EPA-670/2-75-015
April 1975
Environmental Protection Technology Series
PILOT PLANT OPTIMIZATION OF
PHOSPHORIC ACID RECOVERY PROCESS
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
EPA-670/2-75-015
April 1975
PILOT PLANT OPTIMIZATION OF
PHOSPHORIC ACID RECOVERY PROCESS
By
Leslie E. Lancy
Fred A. Steward
James H. Weet
Lancy Laboratories
Division of Dart Industries, Inc,
Zelienople, Pennsylvania 16063
For
Douglas & Lomason Company
Detroit, Michigan 48208
Project No. S802637
Program Element No. 1BB036
Project Officer
John Ciancia
Industrial Waste Treatment Research Laboratory
Edison, New Jersey 08817
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
REVIEW NOTICE
The National Environmental Research Center—Cincinnati
has reviewed this report and approved its publication. Approval
does not signify that the contents necessarily reflect the
views and policies of the U. S. Environmental Protection Agency,
nor does mention of trade names or commercial products con-
stitute endorsement or recommendation for use.
-------
FOREWORD
Man. and his environment must be protected from the adverse
effects of pesticides, radiation, noise and other forms of
pollution, and the unwise management of solid waste. Efforts
to protect the environment require a focus that recognizes the
interplay between the components of our physical environment--
air, water, and land. The National Environmental Research centers
provide this multidisciplinary focus through programs engaged in
studies on the effects of environmental
contaminants on man and the biosphere, and
a search for ways to prevent contamination
and to recycle valuable resources.
The studies for this report were undertaken to optimize
and evaluate the economics of a new acid regeneration process
which permits the recovery of phosphoric acid used in the bright
finishing of aluminum. The unique waste treatment process uses
sorption, a physical process, rather than chemicals, to effect
separation of the valuable acid from contaminating metal salts.
This closed-loop type of approach prevents pollution of the water
environment by purifying the contaminated phosphoric acid for
reuse in the bright finishing operation.
A. W. Breidenbach, Ph.D.
Director
National Environmental
Research Center, Cincinnati
iii
-------
ABSTRACT
A unique process for separating strong acids in aqueous solu-
tions from their associated metal salts has been developed
by Lancy Laboratories. Based upon preliminary pilot plant tests,
the Douglas & Lomason Company applied for EPA support on a
project to demonstrate the application of the new technique
for reclaiming phosphoric acid used in the bright finishing
of aluminum.
Phase I of the resultant project included improvement and optimi-
zation of the process in a pilot plant contactor and detailed
studies of the economic feasibility of the proposed commercial
installation. The pilot work indicated that with feed streams
ranging from 25-35% phosphoric acid content, at least 75% of
the HsPO^ fed to the contactor could be recovered and returned
to the bright dip tank.
Projections for the commercial installation, based upon pilot
plant data, show that the required investment capital can be
returned in 1.5 years or less when both the recovered acid
volume and the obviated waste treatment costs are considered.
For those plants returning 35% waste acid to their supplier, the
additional savings generated by the recovery system will return
the investment in approximately two years.
The new waste treatment process has considerable potential for
controlling environmental problems based on the significantly
reduced waste treatment costs and sludge production achieved in
the pilot plant study, as well as its widespread application to
other strong acid wastes.
This report was submitted in fulfillment of Project S802637 by
the Douglas & Lomason Company under the partial sponsorship of
the Environmental Protection Agency.
iv
-------
CONTENTS
Page
Abstract
List of Figures
List of Tables
Acknowledgements viii
Sections
I Conclusions 1
II Recommendations 2
III Introduction 3
IV Optimization Study 9
V Feasibility Study 19
VI References 27
-------
FIGURES
No. Page
1 Schematic Flow Diagram for Recovery Process 8
2 Photograph of Pilot Plant 10
3 General Arrangement of Recovery System 21
vi
-------
TABLES
No« Page
1 Pilot Plant Operating Data 12
2 Pilot Plant Operating Data 13
3 Pilot Plant Operating Data 14
4 Pilot Plant Operating Data 16
5 Pilot Plant Operating Data 17
6 Summary of Capital Costs 22
7 Estimated Direct Annual Operating Costs 24
vii
-------
ACKNOWLEDGEMENTS
Assistance in gathering data on the operating costs, chemical
consumption, and production loads at the Cleveland works of
Douglas & Lomason was provided by Mr. Harry A. Lomason,
Mr. Charles A. Parker, Mr. G. Lyndon Milek, and Mr. S. D. Cramer,
all of the Douglas & Lomason Company.
The operation of the pilot plant, design of the recovery
system, and preparation of this report were supervised by
Leslie E. Lancy, Ph.D., of Lancy Laboratories, Division of Dart
Industries, Inc., Chemical Group, Zelienople, Pennsylvania.
The pilot plant operation and collection of data, and the design
of the full-scale recovery system were directed by Mr. James H.
Weet of Lancy Laboratories. The report was prepared by
Mr. F. A. Steward, also of Lancy.
The advice and cooperation of Mr. John Ciancia, Project Officer
and Chief of the Industrial Pollution Control Branch, Industrial
Waste Treatment Research Laboratory—NERC Cincinnati, EPA, is
acknowledged with sincere thanks.
This report was submitted in fulfillment of Project No. S802637
under the partial sponsorship of the Environmental Protection
Agency, United States of America.
viii
-------
SECTION I
CONCLUSIONS
1. In an aluminum bright dipping operation, the projected
acid regeneration and recovery system will significantly
reduce the consumption of both process and waste treatment
chemicals and can generate sufficient savings to
return the required investment in a relatively short
time.
