process waste to the
color removal plant had a color exceeding 1, 000 ppm six
operating days in the reporting period. Average process waste
25
-------
color averaged 1006 APH platinum cobalt units with a high of
6, 000 and a low of 250. Treated process waste color averaged
104 color units with a high of 200 and a low of 50. The high of
200 color units is the cut off point for diversion to the holding
ponds.
Average color reduction per 100 ppm of calcium hydroxide
concentration for the range of color and lime concentrations are
shown in Appendix.page 104. Lime usage averaged 37.25 T/D.
The effect of lime treatment for color removal on biochemical
oxygen demand (BOD5) is shown in the following table with
corresponding color units. From this data there appears to be
no correlation between BOD and color. Additional data may
show some relationship.
TABLE 1
BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1968
11/6
11/12
11/13
11/14
11/19
11/20
11/26
UNTREATED
Color BOD
APHA ppm
700
900
800
3000
1500
600
700
245
310
310
335
325
280
335
TREATED
Color BOD
APHA ppm
80
150
100
200
75
75
100
205
260
235
230
235
200
270
% REDUCTION
Color BOD
89
83
88
93
95
88
86
15
17
25
31
27
27
20
Av. 1060 305 110 235 88 23
Av. Ibs.
BOD/Day 17, 100 13, 100
Average process waste flow for the period was 6. 6 mgd. This
is 72. 3 percent of the total mill flow including uncontamiriatek
cooling water discharged to Riceboro Creek. Of the total process
waste 17. 85 mg was diverted to the holding ponds and 27. 19 mg
was returned from the holding ponds for retreatment, giving a
reduction of 9. 34 mg in the amount retained.
26
-------
Operating Period - December 1 through December 31, 1968
Modifications to Equipment:
1. Lime Feed System
Delivery of materials delayed the completion of the system
well into the middle of December. Because of this and with the
pending Christmas mill shutdown and holidays, final
completion was deferred until January 9, 1969.
2. Clarifier Launder
Leveling of the clarifier launder requires outage time on the
waste treatment plant. To minimize flow of process waste
to the holding ponds, this work was rescheduled for January 9,
1969 to coincide with the tie in of new instrumentation of the
lime feed modifications.
Operating Results:
During the month of December, color concentrations were the
highest on record and the percent color reduction the lowest on
record. Of the twenty-seven and one half operating days, color
of the untreated process waste exceeded 1, 000 APHA platinum
cobalt units for fourteen days. Average for the period was 1,200
with a high of 2, 750 and a low of 500. process waste
color averaged 160 units with a high of 250 and a low of 75.
Average percent color reduction was 86. 3.
Average color reduction per 100 ppm calcium hydroxide
concentration for the range of color and lime concentrations is
shown in the Appendix, page 105. Lime usage averaged 42. 9
T./D as 90% calcium oxide (CaO).
The temporary laboratory was destroyed by fire on December 5,
1968. For this reason, there was no testing done other than
color determinations. One set of samples of untreated and
treated process waste was sent to Shuey and Company, Inc.,
Analytical and Consulting Chemists in Savannah, Georgia, for
BOD tests. Results are shown on Table 2 , page 28,with
corresponding color concentrations.
27
-------
TABLE 2
BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1968
UNTREATED
Color BOD
APHA ppm
TREATED
Color BOD
APHA ppm
% REDUCTION
Color BOD
12/22 1500 300 150 310 90.0 neg.
Average process waste flow for the period was 6.26 mgd. This
is 71. 5 precent of the total mill flow including uncontaminated
cooling water discharged to Riceboro Creek. Of the total process
waste 14. 0 mg was diverted to the holding ponds and 26.4 mg
was returned from the holding ponds for re-treatment giving a
reduction of 12.4 mg in the amount retained.
Operating Period - January 1 through January 31, 1969
Modification of Equipment:
The new instrumentation to control lime slurry feed by density
was completed on January 9. At the same time,a new lime
conveyor screw was installed to replace one which had been
bent by pieces of broken lime kiln chain.
While the treating plant was down, the clarifier launder was
leveled and cleaned. Short circuiting due to unbalance has been
eliminated. There is still some intermittent short circuiting due
to windage. With the changes, clarifier performance has been
materially enchanced.
Operating Results:
From January 1 through January 5 the color removal system was
operated with the lime feed controls as originally designed and
installed. The treating plant went down at 10:00 A. M., January 6
to tie in the new instrumentation for lime density control,
sampling and recording lime slurry flow. Treatment was
resumed at 7:00 A. M. January 7. Following is a brief present-
ation of operating results before and after changes to the lime
28
-------
feed instrumentation.
Mill conditions for January 1 through January 5 were normal
with average untreated process waste color of 820 APHA
platinum cobalt units. There were no unusual operating
incidents. Treated process waste color averaged 94 units with
an average lime usage of 36.4 T/D as 90% CaO. Average
process waste flow •was 7. 39 mgd including 1. 01 mgd returned
from the holding ponds for re-treatment.
During the time interval of January 8 through January 31, the
color removal plant was shut down twice for unscheduled mill
shutdowns and seven times for the following repairs and
modifications to equipment.
1. Broken lime conveyor screw.
2. Replacement of variable speed drive bearing on lime
screw conveyor.
3. Replacement of slaker agitator motor.
4. Installation of process waste flow meter flush line.
5. Tie-in of piping to supply evaporator condensate for
slaking lime and diluting lime slurry.
The last mentioned change was made to reduce scaling of the
lime feed system.
Service was also interrupted on thirteen other days. The
major causes were restricted condensate supply, plugging of
process waste flow meter pitot tubes and build-up of lime
density in slaker. Corrective steps were taken to reduce
outages for these reasons.
When the color removal system was in service, lime feed rates
and lime density were varied to calibrate instruments and to
check ranges of treatment. On three consecutive days,
January 27, 28, and 29, controls were set to maintain a
concentration of 1800 ppm of Ca(OH)2 in the effluent. An
average for the three days was 1840 ppm with a variation of
40 ppm. Untreated process waste colors for this run were 1280,
2100, and 1780 APHA cobalt units. Treated waste colors in the
same order were 220, 180 and 200 units representing an
average 87. 6% color reduction.
Average color reduction per 100 ppm calcium hydroxide for the
29
-------
range of color and lime concentrations is shown in the Appendix,
page 106 for the reporting period. Lime usage averaged 35. 8
T/D as 90% calcium oxide (CaO).
Laboratory test work was limited during January because repairs
to equipment damaged in the fire December 8 had not been
completed and chemical reagents had not been completely
restocked.
Available tests for BOD and COD are shown below with corre-
sponding color concentrations.
TABLE 3
COD & BCD REDUCTICN VS COLOR REDUCTION
PROCESS WASTE
Date
1/1/69
1/17/69
1/29/69
UNTREATED
Color
APHA
Units
1000
3600
1780
COD
ppm
909
1440
1Z96
BCD
ppm
360
456
420
TREATE
Color
APHA
Units
1Z5
150
200
COD
ppm
496
574
D
BOD
ppm
220
2 SO
300
% REDUCTION
Color
87.5
95.8
88.7
CCD
51.7
55.7
BOD
38.9
38.6
28. 6
Average process waste flow for the reporting period was 6.49
mgd. This is 73. 4 percent of the total mill flow including
uncontaminated cooling water discharged to Riceboro Creek. Of
the total process waste 27. 1 mg was returned from the holding
ponds for treatment giving an increase of 2.4 mg in the amount
retained. Holding ponds were at their maximum capacity on
January 13.
Operating Period - February 1 through February 28, 1969
Modifications to Equipment:
With the instrumentation for lime feed control installed in
January, hidden operating problems were found and isolated.
The-major one was that there was no control of the lime slurry
specific gravity in the slaker with the existing instrumentation.
Ratios of water to lime were not a straight line function. As a
result, lime slurry specific gravity increased to the point where
the slaker would overload,and the dilution water to the slurry
tank could not maintain the desired specific gravity. A
difference of 0. 01 in lime slurry specific gravity can give a
30
-------
swing of 250 ppm in Ca(OH)2 concentration in the treated waste.
To correct for this, specific gravity controls were added to the
slaker, making the lime screw rpm a slave of specific gravity
instead of to the flow of water to the slaker. With these further
modifications it was necessary to re-verify instrument readings
of specific gravity against weighed samples of lime slurry.
Operating Results:
February was a short operating month with a seven day mill
shutdown for maintenance repairs. The lime color removal
system was operated for one additional day to treat wash up
waste water. During the reporting periods untreated waste
colors averaged 2010 APHA platinum cobalt units. This is
1000 units higher than the normal operating range. Treated
process waste color averaged 169 units with an average lime
usage of 37. 9 T/D as 90% CaO. Average treated process waste
flow for 22 operating days was 7.10 mgd including waste
returned from the holding ponds for re-treatment. Average
process waste flow during the same period was 6. 56 mgd. This
is 68% of the total mill waste flow including uncontaminated
cooling water discharged to Riceboro Creek. Of the total waste
flow for the 28 day period, 19. 58 mg were diverted to the
holding ponds and 21.39 mg were returned for re-treatment
giving a net increase in holding pond capacity of 1.81 mg.
Average color reduction per 100 ppm calcium hydroxide for the
range of color and lime concentration is shown in the Appendix,
page 107, for the reporting period. The wide range of lime
concentrations reflects the lack of lime feed control and illustrates
the effect of the slaker specific gravity on the system.
Part of the laboratory test work included specific gravity
determinations and calcium analysis of the lime slurry feed for
calibration purposes. Available tests for BOD and COD are
shown on page 32 with corresponding color concentrations.
31
-------
TABLE 4
COD 8t BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
3/4/69
2/5/69
2/6/69
2/11/69
2/13/69
2/25/69
2/26/69
2/27/79
UNT
Color
APHA
Units
2000
2000
2250
1200
2500
2000
1000
6800
REAT
COD
ppm
1825
1634
1944
961
1515
985
1464
1496
ED
BOD
ppm
421
663
283
404
339
546
T
Color
APHA
Units
200
300
175
125
250
100
80
125
REA
COD
ppm
682
851
832
685
923
462
223
401 !
TED
BOD
ppm
323
330
238
246
259
244
0
Col
90.
85.
92.
89.
90.
95.
92.
98.
1o
Lor
0
0
2
6
0
0
0
0
REDUC
COD
62.6
47.6
57.0
28.7
39.1
53.0
84.7
73.0
TION
BOD
23.0
50.3
15.9
39.0
24.9
58.0
Operating Period - March 1 through March 31, 1969
Modifications to Equipment:
Three changes were made to the lime feed system during the
reporting period. These were:
1. Pressure diaphram used for specific gravity measure-
ment of the slaked lime in the slurry tank was replaced
with an air bubbler tube in a flow through pot mounted
on the outside of the tank.
2. Slurry tank level control was rearranged as originally
installed.
3. Electric vibrator was installed on lime silo.
The first change was made to reduce outage time. The bubble'r
tube and flow through pot is the same as used for control of
specific gravity of the slaked lime in the slaker. They are so
designed that they can be cleaned and adjusted quickly while the
feed system is on manual control. Until this change was made
the slurry tank had to be drained to remove the pressure
diaphram for cleaning. Total time for cleaning and adjusting
was one and one-half to two hours.
32
-------
The second change -was to revert back to a wider range for
slurry tank level control. The third change is an addition to
level out the flow of lime from the silo to the slaker.
Operating Results:
During March the color removal system was out of service for
one hour and twenty minutes to replace a diaphram with an air
bubbler system for the slurry tank specific gravity control.
During the reporting period, untreated process waste colors
averaged 1150 APHA platinum cobalt units. This is a 42. 8%
reduction compared to the average untreated waste color for the
preceding period. Treated process -waste color averaged 116
units with an average lime usage of 32. 8 T/D as 90% CaO. Total
prpcess waste treated was 195. 1 mg including 30. 8 mg returned
from the holding ponds for re-treatment. During the period, 2. 7
tag was diverted to the holding ponds giving a net increase in
available holding capacity of 28.1 mg. Average process waste
flow from the mill averaged 5. 38 mgd which is 66. 8% of the
total mill waste flow including uncontaminated cooling water
diverted to Riceboro Creek.
On March 2 pulp mill operations changed the pulp washing
sequence. As a result 1;he average total process waste flow
was reduced by 1. 2 mgd. There was a decrease of approximately
2. 7 mgd in the pulp mill sewer. This decrease was offset by an
increased usage in the paper mill. Normally with all systems
operating, there would be no effect on the performance of the
color removal system. "With the paper machine down for a wire
change the process waste flow dropped between 2. 0 and 2. 5 mgd.
In this range of waste flow, the lime feed system was below the
lime feed flow ranjge. This was compensated to maintain lime
feed by changing the ratio setting of slaked lime flow to waste
flow. However, in doing so, the system was taken off the
desired constant feed rate.
To meet the changes in waste flows, it would be necessary to
change the sprockets on the variable screw drive and to drop
the trip out signal on tlje automatic back-flush system. The
pulp washing sequence was on a trial basis and would be
continued until April 13. A third washing sequence would be
tried and run through May 11, 1969.
Average color reduction per 100 ppm calcium hydroxide for
33
-------
the range of color and lime concentration is shown in the
Appendix, page 108, for the reporting period.
Laboratory test work included jar tests to check treatment
results. Specific gravity of slaked lime samples collected
over a 24 hour period were also determined to verify instrument
read-outs. Tests for BOD and COD are shown below with corre-
sponding coloo.- concentration.
TABLE 5
COD & BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1969
3/4
3/5
3/6
3/11
3/12
3/13
3/18
3/19
3/20
3/25
3/26
3/27
Av.
UNTREATED
Color COD BOD
APHA pptn ppm
Units
1020
920
1660
1540
980
1080
1140
460
960
1860
530
2640
1233
1018
800
1443
1217
1023
1018
1019
817
1050
1333
820
1976
1128
411
304
382
332
246
280
314
196
212
351
296
408
311
TREATED
Color COD BOD
APHA ppm ppm
Units
100
80
30
130
130
90
80
90
N.S.
100
90
240
L**T\J
105
592
776
590
589
496
602
N.S.
483
530
582
202
228
115
223
223
204
226
151
230
198
246
£>*±U
204
% REDUCTION
Color COD BOD
90.2
91.3
98.2
91.6
86.7
91.7
93.0
80.4
___
94.6
83.0
90.9
90.9
41.8
30.0
—
51.5
42.4
51.3
26.3
—
63.8
35.4
42.8
26.5
25.0
69.9
32.8
9.3
27.1
28.0
23.0
—
34. 5
33. 1
39. 7
31.7
Comparing the above data with corresponding data for the
preceding period and taking into account decrease in waste flows,
total BOD was decreased by 73. 8%.
Operating Period - April 1, through April 30, 1969
Operating Results:
During the month of April the waste treatment plant was
programmed to treat the process waste with 1500 ppm calcium
hydroxide Ca(OH)2 from April 1 through April 16 and with 1200
ppm calcium hydroxide from April 17 through April 30. There
was a day by day variation. However, for the two periods
during the month, the average for the first period was 1450 ppm
and 1246 for the second period. In comparing the two
34
-------
programmed periods, a higher percent reduction in color was
obtained at the lower lime concentration of 1246 ppm. Lime
usage decreased 0. 58 tons CaO (90%) per million gallons of
process waste treated. This indicates an optimum level of
treatment beyond which additional lime is wasted.
