WATER POLLUTION CONTROL RESEARCH SERIES • 12090ESG01/71
Zinc Precipitation and Recovery
From Viscose Rayon Waste Water
ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFICE
-------
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, develop-
ment, and demonstration activities in the Water Quality
Office, 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 Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Washington,
D.C. 20242.
-------
ZINC PRECIPITATION AND RECOVERY
FROM VISCOSE RAYON WASTE WATER
by
American Enka Company
Central Engineering Department
Enka, North Carolina
for the
ENVIRONMENTAL PROTECTION AGENCY
WATER QUALITY OFFICE
Project No. 12090 ESG
January 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1
-------
EPA REVIEW NOTICE
This report has been reviewed by the Water
Quality Office, 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 recom-
mendation for use.
ii
-------
ABSTRACT
In May, 1968, the Industrial Pollution Control Branch of the Water
Quality Office/Environmental Protection Agency, initiated a research
and development grant with American Enka Company to perfect an improved
process for the precipitation and recovery of soluble zinc in rayon
manufacturing wastewaters.
In the production of viscose rayon, zinc sulfate is used as a com-
ponent of the acid spinning bath. Zinc is lost in a dilute form
when the acid spun yarns are washed with water and at various points
in the spinning bath system. The novel zinc recovery system in-
volves the initial neutralization of the waste stream to pH 6.0,
sedimentation of insolubles, the crystallization of zinc hydroxide
in a high pH environment, the sedimentation of zinc hydroxide and
the solubilization of the zinc with sulfuric acid.
This novel recovery system was operated at a 600 - 1000 gpm rate
with 70 - 120 mg/1 of Zn in the feedwater. The system can maintain
an effluent concentration of Zn less than 1 mg/1, which corresponds
to 98 - 99% removal efficiency. The unique zinc hydroxide sludge
is easily concentrated to 5 - 7% solids by sedimentation and to 107«
solids by centrifugation. The sludge particles obtained by this
process are spheroids of 4 - 8 microns average diameter.
A recovery of 2,000 pounds of zinc daily assures recovery of the
12.5 to 14.0 cents/lb. of Zn operating and maintenance costs. The
cost of zinc oxide purchased by Enka amounts to 15.6 cents/lb. of
equivalent Zn.
This recovery plant was awarded a Finalist prize for achievement in
water pollution control in the 1970 Gold Medal Awards, single-plant
category, by The Sports Foundation, Inc.
This report was submitted in fulfillment of Grant Project 12090-ESG
between the Water Quality Office/Environmental Protection Agency and
American Enka Company.
Keywords: Industrial wastes, textile fibers, heavy metals, chemical
precipitation, flocculation, operating cost
111
-------
CONTENTS
SECTION;
I
II
III
IV
V
VI
VII
VIII
IX
X
XI
XII
PAGE:
Conclusions 1
Recommendations 3
Introduction 5
Purpose and Objectives 5
Background Information 5
Historical Review 9
Description of Pilot Plant Studies 13
Flow Diagrams and Description of the
Recovery Plant 17
Discussion of Operational Problems and
Results 27
Zinc Recovery Efficiency 27
Effluent Water Quality 27
Reuse of Zinc 29
Soluble Zinc Concentration vs. Solution pH 32
Use of Coagulant Aid in the Clarifier 33
Lime System and pH Control 34
Scanning Electron Microscope Studies 41
Costs 51
Capital Investment 51
Operational Costs 52
Acknowledgements 59
References 61
Appendices 65
Detailed List and Description of Main
Items of Equipment 65
iv
-------
PAGE:
XII Appendices (continued) 65
Electrical Power Required for Motors
and Heaters 75
Photographs of Recovery Plant 78
Miscellaneous Data
-------
FIGURES
Figure: PAGE:
1 - Densator Pilot Unit 14
2 - Zinc Recovery Flow Diagram 20
3 - Alkali Supply and pH Control System 21
4 - Lime Slurry Manutacture and Supply System 22
5 - Typical Titration Curve of Acid Waste 23
6 - pH and Soluble Zinc in Hominy Creek During
Recovery of 50% of Zinc Consumed 30
7 - Typical Graph of Plant Feed pH 38
8 - Typical Graph of pH After Lime Neutralization While
Using Complete Feedforward System 39
9 - Typical Graph of pH After Lime Neutralization While
Using Modified pH Control System 40
10 - Electron Micrograph of Plant Sludge (1400 X; 43
11 - Electron Micrograph of Plant Sludge (4buO X) 43
12 - Electron Micrograph of Plant Sludge (5500 X) 44
13 - Electron Micrograph of Laboratory Sludge After 10
Precipitations (500 X) 44
14 - Electron Micrograph of Laboratory Sludge After 10
Precipitations (1075 X) 45
15 - Electron Micrograph of Laboratory Sludge After 10
Precipitations (5400 X) 45
16 - Electron Micrograph of Laboratory Jiudge After 30
Precipitations (550 X) 46
17 - Electron Micrograph of Laboratory Sludge After 30
Precipitations (1100 X) 46
18 - Electron Micrograph of Laboratory Sludge After 30
Precipitations ^200 X) 47
19 - Electron Micrograph of Laboratory Sludge After 60
Precipitations (500 X) 47
VI
-------
PAGE;
20 - Electron Micrograph of Laboratory Sludge After 60
Precipitations (1180 X) 48
21 - Electron Micrograph of Laboratory Sludge After 60
Precipitations (5850 X) 48
22 - Electron Micrograph of Plant Zinc Precipitated Once
and Aged (5200 X) 49
23 - Electron Micrograph of Plant Zinc Precipitated Once
and Aged (10,750 X) 49
24 - Electron Micrograph of Plant Zinc Precipitated Once
and Aged (22,000 X) 50
25 - Zinc Recovery Capaco-cy Versus Acid to Zinc Sulfate
Ratios for Recovery of Cost0 58
26 - View of Acid Feed PuT"ps 78
27 - Air View of Waste Treatment Area 79
28 - View of Neutralization Tank, '"'•larifier and Densator 80
29 - View of Densator from Walkway 81
30 - View of Settling and Dissolving Tanks 82
31 - Interior View of Pump House with Sludge Recirculation
Pumps 83
32 - View of Lime Storage and Slaking Area 84
VII
-------
TABLES
TABLE: PAGE:
1 - Characteristics of Plant Flows 18
2 - Laboratory Analyses of Plant pH, Zinc and Magnesium 28
3 - Soluble Zinc Versus pH in Absence of Sludge 32
4 - Soluble Zinc Versus pH in Presence of Sludge 32
5 - Soluble Zinc Versus pH in the Clarifier 33
6 - Type of Alkali Feed Control Valves 36
7 - Maximum Acid to Zinc Sulfate Ratio at Various Recovery
Capacities to Pay for Operating and Maintenance Costs 54
8 - Maximum Acid to Zinc Sulfate Ratio at Various Recovery
Capacities to Pay for Operating, Maintenance and
Amortization Costs 56
9 - Maximum Acid to Zinc Sulfate Ratio at Various Recovery
Capacities to Pay for Operating}Maintenance and
Amortization Costs if Cheaper Lime is Available 56
10 - Maximum Acid to Zinc Sulfate Ratio at Various Recovery
Capacities to Pay for Operating, Maintenance and
Amortization Costs if Zinc Concentration in Feed is
Doubled 57
Vlll
-------
SECTION I
CONCLUSIONS
The following conclusions can be drawn from this study, which covers
work done in a laboratory, pilot plant and industrial scale over a period
of several years:
1. The efficient and economical recovery and reuse of waste soluble
zinc from viscose rayon plant wastes has been proven under indus-
trial conditions.
2. The efficiency of the precipitation and recovery of soluble zinc
exceeds 95%, which is higher than that of other known methods
such as ion exchange.
3. Dense, easily handled sludges, which are susceptible to simple dis-
posal as a solid, or to recovery for reuse, can be obtained by pre-
cipitation of soluble metals as hydroxides if the operation is
carried out under controlled conditions.
4. The effluent water from the recovery plant is extremely clear water
of constant, slightly alkaline pH and it is essentially free of
toxic metal content. It can be reused to advantage where its high
hardness and salt content is not an impediment, such as water for
flushing and cleaning purposes, pump seal water, etc. Its soluble
zinc content is less than 2 ppm.
5. When lime is used for neutralization, the value of the recovered
zinc easily pays for the operating costs. Depending mainly on
the zinc recovery capacity of the plant, the acid to zinc ratio
in the feed, and the price of lime, the value of the recovered
zinc may also pay part or all of the capital investment costs.
6. Under American Enka Company's plant conditions, it is easy to
obtain a zinc hydroxide concentration of 5 to 7% by settling alone,
and without the use of coagulant aids in the precipitation reactor.
Using centrifugation, a non-flowing solid exceeding 107» zinc hy-
droxide can be obtained. The presence of cellulose floe affects
the settleability and density of the sludge.
7. The particles of the unique dense zinc hydroxide sludge appear
in scanning electron micrographs as spheroidal particles with
a diameter of 4 - 8 microns. They are formed only as a result
of very repeated adsorption or precipitation of soluble metal on
existing hydroxide particles, followed by pH adjustment with
alkali.
8. Addition to the acid wastes of various dilute alkaline wastes
effects a saving in lime consumption,.
-------
9. There is technology already available in the electrolytic zinc in-
dustry to remove impurities which may be found undesirable in the
recovered acid zinc sulfate.
10. The operating and maintenance costs for recovery of the soluble
waste zinc depend on the sulfuric acid/zinc sulfate ratio in the
waste and on the amount of zinc recovered daily. When recovering
2000 pounds Zn daily from a waste with a ratio of 5 to 6, the
operating and maintenance costs are 12.5 to 14.0 cents/lb. Zn.
The cost of purchased zinc oxide is 15.6 cents/lb. of equivalent
Zn.
11. The zinc recovery plant was awarded a Finalist prize in the 1970
Gold Medal Awards, single-plant category, by the Sports Foundation,
Inc. These Awards are presented for achievement in the field of
water pollution control and related water conservation or development.
-------
SECTION II
RECOMMENDATIONS
1. It is advantageous, both economically and process-wise, to collect
the waste flows at their source in a concentration as high as possible.
2. The neutralization plant should be designed to use the cheapest
lime available that is technically acceptable under conditions
of reuse.
3. The clear water t^xuent from the recovery plant should be reused
or otherwise mixed with other plant effluents so that the pH 9.5 -
10 will be reduced before discharge to the stream. If the other
effluents are not acid enough to reduce the pH, the effluent coulu
be contacted with waste combustion gases, such as those generated
by a steam boiler, or by any other acid material.
4. The concentration of the wastes fed to the neutralization plant
should be equalized as much as possible, to aid pH control. An
expensive feedforward pH control system should not be necessary
and may actually create more problems than it solves.
5. All operations should be made as automatic as possible, to save
expense.
6. The clarifier underflow should be thickened sufficiently so that
the impurities can be disposed of as a solid waste.
7. Although apparently simple, a lime slaking system can be a source
of many problems, and therefore, it should be designed carefully
by using all available information from established lime consumers.
8. It appears advantageous to concentrate the hydroxide sludge by using
centrifugation, so as to reduce the soluble calcium and excess water
in the recovered zinc sulfate.
9. The basic principles of tuc precipitation process should be appli-
cable to waste treatment problems other than pollution by soluble
metals. For example, dense iron, aluminum, calcium, and other
sludges could be used to remove suspended impurities, including
dispersed oil particles, or colored compounds in various wastes
from textile, cellulose pulp, and other industries. Previous
attempts to develop this type of process have been frequently im-
peded by the difficult sludges which were obtained.
If waste acid mine waters were treated in a similar fashion, not
only toxic soluble metals could be eliminated but also excessive
sulfate ion could be reduced by gypsum precipitation.
-------
SECTION III
INTRODUCTION
PURPOSE AND OBJECTIVES
The purpose of this project, "Zinc Precipitation and Recovery from
Viscose Rayon Wastewater" was to perfect a process for removing soluble
zinc from industrial effluents in order to protect fish and other aquatic
organisms in the receiving stream.
A new process for precipitation of soluble zinc had been developed by
American Enka Company through the pilot plant stage, and a large-scale
plant had been designed, built and operated for a very short time.
At this point, the Industrial Pollution Control Branch of the Water
Quality Office, Environmental Protection Agency, offered Enka a re-
search and development grant to help perfect the process, in exchange
for making all the process information publicly available.
In order to perfect the process, the following objectives were defined:
1. To install appropriate equipment to collect and channel waste streams
containing soluble zinc into the treatment plant so that a signi-
ficant amount and concentration of zinc could be treated.
2. To perfect the operation by which the excess acid is neutralized
and the soluble zinc is precipitated and removed from the waste
s treams.
3. To improve the quality of the recovered zinc and the water effluent
so that they may be reused in manufacturing processes.
4. To determine the capacity of the plant and its investment and
operating costs.
5. To improve plant efficiency and economics. Operating costs are
to be reduced by the use of rayon plant wastes that are alkaline
in nature and by the use of lime for acid neutralization.
6. To use all of the technology developed, in order to demonstrate
its value.
BACKGROUND INFORMATION
To the best of our knowledge, all viscose rayon producing companies
use zinc in the manufacture of practically all of their rayon products.
The amount of zinc used depends on the type of product which is being
made and the particular company's special knowledge about making its
products. In general, it can be said that textile yarns and most staple
fibers are spun using fairly small amounts of zinc sulfate, while in-
dustrial yarns, such as those used in the manufacture of automobile tires,
conveyor belts, braided hose, etc., are spun using larger amounts.
-------
The zinc which is used in the viscose rayon process is not consumed in any
of the viscose reactions. It is merely Lost, either being carried out by
the spinning filaments and then lost in the subsequent washing operations,
or it is lost by splash and drip from the machinery involved. In addition,
some zinc is lost in the washing of filters and other ".quipment.
It is estimated that more than 50 million pounds of zinc sulfate are con-
sumed annually by the rayon industry while producing more than one billion
pounds of product.
Other industries discharge zinc to the streams. For example, a large
ground-wood pulp mill can use up to five tons of zinc daily (12) because
zinc dithionite is used as a bleaching agent in the production of news-
print. Zinc salts are used commonly now as a chemical treatment in re-
circulating water systems and are discharged to the streams in the blowdown
of the system. And, of course, the electrochemical industry also wastes
a large amount of zinc and other dissolved metals annually.
It has been estimated that in British industrial areas an average of 237= of
the total toxicity of mixed pollution for fish is contributed by zinc and
copper (3).
However, in spite of the importance of pollution by zinc and other soluble
metals, the reduction of inorganic wastes in the U.S. has been running only
about 277,, versus about 577° reduction in organic wastes (1). There are
several reasons for this lower figure, and probably the most important
are the two following.
There are few time-proven, well-developed techniques generally applicable
to inorganic pollution, as compared to organic pollution, where various
established systems for biological digestion have been used for a long
time and are continuously being perfected.
Economic reasons are perhaps even more important. Although the output of
inorganics, and hence the corresponding wastes, has been increasing at
1.5 - 2.0 times the gross national product, the price of inorganics has
been dropping recently at about 2.57, per year (1). In the face of this
reduction in prices, the investment in inorganics waste treatment facili-
ties will have to increase from the present $82 million to at least $135
million by 1974, according to a recent federally sponsored survey.