2. Based upon pilot plant performance, it is anticipated that
a full-scale, 24" dia. contactor can separate the aluminum
from 1,752 gal. (6,624 1) per day of 35% feed liquor, with
at least 75% recovery of the input HsPO^. This level of
recovery corresponds to savings of $166,000/year in raw
purchased acid at the current rate of $218.40/ton of 80%
phosphoric acid.
3. Comparison of the in-house recovery technique with return
to the supplier of 35% waste acid, as currently practiced
at many large plants, indicates that invested capital can
be returned in about two years with the additional savings.
4. Demonstration of the proposed commercial installation will
make available to the metal finishing trade an important
tool for reducing consumption of raw materials and costs,
and greatly reducing the environmental impact associated with
operation of acid solutions used to bright dip, clean, or
pickle metals.
-------
SECTION II
RECOMMENDATIONS
Phase II of the demonstration project should be undertaken to bear
out the efficacy of the recovery system under full-scale plant
operating conditions. This Phase will cover one year of operation
and will result in factual data on the economics of the process.
Potential market opportunities for the by-product aluminum
phosphate solution should be explored and compared. There will
undoubtedly be some industry which can use this material as an
input to their process, thus eliminating waste treatment and
sludge generation, and offering additional economic return.
Investigation on the applicability of the sorption technology
to other commercial acid regeneration processes should be
pursued.
-------
SECTION III
INTRODUCTION
As a manufacturer of hardware, accessories, and trim for the auto-
mobile industry, the Douglas & Lomason Company produces bright
trim for automobile bodies in its Cleveland, Mississippi plant, by
stamping aluminum into shape and giving the surface a bright
corrosion-resistant finish. In the finishing operation, the
parts are immersed in a bright dip which is a mixture of phos-
phoric and nitric acids that chemically polishes the surface of
the aluminum, giving a bright, highly reflective surface.
After rinsing, the parts go through a sulfuric acid anodizing
process to develop a durable protective anodic coating.
This type of bright finishing process is common in the aluminum
industry, and all commercial bright dipping operations utilize
a solution based on phosphoric acid. It is estimated that
approximately 45,000 tons of phosphoric acid are consumed annually
in the United States for such aluminum bright dipping operations.
The operating cost for the bright dip is a significant part of
the total aluminum finishing cost because phosphoric acid is a
relatively expensive chemical, and the process is rather ineffi-
cient. Since the bright dip solution must be operated at a high
acid concentration (60-80%), it has a high viscosity, rather like
a thick syrup. Furthermore, the parts must be moved quickly from
the bright dip solution into a subsequent rinse so as to stop
the attack of the surface film as quickly as possible, to
avoid etching which dulls the surface appearance. The high
viscosity and rapid transfer requirements combine to create
extremely high drag-out of the bright dip solution on the product.
In practice, this means that only 10-15% of the purchased phos-
phoric acid added to the bright dip is actually consumed in
chemically polishing the surface. The remainder is lost to the
subsequent rinses.
In addition to being an expensive operation, the bright dipping
of aluminum generally creates a serious waste water disposal
problem. Since phosphate is a nutrient ion and should not be
discharged to surface waters, lime must be used for neutraliza-
tion of the rinse waters. This neutralization creates large
volumes of calcium phosphate sludge which is low in dry solids.
One pound of H3P04 in a dilute rinse water stream will produce
30-50 pounds of wet, gelatinous sludge requiring dipsosal. A
-------
typical method for disposing of such waste slurries is to build a
series of lagoons on the plant property, but this is obviously
not a long-term solution.
Current Practice
The Cleveland, Mississippi plant reduces their waste discharge
and recoups a portion of the value of the acid lost from bright
dipping by operating a series of three recovery rinses following
the bright dip. These are operated in a countercurrent fashion,
with fresh water input to the third rinse, and a very slight
overflow on the first (most concentratedl rinse going to a
storage tank. The flow rate of the rinse series is controlled
so as to maintain 35% acid in the tank overflowing to storage.
The accumulated 35% acid is hauled by tank truck to a fertilizer
manufacturer and the plant is reimbursed approximately 36% of
the full value of the phosphoric acid content.
This recovery by sale to a by-product user is not uncommon in the
industry, and is an improvement over complete wastage. However,
considerable improvement could be made by returning the acid to
the bright dip solution, thus recovering 100% of the initial
acid value, and eliminating dependence on the needs of a by-
product user. Such a recycling operation requires an economical
method for separating the aluminum phosphate formed by the
chemical polishing reaction from the remaining free phosphoric
acid.
New Approach
Recognizing the requirement in the aluminum finishing industry,
Lancy Laboratories undertook research and development work on
a process which would separate aluminum phosphate and phosphoric
acid, relying on an empirically-observed and theoretically
explainable paradox, called by the inventor, "acid retardation."1
This ingenious observation is based on the fact that acid is
retarded, or held back, when a solution containing both a strong
acid and its salts is passed through a column of strong base
anion exchange resin, but has never been demonstrated earlier
on a pilot plant or commercial scale. The mechanism appears
to be a purely physical adsorption with the entire acid mole-
cule being preferentially adsorbed to the exclusion of the
acid salts. Probably the most significant advantage of the
acid retardation technique is that the acid molecule can be
-------
desorbed from the resin by washing it with water. The resin
surface, now freed of the adsorbed acid, is ready for the next
loading sequence, allowing repeated usage of the resin solid
as an adsorber.