On an overall operation for the month of April,the untreated
process waste color averaged 1084 APHA platinum cobalt
units which,is slightly lower than the average untreated waste
for the preceding period. Treated process waste color averaged
97 units with an average lime usage of 29. 68 T/D as 90% CaO.
Total process waste was 189.27 mg including 11. 84 mg returned
from the holding ponds for retreatment. During the period, 7. 73
mg was diverted to the holding ponds giving a net increase in
available holding capacity of 4. 11 mg. Average process waste
flow from the mill averaged 6. 16 mgd which is 73. 1% of the total
mill waste flow including uncontaminated cooling water diverted
to Riceboro Creek.
Down time of the chemical coagulation plant in April was fifteen
hours and twenty minutes. Ten hours and fifty minutes were for
three periods of spillage of chemicals and kraft liquor. During
this time, the process waste was diverted to the holding ponds
to level out the load on the waste treatment system.
Four hours and thirty minutes were required to clean the inline
mixer, to change the sprocket drive on the lime screw and for
a piping change.
Even taking the above into consideration, performance of the
waste treatment plant had improved materially during the
reporting period. Average color reduction per 100 ppm calcium
hydroxide for the range of color and lime concentration is shown
on page 109.
Laboratory test work on lime slurry specific gravity and calcium
analyses was routine to check lime feed instrument settings.
Tests for BOD and COD are shown on the following page with
corresponding color concentrations.
35
-------
TABLE 6
COD & BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1969
4/8
4/9
4/15
4/16
4/17
4/22
4 23
4/24
Av.
UNT
Color
APHA
Units
1440
460
4200
220
700
960
1140
1320
1305
REAT
COD
ppm
1134
524
1807
688
498
840
1105
1074
959
ED
BOD
ppm
278
212
478
274
305
304
292
367
314
TR
Color
APHA
Units
180
140
160
100
80
90
60
100
114
EATE
COD
ppm
561
755
305
359
330
462
D
BOD
ppm
246
266
267
294
224
202
209
244
% RE
Color
87. 5
69.5
96.2
54. 5
86.6
90.6
94.7
92.4
84.0
DUCT
COD
50. 5
58.2
55.7
57.3
70.1
38.4
ION
EOD
11. 5
44.4
2.6
3.6
26.3
30.8
43.1
23.2
Operating Period - May 1 through May 31, 1969
Operating Results:
During the month of May, the waste treatment plant was
programmed to treat the process waste with 1200 ppm
All during the month, trouble was experienced in holding the
calibration of the specific gravity controls on the slaker and
slaked lime slurry tank. Air lines to and from the D. P. cells
were replaced when it was found that acid, used to clean the
bubbler tubes of the specific gravity controls, had backed up
into these lines. It was not until the 30th of the month that it
was found that acid had also backed up into the D. P. cells.
On inspection both diaphrams were leaking through numerous
cracks and pin holes caused by the acid.
Another problem in holding set specific gravity occurred during
the first part of the month when the water supply to the slaker
and slurry tank intermittently shut off. Water for the lime feed
system was pumped from a surge tank which was also used as
a source of water for the caustic area. High water usage in this
area would drain down the surge tank to a point where there was
not sufficient water for the lime feed system. This was corrected
by resetting the make-up water control valves.
As a result of the above operating difficulties the average
36
-------
concentration of lime to the color removal system was
1418 ppm Ca(OH)_. Variations in treatment -were large, ranging
from 870 to 2160 ppm. Considering the volume of waste treated,
the increase of 1200 to 1418 ppm represents an excess of 112 tons
of lime (90% CaO). Process waste flow was reduced from 189. 27
mg for the month of April to 159. 98 nag for the month of May. As
a net result, lime usage dropped from 29. 69 T/D to 24. 13 T/D as
90% CaO for May. This drop in process waste flow was due to a
trial run on the mill pulp •washing sequence.
Down time of the chemical treatment plant in April -was 39
hours; of this, 19! hours were for work on the inline mixer and
cleaning the lime silo discharge gate. The lime feed system
•was shut down 19f hours while strong -waste was diverted to the
holding ponds to level out the load on the color removal system.
Waste, both treated and untreated, was diverted for a total of
62 hours during May. The strong waste was from a green liquor
clarifier overflow and from washouts of the recovery boiler
cyclone and smelt dissolving tanks.
Other than the overage on lime feed, performance for May was
good. Untreated process waste color, exclusive of waste
diverted, averaged 736 APHA platinum cobalt units. Treated
process -waste color averaged 86 units. Average color reduction
per 100 ppm calcium hydroxide for the range of color and lime
concentration is shown in the Appendix, page 110. The total process
waste of 159. 98 mg treated included 14.43 mg returned from
the holding ponds for retreatment. During the period 12. 26 mg
was diverted to the holding ponds giving a net increase of 2. 17
mg in available holding capacity. Process waste flow from the
mill averaged 5. 11 mgd which is 59. 7% of the total mill waste
flow including uncontaminated cooling water diverted to Riceboro
Creek.
Laboratory test work on lime slurry specific gravity and
calcium analysis were routine to check lime feed instrument
settings. Tests for COD and BOD are shown on page 38 with
corresponding color concentrations.
37
-------
TABLE 7
COD 8c BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
UNTREATED TREATED % REDUCTION
Date Color COD BOD Color COD BOD Color COD BOD
1969 APHA ppm ppm APHA pptn pptn
Units Units
5/6
5/7
5/8
5/13;
5/14
5/15
5/20
5/21
5/27
5/28
Av.
1060
820
1100
620
1140
840
180
160
600
380
690
816
867
955
916
1143
847
667
1039
709
568
853
271
233
288
225
212
NT
NT
235
298
22]
248
100
120
120
50
80
90
50
60
75
80
83
413
578
NT
371
443
NT
356
494
470
NT
446
169
213
251
200
NT
NT
193
161
228
194
201
90.6
85.3
89.1
91.9
93.0
87.8
72,2
62.5
87.5
78.9
83.9
47.1
33.3
m ,
59.5
61.2
46.6
52.5
33.7
• muim— mi -»
47.7
59.5
9.0
12.8
11.0
31.5
23.4
12.2
22. 8
NT - No test
Inline Mixer Evaluation:
Design of the color removal system was based on laboratory
data without verification by pilot plant studies. Laboratory work
clearly indicated that precipitation of color bodies would not
occur without rapid mixing prior to flocculation. The purpose
of this evaluation was to determine the effect of mechanical
agitation vs mixing by baffling.
May 12, 1969 was selected for the date to start the evaluation
of the inline mixer. It was at this time that the pulp washing
sequence in the pulp mill was scheduled to uae the decker as a
fourth stage washer. This trial run was to be for 28 days
corresponding to a mill accounting period. The evaluation test
period was set up to run 14 days with the inline mixer in service
followed by 14 days without the inline mixer in service under
comparable mill operating conditions. Prior to starting the
tests, the inline mixer was removed, cleaned and reinstalled.
At the end of the first 14 days, May 27, the inline mixer was
removed and a blind flange was bolted in place to close the
opening for the agitator. The housing with internal baffles was
not removed because the baffles would prevent laminar flow at
the point of lime slurry injection.
38
-------
Comparing the first five days' performance without the inline
mixer in service with the preceding two weeks with the mixer
in service, no difference has been observed in the degree of
treatment obtained.
Operating Period - Jane I through July 31, 1969
Operating Results:
During the two month periods there have been continuing
problems with instrumentation of the lime feed system and the
process flow metering system. Basically, except for failures
of diaphrams of d/p cell transmitters, and control valve
diaphrams, most of the problems stem from varying water and
air pressures.
With variations in lime feed and malfunctions of the process waste
flow transmitter, it has not been possible to maintain the desired
treatment level of 1200 ppm concentration of calcium hydroxide.
In June the average lime consumption was 4. 81 tons 90% CaO per
million gallons of waste treated or 1, 387 ppm Ca(OH)2- The
average for July was 4. 62 tons of lime or 1, 332 ppm Ca(OH)2-
All during July the lime feed system had to be on partial manual
or full manual control.
In June untreated process waste color averaged 1, 235 APHA plat-
inum cobalt units. Treated process waste color averaged 108
units. Average color reduction per 100 ppm calcium hydroxide
for the range of color and lime concentration is shown on page 111.
The total process waste of 208. 533 mg treated included 29. 20 mg
returned from the holding ponds for re-treatment. During the
period 8. 60 mg was diverted to the holding ponds giving a net
increase of 20. 60 mg in available holding capacity. Process
waste flow from the mill averaged 6. 33 mgd which is 68. 2% of the
total mill flow including uncontaminated water diverted to
Riceboro Creek.
In July untreated process waste color averaged 1388 APHA plati-
num cobalt units. Treated process waste color averaged 127 units.
Average color reduction per 100 ppm calcium hydroxide for the
range of color and lime concentration is shown on page 112. The
total process waste of 169. 94 mg included 23. 83 mg returned from
the holding ponds for re-treatment. During the period 6. 67 mg was
diverted to the holding ponds giving a net increase of 17.16 mg in
available holding capacity.
39
-------
Process waste flow from the mill averaged 5. 88 mgd which is
61. 9% of the total mill flow including uncontaminated water
diverted to Riceboro Creek. Due to a mill shutdown for
maintenance and repairs there were 26 operating days in July.
Comparing the above data for the two periods, process waste
decreased 7. 1% in July. As there was no change in mill process,
some of the decrease can be attributed to errors in flow measure
ment. Average overall color reduction for the two periods was
approximately the same at 91%. Percent reduction is a function
of the untreated waste color. At the lower ranges of color the
percent reduction is approximately 10% less than in the higher
ranges. The percent reduction is shown for each range of color
in the Appendix,page 111, and page 112. Tests for COD and BOE
are shown on page 41 with corresponding color concentrations.
Lime Feed Evaluation:
The lime feed evaluation has not been concluded for the reason
that control has not come up to expectations nor has it given
constant and repeatable results. Problems in holding constant
feed stem from fluctuations in water supply and pressure to the
water control valves to the slaker and to the slurry tank, fluct-
uation in lime slurry flow to the specific gravity measuring pots,
and variations in air pressure to the pneumatic controls.
It was thought that the water supply problem was resolved by
resetting the make-up water control valve. This did not solve
the problem. The make-up valve was changed from a 2" to a
4" to increase availability of mill supply water. A separate
supply pump motor and pressure regulator were ordered.
Installation was scheduled for the week of August 18.
To control lime slurry flow, a constant headbox was installed
ahead of the specific gravity measuring pot of the lime slurry.
This proved very satisfactory. A constant discharge pump
was installed on the slaker to control lime slurry flow to
the specific gravity measuring pot at this point in the system.
The pump was not satisfactory and was to be replaced and
modified. Completion was scheduled for the week of August 11.
Air pressure regulators were in the process of being installed
on the air supply to bubbler tubes of each specific gravity
measuring pot.
40.
-------
TABLE 8
COD & BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1969
6/3
6/4
6/5
6/10
6/11
6/12
6/17
6/18
6/19
6/24
6/25
Av.
7/1
7/2
7/9
7/10
7/15
7/16
7/17
7/22
7/23
7/24
1 / U J.
7/29
7/30
7/31
Av.
NT -
UNTREATED
Color COD BOD
APHA ppm ppm
Units
450
800
250
1250
1000
2500
800
1500
900
2500
1600
1232
850
1000
550
1500
1000
1000
700
1600
5000
1000
2000
1200
1250
1435
No test
620
834
1111
1007
917
1608
760
989
719
1917
1148
1057
844
843
549
NT
1091
NT
555
1070
2055
992
1658
1141
1110
1083
216
302
212
368
328
333
329
398
213
406
382
317
288
280
183
292
405
281
263
447
643
288
356
340
376
342!
T
Color
APHA
Units
60
50
75
100
100
150
85
75
75
200
80
95
80
100
75
80
120
100
95
140
150
125
130
130
100
110
REATED
COD BOD
ppm ppm
286
444
NT
636
500
NT
365
502
NT
696
NT
490
386
NT
368
NT
NT
395
NT
659
715
NT
532
578
NT
519
198
185
173
264
227
265
295
178
208
280
329
236
212
208
183
254
221
233
173
302
268
239
284
265
281
240
% RE DUG
Color COD
86.7
93.7
70. 0
92.0
90.0
94. 0
89.4
95.0
91.6
92.0
95. 0
89.9
90.6
90.0
86.4
94.6
88.0
90.0
86.4
91.3
97.0
87.5
93.5
89.2
92.0
90.5
53.8
46.7
36.8
45.4
51.9
49.2
63.7
49.6
54.3
•^v^«a
32.9
m^^^^
-
-
38.4
65.2
67.9
49.4
51.4
TION
BOD
1.1
38.7
18.4
28.3
30.8
20. 4
1.0
55.2
2.3
31.0
13. 9
21.9
26.4
25.7
0
13.0
45.4
17.1
34.2
32.4
58.3
17.0
20.2
22.1
25.3
25.9
41
-------
Inline Mixer Evaluation:
It was hoped that as many variables as possible could be
eliminated, one of which was the variations in the lime feed to
the system. If this could have been done, comparisons of two
operating periods, one with the inline mixer in service and one
with the inline mixer out of service, would have been sufficient.
This not being the case,two other periods of 28 days each, before
and after the period selected, were compared. Results are
shown in the following tabulation.
TABLE 9
INLINE MIXER EVALUATION
TREATMENT DATA
Operating Period
Inline Mixer
Service
* Waste Color, Untreated
Treated
Treatment- (Average)
Ca(OH)? ppm
Color Reduction
ppm/ 100 ppm Ca(OH)?
Color Reduction %
5/12
5/26
in
804
100
1618
45. 7
87.3
5/28
6/8
out
650
77
1507
35. 1
83.2
4/14
5/11
in
1113
96
1340
77.6
90.3
6/9
7/6
out
1457
118
1277
111.3
90.7
*APHA cobalt units ppm - average
Operating periods -were selected to have corresponding mill
operating conditions. Percent color reduction is less at low
untreated waste colors as shown in the Appendix,page 111. For
this reason the difference in percent reduction between the
periods shown in the above first two columns could be expected
with or without the inline mixer. At the higher untreated waste
colors, as shown in the next two columns, the difference is less.
This is also normal. Treatment above 1200 ppm Ca(OH>2 gives
relative lower reduction of color per 100 ppm Ca(OH) used.
<£
The data is not straightforward as it must be interpreted. The
above comments can be summarized: treatment results were
normal for the combination of untreated waste colors and treat-
ment levels with the inline mixer in service and with it out of
service. In short, the inline mixer does not contribute to the
42
-------
degree of treatment obtained.
Floe Agitator Evaluation:
At the beginning of the program it was planned to vary the degree
of agitation to determine the effect on floe formation and its
effect on color removal. Inspections made during mill shut-
downs indicate that this will not be practical. Full agitation is
required to keep settleable solids in suspension. Even then
there is a build-up of solids around the bottom of the tank at the
outer edges.
Clarifier and Pipe Line Inspection:
During the mill shutdown July 4 through July 8, 1969, the
clarifier was emptied. The concrete bottom was inspected for
cracks and the rakes were checKed for clearance. The bottom
was found in good condition. One rake was dragging at the outer
end. This was corrected by removing one inch from the outer
blade.