The process developed by American Enka Company with EPA aid should have
application to reduce stream pollution in any industry where the waste
water contains metals which form insoluble hydroxides. It is a practical
process which produces a dense, easily handled sludge which can be re-
covered in concentrated form for its metallic value. The process takes
advantage of the fact that acid wastes need to be neutralized anyway be-
fore discharge to a stream, and therefore, neutralization and precipitation
of dissolved metals from acid wastes by this process is more economical
over-all than by other available processes.
-------
It is probable that this process can be used also for removal of dispersed
oil particles and other suspended particles, or removal of color from tex-
tile wastes or wood pulping wastes, by precipitating iron, aluminum or
other hydroxides in a dense, easily concentrated sludge. It could be
used also to neutralize acid mine waters with lime, precipitating their
toxic mineral content and, at the same time, reducing the excessive sul-
fate ion concentration by simultaneous precipitation of calcium sulfate.
The difficult type of sludge obtained in many precipitation processes
has often impeded the successful development or utilization of these
processes in the past (32, 33, 34).
As part of the general background information on zinc recovery, some
of the other available processes found in the literature will be mentioned.
For example, the Dutch Holima process precipitates the zinc as the car-
bonate on inert nuclei which can be screened off and discarded. A
recent patent (9) describes the use of sodium carbonate or bicarbonate
to precipitate zinc carbonate. It is recommended for viscose rayon
solutions in which the ratio of the weight percentages of zinc and
sulphuric acid is at least 1:3.
Precipitation of the zinc using lime is well known (10, 11), and has
been used in plant scale for years in Europe and the U.S. But, a
dense, recoverable zinc precipitate has never been mentioned. Usually,
combined plant wastes are treated with lime, and the resulting sludge
containing zinc, calcium, cellulose, etc., is discarded to lagoons.
Ion exchange processes appear frequently in the literature (5, 6, 8, 10,
11, 15, 16). Some of their disadvantages are:
1. The effluent, still highly acid, has to be neutralized anyway before
discharge.
2. Acid and sodium ions compete for the adsorption of zinc on the resin,
reducing its capacity as much as 50% when the sodium and hydrogen ions
amount to 15 grams/liter (11, 16, 17). A spill of concentrated acid
bath must be collected and diluted with soft water before it reaches
the resin.
3. Certain amines and spinning agents poison the resin and reduce its
capacity.
4. Efficient adsorption of impurities such as calcium, magnesium, iron,
lead, copper, etc., often prevent reuse of the zinc without further
chemical separations.
5. The efficiency of zinc removal is low. By using a recycle re-
generation system, the efficiency of recovery will average 92%
(11).
-------
A Russian process may be mentioned (7). It precipitates and discards
the zinc as zinc sulfide by using hydrogen sulfide from the waste rayon
spinning gases.
A German process (22) uses an activated sludge system to reduce the zinc
content of waste waters to 6 - 10 mg./l. It was more effective in hard
waters. The zinc is discarded together with the biological sludge.
And, of course, if the concentration of zinc and other chemicals warrants
it, the acid waste solutions can be concentrated by evaporation and then
reused. For steam economy, multiple effect evaporators are preferably
employed (11).
-------
SECTION IV
HISTORICAL REVIEW
The toxic effect of zinc on fish and other lower organisms growing in the
stream has been well documented in the literature (2, 3, 4, 31).
Therefore, for a long time the American Enka Company rayon plants sought
to reduce the amount of zinc lost from the spinning processes by col-
lecting those waste flows that were sufficiently concentrated to be
susceptible to evaporation of excess water and reuse for further pro-
cessing of the chemicals dissolved in it. However, no reasonably econo-
mical processes were known which would eliminate or recover the zinc from
very dilute waste solutions.
At the suggestion of personnel from the Stream Sanitation Committee of
the North Carolina Department of Water Resources, a method of treating
the acid and zinc bearing industrial wastes together with the sanitary
sewage waste was tried in 1964, on a pilot plant scale basis. Enka
was told that there were certain extended aeration systems treating
combined sanitary and industrial wastes with an initial pH as low as
3.0 or less. During the aeration treatment, the pH was elevated suf-
ficiently to precipitate and settle out heavy metals such as zinc to-
gether with the biological sludge. A 20-gallon aeration unit provided
with a 1/2 HP mechanical aerator was rented and a small clarifier and
a feed storage tank were manufactured at Enka. During the tests, a
BOD reduction of 80 to 85% was obtained in the presence of 25 to 30
ppm of zinc metal at a pH of 3 to 4. Unfortunately, the biological
environment did not result in a pH increase, there was no appreciable
reduction in the zinc content and the trials were suspended.
During the early part of 1965 Enka's Central Engineering Department
was requested to study all of the information available and to present
a detailed program of action to be taken in the immediate future.
Ion exchange and neutralization with precipitation were the two pro-
cesses mainly considered, and the recommendation given was to proceed
with a neutralization-precipitation method while studying further the
possible attractiveness of ion exchange for certain selected waste
streams. The following considerations were set forth in a report
published in March, 1965:
1. If ion exchange would be chosen, it could not be used to treat
all of Enka's acid wastes. A considerable portion of these wastes
was contaminated with substantial amounts of magnesium, and it
would be strongly adsorbed by the ion exchange resins.
NOTE: It is also known that zinc cannot be recovered efficiently from
solutions relatively high in acid and sodium salts (11,16, 17).
2. The acidity of all zinc bearing wastes treated by ion exchange would
not be reduced, and the effluent would still need to be neutralized
to maintain a satisfactory pH in the receiving stream. This results
in double treatment costs, over-all, before the plant effluent can
be discharged to the streams.
-------
3. Certain impurities in the waste streams to be treated by ion exchange,
in addition to magnesium, such as lead and other metals, calcium hard-
ness, etc., would be retained by the exchange resin and problems re-
lated to regeneration of the resin and reuse of the zinc could be
expected. The Permutit Company had warned Enka certain spinning
agents and organic compounds such as amines would poison the exchange
resin and seriously reduce its capacity. Dilute spin bath samples
were sent to Permutit for testing, and a subsequent report confirmed
the presence of compounds which affected the resin's exchange capacity.
4. The neutralization-precipitation method was judged to be more econo-
mical over-all, more flexible and easier to operate.
5. Other dilute sources of pollution could be treated at the same time,
such as waste lye from the rayon plant dialyzers and dilute waste
viscose from equipment cleaning operations. When these wastes are
acidified, cellulose and hemicelluloses precipitate and a potential
source of BOD pollution can be eliminated. At the same time, the
waste alkali content is utilized to neutralize part of the waste acid.
Company Management accepted the recommendations of this report and several
companies were contacted for their experience handling light hydroxide
sludges.
Infilco, Inc. provided Enka with data on its pilot plant tests precipi-
tating a dense type of ferrous hydroxide sludge using waste pickle liquor
and acetylene sludge in order to obtain a reduction in the oil content
of the waste water from a steel cold reduction mill. The pilot runs
were carried out using a 3-1/2-ft. diameter "Densator" reactor.
Since laboratory tests confirmed that a dense zinc hydroxide sludge
could be obtained using the same principles used for ferrous hydroxide,
American Enka rented a similar "Densator" pilot unit to develop the
process through the pilot plant stage. This unit operated from December,
1965, until May, 1966. The description of the unit and an outline of
the results obtained is given in Section V.
Concurrently with the operation of the pilot unit, an industrial sized
recovery unit was designed.
The North Carolina State Stream Sanitation Committee issued in May, 1966,
a permit for American Enka Corporation to construct and operate the new
plant.
To reduce expenditures to a minimum until the new process could be proven,
it was decided to initially use the zinc waste flow from one of the rayon
plant waste trenches. This trench carried about one-third of the zinc
waste water. When the process was developed, the other zinc bearing flows
would be collected for treatment.
In addition, to simplify the operations and technical problems, it was
decided that initially only caustic soda would be used for acid neutralization,
10
-------
The plant started operating in May, 1967.
When the decision was made to use initially an open trench as the plant
feed, the zinc concentration in the trench was about 100 ppm. By mid-1966,
the concentration had halved; and at the plant start-up date, the average
concentration was about 6 ppm of zinc. This was due to the fact that most
of the production of industrial yarns, which are spun using a high zinc bath,
was transferred to another company plant, as well as a reduction in the market
for this product.
The completed plant was run for a short time and then shut down. The zinc
losses from the rayon plant were so reduced at that time that the zinc con-
centration in the receiving streams was well below the maximum allowable
figure specified by the State of North Carolina.
During the time that the plant was run, essentially all the zinc was pre-
cipitated, but the concentration of impurities was so great in relation
to that of zinc, especially light cellulose floe, that the zinc was not
reused for lack of the means to purify the final zinc sulfate solution.
There were other problems concerning the operation of the plant, chief
among these being a very poor pH control during neutralization, prelimi-
nary to clarification of the feed solution.
In May, 1968, the Water Quality office of the Environmental Protection
Agency, then named Federal Water Pollution Control Administration, offered
American Enka, through its Industrial Pollution Control Branch, a research
and development grant to complete the development of the process. It was
specifically requested that Enka operate the plant at first using only
caustic soda for neutralization. Lime would then be used when the new
equipment designed to prepare and add lime slurry would be available.
A new collection system for acid wastes was built in order to be able to
have a reasonable zinc concentration in the feed solution. A dilute viscose
collection system was also constructed. A new pH control system recommended
by Foxboro Company was installed. Suitable means were provided for adding
a coagulant aid to help clarification. A small rotary belt vacuum filter
was purchased to filter the recovered zinc sulfate solution initially, and
to concentrate further the clarifier underflow, if so desired. Suitable
lime handling, storage and slaking equipment were also provided.
During early September, 1969, the improved plant started operation using
caustic soda for neutralization. By mid-October, the recirculating zinc
hydroxide sludge concentration had increased to 6,500 ppm of zinc and sludge
was being concentrated by further settling and then was being dissolved with
sulfuric acid. The initial recovered product was pumped back to the neu-
tralization tank for reprecipitation because it was dirty, but soon afterwards
the product was sent to the rayon plant for reuse.
11
-------
During the first half of December, 1969, the plant was shut down to
make necessary instrument and other equipment modifications and repairs.
By that time, the recirculating Densator sludge concentration had in-
creased to about 10,000 ppm of zinc. The recovered zinc sulfate con-
centration was approximately 3.5 to 4.1%, after adding excess acid for
dissolution of the sludge.
During the end of February and first part of March, 1970, the plant
was started again in operation. During the shutdown, several pipelines
were left improperly drained and considerable freeze damage resulted
during a period of sustained cold weather.
The plant continued to operate using caustic soda for neutralization
until the end of May, 1970, when the lime equipment became ready for
operation. By that time, the recirculating sludge concentration had
increased to about 17,500 ppm of zinc.
During June and July the lime equipment was giving trouble most of
the time, and many changes had to be made to improve its continuous
operation.
Except for occasional short periods of shutdown for minor repairs
or changes, the plant has operated continuously using lime for
neutralization from August, 1970, through the end of the year.
Operation has been generally satisfactory -
During the period of neutralization with lime, the concentration of
the recirculating Densator sludge has fluctuated between 20,000 and
29,000 ppm of zinc, about 3.0 to 4.4 percent of zinc hydroxide. Sub-
sequent settling, before dissolution, yields a concentration usually
in excess of 5.07= zinc hydroxide. This concentration was the goal
that Enka had set for itself when the plant was being designed.
12
-------
SECTION V
DESCRIPTION OF PILOT PLANT STUDIES
From December, 1965, until May, 1966, American Enka operated a pilot
plant "Densator" reactor rented from Infilco, Inc.
A sketch of this unit is shown in Figure 1. It is 3.5 ft. in diameter
and approximately 10-ft. high. The primary zone, where the acid feed
contacts the recirculated sludge, is located in the top 2.5-ft. of
height. The secondary zone, where caustic soda is added to the mixture
flowing from the primary zone and reacted with it, consists of an inner
cylinder 18 inches in diameter extending from near the top of the Den-
sator to approximately one foot from the bottom. In the lower annular
section the sludge separates from the clear water which overflows to
the sewer. The sludge is pumped out of the bottom and into the primary
zone again. A variable speed agitator is mounted on top of the unit
and provides agitation to the primary and secondary zones and to the
sludge at the bottom of the unit. Enka was told that the unit should
handle a waste water flow of 10 gals./min.
Acid waste from a rayon plant waste water trench, with a pH between
2.0 and 4.0, was pumped to a 3,000-gallon storage tank, the acid
was neutralized with caustic soda to a pH of 6.5 and the solution
was allowed to settle overnight. Additional zinc sulfate was added
when desired.
The concentration of zinc in the feed was varied from about 35 to
165 ppm. There seemed to be somewhat better results with the higher
than with the lower concentrations.
The maximum feed flow used could not exceed 5 to 6 gals./min. Since
on a few occasions there was some loss of zinc sludge in the overflow,
even this flow may have been too high. 5 gals/min. corresponds to
0.635 gals./min./sq. ft. rise in the annular sludge settling section.
The large-scale Densator was subsequently designed for a maximum up-
flow of 0.44 gals./min./sq. ft., but even this flow has never been
reached in practice due to other equipment limitations.
Cellulose floe in the feed affected adversely the density and settle-
ability of the sludge. To reduce this contamination, the Densator
feed was filtered through a filter press dressed with cotton flannel.
The sludge was recirculated to the primary zone at a rate of 2 gals./min.
A "Superfloc" coagulant aid seemed to help the settl.eability of the
sludge at a dosage rate of about 0.5 ppm. Unfortunately, there is no
record of which type of American Cyanamid "Superfloc" was used. "Separan"
NP-10 coagulant aid did not appear to help at dosage rates between 0.1
and 0.5 ppm. "Separan" is sold by Dow Chemical Company.
13
-------
PILOT PLANT DENSATOR
Recirculat ion
Pump
FIGURE 1
14
-------
Using a feed rate of 5 gals/min., a sludge recirculation rate of 2
gals./min., a feed concentration of 165 ppm of zinc and 0.5 ppm of
Superfloc, a sludge with 35 to 44 grams/liter of zinc hydroxide
could be regularly obtained. Once a concentration of 50 grams/liter
was reached.
The volume of sludge could be reduced to about half by subsequent
settling aided by gentle stirring.
No noticeable difference was observed whether the pilot unit was
operated continuously or batchwise.
During most of the tests, the pH in the primary zone was 8.3 - 8.6,
and it was 9.5 - 9.8 in the water effluent.
15
-------
SECTION VI
FLOW DIAGRAMS AND DESCRIPTION OF THE RECOVERY PLANT
The Enka process is different from other similar hydroxide preci-
pitation processes in two respects:
(a) Unlike other companies who have added lime to neutralize acid
and precipitate zinc in one step and have then stored the re-
sulting unusable sludge in lagoons, American Enka has divided
the process in two steps. First, the expensive neutralization
step is carried out with lime and the solution is clarified of
all insolubles. Then zinc is precipitated with caustic soda
in a relatively pure state.
(b) The precipitation is carried out in a manner unique in the field
of metal hydroxides, so as to obtain a new form of dense sludge
with special properties. It can be easily washed to eliminate
or reduce soluble salts, can be centrifuged without difficulty,
stored as a solid or reused as a chemical.