To obtain maximum efficiency from the acid retardation principle,
it is necessary to use a continuous, countercurrent contactor to
expose the resin bed to both feed and stripping solutions. This
permits the metal salt stream (aluminum phosphate) to be drawn
from the column continuously at maximum salt and minimum free
acid concentration. Likewise, the product acid can be continu-
ously withdrawn at its maximum concentration and with minimum
contamination by aluminum. This greatly reduces the equipment
and energy required for evaporating the acid before returning it
to the bright dip. The most important feature of such contactor
design is naturally the avoidance of mixing of the zones of
significantly different solutions of acid and metal salt content.
Another important aspect is the avoidance of undue dilution of
the acid to be reclaimed.
As part of the development program, Lancy Laboratories constructed
a pilot plant contactor device, designed specifically for the
adsorption-desorption process. Early runs on the pilot plant
showed encouraging results and indicated that further development
work was justified.
Background
An earlier approach to recovery of similar bright dip acid
employed a continuous pulsed ion exchange column, but the mechanism
of acid-salt separation was conventional ion exchange. A cation
exchange resin removed aluminum ions from the waste acid stream,
and was subsequently regenerated with sulfuric acid. At least
one commercial installation2 using cation exchange is in opera-
tion, but the chemical, operating, and maintenance costs offset
the value of the recovered acid. The acid retardation approach
has an important advantage in that there is no chemical consump-
tion in the operation.
The requirements for a contactor for the new adsorption process
were different than for the existing ion exchange systems. The
new contactor, aiming for a physical adsorption and desorption,
had to be developed, and is covered by the patent application
-------
bearing U. S. Serial Number 241,639. This system accurately
measures the volume of the various feed streams so as to con-
sistently maintain the necessary balance in the contactor with
a minimum of mixing and dilution in the zones of differential
concentration and no operator attendance.
Construction of the first pilot plant configuration was followed
by an extensive period of evaluation during which various
aspects of the electrical, mechanical, and hydraulic design were
modified to obtain optimum economy in operation. From the
reasonably stable, promising preliminary pilot-scale demonstra-
tion runs, progress was made to a more dependable, stable
operating system with significantly improved economical justi-
fication to the point where the refinements accomplished in the
pilot stage allowed dependable predictions for the consideration
of a full-scale commercial installation.
Demonstration Grant
Since the process is unique, an Application for Federal
Assistance with the proposed prototype installation was filed
with the U. S. Environmental Protection Agency. It was pro-
posed that the demonstration project be conducted in two phases.
Phase I would be a Process Optimization and Feasibility Study.
During this phase, the operation of the pilot plant would be
tailored to a feed stream simulating that which is available
at the Cleveland, Mississippi plant, so as to determine the
operating condition which would give maximum efficiency and
recovery of the wasted acid. After the Optimum Conditions had
been determined, it was planned that a sample of feed liquor from
the Cleveland plant would be processed on the pilot plant over a
period of at least a week to insure consistent and repeatable
results. The Feasibility Study was to be primarily an economic
analysis considering the quantities of waste acid available at
the Cleveland plant, the capital investment required for a
suitable contactor installation, the continuing operating and
maintenance costs for the recovery facility, and the value of
the recovered acid. Based upon the results of this analysis,
it could be decided whether to proceed with the prototype
installation.
-------
Phase II of the proposed demonstration project would cover the
construction and installation of the prototype recovery system
at the Cleveland, Mississippi plant, and documentation of the
first year's operating experience and operating and maintenance
costs. Figure 1 shows a schematic representation of the new
phosphoric acid recovery process.
Because of the potential environmental significance of the process
in permitting recycling of common metal finishing acid solu-
tions, the EPA indicated an interest in participating in the
project. Approval for Phase I of the grant was received cover-
ing a three-month period. A final report on the entire demon-
stration will be written after one year of full-scale operation.
-------
ACID REGENERATION .» RECOVERY
AS APPLIED TO THE ALUMINUM BRIGHT DIPPING PROCESS
r
Figure 1. Schematic Flow Diagram for Recovery Process
-------
IV. OPTIMIZATION STUDY
One of the first tests done during the grant period was to compare
the effectiveness of various resins as an adsorption medium under
laboratory static column conditions. On the basis of these
tests, it was determined that Dowex SBR-P resin showed the best
overall characteristics for the process. The adsorption capacity
was equal to or better than the other resins tested, and the
superior hydraulic and mechanical characteristics which are
extremely important in a moving-bed contactor, were a decided
advantage. Therefore, the pilot plant was charged with the Dowex
resin to begin the grant period tests.
Figure 2 is a picture of the pilot plant contactor used for the
tests. With the exception of the diameter of the resin loop
tubing, and the size of the pumps and reservoir tanks, it is
identical to the commercial unit contemplated. The control
system is for all practical purposes identical. The pilot unit
is constructed with 2" dia. Pyrex glass pipe as the resin loop,
giving a cross-sectional area of 3.14 sq. in. (20.26 m2). The
rate of introducing .both feed liquor and stripping water varied
during the tests between 0.40 gal./hr. (1.5 liter per hour) and
0.66 gal./hr. (2.5 liters per hour), giving a bed flux rate
in the range of 0.13 gal./hr. - square inch (0.074 liters per
hour - square centimeter) and 0.66 gal./hr. - square inch
(0.123 liters per hour - square centimeter). The resin bed
in the sorbtion contactor is periodically pulsed hydraulically
so as to move it an increment (approximately 10 inches -
25.4 cm) in a counterclockwise direction around the resin loop.
Between these pulses, the resin valves close, dividing the con-
tactor loop into a loading and stripping zone. In these areas,
the liquid flow progresses opposite to the direction of resin
movement (clockwise). Thus, the net effect is an approxi-
mation of continuous countercurrent flow of the resin bed and
the liquid feeds. Ideally, the best approximation would be
obtained by very small incremental pulses made frequently.