On July 8, one opening was cut in the steel horizontal run of the
overflow line from the floe tank to the clarifier center feed well,
and two openings were cut in the 36" concrete discharge pipe
line from the clarifier to the lake. The inspection holes in the
concrete pipe line were 1/3 and 2/3 of the distance from the
diversion structure to the outfall.
The interior surface of the steel pipe was evenly covered with
approximately 1/8" layer of light colored scale probably of high
calcium content. At the high water mark on the downward leg of
the overflow pipe, the scale was approximately 1/2 inch.
In the concrete pipe at the inspection hole nearest the diversion
structure, there was approximately 1/4" of hard solid scale and
lumps or rings of a softer scale containing fiber up to 2" thick.
At this point in the line the pipe flows full. At the secord point
of inspection of the concrete pipe line, the same scale and rings
were noted but to a lesser degree of thickness. At this point
the pipe line flows approximately 3/4" full. At the outfall, scale
had formed up to the flow line or about 1/3 of the pipe cross section.
It was less at this point than at the other points of observation.
When scale was removed from the concrete pipe, the exposed
43
-------
surface was smooth and showed no signs of erosion. Measure-
ments of the pipe diameter at the three points -were 36-1/16,
36-1/8, and 36-1/8 inches in the direction of flow.
The pipe lines have been in service for fifteen months. Overall,
there were no signs of erosion. The scale build-up at this time
is not significant. Should the waste be stabilized by carbonation,
the scale build-up will be materially reduced,if not eliminated.
Operating Period - August 1 through August 31, 1969
Modifications to Equipment:
Changes to the water supply system to the lime slaker and slurry
tank were to include:
1. Replacement of 2" water makeup control valve with a 4"
control valve.
2. New pump and motor.
3. Water pressure regulator.
The above equipment was received. The pump foundation and pump
were installed. However, the tie-in to the water surge tank
could not be made until the scheduled mill shutdown, September 1.
The constant discharge pump to control lime slurry flow to the
specific gravity measuring pot on the slaker was rebuilt by the
supplier and was reinstalled. This was not complete until the
last of the month.
Full automatic control operation of the lime feed system is
contingent on the completion of the water supply system.
Other changes made were:
1. Installation of a small booster pump and pressure tank
to maintain a water supply at 30 psi to the purge system
on the process waste flow measuring nozzle.
2. Installation of differential pressure regulators on the
above water purge system.
44
-------
In addition, the waste flow recorder was taken out of service for
re calibration.
In the past there has been considerable difficulty in maintaining
the waste flow measurement instrumentation. Changes in the
system have been made to try to resolve these operating
difficulties. To^further check out the system, the flow nozzle
was scheduled to be removed for inspection during the scheduled
mill shutdown September 1.
Operating Results:
All during the month of August, lime was fed at a constant rate
to give a concentration of 1200 ppm calcium hydroxide for a
calculated average process waste flow of 6. 6 mgd. To do this
the lime screw conveyor rpm was adjusted to give a lime feed
which would maintain a slaked lime slurry specific gravity of
1. 05 with a constant slurry flow rate of 63 gpm.
The actual average concentration of calcium hydroxide maintained
during the reporting period was 1305 ppm. For three days the
concentration averaged 891 because of a high volume of surface
water from heavy rains draining into the process sewer.
Deleting these three days, the average for the period was 1349
ppm Ca(OH)7> Average lime consumption was 4. 59 tons 90%
CaO per million gallons of waste treated. The average percent
reduction in process waste color with this treatment -was 91. 3.
The above color reduction is on the same reporting basis that
has been used since the start of the program. Color is
expressed in APHA platinum cobalt units. The color of
untreated waste samples are at the existing pH of the samples
which for the period averaged 10. 1. The color of the treated
waste samples is at an adjusted pH of 7. 6. Both of these pH
values represent actual pH of waste as they are and will be
observed. The reason for following this procedure is that some
color reduction can possibly occur with the addition of acid
causing precipitation of some of the color bodies. Since June 21,
color of untreated waste has been determined at the unadjusted
pH and at a pH of 7. 6. The difference in color at the lower pH
was 14. 8% less than at the higher pH. With both colors,
untreated and treated waste, at a pH of 7. 6, the overall percent
color reduction for the reporting period was 89. 9 compared to
91. 3 for the color reduction not adjusting the pH of the untreated
45
-------
waste samples. For this period the color reduction per 100 ppm
calcium hydroxide concentration is reported with the untreated
and treated waste both adjusted to 7. 6 and is shown for each
range of color in the Appendix,page 113. It will be noted that the
overall pattern of values has not changed,other than a small
decrease in some instances.
The lime feed system was not put on full automatic control as
consistent performance could not be obtained without first
making changes to the water supply system to the lime slaker,
and the slurry tank. Otherwise, the performance of the waste
treatment plant was at an acceptable level. Total down time was
8^ hours; 5^ hours were for repairs of the lime screw conveyor
variable speed drive,and 3 hours were for inspection of the lime
slurry pump impeller. The lime screw impeller was badly worn
and was scheduled for replacement. The -waste treating plant
was shut down at midnight of August 31, with the mill shutdown
for scheduled repairs.
Total process waste treated was 193. 78 mg including 29. 55 mg
of waste returned from the holding ponds. During the down time
referred to above, 3. 90 mg of waste were diverted to the holding
ponds. Based on the waste in and out of the holding ponds, avail-
able holding capacity should have increased 25. 65 mg. Actually,
the ponds are approximately three-fourths full because of heavy
rainfall during August. Process waste flow from the mill
averaged 5.42 mg, which is 60. 0% of the total mill flow including
uncontaminated water diverted to Riceboro Creek.
Test data for COD and BOD are shown on page 47 with
corresponding color concentrations.
46
-------
TABLE 10
COD & BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
Date
1969
8/5
8/6
8/7
8/12
8/13
8/14
8/19
8/ZO
8/21
8/26
8/27
8/28
AV.
UNTREATED
Color COD BOD
APHA pptn ppm
Units
1150
880
980
1000
1400
1400
1080
900
1220
1200
1500
1140
1154
831
792
931
- 986
1253
1161
928
724
1478
1156
NT
1024
260
243
251
353
334
278
275
192
293
376
349
338
295
TREATE
Color COD
APHA ppm
Units
80
80
125
120
110
110
110
90
100
130
160
130
112
415
401
NT
754
509
NT
451
524
NT
NT
NX
NT
479
D
BOD
ppm
209
184
210
262
177
202
202
150
180
255
259
259
212
% REDUCT
Color COD
93.0
?0.9
87.2
88.0
92.1
92.1
89.8
90.0
91.8
89.2
89.3
88.6
90.2
50.1
49.4
41.8
59.3
51.4
28.0
46.7
ION
BOD
19.6
24.2
16.3
25.8
47.0
27.3
26.5
21.9
38.6
32.2
25.8
23.4
27.4
Note NT - Heating shelf for COD test out for repairs.
Operating Periods - September 1 through October 31, 1969
Modifications to Equipment:
1. Water supply to lime feed system.
Piping, installation of pressure regulator, connection of
pump suction to surge tank and change over from a 2 inch
to a 4 inch make-up water valve were completed on October
22, 1969.
2. Specific gravity measurement of slaked lime from the slaker.
Pumps used to sample lime slurry have not been satisfactory
because of poor performance and excessive maintenance cost.
A new centrifugal pump constructed of a metal equivalent to
Carpenter 22 was purchased. This pump has a capacity of
10 gallons per minute with a 6 ft head at 1050 rpm. A
constant headbox similar to the one used on the specific
gravity measurement of the lime slurry from the slurry
tank was to be used to reduce the flow to the specific gravity
sensing device. Completion of this modification was
47
-------
scheduled for November 5, 1969.
2. Process waste flow measurement:
The Badger-Perm "Lo Loss" flow tube was removed for
inspection on September 18, 1969. Solid material, mostly
lime mud, had deposited in the invert of the pipe and throat
of the tube. The low pressure zone of the tube was
completely filled with solids and the upstream taps were
partially plugged. The buildup of solids was attributed to
low water velocity.
The "Lo Loss" tube had been sized for a maximum flow of
15, 000 gallons per minute with a differential pressure of
114.1 inches of water. Operating flows averaged 4, 600 gpm
and did not exceed 8, 000 gpm. At these low flows the
maximum differential pressure was 32. 5 inches. The control
valve to maintain level in the lift pump sump,located down-
stream from the flow tube, was affecting the differential
pressure read out. Both of these factors decreased accuracy
of flow measurements.
To correct the conditions noted, a major change was indicated.
On October 28, 1969, the flow tube was removed and replaced
with an eccentric orifice plate with vena contracta taps. Up-
stream flush outs were also provided. This orifice is
designed for a range 0 to 8, 000 gallons per minute with a
maximum pressure differential of 100 inches of water.
Performance appears to be satisfactory.
Operating Results:
Difficulties in maintaining the waste flow measurement continued
after all known corrective measures had been made, including
installation of a separate purge water system with pressure
regulators for the flow nozzle. For automatic control of the
lime feed system, the waste flow measuring devices must
function properly. As this was questionable, the lime feed
system was operated on manual control for an average flow of
6. 6 mdg, continuing the operating procedure for the preceding
month.
During September, the color removal system was down one day
for the Labor Day mill shutdown plus another seventeen hours
48
-------
for repairs. The repairs included work on the flocculator
agitator, clean-out of floe tank, work on lime screw drive, and
removing scale buildup in 3, 000 ft of 36 inch concrete pipe line
from the clarifier to the lake. The untreated waste flow nozzle
was also taken out of service for inspection and cleaning.
During these repairs, waste flow was diverted to the holding
ponds. There was an additional nineteen hours and fifteen
minutes of diverting to the holding ponds due to the high color
of the treated waste due to upsets in the mill. Total time
diverted to the holding ponds was thirty-six hours fifteen minutes.
In October waste was diverted to the holding ponds thirty-seven
hours,thirty minutes of which seventeen hours,thirty minutes
were during a mill power failure and for repairs to the lime
screw and replacement of the untreated waste flow nozzle with
an eccentric orifice. The remaining twenty hours were due to
high color.
With manualcontrol of the lime feed system and the excessive
diversion to the holding ponds, the operating results were better
than expected. Lime concentration objective was 1200 ppm
Ca(OH)2. In September the average for the period was 1394 ppm
and in October the average was 1266 ppm. Average lime
consumption for the two periods in sequence as tons 90% CaO per
million gallons of waste treated was 5. 01 and 4. 57. The average
percent reduction in process waste color was 90.4 for September
and 91.4: for October. Color reduction per 100 ppm Ca(OH)2 for
the range of untreated waste and lime concentration are shown
in the Appendix.page 114and page 115,for the two reporting
periods.
i
Waste flows in million gallons were as shown in Table 11.
TABLE 11
Uncontaminated Treated Process Waste
Waste To From From Total Diverted
Riceboro Creek Mill Holding To Holding
Pond Pond
Sept. ,
Oct. ,
69
69
124.63 150.15 29.23 179.38 12.11
112.90 129.17 25.19 154.36 8.33
Tests for COD were not run because of the heating shelf, required
for this test, burned out. Material delivery .and availability of
repairmen delayed its replacement. Some of the BOD tests were
49
-------
depleted of oxygen before completion. To correct this, a new
glass lined water still was ordered. Test data on the completed
BOD tests is shown below with corresponding color concentrations.
TABLE 12
BOD REDUCTION VS COLOR REDUCTION
PROCESS WASTE
UNTREATED
Date
1969
9/3
9/4
9/9
9/9
9/11
9/18
9/23
Av.
10/1
10/7
10/8
10/14
10/15
10/16
10/21
10/22
10/23
10/28
10/29
10/30
Av
Color
APHA
Units
1000
1760
1760
1580
1020
2000
1000
1445
1860
2100
1360
1400
3080
1880
1060
1660
2000
1460
1540
1660
1755
BOD
ppm
236
350
421
296
289
545
253
341
486
472
313
367
607
396
312
564
693
323
462
550
462
TREATED
Color
APHA
Units
130
120
170
160
160
150
80
138
140
130
125
150
150
180
110
125
170
110
100
140
135
BOD
ppm
225
251
281
216
213
273
246
243
361
234
208
267
264
279
293
286
361
267
279
252
279
% REDUCTION
Color
87.0
93.2
90.3
89.9
84.3
92.5
92.0
89.8
92.5
93.8
90.7
89.3
95.1
90.4
89.6
92.5
91.5
92.5
93.5
91.6
91.9
BOD
4.7
28.3
33.2
27.0
26.3
49.9
2.7
24.5
25.7
52.2
33.5
27.2
56.5
29.5
6.1
49.7
47.9
17.3
39.6
54.2
36.6
Operating Period - November 1 through December 31, 1969
Modifications to Equipment:
Specific gravity measurement of slaked lime from slaker.
The centrifugal pump and constant headbox were completed on
schedule, November 5. This pump did not give the flow capacity
required. As it did not come up to purchase agreement it was
returned to the supplier for credit against the purchase of a
Monyo pump. This was the third attempt to obtain a satisfactory
5(X
-------
pump for this service.
Operating Results:
The severe corrosive and abrasive characteristics of lime have
begun to show their effect upon the lime feed equipment after
twenty months of service. The variable speed drive on the screw
supplying the lime to the system has been a continuing source
of trouble. It has been rebuilt twice and there are still problems
in adjusting and maintaining an adequate range of screw speed.
Replacements include the slaker agitator motor, slaker reject rake
bearings, slaked lime slurry agitator and motor, lime slurry
pump impeller, lime slurry recirculation control valve, and
two manual lime slurry dump valves. The lime slurry circul-
ation pump for specific gravity measurement has been replaced
twice and is to be replaced a third time. The lime slurry feed
pump casing has been welded. The pump shaft is worn to such
an extent that packing will not hold more than twnety-four hours.
The pump shaft, impeller and casing, are scheduled for replace-
ment.
In November repairs were made to the lime screw variable drive
on the run,without diverting the waste to the holding pond,in an
effort to hold treatment until the mill scheduled shutdown in
December. On December 11 and 13 it was necessary to divert
to the holding ponds for a total of twenty-two hours because of
low lime feed and high untreated waste color. Because of this
and the amount of work to be done, the lime feed system was
shut down for maintenance and repairs on December 20. Not
including the two-day mill shutdown, December 24 and December
25, untreated and treated waste were diverted a total of 104
hours and 20 minutes to the holding ponds in December.
Waste flows in million gallons were as follows:
TABLE 13
Uncpntaminated Treated Process Waste Diverted
Waste to From to Holding
Riceboro Creek Mill Hold. Pond Total Pond
Nov. ,
Dec. ,
1969 122.05
1969 112.02
172.95
143.24
30.30
8.79
203.25 00.00
170.39 29.39
Attempts to put the lime feed system on automatic control were
51
-------
aborted by the malfunction of the variable speed lime screw
drive. Adjustments to the fixed lime feed were on the basis of
average process waste flow with an objective of 1200 ppm
CafOHU concentration to the floe tank. In November the
average calcium hydroxide concentration was 1189 ppm. In
December the average was 1272 ppm Ca(OH)2. Average lime
consumption for the two periods in sequence was 4. 13 and 4. 24
tons 90% CaO per million gallons of waste treated. The averagv
feed rate of lime was close to the desired rate. However, the
variations in feed were such to give the poorest performance in
color removal since the plant has been in operation. The average
percent color removal for November and December was 89. 3 and
89.1. Color reduction per 100 ppm Ga{OH)2 for the range of
untreated waste and lime concentration are shown in,the Appendix,
page 116 ,?nd page 117 for the two reporting periods.