Briefly, the process can be described as follows:
Sulfuric acid and zinc-bearing wastes are collected at various points
in the Rayon Plant and pumped to one main pit. Various waste alkaline
flows are collected and added to the acid wastes, to reduce other types
of pollution and for reasons of economy. The composite wastes are
neutralized with lime to a pH just below 6.0, the pH at which zinc
begins to precipitate. Some calcium sulfate may precipitate and it,
with or without the aid of a coagulant aid, helps to clarify the
solution as it flows through a clarifier, where the insolubles settle
out and are concentrated for disposal in the plant dump.
The clarified solution now flows to the primary zone of the Densator
reactor where it is contacted with a high concentration of pre-
viously precipitated zinc hydroxide, up to 50 or more times the metal
concentration in the feed solution. During this contact, most of the
dissolved zinc in the feed adds onto the existing particles. Alkali
is added in the reactor's secondary zone to complete the reaction and
to restore the pH back to 9.5 - 10.0. If desired, a coagulant aid can
also be added here to obtain an even greater sludge density and im-
proved settleability.
From the secondary zone of the reactor, the mixture flows to the
annular section, where the solids settle and are separated from the
clear, essentially zinc-free effluent. The settled sludge is scraped
into a center well where it is kept mildly agitated for pumping back
to the primary zone of the reactor or to be removed for further con-
centration in a separate vessel. This next vessel is provided with
even milder agitation and here the sludge settles normally to a con-
centration of 5 to 7% of zinc hydroxide. The zinc is then dissolved
with acid and reused in the rayon plant.
17
-------
By comparison, if the same acid feed solution were treated with caustic
soda to a pH of 9 - 10 without the presence of previously precipitated
hydroxide sludge and then allowed to settle for as much as 18 hours, the
zinc hydroxide concentration in the settled material would be less than
one-half of one percent.
The characteristics of the untreated acid wastewater, the treated
effluent and the recovered zinc sulfate are summarized in Table I.
TABLE 1
CHARACTERISTICS OF PLANT FLOWS
Raw Wastewater
Sources:
Acid yarn wash water, filter backwash water, spin machine pot
spray water, spin bath spillage, etc.
Dialyzer waste lye, containing caustic soda and hemicellulose
Viscose filterpress and storage tank wash water, containing sodium
cellulose xanthate, caustic soda, sulfides, etc.
Characteristics:
Flow: 600 - 1000 gpm
pH: 1.5 - 3.0
ZnS04: 70 - 120 mg/1 as Zn
H2S04: 800 - 2000 ppm
N32S04: 0.20 - 0.25%
MgS04: 60 - 150 mg/1 as Mg
Cellulose and hemicellulose: less than 1%
H2S, Na2S, ZnS, S, Fe > pb, surface active agents, etc: less than 0.1%
Treated Effluent
pH: 9.5 - 10.0
ZnS04= 1 mg/1 as Zn
CaS04' 1£SS than 0-2%,
Na2S04: 0.20 - 0.25%
MgS04: 50 - 150 mg/1 as Mg
Fe: 0.3 mg/1
BOD: 24 mg/1 (mostly from surface active agents)
COD: 60 mg/1 (mostly from surface active agents)
18
-------
Recovered Zinc Sulfate;
ZnS04: 8.5 - 11.0%
less than 1.5%
o
Mg: 0.5%
Fe: 0.1%
Pb: 0.02%
A detailed description of the process follows below. The flow diagrams
for the process are shown in Figures 2, 3 and 4. The underlined numbers
shown in the three figures refer to the equipment item numbers as des-
cribed in Section XII. A description of the acid and alkali collection
systems is also given in Section XII.
The mixture of acid and alkaline wastes, which is preponderantly acid,
is pumped by pump - 1^ through a plastic pipeline for a distance of a
few thousand feet to the neutralization tank, item - 2^.
The zinc concentration in the recovery plant feed is approximately 70 -
120 ppm and the normal flow is 600 to 1000 gals./min.
The feedforward pH control system with feedback trim, item _3, measures
the pH and volume of the incoming flow and adds the required amount of
alkali. If the resulting pH differs from the set point pH, the controls
will correct the flow of alkali accordingly.
Normally, lime slurry is added for neutralization through control valves
CV-1 and CV-2, shown in Figure 3. However, in case of a shutdown of
the lime system, caustic soda solution from the rayon plant can be added
through control valves CV-11 and CV-12. The fluid pressure of either
alkali at the respective control valves is kept uniform by using a
suitable head tank placed above the level of the valves.
The caustic soda head tank also feeds soda solution to the Densator,
through a rotanieter adjusted manually. Because the large amount of hy-
droxide sludge in the Densator acts as a buffer, changes in the pH are
very gradual. Manual adjustment of the soda flow once, or at most twice,
per day has been found to be sufficient. The original plan was to meter
the alkali automatically by using a Hach colorimetric zinc analyzer, but
this has never been tried out because it was considered unnecessary.
It would be less expensive to add lime slurry to the Densator also, but
there would be some calcium precipitation which would contaminate the
zinc sludge. In those cases where this calcium contamination would not
be objectionable, lime slurry should be used for reasons of economy.
Practically all the alkali added to the Densator is consumed for pre-
cipitation of the hydroxide sludge. The preliminary neutralization to
a pH of 5.0 - 6.0 has essentially completed the reaction with acid to
*-he neutral point. See Figure 5.
19
-------
ENKA ZINC RECOVERY
ro
o
PUMPS-9 SLUDGE PUMPS-12 SLUDGE STORAGE
SETTLING TANK ~~ DISSOLVING TANK TANK
1°. il 13
RECOVERED ZINC SULFATE
TO SPIN. BATH
Figure 2
-------
ALKALI SUPPLY AND pH CONTROL SYSTEM
TO
DENSATOR
8 UJ
FROM LIME
SLURRY
STORAGE
TANK
CAUSTIC SODA
SUPPLY
FROM RAYON PLANT
LIME
HEAD
TANK
:AUSTIC
HEAD
TANK
LIME
SLURRY
SUPPLY
TANK
25
NEUTRALIZING TANK
2
PUMPING
PIT
Figure 3
-------
LIME SLURRY MANUFACTURE AND SUPPLY SYSTEM
LIME SUPPLY
FROM
RAILROAD CAR
WATER
From 18
TO LIME
SLURRY
SUPPLY
TANK
LIME
SLURRY
STORAGE
TANK
22
23
Figure 4
-------
TITRATION CURVE OF DILUTE ACID WASTE FROM RAYON PLANT
10.0
9.0
8.0
7.0
PH
N>
6.0
5.0
4.0
3.0
2.0
55
60
65
70
CCS OF 2.5% NoOH
Figure 5
75
80
85
-------
The preparation of the lime slurry is as shown in Figure 4. The rotary
kiln quicklime, with a maximum pebble size of 3/8", is. fed to the slaker -
20, by means of a rotating screw placed under the lime storage tank -
19.
The water feed to the slaker is split into 2 flows, a slaking water flow,
through control valve CV-15 and rotameter Fl-4, and a dilution water
flow through control valve CV-9 and rotameter Fl-3. The slaking water
actually flows through a small heat exchanger placed in the slaker it-
self, where it is heated somewhat by the outflowing slaked lime slurry,
which is at 170 - 190 deg. F. However, because this exchanger supplied
by the slaker manufacturer is quite small, we have added a 24 KW electric
heater with thermostat in the slaking water line. This heater is not
shown in Figure 4, but it is described in Section XII.
The liquid level in the slaker is controlled by level controller LC-10,
which returns the lime slurry pumped by pump - 21, back to the slaker,
or feeds it to the storage tank - 22, through its action on control
valves CV-13 and CV-14.
The setting of the lime screw feeder speed is set manually to provide
more than the average lime slurry consumption rate. The slaking water
rate is adjusted to give a slaking chamber temperature of 170 - 190
deg. F and the dilution water rate is adjusted to give a lime slurry
composition of 10 - 127o calcium hydroxide.
Whenever the liquid level in the storage tank - 22_ reaches a preset
height, the level controller LC-11 turns off the lime screw feeder
and the slaking and dilution water flows. When the liquid level drops
to another preset height, the lime and water feeds to the slaker are
turned on again.
During slaking, if the temperature should exceed 190 deg. F, a high
temperature switch adds additional slaking water during 3 minutes.
Because the lime storage tank - 19, the slaker - 20, and the slurry
storage tank - 22 are placed some distance away from the zinc re-
covery plant itself, beside a railroad siding, the slurry is pumped
continuously by pump - 23 through a long underground pipeline to the
recovery plant site and back to tank - 22_ in a closed loop. The flow
velocity in this loop is kept high to prevent sedimentation and coating
of the inside pipe walls.
To control further the lime slurry concentration, a liquid density
meter and controller - 24 has been placed in the slurry recirculation
loop and it may add dilution water through control valve CV-7 as re-
quired to maintain any desired density level within the range 1.03 to
1.08 at 85 deg. F. In practice, we use this instrument to indicate
the density, but we have not found it necessary to use it as a density
controller. However, we are considering purchasing by-product "carbide
lime" from acetylene manufacture for at least part of our lime require-
ments, if the quality and cost is satisfactory. Since this calcium
24
-------
hydroxide would be in the form of a concentrated slurry, the density con-
troller would be very valuable then for adjusting automatically the slurry
concentration.
The lime slurry is added normally to the first or second compartments of
the wooden neutralization tank - £. Each of the 3 compartments is pro-
vided with vigorous agitation. The solution leaving the tank should
have a pH preferably about 5.5. As it flows out of the tank, coagulant
aid solution is added if required, especially when caustic soda is used
for neutralization. When lime is used and calcium sulfate is precipi-
tated, the subsequent clarification of the solution is much improved and
coagulant aid addition should not be necessary.
Besides sulfuric acid and zinc sulfate, the feed solution also has
sodium sulfate, magnesium sulfate and small amounts of organic cationic
and non-ionic rayon spinning aids. Some other impurities are cellulose
floe, sulfur, zinc sulfide, hydrogen sulfide, and traces of iron and
lead. Except for part of the iron and lead, and some sodium sulfide,
which flow to the Densator, most of the latter impurities are eliminated
in the clarifier - 4_. Of course, when calcium sulfate solubility, about
0.2%, is exceeded during neutralization, calcium sulfate precipitate
also appears in the clarifier underflow. This gypsum precipitate helps
to carry down with it the other suspended impurities during clarification.
In order to exceed the maximum solubility concentration of calcium sulfate,
the sulfuric acid concentration in the feed must be about 0.16% or higher.
To increase the amount of gypsum being precipitated, part of the clarifier
underflow can be pumped back to the acid feed, to dissolve its gypsum
content before lime is added for neutralization. This addition can be
made to the first compartment of the neutralization tank if the lime is
being added to the second compartment, as we prefer doing. The first com-
partment is preferably used as an acid concentration equalizing section.
The clarifier underflow that is not recirculated is sent to a thickener - 6^
for concentration, prior to disposal in a waste dump area. Provision has
been made to concentrate further the thickener underflow, if desired, by
pumping the sludge by means of a Moyno pump - T_ to a rotary vacuum belt
filter - 17..
Some of the suspended impurities are attached to air or gas bubbles, and
tend to float initially when the neutralized feed solution flows into
the clarifier - 4_. To prevent this scum from floating directly to the
clarified effluent weir, a 38-ft. diameter floating baffle, extending
about 4 - 5 ft. down from the water level, retains the impurities until
eventually they lose their bouyancy, settle and are removed in the
underflow.
The clarified solution flows to the Densator reactor - {} where it con-
tacts recirculating sludge in the primary zone, receives caustic soda
solution in the secondary zone and the sludge is separated from the
clear effluent in the annular zone, as already explained.
25
-------
Periodically or continuously, as desired, sludge is removed in the
settling tank - 10, vhere the sludge is retained for several hours
and its density is thereby increased. As the sludge flows to the
settling tank, relatively soft neutral plant water is added to dilute
and wash away much of the calcium dissolved in the water which accom-
panies the sludge. If desired, the sludge in the settling tank could
be rinsed further with soft water to reduce even more the calcium
content of the final sludge recovered, but we find this second step
unnecessary for our requirements. Centrifugation of the sludge would
accomplish a similar purpose and would increase sludge density also.
Settled sludge is removed in batches to dissolving tank - 11, and
concentrated acid is added to it for dissolution of the solids. A
Haveg mixing pot is suspended on top of the tank and serves as a
primary reaction chamber, protecting the tank's rubber lining. The
acid is added with a metering pump. To reduce the attention required
from the operator, the acid pump can be turned off automatically by
means of a preset timer, or even by means of a pH controller set to
turn the pump off when the pH reaches a value of 1 or 2. By experience,
at a pH of about 1.5, there is about 0.5% free acid.
If there are enough impurities in the acidified sludge to warrant it,
the acid solution can be filtered in rotary vacuum belt filter - 17
before being transferred to storage tank - 13 for reuse. This was
actually done when the process was run while using caustic soda for
neutralization, but the filtration rate was quite slow. Later, a new
pipeline was installed between the dissolving tank - 11 and the first
compartment of the neutralization tank - 2_. The zinc sulfate solution
was allowed to settle, and any settled turbid portion was returned to
the start of the process for reclarification and reprecipitation. Since
the acid to zinc ratio is low in this solution, the only added cost of
reprocessing is the caustic soda required for reprecipitation0
26
-------
SECTION VII
DISCUSSION OF OPERATIONAL PROBLEMS AND RESULTS
ZINC RECOVERY EFFICIENCY
The efficiency of the zinc recovery process depends on the concentration
of the soluble zinc in the acid feed. This is due to the fact that un-
less there is an abnormality in the Densator operation, such as an acci-
dental low pH, the soluble zinc in the clear effluent remains essentially
constant. Therefore, the higher the soluble zinc concentration in the
feed, the higher the percent efficiency of zinc recovery.
In Table 2 the final soluble zinc concentration can be observed during
one period in which the more accurate but laborious atomic absorption
technique was being used to determine the zinc concentration in the clear
Densator effluent. This particular period was chosen for the Table be-
cause there were repeated pH problems due to the lime feed tank level
control and clogging of a lime slurry pipeline, resulting in a very low
pH in the clarifier and a moderately low pH in the Densator. Only the
zinc content in the clear Densator effluent was determined by atomic
absorption. A titration with ethylene diamine tetraacetic acid at a
buffered pH of 6.8 was used for the other zinc analyses shown. The
analytical procedure is described in Section XII.
It can be seen that on October 8 and 12, and on November 5, the solution
in the clarifier was very acid. Enough additional caustic soda was added
to the Densator to compensate for the lower pH, and only on November 5
did the Densator pH become too low. At that time, the zinc in the effluent
increased to 10.9 ppm. The rest of the time the zinc remained below 2 ppm
and its concentration did not correlate with the zinc concentration in the
clarifier solution flowing into the Densator, which varied from 64 to 126
ppm.
Using 85 ppm as the average zinc concentration in the clarifier, or the
feed to the Densator, and 1.08 ppm as the average zinc concentration
in the effluent (excluding the November 5 figure), the average percent
loss can be calculated as 1.27%, for a zinc recovery of 98.7%.