However, the frequency of the pulses is limited by the mechanics
of the contactor operation, since a given number of functions,
each requiring time, must occur between each pulse. Thus,
assuming that the pulse frequency is fixed, the only variable
permitting adjustment of the contactor capacity is the stroke
length. Increasing the stroke length moves more resin around
the loop per unit of time, thus tending to increase the capacity
of the unit. However, this tendency is opposed by the facts that:
-------
Figure 2. Pilot Plant
10
-------
(1) The adsorption reaction is time dependent; and
(2) The turbulence during pulsing causes mixing
along the length of the sections, thus disturbing
idealized equilibrium conditions.
Previous pilot plant work had established that the stroke length
employed was a reasonable compromise between capacity and separa-
tion efficiency.
Table 1 shows the data obtained during the initial pilot runs
in the grant period.
A change was made to permit warm water to be used in the strip-
ping section, as some rough laboratory tests had indicated that
it would more thoroughly strip the resin in the same amount of
time, thus giving a small improvement in the capacity of the
resin as it entered the loading section. To avoid exceeding
the temperature tolerance of the resin, the temperature of the
strip water was held at 120° F (49° C). Table 2 gives the opera-
tion data obtained over several days of operation following this
modification. The increase in acid recovery efficiency may be
directly attributed to the increased capacity of the resin.
A study of the previous operating data led to the conclusion
that longer contact time in the loading section should further
increase the amount of acid adsorbed on a unit volume of resin,
thus increasing the rejection of aluminum, and the recovery of
acid at the same aluminum contamination level. Two additional
sections of pipe were added to the loading section increasing
its length by two feet (0.61 m). As can be seen from the data
in Table 3, the anticipated improvement in acid recovery was
realized.
As explained earlier, one of the most important aspects of the
contactor design is maintaining extremely accurate control over
the input of pulse and counterpulse liquid. Since water under
pressure is used to move the resin bed, there is a natural slip-
page of the water through the porous bed, and the volume of slip-
page is dependent on the condition of the bed. Since the amount
of undesirable particulate matter and the compaction of the bed
can vary from hour to hour, it is essential that the volume of
the slippage be measured and precisely opposed by the counter-
pulse volume. The method of accomplishing this is described
completely in the previously-referenced patent application, and
it will be seen that two differential pressure controllers are
key components in the system.
11
-------
Table 1
Pilot Plant Operating Data
Date
Time
Feed Stream
Waste Stream
Product Acid
Acid
8/2/73
8/3/73
8/6/73
8/7/73
8/8/73
S/9/73
0300
0900
1400
0300
0700
2000
0300
0900
1600
0900
2300
0800
0400
0700
0900
1600
2300
0400
1/hr.
1.5
2.0
1.9
2.0
2.0
1.9
1.6
1.5
1.5
1.7
1.5
1.8
1.4
..5
1.9
1.4
1.4
2.0
Al
2/1
16.3
16.3
15.9
16.6
16.6
18.5
15.8
15.8
20.1
17.8
17.8
17.8
17.8
17.8
17.8
17.0
17.0
15.4
H3P0lt
270
270
285
343
343
353
338
338
358
348
348
348
348
348
348
353
353
343
1/hr.
1.75
2.0
2.25
3.25
3.25
2.5
1.75
1.6
1.5
2.75
1.40
1 -5
1.35
1.5
2.3
1.35
1.29
1.9
Al
2/1
16.5
17.0
14.5
10.9
12.1
13.8
13.3
13.8
17.1
11.3
16.4
15.5
17.8
15.7
16-1
16.1
16.7
14.5
H3POi,
88.2
78.5
73.5
49
49
23.5
78.4
64.5
83.0
58.8
78.4
68.5
98
98
87.3
88.2
98
98
1/hr.
1.5
2.0
1.9
1.5
1.5
1.6
1.5
1.5
1.4
1.32
1.42
1.4
1.2
1.25
1.3
1.25
1.30
1.6
Al
2/1
2.3
2.5
1.9
2.0
1.3
0.8
.75
1.25
2.5
0.8
1.6
1.7
2.3
2.3
2.7
1.7
1.8
1.4
^^»*« ****^wh
H aPO^Recovery
g/l
245
244
245 73.0%
245
240
221 67.5%
196
230
235 62.5%
250
235
226 64.0%
281
260
271 58%
230
216
260 72.0%
•r*.k bULI^Al 14AII
Removal
89.0%
93.0%
91.5%
92.0%
89%
90.0%
-------
Table 2
Pilot Plant Operating Data
Date
8/16/73
8/17/73
8/20/73
8/21/73
8/28/73
Time
Feed Stream
Waste Stream
Product Acid
1300
1600
2000
0900
1200
0300
0730
1200
1630
0900
1400
0500
1300
1600
1600
1/hr.
1.9
2.0
2.0
2.2
2.2
2.0
2.0
1.9
2.0
1.8
2.1
2.1
2.1
1.8
2.0
Al
L. g/i
15.7
15.7
15.7
15.4
15.4
16.5
15.4
15.4
15.4
16 5
16.5
16.5
16.5
16.5
18.3
HaPOi*
g/l
343
348
343
343
343
348
343
343
343
348
348
348
348
348
343
1/hr.
1.9
1.8
2.0
2.3
1.9
2.0
1.75
1.80
1.95
2.0
3.4
3.75
1.58
1.83
1.65
Al
g/i
2.6
1.8
1.7
14.7
13.6
10.9
10.5
11.8
12.8
5.3
11.7
12.4
10.1
9.3
15.1
H3P04
g/
15
14
14
93
93
49
49
54
64
63
63
68
54
49
98
1_
.3
.7
.9
.7
.7
.6
1/hr.