Tests for COD were not run in November because the heating
shelf required for the test had not been repaired. BOD tests
were made by the Interstate lab in November and by an outside
commercial laboratory in December. Results are shown below.
The shortage of lab work in December was due to the replacement
and training of a lab technician and due to holidays in this period.
TABLE 14
BOD REDUCTION VS. COLOR REDUCTION
PROCESS WASTE
UNTREATED TREATED % REDUCTION
Date
1969
11/4
11/5
11/6
11/11
11/12
11/13
11/18
11/19
11/20 .
Av.
12/4
12/11
12/18
Av.
Color
APHA
Units
2000
2000
1580
1760
1500
2000
1280.
1100
2000
1691
1400
1600
2400
1800
BOD
ppm
572
573
508
371
437
725
314
257
427
465
450
408
426
428
Color
APHA
Units
200
200
130
130
150
175
120
120
160
154
200
120
200
173
BOD
ppm
376
351
321
305
268
427
219
208
302
309
188
160
168
172
Color
90.0
90.0
91.7
92.6
90.0
91.3
90.6
89.1
92.0
90.8
85.7
92.5
91.7
90.0
BOD
34.3
38.7
36.8
17.8
38.7
41.1
30.3
19.1
29.3
31.8
58.2
60.7
60.5
59.8
52
-------
Clarifier Solids Balance:
Calcium analyses of the clarifier feed, underflow and overflow,
were made over the period of October 20, through November 21,
1969. The data is tabulated below.
TABLE 15
CLARIFIER CALCIUM BALANCE
Calcium as CaCOs
Calculated Values
Date
10/20
10/31
11/3
11/5
11/6
11/7
11/10
11/12
11/13
11/18
11/19
11/20
11/21
Clarifier
Feed
Mil. Gals.
5.98
7.47
7.32
7.44
6.80
7.12
5.65
7.11
6.62
7.04
6.86
6.52
6.40
Clarifier
Feed
ppm
2600
1480
990
2212
1468
1480
2964
1976
1940
1612
976
1592
1552
Clarifier
Underflow
ppm
13,240
12,700
23,793
23,333
14,500
23,333
26,000
17,167
17,333
46, 600
14,333
12,666
12,167
Clarifier
Overflow
ppm
1007
1093
831
959
972
957
1008
1060
1104
972
960
908
972
Clarifier
Overflow
Mil. Gals.
5.20
7.22
7.27
7.02
6.55
6.95
5.07
6.71
6.28
6.94
6.85
6.14
6.07
Clarifier
Underflow
Mil. Gals.
.78
.25
.05
.42
.25
.17
.58
.40
.34
.10
.01
.38
.33
Ca in Over
flow % of
Feed
33.7
71.4
83.4
40.9
63.8
63.1
30.5
50.6
54.0
59.4
98.2
53.7
59.4
Calcium was determined by titrating samples taken with 0. 01
M EDTA (disodium salt of ethylenediamine tetraacetate) using
HHSNN indicator. (Fisher Chemical Co., Cat. No. H-342).
Potassium hydroxide (8M) solution was added to the samples,
allowing five minutes for aging before titration. On November 3
and November 19»the calculated volumes of underflow were 3.5
and 17. 5 percent,re,spectively,of the estimated flow obtained from
the pump curves. These discrepancies can be attributed to the
small sample volume of 0. 3 ml used for titration. It was
necessary to use this sample volume because of the high calcium
concentration in the underflow.
The above data shows that the concentration of lime in the over-
flow is relatively constant regardless of the lime concentration
in the feed. All lime added in excess of 1, 000 ppm as CaCOo
will leave the system as sludge in the underflow. At the level of
lime treatment during the test work,total calcium in the overflow
averaged 52. 8%. If treatment level was maintained at 1200 ppm
Ca(OH)_, 61% of the calcium would be in the clarifier overflow.
53
-------
C. Summary
Equipment Evaluation
Lift Pumps:
The lift pumps are the heart of the system. If they cannot keep
up with the effluent flow, the mill has to shut down. Untreated
process effluent cannot be diverted to Riceboro Creek under any
circumstances.
The pumps initially installed were selected with price as a factor
in an effort to cut cost. No savings were realized as these pumps
had to be replaced by Rust Engineering Company within four
months because of excessive wear and outage time for replace-
ment of bearings, shafts, and impellers.
The Hazleton pumps now in service are for heavy duty at 875 rpm
with forced water bearing lubrication. Impellers are 28% chrome
alloy. Performance has been excellent,with no signs of wear after
eighteen months service.
Chemical Feed System:
For a chemical waste treatment system to operate successfully
under such varying conditions of flow as previously described,
there must be a very flexible chemical feed system which will
respond to accurate flow signals with a minimum of lag time.
To accomplish this, the initial control of lime feed as a function
of the output of the variable screw conveyor to the slaker was
changed to the existing control of slaked lime slurry at a fixed
concentration of calicum hydroxide. The eccentric orifice with
pressure sensor oil seals replaced a flow tube.
The chemical feed system may not be the best that can be
installed,but it does utilize all of the equipment that was avail-
able at startup. Attention has been given to details such as, to
provide constant water pressure at the control valves. A
further improvement would be to maintain a constant temper-
ature in the slurry tank. However, the system can and does
maintain a relatively constant predetermined concentration of
calcium hydroxide in the process waste to the floe tank.
Lime is an aggravating material to handle. Approximately
54
-------
ninety percent of the maintenance cost for the entire waste
treatment system is for lime handling.
Flash Mixing:
An inline flash mixer was provided. It was a continuing source
of trouble,as a hard brown scale would build up on the impeller
and internal baffles to a point where excessive vibration and
head loss would occur. After modifying the internal baffles,
conditions were somewhat improved.
To evaluate the need of flash mixing, the plant was operated for
thirty-eight days with the inline agitator removed. Operating
results were compared with selected periods having correspond-
ing mill conditions. From this it was concluded that flash
mixing does not affect the degree of color removal obtained.
Baffling to break up laminar flow is adequate. Consequently,
the inline agitator has been out of service since the first part of
June, 1969.
Flocculation:
The floe tank was designed for fifty minute retention at a
maximum waste flow of 10 mgd and has a double bladed turbine
agitator with a variable speed drive of a range up to 14 rpm.
Test work by The Rust Engineering Company indicated that the
optimum flocculation time is between fifteen and twenty minutes.
However, the longer retention time was provided for on the
insistence of the Georgia State Water Quality Control Board in
their approval of the .construction drawings. Subsequent removal
of the floe tank overflow funnel to reduce turbulence and the
accumulation of solids has necessarily reduced the minimum
retention time from fifty to thirty-five minutes.
A program to evaluate flocculation was aborted by the excessive
buildup in the floe tank of lime mud and dregs from the mill
recausticizing area with any agitation less than the maximum.
By-passing the floe tank was not tried because of the real
possibility of plugging the underground pipe line to the clarifier
center feedwell. Because of the concern with this factor, the
underground pipe line was opened and inspected after eighteen
months service. There was no scale or buildup of solids.
Apparently flocculation gives a side benefit of effluent
conditioning.
55
-------
Clarification:
The clarifier was sized on a rise rate of 0. 24 gallons per minute
per square foot of clear rise area (gpm/sq ft) for 10. 0 mgd of
effluent. At the present average flow of 5.48 mgd, the rise rate
is 0. 13 gpm/sq ft.
The clarifier is subject to considerable short circuiting because
of the large exposed surface area at the elevated location.
Absorbed calcium carbonate from the atmosphere causes calcium
carbonate scale buildup in the launderer notchweirs and the
discharge pipe line. However, solids carry-over is 10 ppm and
less. Solids concentration in the underflow average 2. 0 percent.
Because of the coagulation of lime and fiber, a very readily
settleable conglomerate is formed. For this reason and for
better flow characteristics, the clarifier could be conservatively
designed for an average 0. 50 gpm/sq ft rise rate.
Lime concentration in the feed to the clarifier affects the quality
of the clarified waste and underflow. At treatment levels below
1, 000 ppm calcium hydroxide, color begins to rise significantly
with an increase in turbidity. With treatment levels above 1, 000
ppm, the decolorized effluent from the clarifier has an average
concentration of 722 ppm calcium hydroxide with a plus or minus
deviation of 14 percent, irrespective of the lime treatment level.
Consequently, there is an increase of free lime in the sludge
underflow in direct relationship to the calcium hydroxide added.
This can be as much as 1, 200 percent with a 500 ppm change of
lime concentration in the feed. Primary sludges from untreated
kraft pulp mill effluent are difficult to dewater because of their
hydroscopic characteristics. Addition of calcium hydroxide
makes the problem of dewatering more difficult and adds problems
to calcining in a calcium recovery process.
Natural Stabilization:
The decolorized and clarified effluent entering the natural stabil-
ization lake is sterilized because of the high alkalinity at a pH of
12.2. The chemical properties are such that there are three
reaction processes required to make it suitable for discharging
to Riceboro Creek, namely, physical, chemical and biochemical.
In the first 100 acres of the lake, carbon dioxide absorbed from
56
-------
the atmosphere precipitates practically all calcium as a
carbonate and the pH drops to 10. 2. Without the buffering effect
of the calcium, the pH drops rapidly as the effluent moves on
from this area. With the decrease in pH to 10. 2, some color
is released from, previously deposited calcium. Color of the
effluent increases as it flows on through the lake. This color is
from natural origin and contributes to the color increase in the
first 100 acre zone because of windage.
Biochemical oxygen demand of the effluent is reduced in two
reaction processes. The first is by chemical oxidation with
oxygen from the atmosphere which occurs in the first 100 acres
of the lake and then by the biochemical reactions under aerobic
conditions in the remaining part of the lake.
Mechanical aeration before discharge plays no part in BOD
reduction. Its function is to assure a minimum of 6 ppm
dissolved oxygen in the effluent.
The stabilization lake area could be reduced by as much as 140
acres to obtain the same quality of effluent with pre-calcium
precipitation.
There are three points of particular note in the biochemical
stabilization of kraft mill waste following lime treatment for
color removal. Foaming is eliminated after calcium
precipitation. This is a distinct advantage if an accelerated
aeration process is to be used. Phosphorus is removed by
precipitation,and nitrogen is converted to the ammonia ion which
can be removed by Stripping '°'. This is an advantage in the
natural process as it makes control of eutrophication unnecessary.
Sterilization at the high alkalinity makes it unnecessary to
separate septic tank effluent.
57
-------
Operating Results
Lime Treated Effluent Color:
Effluent color of lime treated kraft pulp mill effluents is
affected to some degree by sodium alkalinity and by the color
concentration in the untreated effluent.
Taking the background mill operating conditions into considera-
tion, effluent treatment operating data presented is based on
averages. Figure 3,page 59, shows the average treated waste
color for a range of treatment of 1, 000 to 2, 400 ppm calcium
hydroxide with average untreated effluent colors of 460, 840,
1210, 1610, and 2120 ppm. The averages are in grouping of
400 ppm of untreated color and represent 456 operating days of
the first two years of operation. Data for days when there were
mill upsets or when effluent was diverted to the holding lagoon
are not included. Untreated effluent color is in APHA cabolt
units at observed pH.which averaged 10.4. Treated effluent
color is at an adjusted pH of 7. 6.
Even with data averaging there is considerable scattering of points
plotted in Figure 3, page 59. This possibly can be attributed to
the uneven number of days for each point. However, there is
a remarkable close conformity of slope of the lines drawn
through these points for each level of untreated effluent color.
The important relationships to be noted are that there is only
approximately 2 ppm of color improvement in the treated
effluent with each 100 ppm calcium hydroxide increase above
1, 000 ppm; and that there is approximately 100 ppm increase
with an increase of 1, 700 ppm untreated effluent color.
Lime Treated Effluent COD and BOD:
Lime treatment also reduces chemical oxygen demand (COD)
and biochemical oxygen demand (BOD) of kraft pulp mill process
effluents. The overall average reduction through the clai-ifier at
the Riceboro mill has been forty-six percent for COD and thirty-
four percent for BODc on unfiltered composite samples.
Figure 4,page 60,shows these values for the treated effluent
compared to the treated effluent color. COD appears to be a
straight line function whereas the BODg is a well defined
exponental function. In the lower ranges of treated color,
values approach a minimum of 180 ppm.
58
-------
sD
1 UNTREATED EFFLUENT
NO. RUNS AVG. COLOR
1000
1200
1400 1600 1800 2000 2200
CALCIUM HYDROXIDE CONCENTRATION - PPm Ca(OH)2
FIGURES
TREATED EFFLUENT COLOR vs LIME CONCENTRATION
2400
-------
1000
10001
o
o
U
a.
a.
I
O
HI
o
X
o
o
•&.
in
x
o
50
100 150 200 250
TREATED EFFLUENT COLOR - PPm APHA UNITS
FIGURE 4
TREATED EFFLUENT COLOR cs. vs COD and BOD,
300
-------
Natural Stabilization of Lime Treated Effluent
Figure 3 page 62 shows the color and BOD- profile of the lake
in relation to the number of acres in line of flow. The data
obtained from weekly surveys, covers a period of one year
from March of 1969 through March of 1970. During this period
the overall lake color has continued to decline.
The zone of natural calcium precipitation is indicated by the
shaded area to the left of the chart. In this area the color
increases approximately 40 ppm. In the biochemical stabiliza-
tion area, color increase rapidly and then as a straight line
function in relation to the number of acres covered. It is
presently not known whether or not there is a further reduction
of the process effluent color in the final stages of stabilization.
Research work at the FWQA, Southeastern Laboratory in Athens,
Georgia, may give the answer.
The decrease of biochemical oxygen demand following calcium
precipitation follows the trend of maximum performance for a
natural biodegradation system.
In summary, overall chemical treatment in terms of process
color is 90% and biochemical treatment is 98%. Following is a
comparison of average mill discharge with regulatory limitations
as imposed expressed in Ibs per day. The limitation on color in
Ibs per day is based on 30 ppm color and a maximum flow of 20
mgd as defined in the mill construction permit.
TABLE 16
Actual Limitation
#/D #/D
Color 5,020 2,500
BOD 485 800
Because of the color contributed from natural sources within the
biochemical stabilization lake, waste discharges were restricted
to a five percent dilution with fresh water inflow at Half Moon
Landing, based on charts furnished by U. S. Geological Survey
office in Atlanta.
The quality of the treated effluent discharged is comparable to
Riceboro Creek waters with the exception of dissolved inorganic
61
-------
FIGURE 5
STABILIZATION LAKE COLOR
& BOD PROFILE
March 1969 - March 1970
BIOCHEMICAL STABILIZATION
ACRES
62
-------
solids. Color at times is lower than the creek water. BOD
averages 6 ppm with dissolved oxygen at saturation. Because
of the effluent stabilization there is no oxygen sag in Riceboro
Creek below the mill.
Lime Treatment Cost
The actual cost of equipment and facilities for lime treatment
came to $355, 100 exclusive of the clarifier and holding lagoons.