If the average zinc concentration in the clarifier had been 150 ppm
instead, the zinc recovery figure would correspond to 99.3%, as the zinc
concentration in the effluent would not be expected to change.
EFFLUENT WATER QUALITY
The effluent water is normally very clear. It does contain about 2,000 -
2,500 ppm of dissolved sodium sulfate, which is present in the feed to the
plant, and somewhat less than 2,000 ppm of dissolved calcium sulfate, cor-
responding to the solubility of gypsum in the presence of this amount of
sodium sulfate (23). Atomic absorption analysis of this effluent also
indicates 50 - 150 ppm of magnesium, 0.3 ppm of iron and no other metals
except for about 1 ppm of zinc. In addition, there is a small quantity
27
-------
TABLE 2
ro
CO
LABORATORY
PH I
10-7-71 1.9
10-8-71
10-9-71
10-12-71
10-13-71
Shutdown
11-4-71
11-5-71
11-6-71
11-9-71
1.6
1.95
1.7
1.75
10-14
1.7
1.7
1.9
1.8
Feed
jpm Zn
73
80
92
77
77
until
67
154
56
74
ANALYSES OF PLANT pH, ZINC AND MAGNESIUM
Clarif ier
ppm MR p:
127
122
98
133
115
11-4 to
119
91
90
120
5
3
6
3.
5
modify
6
_3
6
6
H
.4
.3
.5
J2
.2
lime
.0
_._!
.0
.5
ppm Zn
85
79
64
81
126
feed tank
71
105
85
67
Dens. effl.
£H
9.8
9.75
9.95
9.65
9.65
level
10.1
8.0
10.1
10.1
ppm Zn
0.88
1.05
1.50
0.83
1.20
control
1,07
10.9
1.20
0.92
Dens, sludge
£H
9.9
9.85
9.95
9.75
9.8
and piping
10.1
8.5
10.1
10.2
ppm Zn
21,580
25,040
26,810
25,110
26,090
17,650
20,600
18,500
23,280
ppm Mj*
3,840
4,620
4,990
3,450
4,430
3,040
2,890
2,970
3,650
-------
of soluble organic rayon spinning agents, non-ionic and cationic, which
are added in the rayon manufacturing process. These agents, plus possibly
some soluble cellulose decomposition products, contribute a small amount
of organic contamination to the effluent. 26 BOD determinations averaged
24 ppm. Five COD determinations averaged 60 ppm, while the corresponding
BOD determinations averaged 27 ppm, or an average COD to BOD ratio of
2.2.
This water is suitable for reuse wherever the high sulfate and high
hardness is not an impediment; for example, as cleaning and flushing
water, pump seal water, etc. It has been used for all water needs in
the zinc recovery plant except for lime slaking and diluting water, and
for the safety shower and eye wash station.
Any unused effluent water should be mixed with other plant effluents to
reduce its pH of 9.5 - 10.0, or acidified in some inexpensive way.
As a matter of interest, Figure 6 has been included to show the effect
on Hominy Creek, the small stream receiving American Enka's effluents,
during the operation of the zinc recovery plant while neutralizing the
acid with caustic soda and recovering about one-half of the zinc consumed
by the rayon plant. The points which are shown correspond to one grab
sample per week and therefore show considerable variation. It is in-
teresting to note though, that the average zinc concentration was re-
duced by more than 50%. We believe that this is due to the well-known
effect of zinc adsorption by the silt solids in the stream (19, 20, 21).
The large pH variations are believed to be caused mostly by the rinse
water of boiler feed water demineralizers. These discharges will be
equalized in the near future.
REUSE OF ZINC
The recovered zinc has been used for many months in the rayon plant.
Because the zinc contains some magnesium, it is used in a rayon bath
that contains both zinc and magnesium. This bath is the same one that
contributes the magnesium to the recovery plant feed.
The average concentration of the magnesium in the recovered zinc sulfate
solution is about 5 percent based on zinc sulfate.
If the caustic soda solution added to the Densator would be considerably
more dilute than 21%, and if the pH in the Densator were kept at -9.0 -
9.5 at all times, there should be no appreciable concentration of mag-
nesium hydroxide in the zinc sludge. Most of the magnesium hydroxide
is precipitated between 9.5 pH and 11.0 pH (27).
Some of the iron in the feed solution is precipitated in the clarifier,
some in the Densator, and there is apparently still a small amount in
the effluent. The average concentration in the recovered zinc sulfate
solution is 2.58% based on zinc or 1.04% based on zinc sulfate. This
amount of iron is apparently of no consequence to the rayon plant operation.
29
-------
pH:
HOMINY CREEK AT BREVARD ROAD BRIDGE
BEFORE AND AFTER RECOVERING ABOUT 50% OF THE ZINC
RRM. Zn:
AVE.:6.1
MONTH: 5
Figure 6
AVE.: 3.0
AVE.; 0.86
11 12
1969
Redrawn: I2-2I- I970
-------
Part of the iron originates from the concentrated sulfuric acid, which may
have as much as 75 ppm when it is used.
It is possible that if a greater percentage of zinc lost by the rayon plant
had been recovered during the development of the process, it might have be-
come desirable to purify the recovered zinc sulfate and reduce its iron content.
The process which would be used is well known and includes oxidation of the
iron to the trivalent state, precipitation using careful pH control, coagu-
lation and separation of the hydroxide. Purification of iron as well as other
impurities from an acid solution of zinc sulfate before the solution is sent
to the electrolytic cells is a normal procedure in the electrolytic zinc
industry (14, 24).
There can be as much as 0.55% of lead in the recovered hydroxide sludge,
based on che weight of zinc, or 0.22% based on zinc sulfate. However, the
lead content in the solution pumped back to the rayon plant is about 0.1570
based on zinc sulfate.
The lead in the hydroxide sludge dissolves more difficultly than the zinc.
For example, if sulfuric acid is added to the sludge until a pH of 3.5 is
obtained, about 92% of the zinc is dissolved but only 15 - 20% of the lead.
At a pH of 2.5, about 95% of the zinc is dissolved, but only about 33% of
the lead.
All the calcium present in the hydroxide sludge appears to be dissolved in
the water which accompanies the sludge, and amounts to about 600 ppm. During
the past tests we found it necessary only to add fresh soft water to the sludge
as it is transferred from Densator to settling tank, diluting thereby the con-
centration of calcium in the water. In the near future, when we plan to
collect and recover most of the zinc lost by the rayon plant, we will pro-
bably find it desirable not only to wash the sludge with soft water, but also
to centrifuge the sludge and increase its solids content. This will reduce
further the calcium content, as well as the water that must be evaporated
when the recovered zinc is added to the rayon spinning baths. Centrifugation
in the laboratory produces a hydroxide sludge of more than 11% concentration.
If a small amount of light cellulose floe carries over from the clarifier,
it is retained by the Densator sludge. If some of this cellulose is sent
back to the rayon spinning bath together with the recovered zinc sulfate, it
can be noticed that the spin bath filters will require more frequent back-
washing, and that the concentration of cationic spinning agent in the bath
will be reduced, requiring a greater flow of make-up solution.
Therefore, if there is appreciable turbidity in the recovered zinc sulfate
solution, the insoluble material is allowed to settle and this impure portion
is returned to the neutralization tank for subsequent clarification and zinc
reprecipitation.
Initially, when caustic soda had to be used for neutralization, there was
more cellulose carry-over and the rotary vacuum belt filter was used for
purification of the final solution. However, filtration rates were very low
and the operation was very undesirable.
31
-------
SOLUBLE ZINC CONCENTRATION VERSUS SOLUTION pH
The "solids contact" process, on which Enka's zinc precipitation step is based,
has been used mostly for lime softening in water treatment plants. When ap-
plied to the precipitation of metal hydroxides in the manner developed by
American Enka, several unusual effects have been discovered.
The unusual shape and density of the zinc sludge particles are discussed in
Section VIII, under the title "Scanning Electron Microscope Studies."
Another unusual characteristic seems to be the special relationship between
pH and soluble zinc in the presence of Enka's dense zinc hydroxide sludge.
For example, caustic soda was added to 285 ppm of soluble zinc sulfate until
a pH of 9.5 was reached and most of the zinc was freshly precipitated. When
more caustic soda was added, some of the zinc began to redissolve already at
a pH of 10.0 This is shown in Table 3 which lists the soluble zinc, as de-
termined by EDTA titration.
TABLE 3
SOLUBLE ZINC VERSUS pH IN ABSENCE OF SLUDGE
PH:
ppm Zn:
3.0
285
9.5
2.0
10.0
7.2
10.5
42.9
11.0
118.3
However, the results are different in the presence of dense sludge, as shown
in Table 4. Densator sludge was taken from the primary zone, with a pH about
9.0. To aliquot samples of this sludge, sulfuric acid or caustic soda was
added to obtain pH values in the range between 8.5 and 10.5 and the samples
were allowed to settle. The soluble zinc in the liquid above the settled
sludge was then analyzed, this time using atomic absorption for greater
accuracy.
TABLE 4
SOLUBLE ZINC VERSUS pH IN PRESENCE OF SLUDGE
£H
8.5
9.0
9.5
10.0
10.5
After settling
ppm Zn
6.1
2.6
2.6
2.2
0.1
2 hours:
Turbidity
turbid
si. turbid
si. turbid
si. turbid
clear
After settling 24 hours:
ppm Zn
0.66
0.22
0.20
0.17
0.13
Considering the samples settled during 2 hours only, it was evident that the
heavier magnesium precipitation obtained at a pH of 10.5 helped to settle the
suspended zinc hydroxide.
The soluble zinc in the samples settled during 24 hours, which were all clear
above the sludge level, shows the real equilibrium condition. Even at a pH
of 10.5, no zinc hydroxide had been dissolved. Indeed, the soluble zinc had
the lowest value at this pH.
32
-------
The table shows that the pH in the Densator effluent should be at least 9.0.
At a pH above 10.0, magnesium hydroxide precipitates heavily and actually aids
the rate of settling of the sludge. However, since the value of the recovered
magnesium is only 5.3 cents per pound, the caustic soda required for its pre-
cipitation costs 24% more than the value of the magnesium recovered. For this
reason, we maintain a maximum pH value of 10.0 in the Densator effluent. The
settling of the Densator sludge can be aided also, of course, by using a coagu-
lant aid such as the "Superfloc" used in the pilot plant tests.
Table 5 below shows that an acid solution of zinc sulfate can be neutralized
up to a pH of 6.0 without precipitation of any of the zinc as hydroxide. The
soluble zinc has been analyzed by both EDTA titration and atomic absorption,
but the latter is considered to be much more exact.
TABLE 5
SOLUBLE ZINC VERSUS pH IN THE CLARIFIER
Atomic absorption
pH ppm Zn (EDTA") ppm Zn ppm Fe
3.0 184 172.0 1.9
4.0 183 172.0 1.9
5.0 182 172.0 1.5
6.0 180 172.0 0.85
7.0 157 150.0 0.26
The table also shows that 55% of the iron has precipitated at a pH of 6.0.
No attempt was made previous to the analysis to make sure that all iron was
in the ferric form and some of it must have been divalent iron.
Although no soluble zinc will precipitate at the normal pH of 5 to 6 maintained
in the clarifier, insoluble zinc sulfide in the acid feed does settle out and
is collected in the clarifier underflow. This underflow normally contains 1.0
to 1.5% of solids. Although there is available only a limited number of under-
flow solids analyses, these solids usually contain 10 - 15% of zinc and 0.5 -
1.5% of iron, analyzed and calculated on a dry basis. It is estimated that
the insoluble zinc amounts to about 5% of the total zinc in the acid feed to
the plant.
The remainder of the clarifier underflow solids consist mainly of gypsum,
sulfur and cellulose floe.
USE OF COAGULANT AID IN THE CLARIFIER
Laboratory tests using a number of different coagulant aids for easier settling
of the cellulose floe and other clarifier impurities indicated that Calgon
No. 227 and Nalcolyte 672 seemed to aid settling. However, plant-scale tests
showed that only the Nalco product worked satisfactorily. It is a slightly
anionic polyacrylamide of very high molecular weight. The average addition
rate is about 3 pounds per million gallons of flow, or a concentration of
about 0.35 ppm.
33
-------
A solution of 12.5 pounds of Nalcolyte 672 dissolved in 700 gallons of water
is metered to the neutralized solution before the clarifier. It is further
diluted with about 5 gals./min. of water flow before the point of addition.
Agglomerates of insoluble material often float on the surface of the clari-
fier and are retained by the floating baffle until they eventually sink.
An analysis of this material showed about 0.4% of zinc sulfide.
The average turbidity of the clarifier effluent is less than 5 Jackson
turbidity units.
LIME SYSTEM AND pH CONTROL
The lime system holds the dubious distinction of having been the most
difficult one to get to operate satisfactorily. The feedforward pH control.
system was the second one most difficult.
The difficulties encountered with the lime system will be outlined first,
as follows:
1. The slaking and diluting water from the rayon plant was found to have
a too low and too variable pressure. A pressure-boosting water pump had
to be installed in the line to the slaker.
2. The control of the slaker was designed originally to operate differently
from that described in Section VI. Referring to Figure 4 the slaking
water flow was continuously measured by an orifice meter FT-3 and the
reading was fed to a controller that continuously adjusted the velocity
of the lime feed screw to give a preset ratio of lime to water. This
ratio could be changed manually. This proved to be a very unsatisfac-
tory way to maintain a constant slaking temperature. The level in storage
tank (22) was measured by a DP cell, and level controller LC-11 was sup-
posed to regulate both water control valves CV-9 and CV-15 to maintain
a constant tank level and an approximately constant lime slurry concen-
tration. Moreover, the control valve CV-15 was not supposed ever to close
completely below a certain flow, so that the lime screw drive reducer
would never operate below its minimum designed speed. These relation-
ships were too difficult to achieve in practice.
3. The lime slaker was supplied to Enka with all the diluting water addition
at the grit removal chamber. Since there is no agitation at that point,
other than the slow revolving of the grit removal screw, the dilution
water did not mix with the creamy slurry from the slaking chamber. The
thick slurry carried out with it most of the grit instead of it being
removed in the grit chamber by the screw. This grit created problems
in pipelines, tanks, and especially in all the control valves. The
second slaking chamber had to be converted to a dilution chamber and
the intercommunication at the bottom of the dividing wall between the
chambers was closed.
34
-------
The sloping bottom in part of the grit removal chamber then began to
collect slowly mounds of grit above the reach of the removal screw.
This collection of grit would interfere with the flow of slurry; and
the liquid would back up and spill out through the grit removal opening in-
stead of being pumped to the storage tank. The return flow pipe for
slurry, through valve CV-13, was then provided with suitable nozzles
inside the grit removal compartment to wash down the grit accumulations.
4. The level control electrodes, placed at the slaker outlet, would bridge
across with lime and stop operating. Each electrode was covered, ex-
cept for tip, with Tygon plastic tubing to prevent electrical short-
circuiting.
5. Because all lime slurry tank outlets were at the tank bottoms, the fine
grit which even now flows with the slurry has no safe place to collect
and be removed. The grit gave problems with the pH control lime feed
valves and would also clog the pipeline whenever the valves temporarily
closed. The outlet of the lime slurry supply tank was changed to the
straight side of the tank and suitable valves and flushing water were
placed at the bottom nozzle to permit flushing out the deposited grit.