1.85
1.80
1.90
1.6
1.6
1.7
1.75
1.80
1.80
1.63
1.54
1.80
1.5
1.73
1.65
Al
g/i
2.6
1.8
1.7
1.3
1.8
2.9
1.8
2.0
2.8
2.9
2.4
1.4
1.3
2.6
1.6
HaPOit Recovery Removal
g/i
230.3
215.6
264 67%
260
304
269 69%
318.5
250
249 71%
305
289 76.5%
225
304
299
294 76.5%
88%
93.0%
87.5%
84.0%
91.0%
-------
Table 3
Date
9/7/73
9/11/73
9/12/73
9/14/73
Time
1900
2200
0800
1000
1200
1030
1130
1330
1530
1230
1430
1500
Pilot
Feed Stream
1/hr.
2.0
2.0
2.5
2.0
2.1
1.5
1.8
2.0
2.0
2.0
2.0
2.4
Al
2/i
18.3
18.3
16.6
16.6
16.6
16.8
17.2
17.2
17.2
16.8
16.8
19.1
U3PO^
343
343
490
490
490
490
490
490
490
490
490
343
Plant Operating Data
Waste Stream
1/hr.
1.65
1.75
2.0
1.9
1.5
1.3
1.8
1.7
1.8
1.3
1.6
1.8
Al
2/1
15.6
15.9
16.8
17.3
16.5
17.0
18.4
18.4
18.1
16.9
17.3
15.9
98
78.5
117.5
122.5
112.5
172
98
98
113
147
117
733
Product Acid
1.8 1.5
1.75 1.3
1.75 1.7
1.25 2.1
1.8 1.7
1.8 1.9
1.8 1.7
1.30 2.2
1.30 2.1
1.85 1.4
Acid Aluminum
Recovery Removal
392
416
441
396
81.5%
77.5%
360
375
368 69%
430
470
2.85 73%
90.0%
91.0%
92.5%
90%
-------
Repeated observation of the liquid level in the measuring tanks
for the pulse and counterpulse water indicated that the desired
precision of measurement was not being obtained. The trouble
was traced to the mechanical differential pressure switches which
were being used to compare liquid levels in these tanks.
After numerous attempts to improve the operation of these
switches, it was concluded that the inherent limitations of a
mechanical switch were an insurmountable obstacle. Therefore,
electronic differential control instruments were installed.
The transducers at the measuring tanks transmit a milliamp sig-
nal to electronic controllers in the master console, allowing
extremely precise and repeatable control points to be held.
This control system change was the most significant improvement
arising from the pilot plant operation, since it gave the
double benefit of improved acid recovery, and greatly reduced
operator attendance. Previously it had been necessary to make
several adjustments each day to ensure that the various liquid
phases were properly positioned in the contactor. Minor varia-
tions between these adjustments obviously led to inefficiencies
and loss of acid recovery. After installation of the electronic
control system, it was found that little or no operator adjust-
ment was required/ even after days of continuous operation. Thus,
it is assumed that one operator inspection per 24-hour period
will be adequate for maintaining the desired balance in the full
scale system. Table 4 shows the results of several days of
operation at an average acid recovery close to 78%.
Close examination of the entire control system indicated that
a number of seconds could be eliminated from many of the func-
tions occurring between pulses. The greatly increased response
time of the electronic differential pressure controllers also
allowed a time savings. As a result, adjustments were made in
the program at the control cabinet/ so as to reduce the amount
of time required between feed cycles. The net result was an
approximate 25% increase in the amount of resin pulsed per hour,
permitting an increase from the previous average feed rate of
0.40 gal./hr. (2.0 liters/hour) to 0.66 gal./hr. (2.5 liters/hour)
with maintaining near 75% efficiency of acid recovery. Table 5
gives the pertinent data from a number of days of operation at
the higher feed rate.
15
-------
Table 4
Pilot Plant Operating Data
Date
9/25/73
9/26/73
9/27/73
10/2/73
10/3/73
10/4/73
10/5/73
Time Feed Stream Waste Stream Product Acid Acid Aluminum
Al
1/hr. g/1
1700
1900
1300
1500
1700
1100
1400
1700
J600
1400
1600
1800
1900
2100
1200
1400
1600
1800
1500
1700
1200
2.3
2.0
2.1
2.1
2.0
2.2
1.8
1.8
2.0
2.0
2.0
2.0
2.2
2.0
2.0
2.2
2.3
2.1
2.1
2.0
2.3
19
19
15
15
15
15
15
15
18
15
15
15
18
18
12
18
18
18
12
12
10
.1
.1
.7
.7
.7
.7
.7
.7
.5
.9
.9
.9
.5
.5
.3
.5
.5
.5
.3
.3
.9
H3POk
343
343
345
345
345
345
345
345
490
490
490
490
490
490
392
490
490
490
392
392
353
1/hr.
1.9
2.0
1.9
1.8
1.9
2.2
1.9
2.0
1.8
1.8
1.7
1.8
1.8
1.8
2.0
2.2
1.9
1.9
2.0
2.0
1.9
Al
2/1
14.5
16.1
12.6
11.9
11.1
12.0
10.7
11.0
14.7
15.6
15.9
16.5
14.9
15.0
11.0
13.6
13.1
13.9
11.4
11.6
8.8
H3POit
73
73
78
73
73
68
68
68
98
117
117
117
98
98
73
78
78
78
73
73
58
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.8
1/hr.