This figure can only be used as a guide because of the possible
reduction in equipment requirements outlined in the commentary
on facilities and because of the escalation of equipment cost.
Chemical lim.6 is the major operating cost. Figure 6 , page 64
shows the data of Figure 3 in terms of percent color reduction
compared to the cost of chemical lime per million gallons of
effluent treated.
Lime is charged to the treatment plant at $15. 35 per ton as
ninety percent calcium oxide. With untreated effluent colors
above 460 average, the increase in percent reduction is
approximately 2. 0 with an increase of 1,400 ppm calcium
hydroxide. This gives a cost of $38 per million gallons for a
one percent change in color above the 1, 000 ppm calcium
hydroxide treatment level.
Labor listed below is in terms of hours per year for conversion
to current cost at any locality.
Operating Labor 8, 760 Hr/Yr
i
Maintenance Labor & Supervision 6,500 Hr/Yr
The above figures are for the entire waste treatment system.
Approximately eighty-five percent of the labor is for lime
treatment.
63
-------
AVERAGE UNTREATED EFFLUENT COLOR ppm
1000
1200
2200
1400 1600 1800 2000
CALCIUM HYDROXIDE CONCENTRATION - ppm Ca(OH)2
FIGURE 6
PERCENT COLOR REDUCTION AND COST vs LIME CONCENTRATION
2400
-------
SECTION VI
CALCIUM RECOVERY - MATERIALS AND METHODS
A. Carbonation Process
The carbonation process can best be described as a physical
process combined with a chemical reaction process.
The transfer of carbon dioxide from a gaseous to a solution phase
is the first step in the physical process. The rate of transfer is
a function of bubble size, time of contact, temperature and gas
solubility. As the carbon dioxide goes into solution, it reacts
with the dissolved calcium hydroxide to form calcium carbonate.
Within pH ranges of 8. 0 and 10. 5, calcium is almost entirely in
the insoluble carbonate form (9).
The succeeding physical process is the crystalline growth of the
precipitated calcium carbonate particles. In a supersaturated
solution, precipitation can occur in two ways - by spontaneous
self nucleation, or by the deposition on solid material. In the
first instance, a large number of colloidal particles are formed
which do not settle. This phenomenon occurs at a high degree of
saturation such as exists -with the lime treated kraft waste. In
general, the rate of precipitation is proportional to both the
degree of super saturation of the solution and the available
surface of previously formed particles. Time is required for
particle growth to take place ' '.
The final phase of the physical process is the particle settling
and thickening into a slurry or sludge. Calcium carbonate does
not follow the reduced solids concentration flux curve for rigid
spheres. Within well defined concentrations, calcium carbonate
goes through a period of free settling followed by aggregation of
particles and finally, compaction settling to maximum density '* *'.
There is an overlapping of both physical and chemical processes
except in the final phase of compaction settling. To design equip-
ment for the process, the system was divided into four sections:
carbonation, flocculation, settling, and sludge concentration.
B. Pilot Plant
Design
65
-------
The continuous pilot plant at Interstate Paper Corporation was
designed to be able to control and vary all factors which would
affect the process. A carbonator was provided for the
formation of calcium carbonate particles by the physico-
chemical and chemical reaction process. A separate unit was
installed for the physico-chemical process of flocculation
settling.and concentration.
The carbonator was built for this specific application. The
second unit is a standard Densator pilot plant obtained from the
Fuller Company/General American Transportation Corporation
on a rental basis. Gross sectional sketchs of both units are
shown on Figure 7,page 67, and Figure 8,page 67. Overall
dimensions are: Carbonator 30 inches ID by 15 feet; Densator
42 inches ID by 10. 5 feet.
Instrumentation was selected so as to have the maximum
application in a future full scale plant and to enable the evalua-
tion of their capabilities for control. Controls are the electronic
type and were obtained from Foxboro.
Auxiliary equipment, including pipe, valves, and pump was sized
to provide a plant hydraulic capacity of 43, 200 gallons per day.
Carbon dioxide gas supply was from two one ton dry ice.
converters. Air was from mill supply.
Figure 9, page 68 , shows the flow diagram of the pilot plant.
Included are the modifications to the gas supply system,
carbonator piping system for counter flow of water and gas,
and slurry withdrawal from the Densator.
66
-------
TO DENSATOR
RECYCLE
FLOW FROM
CARBONATOR
RECYCLE
V
TREATED
WASTE
SLUDGE
FIGURE?
CARBONAtOR
FIGURES
DENSATOR
-------
DENSATOR SLURRY RECYCLE
oo
FIGURE 9
PILOT PLANT FLOW DIAGRAM
-------
Layout and Orientation
Figure 10
In the above photograph, the mill's 200 ft. diameter primary
clarifier is shown in the background,with the sloped 36 inch over-
flow line in the foreground. The tall elevated tank with the
perpendicular manifold to the left is the carbonator. The two
horizontal cyclinders in the left foreground are the dry ice
storage for the carbon dioxide system. To the far right is the
Densator. The small, sloped, elevated pipe line connects the
carbonator and Densator. In this line just ahead of the Densator,
is the magnetic flow sensor, flow control valve and pH electrodes.
The horizontal, small pipe line below the magnetic flow sensor is
the overflow from the Densator. Just below this line are the
flow-through reference cell, turbidity meter,and pH indicator
mounted in the enclosed panel board. Control and recording
instruments, motor switch gear, and field laboratory are in a
small building to the right of the Densator. This building is not
shown in the above picture.
In the order of water flow, lime treated mill process effluent is
withdrawn from the primary clarifier overflow line by a. pump
mounted on the base below the carbonator. Flow is then into the
perpendicular manifold to the carbonator at one of the selected
levels and then out of the base of the carbonator by gravity flow
through the sloped, small pipe line to the Densator. Flow is
69
-------
controlled by the automatic valve ahead of the Densator. A d/p
cell mounted on the lower part of the carbonator shell and a
control valve on the discharge side of the pump are part of the
system that maintains a set water level in the carbonator.
A pump at the base of the Densator returns partially settled
calcium carbonator slurry through a rubber hose back to the
lime treated effluent feed line to the carbonator. An in-line
booster pump mounted on the carbonator elevated platform
increases flow as needed. Heated carbon dioxide and air enter
the carbonator near the bottom. Not shown is a rectangular
removable plate to facilitate inspection and interchanging of
diffusers. The cylindrical housing on top of the carbonator is a
cover for the variable speed motor on the carbonator agitator.
The settled carbonate slurry is discharged by gravity to collection
drums behind the Densator. Control is by a timer operated
solenoid valve.
The flow pattern outlined above is the one used for the experi-
mental runs. Referring back to Figure 9 page 68, it will be noted
that there are several variations. Water flow can be concurrent
to gas flow and withdrawal from the manifold. Settled slurry can
be recirculated back to the flocculation zone of the Densator,or
a split flow of slurry can be used.
Startup and Shakedown
Each programmed test run was scheduled to have a stabilization
period of not less than fifteen hours with all systems functioning
properly. Change-overs were to be made by 5:00 P. M. each
operating day with the test work starting at 8:00 A. M. the
following morning for a period of one hour. Equipment main-
tenance and modifications were to be done by the mill maintenance
department on the first shift.
On starting up, the first problem was with the gas control system.
As originally planned, air and CC^ were to be blended in a mixing
tee, heated and controlled by one valve actuated by the pH
controller. Differences^ gas and air supply pressure gave a
wide swing in carbon dioxide concentrations. This was resolved
by installing a separate control valve for the air supply. Both
valves were actuated by the pH controller. Th'e mixing tee was
relocated down stream from the control valves with the heating
70'
-------
tapes next in line. To increase gas transfer, piping changes
were made for counter flow of gas and waste.
There were also difficulties with the carbon dioxide gas pressure
regulators freezing, particularly so in extreme cold weather.
Additional heating tapes finally gave fair control. However, a
thermostat would have been of great help.
Sludge recycle to the carbonator was difficult to control. Both
the pump suction and blow-down solenoid valve were on the same
manifold connected to a twelve inch riser from the bottom of the
Densator. The riser would plug as the density of the slurry
increased to the desired consistency. First, the booster pump
was added to increase and maintain the slurry flow. Then, it
was found that the settled slurry was being transferred to the
carbonator where it was building up and not getting out of the
system through the Densator blow-down.
With the slurry recycle problem, there was also a severe
foaming problem in the carbonator. At first it was thought to
be caused by abnormally high organic color following mill upset
conditions. Other possible explanations •were a high solids to
air ratio, low gas temperature and high pH. To eliminate the
first possibility, the slurry recycle pump suction was changed
to a point 18 inches above the bottom of the Densator. Next, the
slurry blow-down riser and manifold were eliminated as plugging
at this point became more severe.
Practically all of the problems with slurry recycle and foaming
were eliminated with the above changes. Counter flow of waste
and gas in the carbonator, higher gas temperatures and operating
at the lower pH range of 9. 0 apparently improved performance.
Because of the difficulties encountered, many test runs were not
acceptable. It was not until November 4, 1969 that the test
program actually got under way. This was thirty days behind
schedule. The delay was not a complete loss as much was
learned about the process from observations.
Test Procedure
Following the plant stabilization period, all charts were checked
for consistency of control. If satisfactory, three sets of samples
of ;feed and Densator overflow were collected at 20 minute inter-
vals. Accumulated slurry in blow-down barrels -was measured
and sampled. A composite sample of the feed was also taken.
71
-------
Analytical and test procedures were as follows:
1. Calcium
a. Soluble calcium concentration of the samples were
determined by titrating with 0. 01 M EDTA (ethylene
diamine tetracetate) after the addition of 8 M potassium
hydroxide and aging for five minutes. HHSNN (2-
Hydroxy- l-(2-Hydroxy-4Sulf-l-Naphthyl -AZO-3-
Naphthoic Acid) was used as an indicator.
b. Total calcium was obtained by following the above
procedure after all calcium had been solubilized by
acidification to a pH of about 4. 0.
c. Suspended calcium was by difference of total and soluble
calcium.
2. Carbon Dioxide for Sodium Alkalinity
Composite samples of feed were titrated with N/50 E^SO.
to the pH end point of the carbonated effluent using a pH
meter. This gave the total alkalinity reacted express as
calcium carbonate. Calcium alkalinity reacted was obtained
by the decreased calcium in solution, also expressed as
calcium carbonate. The difference between these two
figures was converted to CC2 and expressed as a percent
of the total CO2 used for sodium alkalinity.
3. Percent Carbon Dioxide in Gas Feed
Percent carbon dioxide was calculated from the measured
COo and air volumes, temperatures and pressures. A
Fyrite CC"2 indicator was used for verification.
4. Densator Slurry Underflow
a. Specific gravity was obtained by weighing a 1, 000 ml
sample.
b. Calcium content was obtained by titration using the
procedure for calcium above.
5. Filter Leaf Test
72
-------
An Eimco Filter Leaf Kit ' °' was used to make filtration
tests on the Densator slurry. The 1/10 square foot filter
was covered with an Eimco polyethelene Style No. P O -
801 RF monofilament wire, thread count 105 x 40. Most
tests were made with a formation time of 15 seconds and
a drainage time of 1 5 seconds. In some instances, the
formation time had to be reduced. All tests were made
at 15 inches vacuum, unless other-wise noted.
There was one deviation from the Eimco procedure -
there was no drying time. Thus, a test made with the
above conditions was calculated to have 120 cycles per
hour. Each filter rate then was multiplied by a factor of
0. 8 as recommended by Eimco.
6. Slurry Carbonate Purity
Slurry carbonate purity was determined by analysis in the
Continental Can Company's research laboratory. Calcium
was determined by precipitation as the oxalate and carbon
dioxide was measured by loss of weight on ignition.
C. Methodology Used in Computer Program
Program
A statistically designed experminental program based on the
function block diagram, Figure 11,page 74,was used to minimize
the number of experimental runs and yet gain maximum design
and operation information.
A two-level factorial design of sixteen (16) runs was used to
examine four design and operating parameters. Table 17,page 75,
shows this experimental program and the performance response
measured and studied. These parameters and responses were:
Parameters
Code
1) pH Carbonation pH, dimensionless.
2) Depth Carbonation depth, in feet.
3) Rate Carbonation feed rate, in gallons per minute.
4) Cycle Densator recycle rate, in gallons per minute.
73
-------
VARIABLES
SLUDGE LEVEL
SLUDGE RECIRCULATION
AGITATOR SPEED
NON
CONTROLLABLE
VARIABLES
PH
ALKILINITY
COLOR
FLOW
TEMP.
pH
WASTE DEPTH
WASTE FLOW
SLUDGE RECYCLE
AGITATOR SPEED
C02 EXCESS GAS
C07 GAS CFM
CA CONC.
TURBIDITY
FLOW
RISE RATE
OVERFLOW RATE
CA CONC.
WASTE COLOR
pH
COLOR
REMOVAL
PLANT
CLARIFIER OVERFLOW
CARBONATOR
SLUDGE
RECIRCULATION
C02
TEMP.
SPARGER
SLUDGE
BLOW DOWN
SLUDGE RECYCLE
SLUDGE BLOW DOWN
RESPONSE
C02
FEED
SYSTEM
FLOW
CA CONC.
SLUDGE FILTER RATE
CO
FIGURE 11
FUNCTION BLOCK DIAGRAM CARBONATION PILOT PLANT
-------
TABLE 17
CARBONATIQN EXPERIMENTAL PROGRAM
Run #
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
1Z.
13.
14.
15.
16.
j>H
+ = 10
- = 9
Depth
+10
- 5
+
+
PARAMETERS
Feed
Rate
+ = ZO
- 10
Recycle
Rate
+ + +
+ + + +
Notes:
1. Run experimental prograrr in radom order.
Responses
To Be Measured
CO, Transferred to CaCOj
Total CC2 Transferred
Dissolved Calcium loss
Suspended Calcium loss
Total Calcium loss
75
-------
Responses
Code
1) CO-CA CCstransferred to CaCOo, in percent.
2) CO-TO Total CCU transferred, in percent.
3) CA-DI Dissolved calcium loss, in percent.
4) CA-SU Suspended calcium loss, in percent.
5) CA-TO Total calcium loss, in percent.
CO-XX 2) - 1)
Table 18, page 77, shows a list of other feed and operating
variables which were monitored during the experimental
program and included in the computer study.
Tables 19, 20, 21, and 22* pages 78, 79, 80, and 81, show the experi-
mental data generated by the sixteen (16) run statistically
designed experimental program and duplicate runs during check-
out of the equipment.
Data Analysis System
Certain key assumptions regarding the normality of the above
raw data were checked to assure a proper interpretation of the
calculated statistics. Although the number of degrees of freedom
(total observations) available for calculating statistics and their
confidence limits was adequate, the design used guaranteed
orthogonality (independence) of the parameters.
The data, collected from the above statistically designed experi-
mental program and duplicate runs were analyzed, using a
computer-based Data Analysis System (*^). This system is a
coordinated set of user-oriented computer programs to statist-
ically extract information from data, to develop mathematical
models and to test their validity. It has available the usual
mathematical modeling routines of simple correlation, simple
and multiple regression. Also provided is an effective tear.-
down regression " "' and the new powerful heuristic regression
(17, S
Study Results
All of the experimental data were studies in total. That is,
twenty-one runs, four parameters, twenty-one other variables
76
-------
TABLE -18
LIST OF OTHER VARIABLES MEASURED
FEED STREAM CHARACTERISTICS
Code
1.