6. It was found initially that the lime feeder screw was really too large
for our lime usage and a sprocket and chain reduction was added to the
drive.
7- A most difficult problem occurred, while operating the initial slaker
control system, when the slaking mixture would get too thick, or start
to boil, and the slaking compartment would bridge over and fill up with
lime. The lime flow would back up and overflow, spilling over the whole
slaker and surroundings.
8. A high-temperature alarm and switch were added so that whenever the
slaking compartment temperature exceeded a preset value, usually 195
deg. F, a solenoid valve would add an additional flow of water to the
slaking compartment during a period of 3 minutes, thereby reducing the
temperature and diluting the thick slurry. This control was found very
useful.
Compounding the problems during the production of lime slurry were the pro-
blems with the feedforward pH control system, which required excessive
attention and never operated really satisfactorily.
Although the concentration of acid in the feed varied continuously; within
a maximum pH range of 1.5 to 3.0, the flow was very constant over the short
term, varying slowly from one day to the next. To prevent initial diffi-
culties this was done on purpose, allowing any excess acid being collected
to overflow at the pumping pit. A typical graph of the feed pH is shown in
Figure 7 showing that the normal pH changes were within 0.5 pH unit. To be
able to operate correctly, the recording pen had to be adjusted so as to
read in the middle of the chart, and the pH of 6.0 in the chart is really
1.5.
35
-------
The initial recommendation to add the neutralizing alkali at the inlet box to
the first tank compartment had to be changed due to excessive lag time. The
alkali had to added very close to the compartment mixer.
The feedback trim could not correct the alkali addition in such a way as to
prevent continous cycling of the pH, sometimes over several pH units, and
still be able to respond satisfactorily when there was a sudden greater pH
change.
A typical pH graph, as measured by the second set of electrodes after lime
neutralization, is shown in Figure 8.
It was found that the pH graph was similar whether the first set of elec-
trodes for the feedforward control were operating or not, so that they were
evantually removed and the corresponding pH transmitter switch placed on
"check," so as to "read" artifically a constant feed pH.
The only way to obtain a much improved pH control was to use the first tank
compartment as an equalizing tank, so as to feed to the second compartment
a nearly constant acid composition and then add the alkali to this second
compartment. The feedforward electrodes remain inoperative and the feed-
back trim electrodes, placed in the second compartment, control the alkali
addition. This controller is provided with reset and derivative action.
A typical graph using lime neutralization is shown in Figure 9.
Our conclusion is that although it may be possible to make our feedforward
pH control system work if sufficient specialized and experienced personnel
would work on it, it appears to us to be too difficult to adjust and control
to be satisfactory for our purposes.
If sufficient feed concentration equalization is provided, a feedback con-
troller is satisfactory if there is good agitation for a short lag time.
The neutralization can be carried out in one step, as done at Enka, or in
more than one step (25), for a greater factor of safety.
It is very desirable to have an agitated compartment or tank, subsequent
to the neutralization step, to even out any pH fluctuations.
For the sake of the record, it should be mentioned that each neutralizing
alkali is added at Enka with two control valves in cascade so that the
smallest opening of the larger one equals the largest opening of the smaller
one. The Foxboro valves are as described in the following table:
TABLE 6
TYPE OF ALKALI FEED CONTROL VALVES
Use: Lime Lime Soda Soda
Diameter: 2 in. 1 in. 1-1/2 in. 1/2 in.
Type: AS split body F7 parabolic AS split body F7 needle (J)
needle (K)
Cv : 46 2.25 25 1.13
36
-------
The large caustic soda valve was adjusted so that it could never open more
than halfway.
It should be mentioned here that only high calcium lime is suitable for good
pH control. Limes with high magnesium content react more slowly and the pH
keeps increasing over a long period of time.
37
-------
Figure 7
Typical Graph of Plant Feed PH (6 in scale = 1.5 pH)
6 DAY'
38
-------
Figure 8
Typical Graph of PH After Lime Neutralization
While Using Complete Feedforward System
•+
'umitex Phantom Chart
UNIFORM
^•^^•:>-S'''--'X--
>
-------
Figure 9
Typical Graph of pH After Lime Neutralization
While Uaing Modified pH Control jiystern
6 DAY'
P:H«
'tillfafy/ Humitex Phantom Chart '^AMflfe. I
imtf UNiroRM wBB»\\^ 1
//////'// __ \\\\\\\\\\\H\\U\\\\\\\\\\l55
III ll X" "X \i\i\HU>lll\U\1.\IH!!!l'j? »
sw^^Mfe^//wi,
40
-------
SECTION VIII
SCANNING ELECTRON MICROSCOPE STUDIES
As part of an attempt to understand better the mechanism by which Enka's
unique form of dense zinc hydroxide is formed and to be able to visualize the
actual appearance of the individual sludge particles, the Physics Research
Department of American Enka took many photos of the recovery plant sludge and
various laboratory-made sludges using a scanning electron microscope. A
selection of the photos is described and shown below.
All samples of sludge were first washed well with distilled water, to
eliminate dissolved salts which would become solid when the sludge samples
were air dried. Therefore, the particles shown should be only the sludge
particles. It is true, of course, that the appearance of the sludge
particles is susceptible to change when drying. However, since all sludges
were washed and dried in the same way, it should be possible to detect
different results by comparison.
The appearance of the actual recovery plant sludge from the Densator is
shown in Figures 10, 11 and 12 at 1,400, 4,800 and 5,500 magnification,
respectively. The particles can be seen to be fairly rounded, smooth
and with a very dense appearance.
A series of laboratory precipitations was carried out using pure chemicals,
in an attempt to observe any gradual changes in the appearance of the pre-
cipitate. Chemically pure zinc sulfate was dissolved in distilled water
to give a solution with 500 ppm of zinc. This high concentration was chosen
to try to accelerate the production of the dense sludge. Pure caustic soda
solution was added to obtain a pH of 9.5. The suspension of zinc hydro-
xide was centrifuged very gently to accelerate settling, but not enough
to agglomerate the precipitate. The clear solution was decanted, additional
zinc sulfate solution was added to the sludge, and the process repeated
again and again. After 10 precipitations, the dried sludge appeared as
shown in Figures 13, 14 and 15 at 500, 1,075 and 5,400 magnifications,
respectively. The two higher magnifications show the sludge particles as
loose leaflets. The photo at 500 magnification shows irregular chunks which
probably resulted from drying a thick layer of sludge. These chunks are not
rounded or smooth as are the particles found in the recovery plant sludge.
After 30 successive precipitations, the dried sludge appears as shown in
Figures 16, 17 and 18 at 550, 1,100 and 6,200 magnifications, respectively.
The photos of the two smaller magnifications show that the leaflet particles
appear to be agglomerating together, while the photo at the greatest magni-
fication shows that the sludge particles are indeed composed of leaflets.
After 60 precipitations, the dried sludge appears as shown in Figures 19,
20 and 21 at 500, 1,180 and 5,850 magnifications, respectively. The photos
of the two smaller magnifications show that the agglomerates are now be-
coming more rounded; and, the photo at the greatest magnification shows how
the surface of the sludge particles is becoming very dense, with the leaflets
barely visible now.
41
-------
It is possible to imagine that after many more precipitations, the sludge
particles can become as rounded and dense as shown in the photos of the
sludge from the recovery plant.
Three more photographs are worthy of showing, although we are not certain
of the proposed explanation for the unique appearance of these sludge par-
ticles. They are shown in Figures 22, 23 and 24 at 5,200, 10,750 and 22,000
magnifications, respectively. Two or three months before the photos were
taken, a sample of the recovery plant solution was taken from the clarifier.
Caustic soda was added until the zinc precipitated, and then the sample was
kept in a stoppered bottle for occasional exhibition to demonstrate graphi-
cally the difference between freshly precipitated sludge and Enka dense
sludge. It is speculated that during the long storage at room temperature,
the precipitated zinc hydroxide rearranged its shape into the special crystals
with pointed needles. Since the experiment has not been repeated for con-
firmation, we cannot be certain of the explanation proposed, especially con-
sidering that the original solution was the impure plant feed, instead of
pure zinc sulfate as was used for the 60 successive precipitations. However,
if the explanation is correct, as we are inclined to believe, it could be
said also that the dense round shape of the particles in the Enka plant sludge
cannot be obtained simply by long sludge storage time, since the shape of the
crystals obtained would be entirely different. Indeed, over the 2 to 3 months
of storage, the single-precipitation sludge obtained from the plant solution
did not appreciably increase in density.
From Figures 10, 11 and 12 it can be calculated that the dense sludge particles
average 4-8 microns in diameter. Based on the size of the particles shown
in Figure 15, it appears that fresh, normally precipitated zinc hydroxide is
in the form of thin curved platelets about 2 microns in diameter. This con-
figuration explains the slower settling velocity and the lower final density
normally obtained with zinc hydroxide precipitates.
42
-------
Figure 10: Recovery Plant Sludge (1,400 X)
Figure 11: Recovery Plant Sludge (4,800 X)
43
-------
Figure 12: Recovery Plant Sludge (5,500 X)
Figure 13; Laboratory Sludge after 10 Precipitations (500 X)
44
-------
Figure 14: Laboratory Sludge after 10 Precipitations (1,075 X)
Figure 15: Laboratory Sludge after 10 Precipitations (5,400 X)
45
-------
Figure 16: Laboratory Sludge after 30 Precipitations (550 X)
Figure 17: Laboratory Sludge after 30 Precipitations (1,100 X)
46
-------
fi.
Figure 18: Laboratory Sludge after 30 Precipitations (6,200 X)
Figure 19: Laboratory Sludge after 60 Precipitations (500 X)
47
-------
Figure 20: Laboratory Sludge after 60 Precipitations (1,180 X)
Figure 21: Laboratory Sludge after 60 Precipitations (5,850)
48
-------
Figure 22: Plant Solution. Precipitated Once and Aged (5,200 X)
Figure 23: Plant Solution. Precipitated Once and Aged (10,750 X)
49
-------
Figure 24: Plant Solution. Precipitated Once and Aged (22,000 X)
50
-------
SECTION IX
COSTS
CAPITAL INVESTMENT COSTS
The investment of equipment for installation of the acid and alkali collection
systems will be shown separately from that corresponding to the recovery plant
proper, since there should be a collection system for any comparable kind of
zinc recovery or treatment plant.
Recovery Plant Collection Systems
Period before R and D grant $370,000
Period during grant 240,QOO 120,000
Total $610,000 $120,000
The engineering cost, or the cost of technical services related to the design
and construction, of the recovery plant and the collection systems was not
recorded separately.
Period before R and D grant $ 65,000
Period during grant 130,000
$ 195,000
If the design and construction had not been carried out during two separate
periods, each with different purposes, and if this project had not been pri-
marily a research and development project, the engineering cost should have
been appreciably lower. In addition, because of lack of experience with the
handling and slaking of lime, the design and construction of parts of the
lime installation and its instrumentation was changed more than once, adding
to engineering and construction costs.
Capital investment costs for a new similar future plant could be reduced by
doing the following:
(a) Instrumentation costs which mounted to about $44,000, can be reduced
considerably using a properly designed feedback controller and two
small neutralization tanks, together with preliminary acid feed con-
centration equalization.
(b) The rotary vacuum belt filter installation, which was purchased partly
as a safety measure, to be able to filter out cellulose floe when
using the caustic soda neutralization system, or to dry further the
clarifier underflow mud after suitable thickening, could be dispensed
with. The thickened mud can be dumped directly without further fil-
tration, saving approximately $40,000.
51
-------
(c) If neutralization with lime is carried out under such conditions that
there is calcium sulfate precipitation, probably the coagulant aid
addition would be unnecessary and the corresponding make-up and addition
system, costing about $8,000, could be saved. Of course, operating
costs would be reduced also to the extent of the daily cost for the
coagulant aid.
(d) There may be some additional cost savings if instead of a Densator,
which includes a reactor core section and an annular settling section
with expensive scraper design, a separate small reactor tank and a
clarifier of standard design would be substituted.
Some of these savings would be offset, however, by increased costs for the
remainder of the equipment, due to escalation.
It should be pointed out here that the investment in a similar recovery
plant depends mostly on the magnitude of the acid flow to be treated, rather
than the amount of zinc to be recovered. Therefore, it is very worthwhile
to collect the waste zinc flows at their source with the highest concentration
feasible.
OPERATIONAL COSTS
The cost to operate the zinc recovery plant depends mostly on the amount of
acid that must be neutralized per unit of zinc recovered. The ratio of sul-
furic acid to zinc sulfate in the plant feed, called ratio R in this report,
depends on the usually secret composition of the rayon spinning and regenerating
bath formulas used by the rayon spinner, and the actual proportion of the dif-
ferent types of rayon spun in each plant, as each type usually employs a dif-
ferent formula. For example, textile filament yarns are usually spun in baths
with a high acid to zinc sulfate ratio, while high-tenacity industrial yarns
and some types of specialty fibers are spun using a lower ratio. Further-
more, for each type of rayon product the ratio varies depending on the source
of the waste. For example, the ratios are different for the spin bath filter
backwash, the spinning pot spray water and the first yarn cake wash.
The over-all ratio R for this plant has usually been about five to six when
adding waste alkaline flows to the waste acid flow. Otherwise, the ratio has
been about six to seven.
The cost of chemicals is the following:
(a) High-calcium rotary kiln quicklime, 95% available lime, costs about
$20 per ton.
(b) Caustic soda solution, available to the recovery plant as a 21%
solution costs about 2 cents per pound of 100% NaOH.
(c) Rayon grade zinc oxide, available to the rayon plant at 12.5 cents
per pound.
(d) Nalcolyte 672 coagulant aid costs $2.32 per pound. It was added at
the rate of 3 pounds per million gallons.
52
-------
For the consumption of electrical power see Section XII, Appendix. The cor-
responding average electrical cost, at 0.7 cent per kilowatt-hour, is $22.40
per 24 hours.
The cost of labor and supervision under normal operation is assumed to be
one-half man, about $5,000 per year. This amounts to $13.70 per day. His
duties would be to check the equipment and instrumentation, to acidify a
batch of zinc hydroxide sludge and to pump the recovered zinc sulfate to
the rayon plant. The latter two operations can be made so as to turn off
themselves automatically.
Because there appears to be little maintenance required for this plant,
maintenance costs are assumed to be one percent of capital costs. If
the latter are taken as $600,000, maintenance costs would amount to $16.40
per day.
Using the given figures, it can be easily shown that if only caustic soda
is used for neutralization, instead of lime, the value of the recovered
zinc would not even pay for the cost of the caustic soda used. Per pound
of zinc recovered:
Cost of soda for neutralization:
00 (161.4) (80) (0.02) = ($0.0404) (R)
(65.4) (98)
Cost of soda for zinc precipitation:
(80) (0.02) = $0.0245
(65.4)
Value of the zinc recovered:
(81.4) (0.125) = $0.1555
(65.4)
To be able to pay for just the cost of the caustic soda, the ratio R for
sulfuric acid to zinc sulfate in the feed to the plant cannot exceed the
following value:
(0.0404) (R) + (0.0245) = (0.01555)
or,
R = (0.1310)7(0.0404) = 3.24
As mentioned before, the ratio R for our plant is 5 or higher.