2.0
2.0
1.8
1.8
1.8
2.0
1.8
2.0
2.0
2.0
1.8
1.9
2.0
2.0
2.0
1.85
2.0
2.1
2.0
2.0
1.9
Al
9/1
1.8
2.1
1.4
1.4
1.4
1.4
1.7
1.9
1.4
1.5
1.7
1.8
1.4
1.6
1.5
2.2
1.9
1.9
1.5
1.4
1.4
HSPOI+ Recovery Removal
324
304
294 81.0% 89.0%
294
294
270
270
270
392 80.5% 89.0%
345
345
357 69% 90%
392
392
392 82% 90%
392
392
392 77% 90%
245
250
216 77% 88%
-------
Table 5
Pilot Plant Operating Data
Date
10/11/73
10/15/73
10/16/73
10/17/73
10/18/73
Time
1400
1600
0900
1000
1400
1800
1100
1300
1200
1000
1300
1600
1400
1800
Feed Stream
1/hr.
2.7
2.5
2.5
2.5
2.5
2.5
2.5
2.6
2.5
2.6
2.4
2.5
2.5
2.5
Al
10.9
10.9
10.0
10.4
10.4
10.4
10.0
10.0
10.5
10.0
10.0
10.0
10.5
10.5
HsPOt,
353
353
343
363
363
363
343
343
363
363
363
363
363
363
Waste Stream
1/hr.
2.3
2.2
2.3
2.2
2.2
2.4
2.3
2.3
2.3
2.4
2.3
2.5
2.3
2.3
Al
2/1
8.9
8.9
8.5
8.5
8.4
8.5
8.6
8.1
9.6
8.6
8.6
8.6
9.5
9.4
58.8
68.6
78.4
117.6
112.5
112.5
78.4
78.4
78.4
98
98
98
88.2
88.2
Product Acid
1/hr.
2.0
2.1
1.9
2.1
2.0
2.1
2.0
2.0
2.35
2.0
2.0
2.2
2.30
2.50
Al
2/t
1.3
1.4
1.6
1.5
1.5
1.5
1.5
1.3
1.7
1.6
1.5
1.5
1.4
1.5
q/1
245
260
274
289
289
289
274
274
304
309
309
309
284
284
Acid
Recovery
79%
68%
75%
72%
79%
Aluminum
Removal
87%
88%
87%
87%
86%
-------
One recurring mechanical problem with the operation of the
pilot plant was failure of the automatic valves in the various
feed and take-off lines connecting to the resin loop. Precise
control of the volumes entering and leaving the pressurized
resin loop, and thus the overall process balance, requires
that these control valves open and close rapidly, and that their
shut-off be absolutely leak-tight.
PVC-body, Teflon-faced diaphragm-type valves were used in the
original pilot plant construction. In order to assure the neces-
sary rapid, leak-tight shut-off of the diaphragm valves, a rather
high operating air pressure was required. Because of this high
pressure, a valve which remained in the closed position for
any period of time developed a deep imprint on the diaphragm
face, resulting in premature failure. Tests with a number
of alternate diaphragm materials gave only marginal improve-
ment. As a result, early in the grant period, the automatic
valves were replaced with stainless steel body, Teflon-sealed
plug valves with pneumatic operators. Shut-off on these was
extremely rapid, and completely leak free, and inspection of
^ne units after three months of nearly continuous operation
showed no noticeable wear. In view of the fact that the con-
tactor system includes twenty automatic valves, evaluation of
plug valves was an important step in assuring reliable per-
of the full-scale installation.
18
-------
V. FEASIBILITY STUDY
Determination of the feasibility of constructing the full scale
contactor system required that accurate cost figures be obtained
for the assembly and installation of the equipment. Such figures
could only be obtained by doing a detailed design on the full
scale contactor and obtaining quotations on the supply of the
raw materials and on the necessary welding and fabrication work
required for the stainless steel contactor loop, the evaporator,
and the structural steel support tower. Quotations were also
required on all accessory equipment, such as tanks, pumps,
heat exchangers, controls, valves, and piping.
Early in the grant period, design was started on a full scale
contactor, based on a resin loop of standard 24" dia., Schedule
10 stainless steel pipe, having an actual inside diameter of 23.5"
(59.69 cm). This size was chosen because the single existing
anodizing line at the manufacturing plant was generating 1350 gal-
lons (5,110 liters) per day of 35% waste phosphoric acid. The
24" column has a cross-sectional area of approximately 434 sq. in.
(2,800 cm2) which is 138 times that of the pilot unit. Based
upon a normal capacity of 2 liters per hour for the pilot plant,
the indicated capacity of the 24" unit is 73 gal./hr.
(276 liters/hr.) for a total daily throughput of 1,752 gallons
(6,624 liters). Use of the next smaller standard size pipe
(22") would have given only marginal capacity when calculated
by this method. Conservative design requires that a safety
factor be allowed when scaling up from the pilot plant results.
For example, distribution is one of the problems encountered when
going to larger cross-sectional areas. Design of the in-flow
distributors must aim for uniform distribution across the entire
cross-section of the resin bed, and yet avoid excessive inter-
ference with movement of the compacted resin bed during pulsing.
The excess capacity indicated by the calculation above, combined
with the fact that the concluding runs during the pilot plant
optimization study were at 2.5 1/hr., should be more than ade-
quate provision for the uncertainties of scale-up.
A square tower of structural steel was intended to support the
resin loop, and provide mounting supports for all auxiliary
equipment, such as tanks, pumps, piping, and valves. This tower
would also include several grating platforms to provide access
to the entire length of the resin loop. Since the overall
height of the tower was approximately 47', some provision for
19
-------
housing and weather protection was necessary. Therefore, the
support tower was designed large enough to completely enclose
all fittings and connections to the contactor, and to provide
space for an access ladder to each of the platform levels.
Since appropriate wind loading factors were considered, sheet
metal siding and a roof can be applied directly to the support
structure, forming a convenient enclosure.