2.
3.
4.
5.
6.
FD-PH
FD-TP
COLOR
FD-DI
FD-SU
FD-TO
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Feed stream pH, dimensionless.
Feed stream temperature, in degrees Fahrenheit.
Feed stream color, in APHA units at 7,6 pH.
Feed stream dissolved calcium as CaCO,, in pounds per 1000 gallons.
Feed stream suspended calcium as CaCO~, in pounds per 1000 gallons.
Feed Stream total calcium as CaCC,, in pounds per 1000 gallons.
CARBONATOR OPERATING VARIABLES
ST-IR Carbonator agitator speed, in revolutions per minute.
Flow Densator recycle flow, as percent of feed.
SOLID Densator recycle solids, as percent of feed.
CC"2 Gas feed CO2 concentration, in percent.
TEMPC Gas feed temperature in degrees Fahrenheit.
TIME Carbonator retention time, in minutes.
DENSATOR OPERATING VARIABLES
* Densator agiator speed, in revolutions per minute,
FLOC Densator floculation time, iri minutes.
RISE Densator rise rate, in gallons per square foot per minute.
TEMPD Densator temperature, in degrees Fahrenheit.
DE-DI Densator overflow dissolved calcium as CaCO3> in pounds per
1000 gallons.
DE-SU Densator overflow suspended calcium as CaCOj, in pounds per
1000 gallons.
DE-TO Densator overflow total calcium as CaCOj, in pounds per 1000
gallons .
SP-GR Densator underflow specific gravity.
SYSTEM TIME EFFECT
DAYS Time, as days since start-up..
* Held constant.
77
-------
TABLE 19
CARBONATION PILOT PLANT OPERATING CONDITIONS
o
JZ
I
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Feed Carbonator Densator
ffi
ft
12.2
12.1
12.2
12.2
12.0
12.2
12.2
12.2
12.1
12.2
12.1
12. 1
11.9
12.2
12.1
12.2
fe
o
H
95
103
70
70
98
104
94
63
102
98
99
97
102
106
102
100
|
O©
180
140
180
200
140
200
140
130
175
200
180
160
120
300
180
120
Ca as CaCO,
Ibs / 1000 gal.
Dissolved
8.677
7.844
8.487
8.045
10.148
8.211
9.238
7.677
9.113
8.045
8.011
7.477
~ 8.045
8.645
7.611
8.946
Suspended
.067
.237
.184
.100
.067
.526
.042
.559
.100
.067
.200
.100
.067
.234
.067
.134
3
o
EH
8.746
&. 078
8.670
8. 145
10.214
8.737
9.288
8.153
9.213
8.111
8.211
7.577
8. Ill
8.879
7.677
9.079
Agitator - rpm
64
61
64
63
64
64
64
64
63
64
64
63
63
64
64
64
Densator Gas
Recycle Feed
•u
r*
k
jg <*4
0
0
0
0
0
0
0
0
12.2
20.0
24.3
45.0
10.8
14.3
23.1
20.0
•B
V
4!
Tf »M
3 o
0
0
0
0
0
0
0
0
-
-
-
-
131
-
-
-
(M
O
O
21.0
28.5
14.3
24.2
19.8
50.5
16.2
36.5
15.8
18.8
21.5
17.2
27.6
18.3
14.7
17.6
b,
0
-------
TABLE 20
RESULTS OF STATISTICALLY DESIGNED
CARBONATION EXPERIMENT
Experiment
•
o
2
tf
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
(0
•4->
&
12/12/69
11/26/69
12/23/69
11/21/69
1/7/70
11/25/69
1/6/70
11/24/69
11/13/69
12/10/69
12/9/69
11/20/69
11/18/69
12/19/69
12/8/69
12/11/69
Variables
&
8.7
10.2
8.7
10.0
9.0
10.3
8.7
10.4
8.8
10.2
8.8
10.2
8.9
10.2
8.8
10.0
Carbonation Depth - Ft.
5.0
5.4
10.0
10.0
5.0
5.0
10.0
10.5
5.0
5.0
10.0
9.Q
5.0
5.0
10.0
10.0
1
SO
1
T3
HI
V
fe
10.00
9.30
10.00
10.00
18.25
19.00
18.50
20.00
9.80
10.00
9.25
10.00
19.40
21.00
19.50
20.00
Recycle - gpm
0
0
0
0
0
0
0
0
1.20
2.00
2.25
4.50
2.10
3.00
4.50
4.00
Responses
C02
Transfer
%
CO
O
O
R)
u 3
4J
O O
H H
20.5 22.0
19.9 24.0
27.1 34.7
16.3 20.8
21.0 26.4
20.6 24.6
25.6 33.0
25.9 33.2
20.0 25.5
31.7 33.7
35.5 46.0
38.7 47.8
20.4 25.5
26.1 27.8
33.4 42.1
42.3 45.0
Calcim Loss
%
Dissolved
6.9
5.6
5.6
5.7
4.2
6.0
4.9
6.5
1.1
3.7
3.6
2.2
2.9
5.3
5.2
4.8
Suspended
11.6
10.4
1.3
24.9
1.6
17.1
3.6
15.0
6.5
38.0
21.3
13.6
16.0
26.1
23.7
9.5
•— i
3
o
H
18.5
16.0
6.9
30.6
5.8
23.1
8.5
21.5
7.6
41.7
24.9
15.8
18.9
31.4
28.9
14.3
79
-------
TABLE 21
CARBONATION PILOT PLANT OPERATING CONDITIONS
Duplicate Runs
Feed
•
o
c
PS
9
9A
9B
9C
9D
9E
0.
12.1
12.0
12.1
12.1
12.3
12.2
fa
o
1*
H
102
100
99
101
97
98
si
Bltn
1?,
8^
o©
175
150
240
160
130
200
Ca as CaCO-
Ib. s/1000 gal.
•o
i*
"3
Q
9.113
8.779
8.378
7.936
8.111
7.953
•o
s
8.
CO
.100
.067
.033
.050
.100
.050
fH
o
9.213
8.846
8.412
7.986
8.211
8.003
Carbonator
i
•
0
1
DO
63
63
64
64
33
35
Densator
Recycle
•o
0)
PN
1 *8
fa^
12.2
10.0
26.5
21.1
25.6
38.9
•0
«
fa
n
13 °
**
263
123
141
110
3000
3600
Gas
Feed
N
0
o
*
15.8
15.2
23,2
_
15.5
22.2
o
1
H
«
• -
145
191
185
180
1
fl
O
fl
0)
*
18.4
18.4
19.3
17.6
15.0
16.2
Densator
6
1
i
o
1
*
10
L 10
10
10
10
10
.s
6
.2
V*
r-1
3
O
o
fa
11.6
10.6
14.1
12.3
10.9
13.0
1
1 *
PS o4
IB
4) -^
.3 a
OS O
1.26
1.28
1.06
1.22
1.38
1.15
o
t
V
EH
97
94
97
93
90
86
Overflow
Ca as CaCO,
Jba. /1 000
•O
0
•
rt
LI
^0
M 'M
fi (^
D w
1.0689
1.0260
1.0522
None
1.0526
1.0658
00
o
-------
TABLE 22
RESULTS OF STATISTICALLY DESIGNED
CARBONATION EXPERIMENT
Duplicate Runs
Exper iment
0
2
rf
9
9A
9B
9C
9D
9E
4)
fl)
°
11/13/69
11/10/69
11/7/69
11/6/69
11/5/69
Variables
£
P,
8.8
8.8
8.7
8.7
9.3
8.9
,
£
i
4
V
Q
g
•H
52
0
rO'
n
ri
U
5.0
5.0
5.0
5.0
5.0
5.0
6
ft
on
i
13
-------
and five responses. This study eliminated many variables and
indicated that an upset condition occurred in the pulp mill
during four runs. These runs were:
Run # Date
10 12/10/69
16 12/11/69
1 12/12/69
14 12/19/69
Consequently, a second study was made using seventeen runs by
eliminating these four runs. These two studies then allowed
some contrasts to be made on the ability of the system to handle
both "normal" feeds and "abnormal" feeds.
Table 23 page 83 shows the means and standard deviations of all
the variables included in the twenty-one run study. Note that
each parameter was studied over a wide range of values and that
the responses also varied widely. This wide range illustrates
the efforts that were made in the study to thoroughly examine the
carbonation process system.
The simple correlation coefficient is a dimensionless measure of
the amount of correlation between two variables. A zero value
means no correlation. Direct correlation is indicated by a plus
one, while indirect correlation is a minus one.
The correlation coefficient of the various independent variables
with the dependent response ranks their importance when used as
single variables to predict the response. The correlation
coefficient for each pair of independent variables measures the
degree of redundancy existing among the independent variables
which can influence the study.
The significant correlations in this study were examined. Table
24 page 84 shows a list of variables that -were eliminated from
subsequent studies due to their high correlation with the primary
design and operating parameters.
Table 25 page85 shows the means and standard deviations of the
remaining eleven independent variables and parameters and the
five performance responses for all twenty-one runs.
82
-------
Variable
TABLE 23
MEANS AND STANDARD DEVIATIONS OF ALL VARIABLES
Mean
Deviation
High
Low
Ratio DEV/XBAR
FD-PH
FD-TP
COLOR
FD-DI
FD-SU
FD-TO
PH
DEPTH
RATE
CYCLE
STIR
FLOW
SOLID
C02
TEMPC
TIME
FLOC
RISE
TEMPO
DE-DI
DE-SU
DE-TO
SP. GR
DAYS
CO-CA
CO-XX
CO-TO
CA-DI
CA-SU
CA-TO
12. 1381
95.1438
172.6212
7.9163
0.1452
8.0549
9. 3476
6. 9000
13.4071
1.6666
60.7620
13.8952
408.0029
21.8380
175.0969
19.2714
9.1761
1.7157
90.3814
0.3377
1.2203
1.5581
1.0828
24.2382
25.2380
5.4380
30.6761
3.8761
14.5857
18.4619
0.0920
11. 9181
43.1166
1.6552
0.1455
1.6821
0.6925
2.4556
4.9139
1.6082
8.9325
13.7485
970.5128
8.5189
83.8884
7.8743
3.1429
0.6274
12.5839
0. 1796
0.7527
0.7752
0.0434
18.7960
7.1617
2.5218
8.2583
1.9929
9.2957
9.6376
12.3000
106.0024
300.0022
9.2379
0.5590
9.2879
10.4000
10.5000
20.9999
4.5000
64.0002
45.0000
3600.0214
50. 5000
350.0021
38.8000
14.1000
2.6899
104.0024
0.6340
3.1039
3.4049
1.1649
63.0002
42.3000
10.5000
47.8000
6.9000
38.0000
41.6999
11. 9000
63.0002
120.0024
1.0149
0.0330
1.0209
8. 7000
5.0000
8.3000
0.0000
33. 0002
0. 0000
0.0000
14.3000
50.0002
8.5000
4.9000
1.0600
62.0002
0.0500
0.1170
0.6090
1.0109
0.0000
16.3000
1.5000
20.8000
0.6000
1.3000
5.8000
0.96
0.59
0.40
0.10
0.05
0.10
0.83
0.47
0.39
0.00
0.51
0.00
0.00
0.28
0. 14
0.21
0.34
0.39
0.59
0.07
0.03
0.17
0.86
0.00
0.38
0.14
0.43
0.08
0.03
0.13
0.
12.
24.
20.
100.
20.
7.
35.
36.
96.
14.
98.
237.
39.
47.
40.
34.
36.
13.
53.
61.
49.
4.
77.
28.
46.
26.
51.
63.
52.
83
-------
TABLE 24
LIST OF VARIABLES ELIMINATED OUE TO CORRELATION
FEED STREAM CHARACTERISTICS
Code
1. FD-PH Feed stream pH, dimensionless.
2. FD-TR Feed stream temperature, in degrees Fahrenheit.
3. FD-DI Feed stream dissolved calcium as CaCO3, in pounds per 1000
gallons.
4. FD-SU Feed stream suspended calcium as CaCOj, in pounds per 1000
gallons.
CARBCNATOR OPERATING VARIABLES
5. STIR Carbonator agitator speed, in revolutions per minute.
6. FLOW Densator recycle flow, as percent of feed.
7. SOLID Densator recycle solids, as percent of feed.
DENSATOR OPERATING VARIABLES
8. * Densator agitator speed, in revolutions.per minute.
9. FLOG Densator floculation time, in minutes.
10. RISE Densator rise rate, in gallons per square per minute.
11. TEMPD Densator temperature, in degrees Fahrenheit.
12. DE-DI Densator overflow dissolved calcium as CaCO,, in pounds per 1000
gallons.
13. DE-SU Densator overflow suspended calcium as CaCO , in pounds per 1000
gallons.
14. DE-TO Densator overflow total calcium as CaCO,, in pounds per 1000
gallons.
* Held constant.
84-
-------
TABLE 25
MEANS AND STANDARD DEVIATIONS OF SELECTED VARIABLES
Code
Mean
Deviation
High
Low
Ratio Dev/XBAR
COLOR
FD-TO
PH
DEPTH
RATE
CYCLE
CO2
TEMPC
TIMEC
SP. GR.
DAYS
CO-CA
CO-TO
CA-DI
CA-SU
CA-TO
172,6211
8.0549
9.3476
6. 9000
13.4071
1.6666
21.9285
179.8589
19.2714
1.0843
24. 2382
25.3714
30.6523
3.8761
14.5857
18.4619
43.1166
1.6821
0.6925
2.4556
4.9139
1.6082
8. 5085
82.9910
7. 8743
0. 0428
18.7960
7. 1002
8.2716
1.9929
9.2957
9.6376
300.0022
9.2879
10.4000
10. 5000
20.9999
4.5000
50. 5000
350.0021
38.8000
1.1649
63.0002
42.3000
47.8000
6.9000
38.0000
41.6999
120.0023
1.0209
8.7000
5.0000
8.3000
0.0000
14.3000
50.0002
8.5000
1.0109
0.0000
16.3000
20.8000
0.6000
1.3000
5. 8000
0.40
0.10
0.83
0.47
0.39
0.00
0.28
0.14
0.21
0.86
0.00
0.38
0.43
0.08
0.03
0.13
24.
20.
7.
35.
36.
96.
38.
46.
40.
3.
77.
27.
26.
51.
63.
52.
85
-------
Table 26,page 87, shows the simple correlation coefficients
between the variables. Note that nearly all these coefficients
are close to zero,indicating independence (orthogonality) of the
variables. Note also that some responses are correlated with
other responses,meaning that they characterize similar things.
This redundancy is illustrated in Figures 12 and 13 pages 88 and
89 and indicates that the system's performance can be
characterized by Total CO^ Transferred (reaction rate) and
Total Calcium loss (efficiency).
Mathematical Models of System
Heuristic Regression Analysis was used to obtain the mathemat-
ical models of the dependent responses. A residual study was
included for each model, questionable observations eliminated,
and additional models developed. The previously mentioned four
upset runs were eliminated and models developed. A study of
data indicates that even though the individual data were variable,
good and similar mathematical models were developed for the
responses characterizing the carbonation pilot plant's perfor-
mances during each run of both the twenty-one run study and
the seventeen run study. It is beyond the scope of this report
to explain the many statistical and computer concepts used.