If lime is used for neutralization, as is being done under normal operation,
the cost of the lime, per pound of zinc recovered would be:
(R) (161.4) (56) (0.01) = ($0.01485) (R)
(65.4) (98) (0.95)
or,
53
-------
2.7 times less than the cost of caustic soda for the same purpose.
When using lime, it is assumed that the precipitation of calcium sulfate after
neutralization makes the use of coagulant aid unnecessary to obtain satisfac-
tory clarification.
If the amount of zinc recovered daily is called Z, the ratio R at which the
value of the zinc pays for the operating and maintenance costs in the Enka
Plant is given by the following formula:
(cost of lime) + (cost of soda) + (cost of labor) + (cost of electrical
power) + (maintenance cost) = (value of zinc)
(0.01485) (R) (Z) + (0.0245) (Z) + (52.50) = (0.1555) (Z)
The corresponding values of Z and R are shown in the table below:
TABLE 7
MAXIMUM ACID TO ZnS04 RATIO AT VARIOUS RECOVERY
CAPACITIES TO PAY FOR OPERATING AND MAINTENANCE COSTS
Lbs. zinc/day (Z): 1000 2000 3000 4000
Max. R to pay 0. & M. costs: 5.28 7.05 7.64 7.95
A minimum recovery of about 1000 pounds of zinc daily assures payment of the
operating and maintenance costs.
The maximum flow that can be handled by the Enka Plant is believed to be about
1200 gals./min. The plant bottleneck is the size of the clarifier, which is
74 feet in diameter. To obtain good settling of the suspended impurities, in-
cluding cellulose floe, the upflow velocity should not exceed about 0.28 gal./
min./sq. ft., and preferably even lower, 0.24 gal./min./sq. ft.
Based on the pilot plant tests, the Densator should be able to operate satis-
factorily at about 2000 gals./min., equivalent to 0.425 gals./min./sq. ft.
To handle this flow, the Enka clarifier should have a diameter of 95 feet,
or another additional clarifier with a diameter of 60 feet should be provided.
Due to the fact that even the very dilute zinc wastes must be collected and
treated, it is doubtful that the Enka Plant's concentration of zinc in the
composite collected wastes flows would ever exceed about 150 ppm. Therefore,
the maximum present capacity for zinc recovery of the plant is about 2,160
pounds of zinc per day, corresponding to a flow of 1200 gals./min. and 150
ppm of zinc. If a larger diameter clarifier were available, the plant capa-
city for zinc recovery would be about 3,600 pounds of zinc per day.
At a zinc recovery capacity of 2000 pounds daily, operating and maintenance
costs would total 12.5 and 14.0 cents/lb. of zinc at a ratio R of 5 and 6,
respectively. At a plant capacity of 3,500 pounds daily, these costs would
be 11.5 and 12.9 cents/lb.,respectively. The cost of purchased ZnO is 15.6
cents/lb. of equivalent zinc.
54
-------
To pay for amortization of capital, greater plant capacity would be required,
as shown in the following calculations.
Amortization payments corresponding to a capital investment of $600,000, at
20 years and 7% interest, would amount to $155 per day. To include this cost,
based on a larger sized plant, the required capital investment will be estimated
to increase as the 0.6 power of the increase in plant size as given by the in-
crease in the plant feed flow. The concentration of zinc in the feed flow is
assumed constant at 150 ppm.
Therefore, the amortization cost, expressed in terms of the amount of zinc
recovered (Z), would be:
f(z) go6) 1 °-6 ($155 00) =
[(150) (8.3) (1200) (1440) J
(Z/2160)0.6 (155) = (1>55) (z)0.6
To determine the electrical costs, assume that they are directly proportional
to the flow, or to the amount of zinc recovered:
(Z/2160) ($22.40) = (Z/96.5)
Maintenance costs of 1% would increase as the capital investment increased,
and would amount to:
(Z/2160)0'6 ($16.40) = (0.164) (Z)0'6
The cost of labor is assumed constant.
Cost of insurance and taxes, as well as general overhead items such as research,
etc., are not included.
Therefore, the corresponding formula for this case would be:
(cost of lime) + (cost of soda) + (cost of labor) + (cost of elec. power)
+ (maintenance cost) + (amortization cost) = (value of zinc recovered)
(0.01485) (R) (Z) + (0.0245) (Z) + (13.70) + (Z/96.5) +
(0.164) (Z)0-6 + (1.55) (Z)0-5 = (0.1555) (Z)
The corresponding values of plant flow, Z and R are shown in the table below
and in Figure 25.
55
-------
TABLE 8
MAXIMUM ACID TO ZnSC>4 RATIO AT VARIOUS RECOVERY CAPACITIES
TO PAY FOR OPERATING, MAINTENANCE AND AMORTIZATION COSTS
Flow Z_ Max. R to pay
(million gals./day) Ibs./zinc/day Q., M. and Amortiz. Costs
1.6 2,000 2.14
3.2 4,000 3.70
4.8 6}000 4.41
6.4 8,000 4.85
8.0 10,000 5.13
If lime would be available at a lower cost, the economics of the process would
improve greatly. For example, a kraft pulp mill (28) purchases limestone and
calcines it to quicklime in a rotary kiln at the rate of 120 tons per day. A
portion of the quicklime, with 907» calcium oxide, is available to its waste
treatment plant at cost, $15.35 per ton. If quicklime were available at this
cost for Enka's zinc recovery- the new values of R would be as shown in the
following table and in Figure 25.
TABLE 9
MAXIMUM ACID TO ZnS04 RATIOS AT VARIOUS RECOVERY
CAPACITIES TO PAY FOR OPERATING, MAINTENANCE AND
AMORTIZATION COSTS IF CHEAPER LIME IS AVAILABLE
Flow Z_ Max. R. to pay
(million gals./day) Ibs./zinc/day 0., M. and Amortiz. Costs
1.6 2,000 2.65
3.2 4,000 4.57
4.8 6,000 5.45
6.4 8,000 6.00
8.0 10,000 6.35
Although a minimum recovery plant of about 10,000 pounds of zinc daily capacity
is required to pay all operating, maintenance and amortization costs using our
present lime costs, only about half that capacity is required if the lower priced
lime were available to Enka, although the price reduction is only about 1970
based on available calcium oxide„
In an attempt to lower lime costs, Enka is investigating the possibility of
using "carbide lime," the by-product of acetylene manufacture.
In the case of other industrial plants, it may be easier to collect acid wastes
with a higher concentration of zinc than is the case at Enka. To show the ef-
fect of increased zinc concentration on the economics of this process, the maxi-
mum R ratio in the feed that will allow recovery of all operating, maintenance
and amortization costs at different plant capacities has been calculated again.
Capital investment costs have been assumed to vary as the 0.6 power of the
56
-------
increase in plant size as given by the increase in the plant feed flow. The
new values of R are as shown in the following table and in Figure 25 and
correspond to a zinc concentration of 300 ppm in the feed.
TABLE 10
MAXIMUM ACID TO ZnS04 RATIOS AT VARIOUS RECOVERY
CAPACITIES TO PAY FOR OPERATING, MAINTENANCE AND
AMORTIZATION COSTS IF Zn CONCENTRATION IN FEED IS DOUBLED
Flow Z Max. R to pay
(million gals./day) Ibs./zinc./day 0., M. and Amortiz. Costs
0.8 2,000 4.38
1.2 3,000 5.07
1.6 4,000 5.50
2.4 6,000 5.98
3.2 8,000 6.28
It can be seen that the economics improve greatly in this case. Therefore,
in certain instances it may be better not to dilute the acid wastes with
alkaline wastes, so as not to increase the recovery plant flow, even if
the resulting R ratio becomes greater.
The use of a cheaper lime improves the economics to a greater degree at the
larger plant capacities. On the other hand, when the amount of zinc to be
recovered is relatively small, it is then most important to treat the zinc
in the most concentrated form.
57
-------
oo
6.0
Max.
R
Ratio
5.0
4.0
3.0
2.0
2000
3000
4000
5000 6000
Lbs. Zinc/Day
7000
8000
9000
C
i-i
(D
Ul
10,000
-------
SECTION X
ACKNOWLEDGEMENTS
As is usual with a study project of this magnitude, the assistance and
advice of many persons was essential for its successful completion.
The foresight and encouragement of the top management of American Enka
Company, who sought for years a solution to the zinc pollution problem,
is gratefully acknowledged.
We acknowledge the aid and encouragement of the personnel of the Water
Quality Office of the Environmental Protection Agency, mainly Mr. William
Lacy, head of the Industrial Pollution Control Branch, Mr. Charles Ris,
the Project Manager, and Mr. Edmond P. Lomasney, the Project Officer
who patiently followed and helped for so long with the many details of
the project.
On the American Enka side, the Project Director was Mr. Frans van Berkel,
Enka Vice President and Technical Director.
The Project Coordinator for the grant, and Enka Project Engineer, was
Mr. David M. Rock.
The Management of Central Engineering, the department with the major
responsibility for the development of the project, always offered all
possible aid for its successful completion. Mr. Grady Allman was respon-
sible for the design of the plant. He also substituted for the project
engineer whenever Mr. Rock was absent.
The management and personnel of the Enka rayon plant were cooperative
and helpful at all times. Mr. B. V. Hill, Manager of the Chemical Labo-
ratory, ran all the tests with the Densator pilot plant. Subsequently,
he was responsible for the analytical work, and lately also for running
the recovery plant. Mr. Fred Walker operated the recovery plant most of
the time, including many nights and weekends and showed much ingenuity in
solving many starting-up problems.
We gratefully acknowledge also the valuable work done with the scanning
electron microscope by Dr. J. Parker, Manager of Physical Research, and
Dr. Nancy Watkins. For the many indispensable analyses made by using
atomic absorption techniques, grateful thanks are due to Mr. J. P. Price.
And last, but certainly not least, we acknowledge the efficient and
patient typing of Mrs. Ruth Ownbey, who typed this report as well as all
the voluminous correspondence concerning the zinc recovery project.
Space limitations prevent us from mentioning many more who rendered valuable
assistance to this project.
59
-------
SECTION XI
REFERENCES
lo Chemical Week; June 17, 1970; pp. 90 - 102.
2. J. Cairns and A. Scheier; Notulae Naturae of the Acad. of Nat.
Sciences of Phila.; No. 299 (June 21, 1957).
3. Sprague, J. B.; Nature 220. 1345 (December 28, 1968).
4. F. Doudoroff and M. Katz; Sewage and Industrial Wastes, 25, 1953;
pp. 802 - 839.
5. A. B. Mindler, The Permutit Co.; "The Economics of Base Metal Recovery
by Ion Exchange;" Preprint from the AIMME Meeting, Dallas, Texas,
February, 1963.
6. R. Kunin, "Applications of Ion Exchange, Part XIV;" "Amber-Hi-
Li tes," Rohm and Haas Co., May 1968.
7o Bo Barochina, Golyand, and Zakharyina; Khimicheski Volokna (Moscow),
5 (1963) No. 5, pp. 46 - 51.
80 C. Sharda and K. Manivannan, "Viscose Rayon Factory Wastes and
their Treatment;" Technology, Vol. 3, No. 4, Spl. Issue: Seminar
on Wastes and Effluents in Chemical Industries (1965), pp. 56 - 60.
9. Fr. Patent 2,000,930; Courtaulds Ltd.
10. D. Kantawala and H. Tomlinson (Washington Univ.); Water and Sewage
Works, Reference Number (1964); pp. R-280 - R-286.
11. R. Aston; "Recovery of Zinc from Viscose Rayon Effluent," Proceedings
of the 23rd Industrial Waste Conference, May 1968; Engineering Bulletin
of Purdue University, Part one; Engin. Extension Series No. 132, pp.
63 - 74.
61
-------
12. Pulp and Paper, p. 149 (October 1969).
13. Industrial Water Engineering p. 34 (June 1969).
14. C. H. Mathewson et al.; "Zinc, The Science and Technology of the
Metal, its Alloys and Compounds" (ACS Monograph Series), Reinhold
Pub. Co. (1959); pp. 183 - 199.
15o A. Mindler and C. Paulson; "Ion Exchange Finds Wider use in Concen-
tration and Recovery of Metals from Dilute Solutions;" Journal of
Metals, pp. 2-7, August 1953.
16. A. Mindler et al; "Metal Recovery by Cation Exchange;" Ind. and Eng.
Chem., pp. 1079 - 1081; May 1951.
17. W. Blake and J. Randle, "Removal of Zn from the Ternary System Zn -
Na - H by Cation-Exchange Resin Column;" J. Appl. Chem., 17, pp. 358 -
360 (December 1967).
18. J. O'Connor and C. Renn, "Evaluation of Procedures for the Deter-
mination of Zinc;" J. of the Amer. W. W. Assoc., 55. (1963),
631 - 638.
19. C. E. Renn, Discussion; Sewage and Industrial Wastes, 25, No. 3
(March, 1953), pp. 302 - 304.
20. J. T. O'Connor and C0 E. Renn, "Soluble-Adsorbed Zinc Equilibrium in
Natural Waters;" J. of the Amer. W. W. Assoc., 56 (1964), 1055 - 1061.
21. G. C. Bratt, "Effect of pH on Ion Adsorption;" J. of the Amer. W. W.
Assoc., 5_8 (1966), 264 - 266.
22. K. Offhaus, "Zinkgehalt und Toxizitat in den Abwassern der Chemiefaser-
industrie," Wass. Abwass, Forschung, 1968, No. 1, 7 - 21.
23. K. Tanji; "Solub. of Gypsum in Aq. Electrolytes as Affected by Ion
Asso. and Ionic Strengths up to 0.15 M and at 25° C;" Environm. Sc.
and Techn.; 3>, No. 7; July 1969, 656 - 6610
62
-------
24. Kirk and Othmer Encyclopedia; Vol. 15, 252 - 262.
25. B. Dickerson, R. Brooks, "Neutralization of Acid Wastes," Ind. and
Eng. Chem., 42, No. 4, April 1950, 599 - 605.
26. H. Jacobs, "Neutralization of Acid Wastes;" Sewage and Industrial
Wastes, 23, No. 7, July, 1951, 900 - 905.
27. E. Berg, C. Brunner, R. Williams, "Single-stage Lime Clarification
of Secondary Effluent," Water and Wastes Engin., March, 1970, 42 - 46.
28. C. Davis, Jr., "Lime Precipitation for Color Removal in Tertiary
Treatment of Kraft Mill Effluent;" Preprint, 63rd Annual Meeting,
Amer. Inst. of Chem. Eng., Nov. - Dec., 1970, Chicago, 111.
29. F. Shinskey, "Feedforward Control of the Neutralization of Process
Waste Streams;" The Foxboro Co., Foxboro, Mass.
30. E. Williams (The Foxboro Co.) "Optimizing Waste Treatment Control
Systems;" Nat. Pollution Control Conf., San Francisco, Calif., April
1970.
31. E. Barth, M. Ettinger et al.; "Summary Report on the Effects of Heavy
Metals on the Biol. Treatm. Processes," J. Water Poll. Control Fed.,
37, 1965, 86 - 96.