It is planned that the existing 35% waste acid storage tank be
used as a source of feed to the recovery system. Figure 3 shows
the intended layout of the recovery system and associated tanks
adjacent to this existing storage tank.
Upon completion of the engineering plans for the complete re-
covery system, firm quotations were obtained on all of the re-
required materials and equipment, as well as the necessary labor
for assembly and welding of the stainless steel resin loop, and
the support tower. It is intended that the support -tower
be constructed in modules, which can readily be stacked and
connected on site, so as to minimize the field erection time and
labor. Quotations were also received from contractors in the
Cleveland, Mississippi area for the erection, siding installa-
tion, and piping and electrical connection work. Table 6
itemizes the various items of capital cost, and indicates the
total investment required to have a completely operational
phosphoric acid recovery system. The costs associated with
conducting, and report on, the demonstration project are not
included in this listing, since they have no influence on the
economics of the recovery scheme.
Calculations on the economic feasibility of the proposed recovery
process can be based on either the total capacity for the 24"
unit—1,752 gallons (6,624 liters) per day—or on the present
quantity of water—1,350 gallons (5,110 liters) per day. The
former approach is applicable, since a second, parallel ano-
dizing line is being installed, so that sufficient waste
phosphoric acid will be available to fully utilize the capacity
of the recovery system. Thus, at this higher rate of feeding
the 35% waste acid, 6,500 Ibs. (2,954 Kg) of E^PO^ will be fed
to the contactor per day. Based on a very conservative 75%
recovery of the feed acid, 4,875 Ibs. (2,216 Kg) of H3POtt will
be fed to the evaporator for concentration into 448 gallons
(1,696 liters) per day of 80% phosphoric acid. The plant is
currently paying $218.40/ton for 80% phosphoric acid solution,
giving the 3.04 tons of recovered solution a value of $664.00
per day, or $166,000 per year.
20
-------
: PLANT BUILDING
GO
OQ
PRODUCT
'-y '
LCOLUMN
^- CONCRETE -
OSOLN. [ ACID \ f ACID \
1 WASTE I ( SUPPLY 1
O BYPRODUCT \^_^X \^__^/
v_
v/gvv
PLANT
BUILDING
1
^-SUPPORT TOWER 4 ENCLOSURE
! T-r7
1 L_ '
•4^K »* LuB _^_
.!:• * T
i
— \
N, f
—i
I
i i
1 V-'' /
"*• . /'
•CONTACTOR1 COLUMN
^-WORKING LEVELS(4)
•i ill
i
ELEVATION
-REINFORCED CONCRETE
Figure 3. General Arrangement of Recovery System
21
-------
Table 6
Summary of Capital Costs
1. Equipment - Contactor Resin Loop. ... $ 37,166
Support Structure 19,049
Pipe and Fittings ..... 5,805
Pumps 16,327
Valves. 39,023
Tanks 17,884
Control System 16,106
Miscellaneous 14,950
Evaporator 33^630 $ 199,940
2. Supplies - Resin, 210 ft.3 - Dowex SBR-P ...... 13,230
3. Installation* - Site Preparation and
Erection $ 9,000
Building Construction and
Siding Installation. . . 7,000
Piping and Electrical
Installation 12,000 28,000
4. Engineering Services - System Design. . $ 35,000
Start-up Supervision. . . . 7,800 42,800
Total Invested Capital $ 283,970
*The installation cost figures are estimates based upon actual
quotations from local contractors. They are, of course, sub-
ject to some variation, but are considered conservative.
22
-------
In many cases, waste treatment costs will be reduced by recovery
and reuse of the bright dip drag-out. Associated with the re-
cycle of 4,875 Ibs. (2,216 Kg) of acid are savings of 2.76 tons
(2.5T) per day of lime required for neutralization. This amounts
to an annual savings of $15,180 at $22.00 per ton. Furthermore,
sludge handling costs of approximately $38,500 per year are
avoided, based upon 7,700 gallons (29.2 m3) per day of slurry
containing 10% dry solids, and handling costs of 2<:/gallon.
Adding these savings in waste treatment costs gives a total
return value of $204,500 per year.
Table 7 lists the various items of operating cost for the
recovery system, and indicates a total annual operating cost of
$11,905 per year. This gives a net recovery value of $192,595
per year, allowing amortization of the investment capital in
approximately 1.5 years.
As mentioned earlier, the Cleveland, Mississippi plant presently
collects their recovery rinse overflow and has the 35% acid hauled
away by the supplier of the fresh acid. This is a rather common
practice for the larger aluminum bright dipping operations. Be-
cause of equipment failures and/or operator mistakes, the value
obtained with this approach inevitably falls short of the calcu-
lated amount. In 1973 the plant received a credit of $32,760
for return of the by-product. Adjustment of the 1973 plant capa-
city to a production level corresponding to full contactor capa-
city would give a projected return value of $42,515/year.
The preceding calculations are all based simply upon the value
of the phosphoric acid and the associated waste treatment costs
and ignore several other significant savings which will accrue
because of reduction in the overall operating costs of the ano-
dizing plant. These are due to the fact that the recovery rinses
may be run considerably faster with the in-house recovery system
than with the haul-away approach. While a nominal figure of 35%
has been referred to as the strength of the contactor feed acid,
the sorption process can work effectively with a feed of con-
siderably lower acid concentration. The pilot plant operating
data given in Table 5 were obtained with a feed stream of 26-27%,
and very little efficiency will be sacrificed by going to even
lower concentrations.