Only the conclusions are discussed in the subsequent sections
and are primarily based on models using all the data.
A study of the resulting mathematical models indicates that
several variables thought to be important actually had little
or no effect on the system. Neither carbonation feed rate or
carbonator retention time had an influence on the system's
performance an indication that the pilot plant was not run
at full design rate. Feed stream total calcium and gas feed
CO, concentration had little or no effect on the system.
Lf
Reaction Rate Model
One of the major objectives of this pilot plant experimental
program was to establish the best design conditions for promoting
the reaction rate of CO£ transferred to CaCOj. A simple math-
ematical model was developed based on just two variables.
Figure 14,page 90, shows a direct reading response plot of the
effect of carbonation depth and Densator recycle rate on reaction
rate measured as Total CO? transferred. A high rate of 46
percent can be obtained at 10. 5 feet carbonation depth and a
86
-------
Code
TABLE 26
SIMPLE CORRELATION COEFFICIENTS
Color FD-TO PH Depth Rate Cycle CO2 TEMPC TIMEC SP. GR
COLOR
FD-TO
PH
DEPTH
RATE
CYCLE
CO2
TEMPC
TIMEC
SP. GR
DAYS
CO-CA
CO-TO
CA-DI
CA-SU
CA-TO
1.00
0.00
0.00
0.00
0.00
0.00
0. 00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.18
1. 00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Days
-0.05
.0.35
-0.03
0.38
0.43
-0.29
-0.18
0.37
0.08
0.09
1.00
0.00
0.00
0.00
0.00
0.00
0.10
0.06
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
CO-CA
-0.12
0.10
0.21
0.55
0.20
0.59
-0.30
-0.06
0.00
-0.18
0.30
1.00
0.00
0.00
0.00
0.00
-0.22
0.14
0.12
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
CO-TO
-0.19
0.05
0.12
0.66
0.17
0.57
-0.30
-0.03
0.09
-0.17
0.28
0.95
i.oo
0.00
0.00
0. 00
-0.10
-0.15
0.27
0.22
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
CA-DI
-0.01
0.00
0.40
0.42
0.47
-0.45
0.34
0.03
0.21
0.59
0.59
0.09
0.03
1.00
0.00
0.00
0.13
0.11
0.02
0.06
0.05
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0. 00
0.00
0.00
CA-SU
0.43
0.17
0.43
-0.04
0.00
0.38
0.13
-0.38
-0.14
-0.00
-0.20
0.13
0.08
0.06
1.00
0.00
0.01
0.02
0.46
-0.12
0.27
-0.38
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.14
-0.20
-0.15
0.13
0.28
-0.04
-0.40
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
-0.
0.
-0.
0.
-0.
-0.
-0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
02
20
03
57
50
31
19
10
00
00
00
00
00
00
00
00
-0.21
-0.23
0.16
0.22
-0.01
-0.38
0.38
-0.33
0.36
1.00
0.00
0.00
0.00
0.00
0.00
0.00
CA-TO
0.41
0.17
0.49
0.04
0.10
0.27
0.19
-0.36
-0.09
0.12
-0.07
9.15
0.09
0.27
0.97
1.00
87
-------
00
00
60-,
50-
40-
tl
m
73
m
z
n
rn
73
-n
m
73
73
m
Q
o
n
a
O
O
10
20 30 40
C02 TRANSFERRED TO CaC03/ %
FIGURE 12
REACH ON RATES
50
60
70
-------
oo
50
40
S-30
u
<
u
20
O
10
~7\
o
10
20 30
SUSPENDED CALCIUM LOSS, %
FIGURE 13
EFFICIENCY
40
-50
-40
-30
IS*
O
m
o
n
>
o
-20
10
50
-------
sO
O
2 3
DENSATOR RECYCLE RATE, GPM
FIGURE 14
REACTION RATE (TOTAL C02 TRANSFERRED)
-------
Densator recycle rate of 4. 5 gpm. A more economical design
from a capital investment view point would be 9. 0 feet carbona-
tion depth at a densator recycle rate of 6. 5 gpm. This condition
would give about 50 percent transfer.
System Efficiency Model
The carbonation-densator pilot plant efficiency and operating
economics were influenced by three variables. High feed
stream color, especially during the four runs in December 1969,
when the pulp mill was upset, caused high calcium losses.
Figure 15,page 92, shows a direct reading response plot of the
effect of carbonation gas feed temperature and carbonation pH
on the System Efficiency as measured by Total Calcium Loss.
A low loss (high efficiency) of 5 percent can be obtained at 8. 7
carbonation pH and a carbonation gas feed temperature of 325°F.
At a lower gas temperature of about 200 F. , the loss would rise
to about 13-14 percent.
D. Carbonate Slurry
During the pilot plant shakedown period, it was found that the
optimum setting for the Densator variable speed drive was for
a shaft rpm of 10. This gave a peripheral speed on the thickener
raRe of 200 ft per minute. Agitation in the flocculation zone was
also adequate at this speed to prevent settling in this area.
Blow-down times were set to give an underflow in the order of
0. 33 percent of feed. This gave a slurry in the order of 20
percent solids or 2 Ibs per gallon, and held a slurry depth of
approximately 18 inches in the Densator.
The above settings were somewhat critical in that with higher
settings the slurry bed would rise above the Densator down draft
tube, carryover would increase and slurry underflow concentrat-
ions would decrease. With lower settings, the slurry blowdown
would be too highly concentrated to flow by gravity.
Filtration test on the carbonate slurry gave filtration rates
comparable to lime mud in the mill causticizing system.
Figure 16, page 93, compares filtration rates vs percent solids.
The carbonate solids have a distinct tannish color. However, the
carbonate purity averaged 99.49 percent calcium carbonate on
two samples collected during different weeks.
91
-------
xO
to
9.5 10.0
CARBONATION PH
FIGURE 15
SYSTEM EFFICIENCY (TOTAL CALCIUM LOSS)
10.5
11.0
-------
U>
720
640
560
480
•^400
Of.
CD
320
240
160
80
NOTE: VALUES HAVE BEEN MULTIPLIED
BY FACTOR OF 0.8.
10
20
30
% SOLIDS FEED
40
50
FIGURE 16
EIMCO FILTER LEAF TESTS DENISATOR SLUDGE
-------
Based on the filterability and purity, there would be no problem
of calcining the carbonator solids in the mill lime kiln.
E. Instrumentation Evaluation
The pilot plant instrumentation performed satisfactorily after
adjustment and final tune-ups with exception of the turbidity
measurement device. In this one exception, the instruments were
well designed. The unsatisfactory performance was due to
misapplication. If the horizontal flow through reference cell
could be replaced with one of perpendicular flow through design,
the problem of calcium carbonate disposition would be eliminated.
The instrument could then be used in controlling the carbonate
slurry level in the clarification zone of the system as well
record the quality of the treated waste.
With the above modification, all of the instruments and controls
can be used with excellent results in a full scale carbonation
plant.
F. Summarization of Carbonation Process
The pilot plant adequately served its purpose in that it made it
possible to study the many variables of the carbonation process
under actual mill operating conditions. Flows were such that
scale up to handle the total mill -waste can be made on an
economical design to meet the full range of mill demands.
The operating problems and modifications experienced in the
startup were realistic in that they simulated what could have
been experienced with a full scale plant without the benefit of
the prior pilot runs.
There was a close agreement with other investigators of the
carbonation process. However, the computer analysis
emphasized the importance of keeping the formation of colloidal
calcium carbonate to a minimum. The five variables and ;
efficiency all pertained to the physical and chemical reaction
process of solid calcium formation. Variables such as carbona-
tion retention time, Densator flocculation time and rise rates
which were thought to be of major consideration -were found to
have no influence on the process within the operating range of the
pilot plant.
94
-------
The calcium carbonate product is readily concentrated and has
filtration characteristics comparable to mill causticizing mud.
As it has a high purity, there are not foreseeable problems in
its reuse in the mill system.
Some "do's" and "don't's" which are based on operating
experience are:
1. Maintain calcium carbonate solids concentration in the
carbonation zone in the range of five to six thousand parts
per million pounds of total treated effluent. This may be
conservative as this was the solids concentration at which
the ratio of solids to unreacted gases gave a minimum
foaming tendency in the pilot plant test runs.
2. Do not use orifice type diffusers as openings up to 3/8 inch
will plug within a few days. There may also be problems
with turbine type diffusers as calcium deposits could cause
imbalance of the turbine blades. The most practical type of
diffusers appears to be of a slotted design similar to the one
used in the pilot plant. In any case provisions should be
made for cleaning and servicing without having to shut the
plant down.
3. Instrumentation for turbidity measurement should have a
perpendicular flow through reference cell with provisions for
intermittent cleaning.
4. With recycle to the carbonator, a combined carbonation and
clarification unit would have a definite advantage over
separate units in that one mechanical rake for solids removal
would serve both steps in the process.
95
-------
SECTION VII
ACKNOWLEDGMENT
The color removal system was developed, and designed as an integral
part of the tertiary treatment process for kraft mill effluent by
Mr. W. J. Verross, Vice President and General Manager of Inter-
state Paper Corporation and Mr. J. G. Vamvakias of The Rust
Engineering Company, Pittsburgh, Pa. It was through their efforts
and the cooperation of Mr. R. S. Howard, Director of the Georgia
State Water Quality Control Board, and Mr. Charles H. Starling,
Director of Industrial Waste Service, that the treatment system
became a reality.
The assistance of Mr. E. P. Lomasney, Project Officer of the Federal
Water Quality Administration, Atlanta, Georgia, and the guidance given
by Mr. George R. Webster, Federal Water Quality Administration,
Washington, D. C. , has contributed materially to the successful
operation of the color removal system.
The very necessary environmental guidelines have been provided by
Mr. H. B. Counts of the U. S. Geological Survey and his staff and
Mr. M. D. Dahlberg and co-workers of the University of Georgia
Marine Institute. Their findings in their follow up investigations have
provided a much needed encouragement.
The carbonation pilot plant study was a joint project of Interstate Paper
Corporation and Continental Can Company. Mr. R. L. Scoville,
Manager Special Projects, Continental Can Company,'provided the
necessary manpower from his technical staff and contributed materially
in his close association with the project.
Mr. Floyd A. Miller, of Glidden-Durkee Division of SCM Corporation,
assisted in setting up the statistically designed program and provided
the computer analyses of the operating variables and responses
monitored during the program. His report of the computer analyses
is incorporated in this report as submitted.
The program was under the direction and supervision of
Charles L. Davis, Jr. , Pollution Control Director, Interstate Paper
Corporation. The responsibility for the program has been with
Mr. Edgar L. Hart, Jr., Utilities Manager, Interstate Paper
Corporation. His support and administrative assistance provided the
necessary team effort.
97
-------
SECTION VIII
REFERENCES
1. Dahlberg, Michael, "Inventory of a Natural Estuarine Fish
Community on the Georgia Coast", presented at TAPPI
Air & Water Conference, Jacksonville, Fla. , April 1969.
2. Dahlberg, Michael, "Annual Cycle of Diversity in Georgia
Estuarine Fish Populations", presented at TAPPI Air &
Water Conference, Jacksonville, Fla., April 1969.
3. Dahlberg, Michael, "Aspects of the Hydrography of a Georgia
Estuary", presented at TAPPI Air & Water Conference,
Jacksonville, Fla., April 1969.
4. Southern Pulp and Paper Manufacturer, May 10, 1968.
5. Pulp & Paper, July 1, 1968, Page 25.
6. Institute of Paper Chemistry, Appleton, Wisconsin, Grant
12040 DKD.
7. Davis, Charles L. (Jr.), "Tertiary Treatment of Kraft Mill
Effluent Including Chemical Coagulation for Color Removal"
presented at the TAPPI Air & Water Conference, April 30,
1969, TAPPI Volume #52, Page 2132.
8. Davis, Charles L. (Jr. ), "Lime Precipitation for Color Removal
in Tertiary Treatment of Kraft Mill Effluent at the Interstate
Paper Corporation, Riceboro, Georgia11, presented at the
American Institute of Chemical Engineers Chemical Engineering
Progress Symposium Series 1971, No. 107, Vol. 67.
9. National Council for Stream and Air Improvement, Inc., "A
Process for Removal of Color from Bleached Kraft Effluents
Through Modification of the Chemical Recovery System",
Technical Bulletin No. 157, June 1962.
10. U. S. Department of the Interior Federal Water Pollution
Control Administration, Cincinnati, Ohio, September 1969.
"Mathematical Model of Tertiary Treatment by Lime Addition",
Report No. TWRC-14.
99
-------
11. Scott, Keith, J. , "Thickening of Calcium Carbonate Slurries",
Ind. Eng. Chem. Fun. 3_ 484 (1968).
12. Haney, Paul D. & Hanann, Carl L., "Recar Donation and Liquid
Carbon", Paper presented at the AWWA Annual Conference,
San Diego, Calif. , (May 1969>
K
13. Eaton, C. D. & Martin, J. H., "New Developments in
Sedimentation Units for Water Clarification and Softening",
Paper presented at TAPPI 3rd Water Conference, Mobile, Ala.
(April 1966).
14. National Council for Stream and Air Improvement, Inc. ,
"Measurement Control and Changes in Foaming Characteristics
of Pulping Waste During Biological Treatment", Technical
Bulletin No. 195.
15. Miller, F. A., "Users Manual of the Data Analysis System"
Corporate Systems Department, The Glidden Company
(September 1967).
16. Miller, F. A., "Strengthening Stepwise Regression11,
Association for Computing Machinery Conference, Palm Beach,
Fla., (June 1965).
17. Miller, F. A., "Huristic Regression Analysis", Operations
Research Society of American Conference, Houston, Texas
(November 1965).
18. Miller, F. A., "Improving Heuristic Regression Analysis",
Association for Computing Machinery Conference, Chapel Hill,
N. C., (June 1967).
19. The Eimco Corporation, "Eimco Filter Test Leaf Kit-Instruction
Booklet".
100
-------
SECTION IX
APPENDIX
101
-------
TABLE I
COLOR REDUCTION VS. LIME FEED
JUNE 27, 1968 - SEPTEMBER 30, 1968
Untreated
Waste
Color Range
Av. Color
O
lif
U
1
P.
•v
a/
V
I
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
Treated color
Av
Max.
Min.
Average Color Reduction (ppm) per 100 ppm Ca(OH)2
200-500
423
(1) 15.3
(1) 7.4
(5) 20.0
(1) 17.4
(1) 21.4
(1) 18.9
(1) 19.3
(11) 72
150
40
600-1000
795
(1) 49.8
(2) 54.7
(7) 41.1
(6) 39.1
(16) 39.2
(10) 39.6
(3) 31.4
(3) 39.0
(2) 30.8
(1) 21.8
(51> 86
125
50
1100-1500
1320
(2) 79.8
(3) 68.3
(1) 61.3
(3) 69.7
(1) 44.4
(10) 118
200
60
1600-2000
18-00
(1) 111.3
(1) 81.5
(1) 83.2
(1) 72.0
(4) 133
150
100
2100-2500
3,000
3,000
(1) 192.7
(1) 151.4
(2) 108
125
90
7,000
(1) 311.9
(1) 450
450
450
8,000
(1)396.0
(1) 80
80
80
(-) Number of days.