32. "Water Pollutant or Reusable Resource?," Environmental Science and
Technology, Vol. 4, No. 5, May 1970, 380 - 382.
33. "States Make Headway on Mine Drainage," Environmental Science and
Technology, Vol. 3, No. 12, December 1969, 1237 - 1239.
34. N. E. Tejera, M. W. Davis, "Removal of Color and Organic Matter from
Kraft Mill Caustic Extraction Waste by Coagulation," TAPPI,. 53. No.
10, Oct. 1970, 1931-4.
63
-------
SECTION XII
APPENDICES
DETAILED LIST AND DESCRIPTION OF MAIN ITEMS OF EQUIPMENT
EQUIPMENT USED IN WASTE COLLECTION SYSTEMS
Acid Wastes
The waste acid flows from various sources are collected as follows:
The spinning pot spray water applied to the outside of the centrifugal
pots in rayon cake spinning machines is collected in underground sumps
or pits and pumped to the main outside pumping pit. The pumps are
actuated by a liquid level switch.
When the rayon spinbath filters are backwashed, the initial flow carries
with it some spinbath which is impregnating the filter medium, as well
as much of the trapped suspended impurities. By using timers, switches,
and effluent flow-diverting valves, the first six to ten minutes of
backwash flow is collected and sent to the main outside 'pumping pit.
The first water wash applied to the rayon yarn cakes after spinning
also carries with it acid and zinc solution impregnating the cakes„
By using timers, switches, and effluent flow-diverting valves, the
first six to ten minutes of wash flow is collected and sent to the
main outside pumping pit.
Other sources of acid and zinc waste solutions, such as drains and
losses from the spinning machines and the spinbath storage tanks are
also collected in sumps and pumped with pit pumps to the main outside
pumping pit.
The wetted parts of all pumps and valves used in this acid service are
all rubber, neoprene or similar acid resisting elastomers, Karbate, PVC
or PVDC, or alloys such as Hastelloy-C, Carpenter-20, Durimet-20, or
Worthite. Most of the piping is PVC or Fiberglass-reinforced Atlac-382
polyester resin.
The main outside pumping pit is made from concrete lined with acid-
resisting brick. Screens made from thin redwood slats are provided
to remove any large material, such as rags, pieces of wood, etc., that
may be present accidentally.
Alkaline Wastes
Certain dilute alkaline wastes are collected and added to the acid wastes
at the main pumping pit. This is done to eliminate an additional source
of stream pollution and to neutralize partly the acid in the zinc re-
covery plant feed for reasons of economy, since the lime consumption is
reduced thereby.
65
-------
Any available waste lye from the rayon plant caustic soda dialyzers is added
to the acid in the main pumping pit. This waste lye normally contains one
to two percent of caustic soda and a similar concentration of hemicellulose.
Alkaline wash water generated from the periodic washing of rayon viscose
tanks, filter presses, etc., is collected in a common pit tank provided
with steel screens removable for cleaning. The water and viscose mixture
is pumped to a concrete pit by means of a Viking heavy duty Model LQ-288
rotary pump provided with a Viking B-size, 6.27/1 helical gear reducer
and a 7-1/2 HP - 1800 RPM motor.
In this pit, the water-viscose mixture is further diluted with waste water
by using an Eastern Model VG-7 mixer with a 14-inch diameter propeller
driven at 420 RPM by a 2 HP motor.
The diluted mixture is now pumped with ordinary centrifugal pumps to an
outside buffer tank of 12,000 gallons capacity which is placed close be-
side the main acid pumping pit. The alkaline solution flows by gravity at
a nearly constant rate into the acid pit. The large 14-foot diameter by
12-foot high steel buffer tank is provided with a Chemineer agitator Model
MDP-75-426 consisting of a 42-inch diameter turbine impeller placed on the
2-1/2-inch diameter x 142-inch long shaft at a level 36 inches from the
tank bottom and another 33-inch diameter impeller placed at the end of the
shaft. The 304 SS impellers and shaft are driven at 68 RPM by a 7-1/2 HP
motor and reducer.
During the time when the sodium cellulose xanthate from the viscose
solution is in contact with acid, while in the main pumping pit or in
the long 10-inch diameter FRP pipeline from the pumping pit to the re-
covery plant itself, the xanthate has sufficient reaction time to decom-
pose to insoluble cellulose floe, sodium sulfate and sulfur by-product
compounds. This reaction should be completed before the first pH ad-
justment, when lime is added to neutralize the acid.
All the pipelines for dilute waste viscose collection inside the rayon
plant, and up to the large outside buffer tank, are provided with Victaulic
couplings at frequent intervals, for easy disassembly in case of plugging.
Recovery Plant Equipment
Item
1 Two 10" x 8" Denver SRL Frame Four pumps, soft-rubber-lined,
~ centrifugal, closed impeller, with overhead hinged motor base,
water gland type, with replaceable pressure molded impeller and
casing liners, cast iron split case and flanged ring clamps.
All wetted metal parts other than ceramic sleeves are Carpenter-
20 alloy. Slide type base, drive guard, an 8-groove V-belt drive
and 50 HP motors. All designed originally to pump 2,000 GPM at
50 feet of TDK. However, V-belt pulleys were changed subsequently,
to pump about 1,000 GPM at a similar head.
66
-------
Item
_2 One open top, pressure creosoted Douglas Fir tank 32-feet I.D. and
20-feet nominal stave length. Staves and bottom are machined from
3" lumber and given an 8-pound pressure treatment of creosote oil.
The tank is banded with 3/4-inch diameter mild steel rods and mal-
leable iron straight pull draw lugs; the bands are epoxy-coated
and mastic was used to coat the lugs, nuts and threads. One set
of 4-inch x 8-inch pressure treated Douglas Fir chine joists are
spaced at the bottom at approximately 16 inches center-to-center
distance. Approximate tank volume is 100,000 gallons. The tank
is subdivided by radial partitions into three approximately equal
sections. Each section is provided with an agitator, described
as follows:
Denver Equipment turbine-type agitator with 5-blade, rubber-
covered impellers, driven by a 25 HP TEFC motor
NOTE: Initially, the alkali required for neutralization was added in
the first compartment of the wooden tank. Subsequently, the
first compartment was allowed to function as an acid concentration
equalizer, and the alkali addition is now made into the second
compartment.
^ One complete Foxboro feedforward pH control system. The influent
flow is measured and characterized to a fractional power relation-
ship. The characterized flow signal is combined with the influent
pH measurement and the reagent valve characteristics to produce
an equation which is the process model:
1
MB = f (F) + ( log R ) (rB - PH), where
M = reagent valve stroke, % of max.
B
f (F) = characterized influent flow signal
R = rangeability of reagent valve
r = feedback term, supplied from a non-linear controller
B
pH = influent pH
The influent flow and pH position the reagent flow control valve
to deliver the precise amount of reagent required to neutralize
the acid in the influent. As the influent conditions change,
the amount of reagent is adjusted accordingly. The reagent flow
computation (actually, valve position) is altered by the output
of a feedback controller where gain is automatically adjusted as
a function of the error existing between the set pH and the final
measured pH.
67
-------
Item
The main items of equipment purchased from Foxboro are:
(a) Model 613 DM d/p cell transmitter with Model 610 AR Power
Supply Unit.
(b) Flow Characterizer Model 66 N.
(c) Two Model 69TA-1 Current-to-air Transducers.
(d) Two Model 40 PH Recorder Controllers with Type JJ Pneu-
matic Transmitter.
(e) Model 52A-SP4 Pneumatic Indicating Controller with
External Reset.
(f) Model 62H-4E Special Non-linear Controller.
(g) Model 66F Pneumatic to Current Convertor.
More details on the pH control instrumentation are given in the
main body of the report.
One Infilco BF Clarifier drive mechanism size BF-100, with
Philadelphia reducer and pinion assembly and 1/2 HP, 900 RPM
motor. Scraper RPM is 3 RPH.
One clarifier tank with steel sidewalls, 74-foot diameter and
14 feet, high set on a concrete sloping-bottom pad. Inside
surfaces were coated with a thin layer of type H Fiberglass
mat and bisphenol-A (Atlac 382) polyester resin..
One clarifier floating baffle approximately 38-foot, 7-inch
diameter, made from eight straight FRP (Atlac 382) panels,
each 16 feet long. Polyurethane foam blocks are attached
to provide bouyancy. Panel height is about 5 feet.
One clarifier underflow centrifugal pump: Dean Bros., 1-1/2" x
3" - 6-DL-200 inline pump, 316 SS, rated for 50 GPM at 35 ft.
TDK. Motor is 1 HP, 1750 RPM.
One thickener mechanism: Type B Eimco Heavy Duty to handle
6 tons/day of solids. Continuous operation at 8,000 ft.-lbs.
of torque and peak loading of close to 12,000 ft.-lbs. Rake
speed of 0.33 RPM. All wetted parts are mild steel with 3/16-
inch thick natural rubber covering, or FRP-
One 20-foot diameter x 8-1/2-foot high thickener steel tank with
flat bottom. Inside surfaces coated with coal tar epoxy.
68
-------
One Type SSQ Moyno sludge pump, to pump the thickener discharge.
Pump has 316 SS body and internals, Buna N lined stator. Rated
for 10 GPM of 50% gypsum slurry at 25-ft. TDH. Motor is 1-1/2 HP,
900 RPM.
8 One Infilco Densator Reactor: outside tank is 80-ft. inside dia-
meter by 13-ft. 6-in. side depth, with concrete bottom. Inside
tank and concrete bottom centerwell (4-ft. deep) are 20-ft.
diameter. All inside surfaces coated with Bitumastic No. 33 and
all outside surfaces with Bitumastic No. 50.
Center mixer drive is 1 HP U.S. Varidrive with a #9 CVD Winsmith
reducer. Top section with 2 paddles 7 feet long by 1 foot high.
Middle section with 4 paddles 4 feet long by 9 inches high. There
is also a cutter bar rotating 1 inch above the bottom well surface.
Normal shaft RPM is 2.5.
Scraper drive is 1-1/2 HP U.S. Varidrive with a #10 CTDO Winsmith
reducer provided with torque limit device. Normal RPH is 2.
J3 Two Fairbanks Morse "Paper Stock" sludge recirculating cen-
trifugal pumps, Fig. 5460 P, 4-in. horizontal end suction, for
400 GPM maximum flow at 32 ft. TDH. Motor is 5 HP and 1200 RPM.
10 One 4,000-gallon sludge settling steel tank 8-ft. diameter by
11 feet high, covered on the inside with 1/4-inch thick "Triflex"
rubber. Provided with a Nettco agitator having two flat blades
each 3 inches wide, 3 feet, 7-11/16 inches long and 5/8 inches
thick, rotating 3 inches above tank bottom at 3 RPM. Drive has a
1/2 HP-900 RPM motor, 3 V-belts, and a Nettco Model T-H5 reducer
with 100/1 ratio, and a 2-3/6-in. diameter steel shaft.
JLJL One 4,000-gallon sludge dissolving steel tank similar to item
10 above, but without agitator. Provided on top with a Haveg
mixing pot to react the recirculating sludge with the concen-
trated sulfuric acid being added by means of a Milton Roy
diaphragm type metering pump.
12^ Two LaBour sludge recirculating and dissolving pumps, rated to
pump 50 GPM, at 80 ft. TDH, of a sludge with 1.2 - 1.8 sp. gr.
and 70 - 700 centipoise viscosity. The pumps are size 15, type
DZT, made from Elcomet-K alloy.
JL3 One recovered zinc sulfate storage tank of 1,000 gallons capacity,
constructed of Fiberglass reinforced polyester resin (Atlac 382).
_14 One acid-resistant centrifugal pump, used to transfer recovered
zinc sulfate to the rayon plant for reuse, capable of pumping
20 GPM at 156-ft. of TDH. Pump was manufactured by Worthington,
with all wetted parts made from Worthite alloy, and is Model
1-1/2 CNG 104. Motor is 10 HP, 1800 RPM.
69
-------
Item
15 Two coagulant aid solution preparation and feed tanks, each with
750 gallons capacity, constructed of Fiberglass reinforced poly-
ester resin (Atlac 382). Each tank is provided with an agitator
having one 12-inch diameter propeller revolving at 350 RPM and a
1 HP - 1750 RPM electric motor.
The coagulant aid powder is added, for dissolution, by means of
a funnel and a Schutte-Koerting mixing-eductor Type 227.
16 One coagulant aid solution metering pump, a Model FR-131A, Milton
Roy Simplex pump designed for a maximum flow of 18.3 gallons per
hour against a discharge pressure of 6.5 Ibs./sq. in. Wetted
parts are 316 type SS with Teflon diaphragm. Drive motor is 1/4
HP.
17 One Eimco pilot plant filtering equipment assembly mounted on a
common platform and consisting of:
(a) One 3-ft. diameter x 1-ft. face rotary vacuum belt filter
with 9.4 sq. ft. of filtering area, made of Eimco-Met plas-
tic. The drum has 8 sections. The belt is polypropylene
cloth material.
(b) Eimco-Met plastic tank for 33-1/3% submergence of the filter
drum.
(c) One mechanical variable speed drive connected to a chain
drive, and designed to allow drum speeds of 1.0 to 10 minutes
per revolution. Motor is 1/4 HP-
(d) One swing type, pin mounted oscillating arc agitator with
rake assembly driven through arcs by a crank disc and con-
necting rod arrangement. Drive is with a 1/2 HP gearmotor
and chain drive, designed for an agitation speed of approxi-
mately 19 cycles per minute.
(e) Two cake wash pipes with spray nozzles.
(f) One cake discharge roll and one take-up and wash roll.
Two cloth wash pipes with spray nozzles. One cloth
tracking mechanism, an Edge-Track alignment system.
(g) One vacuum receiver tank, 15-in. diameter x 5-ft. high, made
of Eimco-Met plastic.
(h) One Worthington 1-1/2 - CNG - 74 Worthite alloy filtrate
pump rated for 20 GPM at 60' TDH. Motor is 1-1/2 HP, 1800
RPM, TEFC.
70
-------
Item
(i) One Nash CL-402 water sealed rotary cast iron vacuum
pump with V-belt drive and water trap silencer, rated for
362 CFM at 20-in. Hg and 1170 RPM. Motor is 20 HP, 1800 RPM,
TEFC.
NOTE: In addition to the above assembly, there is also an in-
clined belt conveyor to collect the filter cake discharge
and to drop it into a 3 cu. yd. Dempster Dumpster con-
tainer, provided with a T-2 bail, which is placed outside
the pump and filter house.
The belt conveyor is 14-in. wide and 8-1/2-ft. long. The
structure is of 304 SS and the belt is Goodyear Ply Ion 140.
Belt speed is 50 ft./min. The drive motor is 3/4 HP.
JLfi One Dean Brothers plant water pressure booster pump, Type PH-231,
size 1-1/2" x 3" - 11-1/2, rated for 65 GPM at 140 ft. TDK. Motor
is 10 HP, 1750 RPM.
19 One Sprout-Waldron and Company 6-in. Schedule 40 "Pneu-Vac"
negative pressure system designed to unload pebble quicklime
weighing 55 to 60 Ibs./cu. ft. and 3/16-in. to 3/8-in. dia-
meter, dry and free-flowing. The material is to be delivered
directly from the railroad car to the intake of the system under
uniform feeding conditions by an adjustable opening in the hopper
car manifold, and conveyed over a maximum combined vertical and
horizontal distance of 100 ft., with not more than two 90 degree
rigid elbows, or their equivalent, on the negative side of the
conveying line. The system is designed to unload the specified
lime at about 50,000 to 60,000 Ibs. per hour. The components
of the system are listed below:
(a) One tandem portable hopper car unloader for a 6-in. diameter
line.
(b) Three sections of 6-in. diameter x 10-ft. long non-toxic
rubber hose.
(c) 6-in. Schedule 40 pipe, elbows, couplings, etc.
(d) One filter-receiver, Flex-Kleen Model 84-CT-30, 17-in.
mercury design, continuous and automatic, with 300 sq.
ft. of cloth area, to handle 60,000 Ibs./hr. of 3/8-in.
pebble quicklime conveyed by 1400 CFM of air at ambient
temperature. Includes suitable connections and instru-
mentation.
(e) One rotary valve, 16 x 14, square inlet and outlet, cast
iron construction, with inspection panel. Capacity is
1.2 cu. ft. per revolution, with design clearance of
0.002/0.003 inches radially. Provided with a 1-1/2 HP -
155 RPM motor and a guarded roller chain drive to operate
the airlock at 45 RPM.
71
-------
Item
(f) One Butler bolted steel tank, 12-fto diameter x 56-ft. high
to store 4350 cu. ft. of 60 Ibs./cu. ft., 3/16-in. to 3/8-in.
pebble quicklime, based on a 45 degree angle of repose. In-
cludes a 60 degree hopper with 12-in. diameter opening, rack-
and-pinion operated slide gate, two Bin-Dicators, etc.
(g) One conveyor screw, 6-in. diameter x 12-ft. long, with
sprocket and chain drive, to deliver about 86 to 860 Ibs.
of lime per hour. Driven by a Uo S. Varidrive motor of
1 HP.
(h) One blower assembly, rotary positive displacement single
stage unit, Roots-Connersville 816-RAS-60, with 8-in.
diameter inlet and outlet, to operate at 1390 RPM and
inlet flow of 1320 CFM ambient air at 16-in. Hg vacuum,,
Provided with guarded V-belt drive and a 75 HP motor,
snubber, etc.
20 One BIF detention type lime slaker, with 77.4 gallons total
volume in the two main slaking chambers, each chamber pro-
vided with one 1-1/2 HP and V-belt driven propeller agitator,
10 gallon grit removal chamber provided with a removal screw
driven by a 1/4 HP motor and a gear reducer to rotate at
4.4 RPM, a small heat exchanger to preheat the slaking water
with the hot lime slurry, and two rotameters to measure the flow
of slaking and dilution water. Slaking capacity is approximately
700 Ibs. lime/hr.
NOTE: The grit is discharged to a 3 cu. yd. Dempster Dumpster
container provided with a T-2 type bail.
A Chromalox NWH-3 electric heater with 24 KW capacity
(50 watts/sq. in.) has been provided to preheat addi-
tionally the slaking water.
21 One Goulds centrifugal pump, Model 3196, size 1x2 - 8, 316 SS,
to transfer the slaked lime slurry from the slaker to a storage
tank, rated for 35 GPM of 1270 slurry at 180 degrees F maximum
and 40 ft. TDK. The drive is a 1-1/2 HP- 1800 RPM motor.
22_ One 8,000-gal Ion capacity lime slurry storage tank provided with
a Chemineer turbine type agitator driven at 87 RPM by a 5 HP
electric motor and reducer.
2_3 Two Goulds centrifugal pumps, Model 3196, size 1-1/2 x 3-13,
Group M, 316 SS construction, rated for 50 GPM of 12% lime
slurry at 270 ft. TDK, and driven by a 10 HP, 1800 RPM motor.
72
-------
Item
24 One lime slurry density meter, Hallikainen Instruments' Model
No. 1373 Gravitrol Density Analyzer, with 316 SS sample loop
and SS flex bellows, 1.4 inch I.D., density range of 1.03
to 1.08 at 85 degrees F and a maximum of 80 PSI, linearly
related to a pneumatic output signal.
25 One lime slurry feed tank with 750 gallons capacity, provided
with a turbine type agitator driven by a 1 HP motor and re-
ducer. RPM of mixer is 30.
26 One Dean Brothers lime slurry feed centrifugal pump, size 1" x
2" - 8-1/2, type PH-231, rated for 32 GPM of 12% slurry at
65 ft. TDK, and driven by a 3 HP, 1800 RPM motor. Pump material
is 316 SS.
27 Two 434-gallon (each) Dempster Dumpster portable carbon steel
tanks, for concentrated sulfuric acid storage and transpor-
tation, 40-in. diameter, designed for 98 PSIG pressure.
28 One Wayne 2-stage tank mounted air compressor, to supply air
for instrumentation use. One B-6-A Lectrodryer has been pro-
vided also, to dry the compressed instrumentation air.
NOTE: All pumps for lime slurry operation (items Nos. 21, 23
and 26) had to be equipped with Rott Durametallic mecha-
nical seals with internal water lubrication injected
between the seal rubbing faces.
73
-------
SECTION XIII
APPENDICES
ELECTRICAL POWER REQUIRED FOR MOTORS AND HEATERS
(EXCLUSIVE OF IN-PLANT COLLECTION SYSTEMS)
Breakdown by use
Item Rated Cap./Unit Continuous Intermittent
JL Recovery plant acid feed pumps at main pit: (2) ea. 50 HP 50
"i_ Lime System:
(a) Water Heater 24 KW 24 KW
(b) Vacuum Pump for Lime Unloading 75 HP -- 75
(c) Rotary Valve for Lime Unloading 1-1/2 HP -- 1-1/2
(d) Lime Feeder Screw 1 HP 1
(e) Slaker Mixers (2) ea. 1-1/2 HP 3'
(f) Grit Removal Screw Drive 1/4 HP 1/4
(g) Slurry Pump, Slaker to Storage Tank 1-1/2 HP 1-1/2
(h) Slurry Storage Tank Agitator 5 HP 5
(i) Slurry Recirculation Pumps (2) ea. 10 HP 10
(j) Slurry Feed Tank Agitator 1 HP 1
(k) Slurry Feed Pump 3 HP 3
_3 Water Pressure Booster Pump 10 HP 10
4 Neutralization Tank Agitators (3) ea. 25 HP 75
f> Clarifier:
(a) Scraper Drive 1/2 HP 1/2
(b) Underflow Pump 1 HP 1
-------
Breakdown by use
Item Rated Cap./Unit Continuous Intermittent
5^ Densator
(a) Scraper Drive 1-1/2 HP 1-1/2
(b) Mixer 1 HP 1
(c) Sludge Recirculation Pumps (2). ea. 5 HP 5
1_ Sludge Settling Tank Agitator 1/2 HP 1/2
8 Sludge Circulation and Dissolving Pumps (2) ea. 3 HP -- 3
9 Concentrated Acid Metering Pump 1/2 HP -- 1/2
_10 Clarifier Underflow Thickener
(a) Scraper Drive 1-1/2 HP 1-1/2
(b) Mud Pump 1-1/2 HP 1-1/2
_11 Rotary Vacuum Belt Filter
(a) Drum Drive 1/4 HP -- 1/4
(b) Agitator 1/2 HP -- 1/2
(c) Vacuum Pump 20 HP -- 20
(d) Filtrate Pump 1-1/2 HP -- 1-1/2
(e) Belt Conveyor 3/4 HP -- 3/4
_1_2 Coagulant Aid System
(a) Agitators (2) ea. 1 HP 2
(b) Metering Pump 1/4 HP 1/4
_13 Zinc Sulfate Return Pump 10 HP -- 10
-------
Item
_14 Air Compressor
15 Pump Room Heaters
Rated Cap./Unit
I HP
10 KW
15 KW
5 KW
Breakdown by Use
Continuous Intermittent
10 KW (winter)
15 KW (winter)
5 KW (winter)
175-1/2
24
155
113
30
114
Total HP (motors)
Total KW (heaters)
Total equiv. KW
Neglect area lighting and other minor miscellaneous loads.
Assuming average continuous operating load is 75% of the rated load: (155) (0.75) = 116 KW
Assuming the average intermittent load operates only one-fifth of the time, at 75% of the rated load:
(114) (0.20) (0.75) = 17.1 KW
At a cost of 7.0 mils/KWH, the average daily electrical cost is: (133.1 KW) (0.007) (24 hrs.) = $22.40
-------
Figure 26: Main Acid Feed Pumps (Item 1) by the Collection and Pumping Pit
-------
VO
Figure 27: Air View of Waste Treatment Area. Zinc recovery Equipment can be seen at lower left,
-------
00
o
Figure 28: The Neutralization Tank (Item 2), Clarifier (Item 4) and Densator (Item 8) can be
seen from left to right. *
-------
00
Figure 29: View of the Densator from its walkway
-------
00
Figure 30: Interior of Pumphouse. The two sludge recirculating pumps (Item 9) in the center,
-------
00
Figure 31: The Sludge Settling Tank (Item 10) at the left, and the sludge dissolving tank (Item il)
at the right, as viewed from the roof of the Pumphouse.
-------
Figure 32: View of Quicklime Handling and Storage Equipment (Item 19),
Slaker (Item 20) and Slurry Storage Tank (Item 22).
84
-------
SUBJECT:
PROCEDURE FOR THE DETERMINATION OF ACID, ZINC, AND MAGNESIUM IN WASTE WATER
NUMBER:
028.3
DATE:
January 25, 1971
ACID:
Measure 100 ml. of sample into a 250 ml. Erlentneyer flask, add 3 or
drops of 35$ hydrogen peroxide, and let stand for 3 or h minutes. Add
2 or 3 drops of Methyl Red indicator, and titrate with N/10 NaOH.
Calculation; Titration x 0.1 x 0.0*4-9 x 100 _ „
~
ZINC:
Add 10 ml. of maleic acid - sodium maleate buffer solution (pH 6.8),
and 4 to 8 drops of Eriochrome Black T indicator to the titrated solution
used for acid determination. Titrate with 0.02 disodium versenate solu-
tion to a color change .from purple-red to purple-blue.
Calculation: Titration x 6.5 = p. p.m. as Zn.
MAGNESIUM:
Add 15 ml. of ammonia - ammonium chloride buffer solution (pH 100)
to the titrated solution from the acid and zinc determinations. Titrate
with 0.02 N disodium versenate solution to approximately 80$ of the
expected titration. Add 2 to k drops of Eriochrome Black T indicator
and continue the titration to a sky-blue color
Calculation: Titration y.2.k = p.p.m. as Mg.
BY: CHEMICAL-MICROSCOPICAL LABORATORY COPIES:
B. V.
ORU 201* fl-IT-STI ~ ~ 85
-------
CORRECTED -
REVISED -
REAPPROVED -
of
lllk
me Soutio
Degrees
1.23
2,43
3.47
4.43
5.44
6.375
7.36
8.275
9.17
10.11
11.05
11.91
12.7G
13.66
14.5
15.36
16.17
16.97
17.81
18.63
19.40
20.16
20.91
21.70
22.48
23.20
23.91
24.67
25.41
26.10
Specific gravity
at 20 deg. C.
1.0085
1.017
1.0245
1.0315
1.039
1.046
1.0535
1.0605
1.0675
1.075
1.0825
1.0895
,0965
,104
111
,1185
,1255
1325
,140
,1475
,1545
1.1615
,1685
176
1835
,1905
1975
205
2125
1.219?
1 liter
contains
grams CaO
10
20
30
40
50
60
70
80
90
100
110
120
130
140
,50
160
170
180
190
200
210
220
230
240
250
260
270
280
290
300
Per cem
by weigh*
CaO
0.99
1.96
2.93
3.88
4.81
5.74
6.65
7.54
8.43
9.30
10.16
11.01
11.86
12.68
13.50
14.30
15.10
15.89
16.67
17.43
18.19
18.94
19.68
20.41
21.12
21.84
22.55
23.24
23.92
24.60
1 liter
contains
grams Ca(OH)3
13.2
26.4
39.6
52.8
66.1
79.3
92.5
105.7
118.9
132.1
145.3
158.6
171.8
185.0
198.2
211.4
224.6
237.9
251.1
264.3
277.5
290.7
303.9
J17.1
330.4
343.6
356.8
370.0
383.2
396.4
Noter To convert grams per liter to puunds per cubic foot multiply by .06243
To convert grams per liter to pounds per Imperial gallon multiply by .01
To conver.t grams per liter tr> nounds per U. S. gallon multiply by .000345
Per cent
y weight
Ca(OH)2
1.31
2.59
3.87
5.1J
6.36
7.58
8.79
9.96
11.14
12.29
13.43
14.55
15.67
16.76
17,84
18.90
19.95
21.00
22.03
23.03
24.04
25.03
26.01
26.96
27.91
28.86
29.80
30.71
31.61
REFERENCE
C.P.P.A. Technical Section Data Sheet 98
86
-------
1
Accession Number
w
5
n Subject Field & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
(Organization
AmpT'T nan 'Enlra Hrim-namr
Central Engineering Department
Title
Ainc Precipitation and Recovery from Viscose Rayon Waste Water
10
Author(s)
Rock, David M.
Allman, Grady
16
Project Designation
EPA, V&O, Grant 12090 ESG
21
Note
22
Citation
Descriptors (Starred First)
25
Identifiers (Starred First)
Industrial wastes, heavy metals, textile fibers flocculation, chemical
precipitation
27! Abstract In May 1968, the Industrial Pollution Control Branch of the Water Duality Office/
—fiwlronmental Protection Agency, initiated a research and development grant with American
Enka Company to perfect an improved process for the precipitation and recovery of soluble
zinc in rayon manufacturing wastewaters. In the production of viscose rayon, zinc sulfate
is used as a component of the acid spinning bath. Zinc is lost in a dilute form when the
acid spun yarns are washed with water and at various points in the spinning bath system.
The novel zinc recovery system involves the initial neutralization of the waste stream
to pH 6.0 sedimentation of insolubles, the crystallization of zinc hydroxide in a high
pH environment, the sedimentation of zinc hydroxide and the solubilization of the zinc
with sulfuric acid. This novel recovery system was operated at a 600-1000 gpm rate with
70-120 mg/1 of Zn in the feedwater. The system can maintain an effluent concentration of
Zn less than 1 mg/1, which corresponds to 9#-99$ removal efficiency. The unique zinc
hydroxide sludge is easily concentrated to 5-1% solids by sedimentation and to 10$ solids
by centrifugation. The sludge particles obtained by this process are spheroids of 4-8
microns average diameter. A recovery of 2,000 pounds of zinc daily assures recovery of the
12.5 to 14.0 cents/lb. of Zn operating and maintenance costs. The cost of zinc oxide
purchased by Enka amounts to 15.6 cents/lb. of equivalent Zn. This report was submitted
in fulfillment of Grant Project 12090 ESG between the Water Quality Office/Environmental
Protection Agency and American Enka Company.
Abstractor
Institution
American Enka Company
WR:102 (REV. JULY 1969)
WRSIC
SEND WITH COPY OF DOCUMENT, TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
* SPO: 1 970-389-930
------- |