-------
TABLE 2
COLOR REDUCTION VS. LIME CONCENTRATION
OCTOBER 1, 1968 - OCTOBER 31, 1968
Untreated
Waste
Color Ra nge
Av. Color
Lime Concentration ppm Ca(OH)£
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
Treated Coloi
Av.
Max.
Min.
Average Color Reduction (ppm) Per 100 ppm Ca(OH)2
200-500
400
(1) 17.6
(1) 50
50
50
600-1000
796
(2) 35.6
(5) 40.9
(2) 35.0
(2) 29.7
(1) 41.6
(1) 29.1
(1) 29.6
(14) 87
200
50
1100-1500
1225
(1) 62.8
(1) 53.4
(2) 46.6
(4) 114
150
75
1600-2000
2000
(1) 75.4
(1) 150
150
150
2100-2500
3000
(1) 121.5
(1) 106.6
(2) 200
250
150
4000
(1) 145.3
(1) 300
300
300
5000
(1) 232.0
(1) 300
300
300
o
W
en
H
O
H
-------
TABLE 3
COLOR REDUCTION VS. LIME CONCENTRATION
November 1, 1968 - November 30, 1968
Untreated
Waste
Color Range
Av. Color
N
§
0
t
P<
Pi
o
*rt
a
*•<
Lime Concent
1200
1300
1400
1500
1600
1700
1800
Treated Color
Av.
Max.
Min.
Average Color Reduction (ppm) Per 100 ppm Ca(OH)
200-500
388
(1) 32.0
(2) 22.1
(3)
77
90
70
600-1000
764
(1) 57.6
(5) 52.1
(3) 46.5
(5) 41.3
(5) 41.8
(1) 40.8
(1) 32.8
(21)
104
200
50
1200
(1) 65.7
(1)
150
150
150
1500
(3) 93.1
(3)
108
150
75
3000
(1) 215.4
(1)
200
200
200
6000
(1) 372.8
(1)
125
125
125
-------
TABLE 4
CCLOR REDUCTICN VS. LIME CONCENTRATION
DECEMBER 1, 1968 - DECEMBER 31, 1968
Untreated
Waste
Color Range
Av. Color
03
•j
o
rt
0
g
&
a
o
rt
M
8
u
A
o
O
0)
I
J100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
Treated Color
Av.
Max.
Min.
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca (OH)2
200-500
500
(I) 32.9
(1) 75
600-1000
818
(1)80.0
(2)45.1
(2)42.8
(4)44.2
-
(1)33.6
(1)42.8
(11) 135
250
75
1100-1500
1261
(1) 78.4
(2) 69.1
(2) 68.0
(1) 66.8
(2) 61.1
(8) 158
250
80
1600-2000
1890
(1) 72.5
(2) 91.5
(2) 77.5
(1) 71.7
(6) 178
250
90
2750
2750
(1) 197.7
(1) 200
200
200
o
(J\
( ) - Number of days
-------
TABLE 5
COLOR REDUCTION VS. LIME CONCENTRATION
JANUARY 1, 1969 - JANUARY 31, 1969
Untreated
Waste
Color Range
Av. Color
PJ
K
9-
n)
U
S
CL,
fi
O
• H
la
1
u
a
o
U
1
800
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
Treated Color
Av.
Max.
Min.
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca(OH)
600-1000
820
(2) 57.2
(2) 50. 5
(1) 42.1
(5) 94
125
75
600-1000
880
(1) 58.6
(2) 66.5
(1) 56.8
(1) 53.5
(2) 47.3
(1) 40.4
Modifi
(8) 127
190
100
1100-1500
1300
(1) 82.1
(3) 58.5
(1) 56.1
sd Lime Fe
(5) 138
220
90
1600-2000
1890
(1) 112.9
(1) 85.9
=>d System
(2) 188
200
175
2100-2500
2270
(1) 159.3
(1) 107.9
(2) 155
180
130
2860
(1) 165.1
(1) 200
3600
(1) 273.8
(1) 150
( ) Number of days
-------
TABLE 6
COLOR REDUCTION VS. LIME CONCENTRATION
FEBRUARY 1, 1969 - FEBRUARY 28. 1969
Untreated
Waste
Color Range
Av. Color
^pq
O
*tt
O
6
S
n
o
• H
j»
•a
V
o
c
o
O
2
a
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
Av.
Max.
Min.
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca(OH)2
600-1000
920
(2) 66.7
(1) 51.4
(1) 47.3
(1) 46.6
(5) 141
200
80
1100-1500
1275
(1) 104.4
(1) 72,7
(1) 62.6
(1) 60.7
(4) 143
180
90
1600-2000
1914
(1) 176.0
(1) 113.5
(1) 118.8
(1) 133.8
(2) 115.1
(1) 70.8
(7) 191
300
100
2100-2500
2313
(1) 238.8
(1) 177.4
(1) 141.3
(1) 104.4
(4) 193
250
170
5000
(1) 227.1
(1) 200
200
200
6800
(1) 389.2
(1) 125
125
125
o
-vj
( ) Number of days.
-------
TABLE 7
COLOR REDUCTION VS. LIME CONCENTRATION
March 1, 1969 - March 31, 196'9
Untreated
Waste
Color Range
Av. Color
' 5s
0
a
U
Qj
8
g
o
• r-4
-M
rt
h
Lime Concent
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
Av.
Max.
Min.
Average Color Reduction (ppm) Per 100 ppm Ca(OH)2
200-600
410
li \
I1' 13.2
t1' 15.8
22.4
-------
TABLE £
COLOR REDUCTION VS. LIME CONCENTRATION
APRIL 1, 1969 - APRIL 30, 1969
Untreated
Waste
Color Range
Av. Color
Lime Concentration ppm Ca(OH)2
800
900
1000
1100
1200
1300
1400
1500
1600
1700
Treated Waste
Color
Av.
Max.
Min.
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca(OH)2
200-600
389
(2) 19.9
(2) 33.8
(1) 23.0
(1) 17.7
(1) 7.0
(1) 30.1
(8) 65
140
30
600-1000
839
(1) 1CF9. 3
(1) 63.8
(3) 66.3
(2) 52.4
(1) 51.8
(2) 39.0
(10) 81
120
40
1000-1600
1344
(1) 123.7
(1) 114.4
(1) 107.0
(2) 87.0
(2) 86.1
(7) 123
180
60
1600-2000
1707
(1) 108. 1
(1) 99.9
(1) 104.6
(3) 160
200
100
2280
2280
1} 134.0
(1) 80
80
80
4200
4200
(1) '311.5
(1) 160
160
160
o
vO
( ) - No. Days
-------
COLOR REDUCTION VS. LIME CONCENTRATION
Mayl, 1969 -May 31, 1969
Untreated
Waste
Color Range
Av. Colo r
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
Treated Waste
Color
Av.
Max.
Min.
Average Color Reduction (ppm) Per 100 ppm Ca(OH)
t*
200
(1) 6.2
(2) 8.4
(3)
50
60
~- 40
200-600
(1) 23.0
(2) 26.8
(1) 18.2
(4) 75
100
50
600-1000
(2) 68. 8
(3) 63.4
(1) 75.9
(1) 75.8
(3) 47.3
(2) 40.8
(2) 49.1
;
(2) 29.0
(16)
92
150
50
1000-1400
(2) 87.2
(2) 75.3
(2) 75.2
(1) 60.3
(7)
96
120
60
o
( ) number of days.
-------
TABLE ln
COLOR REDUCTION VS LIME CONCENTRATION
JUNE 1, 1969 - JUNE 30, 1969
Untreated
Waste
Color Range
Av. Color
CO
£
o
~ctf
o
1
A
c
0
-i-t
-*->
«J
M
4->
A
u
o
fi
o
(J>
V
£
•r-i
J
1000
1100
1200
1300
1400
1500
1600
1700
1800
Treated Waste
Color % Red.
Av.
Max.
Mill*
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca(OH)2
200-600
489
(2) 41.3
(1) 32.4
(1) 35.7
(2) 22. 8
(2) 21.5
(1) 25.4
81.5
(9) 91
100
65
600-1000
937
(1) 78.9
(1) 73.7
(2) 78.2
(2) 65.1
(1) 51.0
90.5
89
110
60
1000-140C
1180
(1) 91.6
(4) 73.3
(1) 76.1
90.2
(6) 115
150
90
1400-1800
1607
(2) 117.2
(1) 123.7
94.4
(3) 90
100
80
2400-2500
2450
(2) 169.4
93.5
(2) 160
170
160
3000-3200
3067
(1) 262.6
(1) 232.8
(1) 166.8
94. 2
(3) 175"
200
150
j
' j
j
j
*
j
ii
i
(-) No. days
-------
TABLE 11
COLOR REDUCTION VS LIME CONCENTRATION
JULY 1, 1969 - JULY 31, 1969
Untreated
Waste
Color range
Av. Color
Lime Concentration ppm Ca(OH)2
700
800
900
iuOu
1100
1200
1300
1400
1500
1600
1700
1800
Treated Waste
Color % Red.
Av.
Max.
Min.
AVERAGE COLOR REDUCTION (ppm) PER 100 ppm Ca(OH)2
600-1000
853
(1) 84.7
!2) 59.8
2) 58.2
2) 50.6
;i) 54.7
87.6
8) 105
108
89
1000-1400
1146
1) 127.5
2) 89.8
1) 99.2
3) 76.0
1) 65.4
1) 54.7
1) 54.4
89.1
(10)125
180
85
1400- 1800
1603
(1) 207.8
(1) 1OU. 4
1) 117.4
(1) 118.8
(1) 84.5
(1) 91.5
90. 7
(6) 148
180
110
2200
(1 ) 136. 3
93.2
(1) 150
150
150
6000
(1)454. 0
97.1
(1)175
175
175
<-) No. days
-------
TABLE 12
COLOR REDUCTION VS. LIME CONCENTRATION
August 1, 1969 - August 31, 1969
Untreated
Waste
Color Range
Av. Color
X
O .
o
g
P
p<
1
.*->
a
(-) No. Days
* Diluted by surface run-off from heavy rains.
Note - Both untreated & treated waste colors reported at pH 7. 6
-------
TABLE 13
COLOR REDUCTION VS.. LIME CONCENTRATION
September 1, 1969 - September 30, 1969
Untreated
Waste
Color Range
Av. Color
(M
ffi
O
O
8
a
o
rt
•g
o
a
0
O
4)
•*>4
J
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
Treated Waste
Color % Red,
Av.
Max.
Min.
Average Color Reduction ppm per 100 ppin Ca(OH)
1000-1400
1216
(1) 135.5
(1) 76.4
(1) 70.0
(2) 79.2
(3) 81.0
(2) 63.4
(1) 54.3
(11)
89.3
126
190
90
1400-1800
1636
(1)161.7
(6)118.4
(1)103.7
(8)
91.1
146
180
110
1800-2200
1993
(2) 145.5
(1) 71.4
(3)
91.8
157
240
100
1000-2200
1475
(22)
90.4
138
240
90
(-) No. of days
Note - Untreated waste color unadjusted pH 12.3
Treated waste color adjusted pH 7. 6
-------
TABLE 14
COLOR REDUCTION VS. LIME CONCENTRATION
October 1, 1969 - October 31, 1969
Untreated
Waste
Color Range
Av. Color
O
"
-------
TABLE 15
COLOR REDUCTION VS. LIME CONCENTRATION
November 1, 1969 - November 30, 1969
Untreated
Waste
Color Range
Av. Color
5J
8
It
o
p.
a
O
4*
S
o
a
0
u
1000
1100
1200
1300
1400
1500
1600
Treated Waste
Color % Red.
Av.
Max.
Average Color Reduction (ppm) Per 100 ppm Ca(OH)
1000-1400
1214
(2) 112.1
(2) 96.2
(2) 87.9
(4) 80. 0
(10)
87.2
149
200
120
1400-1800
1534
(3) 147.2
(3) 120.2
(4) 113.9
(1) 113.0
(1) 93.8
(1) 91.0
(13)
90.4
147
190
110
1800-2200
2000
(6) 159.0
(2) 140. 5
(1) 124.1
(6)
90.2
196
240
160
1000-2200
1520
(29)
89.3
160
240
110
(-) No days Note untreated waste color at unadjusted pH 12.2
Treated waste color at adjusted pH 7. 6
-------
TABLEI6
COLOR REDUCTION VS. LIME CONCENTRATION
December 1, 1969 - December 31. 1969
Untreated
Waste
Color Range
Av. Color
JM
1
id
O
g
a
a
o
• |H
4J
ti
u
•a
4>
o
B
o
O
.1
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
Treated Waste
Color % Red.
Av.
Max.
Min.
Average Color Reduction (ppm) per 100 ppm Ca(OH)2
1000-1400
1195
(1)124.3
(1) 89.8
(2) 85.8
(1) 35.7
86.1
(5)170
200
130
1400-1800
1766
(1)120.7
(5)135.6
(1)116.4
(1)114.2
_
(1) 75.9
90.2
(9)174
200
110
1800-2200
2050
(1)189.9
(1)167.6
(1)148.1
(1)151.8
91.4
(4)175
220
160
2200-2600
2467
(2)179.4
(1)183.6
87.5
(3)253
300
200
3000
3000
(1)188.3
90.0
(1)300
300
300
1000-3000
1849
89.1
(22) 190
300
110
(-) No days
Note untreated waste color at unadjusted pH 12. 2
Treated waste color at adjusted pH 7. 6
-------
Accfxxion Number
Subject [-it-Id & Croup
SIC 056
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Interstate Paper Corporation
Riceboro, Georgia
6.
Title
Color Removal From Kraft Pulping Effluent by Lime Addition
10
— ....
Authorfs)
Davis, Charles
L.
16
21
Project Designation
Program #12040 ENC
Grant #WPRD 183-01-68
Note
22
Cita tion
23
Descriptors (Starred First)
Color-Removal, Lime Precipitation, Tertiary Treatment, Waste Water
Treatment
25
Identifiers (Starred First)
Color Removal Process, Waste Water Quality.
27
Abstract A prototype color removal system, was designed, constructed and operated as
an integral part of a tertiary treatment system for total process effluent from a kraft
linerboa-rd -mill. T-he ba-sie system includes a lime-precipitation process for the removal
of color combined with primary clarification foiled by'natural biocliemical lake stabilization
and mechanical aeration.
Operating results show that" the color removal system can ope-rate successfully under
widely varying conditions to give a relatively constant effluent color in the range of 125
ppm APHA color units at treatment levels of 1000 (+ 50) ppm of calcium hydroxide with
untreated effluent colors in the range of 1200 (±200) ppm. Treatment at this level
reduces-lime-cost to-$53.73 per million-gallons with lime-a±_$l5..35/ton._(9JD% CaO)_
Performance is directly related to control of lime feed. Equipment evaluation indicates
substantial savings in capital cost for future installations.
Recovery of calcium used was carried out under mill conditions on a continuous
basis following a statistically designed program. Results and full size design factors
are given.
Performance of natural biochemical stabilization following lime treatment is shown
graphically. Overall BODi; reduction for the tertiary treatment system is 98% with a.
.final
Abstra
discharge averaee BODt; of 6 room.
Charles L. Davis
Institution
Interstate
Pape
r
Corporation
WR:tOZ (REV. JULY 1969)
WRSIC
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D, C. 2O240
* £»0: >9S»-S5»-S»»
------- |