By running the recovery rinses at a higher rate, the following
advantages are realized:
23
-------
Table 7
Estimated Direct Annual Operating Costs
1. Personnel - Operator - 2 hrs./day, 250 days
@ $6/hour $ 3,000
Maintenance - 5 hrs./week,
$6/hr 1,872 $ 4,872
2. Supplies - Resin - 42 ft.3 @ $63.00/ft.3 2,646
3. Utilities - Electric - 22.4 KW, 24 hrs./day
250 days, l.OC/KW hr $ 1,344
Fuel for Heating - 15 Boiler
Horsepower, 250 days,
24 hrs./day 3,043 4,387
$ 11,905
24
-------
1. The amount of acid dragged out of the final recovery rinse
into subsequent fast-flowing rinses is reduced, resulting
in less loss and more recovery of acid value. For example,
at a production rate corresponding to full contactor
capacity, running the recovery rinses to give a concen-
tration of 25% instead of 35% will result in recovery of
approximately 72.3 tons (65.6T) per year more acid, with
a value of $15,790.
2. Since waste treatment will inevitably be required for the
subsequent fast-flowing rinses, the savings of acids men-
tioned in (1) above will also result in savings of 81.9
tons (74.3 T) per year of lime, with a value of $1,800 and
will eliminate the handling of 228,700 gallons (865,630
liters) per year of sludge, saving an additional $4,574 per
year. Thus, the total acid and waste treatment savings due
to faster flow in the recovery rinses are $22,164 per year.
3. A major problem in producing high-quality anodized parts is
the build-up of phosphate ion in the anodizing bath, since
it interferes with subsequent sealing operations and re-
duces the corrosion resistance of the surface. A typical
maximum concentration permitted in the bath by automotive
specification is 1,000 mg/1. Carry-over of phosphate from
the bright dip often causes the anodizing bath to be dis-
carded more frequently than would otherwise be required.
Thus, it is obvious that a reduced acid concentration in
the recovery rinses can also produce savings in sulfuric
acid and lime consumption, as well as sludge handling
costs for this secondary waste source. As an example, re-
placement of a 2,000 gallon (7,570 liter) anodizing bath
will cost $75.00 for sulfuric acid, $33.00 for lime, and
$40.00 for sludge handling—a total of $148.00, not
counting the labor for dumping and preparing the new bath.
4. The work pieces are generally moved rapidly through the
recovery rinses to minimize the etching or dulling which
takes place when the surface is immersed in dilute phos-
phoric acid solution. Because of the higher flow rate
employed with the recovery system, longer immersion and
drainage time can be used, further reducing loss of acid.
The product quality can also be appreciably improved.
25
-------
At full contactor capacity, these additional savings, due to the
increased recovery rinse flow rate, may be assumed to be at
least $23,000 per year. Adding this to the previously-mentioned
acid value ($166,000) gives a total return of $189,000/year. In
the comparison with a haul-away approach, this would be reduced
by $42,515 of returned acid credit and $11,905 of operating cost,
to give a net return of $134,580/year. This would completely
return the invested capital in about two years.
One additional factor affecting the economics is the value
of the by-product aluminum phosphate/phosphoric acid stream
exiting the contactor. Investigations are under way to find the
best approach to marketing this material, but since the value has
not been firmly established, no credit has been assumed in the
foregoing calculations.
More accurate figures will be generated by operation of the
prototype unit under actual plant conditions during the second
phase of the demonstration project.
26
-------
SECTION VI
REFERENCES
1. Hatch, Melvin J., and Dillon, John A., "Acid Retardation,"
I & EC Process Design and Development, 2, 4, pp 253-263
(Oct., 1963)
2. Church, F. L., "Bright Dip Breakthrough," Modern Metals,
XIX, 7 (Aug., 1963)
27
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-670/2-75-015
3. RECIPIENT'S ACCESSION>NO.
4. TITLE AND SUBTITLE
PILOT PLANT OPTIMIZATION OF PHOSPHORIC ACID
RECOVERY PROCESS
5. REPORT DATE
April 1975; Issuing Date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Leslie E.
Weet
8. PERFORMING ORGANIZATION REPORT NO
Lancy, Fred A. Steward, and James H,
9. PERFORMING ORG MSIIZATION NAME AND ADDRESS
Lancy Laboratories Douglas & Lomason Company
P. 0. Box 98 for 5800 Lincoln Avenue
Zelienople, Pa. 16063 Detroit, Michigan 48208
10. PROGRAM ELEMENT NO.
1BB036;ROAP 2lAZO;Task 2
11. CONTRACT/GRANT NO.
S802637-01
12. SPONSORING AGENCY NAME AND ADDRESS
National Environmental Research Center
Office of Research and Devleopment
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT "
A pilot plant study was carried out which demonstrated the effectiveness
and economic feasibility of a unique ion exchange process referred to as
"acid retardation" for purifying spent phosphoric acid used in bright
finishing aluminum parts. A continuous ion exchange system was employed
to separate the aluminum contamination from a spent phosphoric acid
waste obtained from a manufacturing plant. The anion resin accomplishes
the separation by retarding the phosphoric acid as the processing solu-
tion flows through the bed. The aluminum remains in the waste solution
and passes out of the column in the effluent. The acid is then eluted
from the bed with water, eliminating the use of chemicals which are
needed to regenerate the resin in conventional ion exchange systems.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Industrial wastes, Waste water,
Waste treatment, Metal finishing,
*Ion exchanging, Bright plating
*Waste recovery, *Nonferrous
metal industry, *Aluminum fabri-
cating wastes, Bright finishing
wastes, *Phosphoric acid wastes,
Wastewater disposal, Wastewater
reuse, Water pollution control,
*Phosphoric acid purification
13B
8, DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
36
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
28
T!!rU.i60WMUIfllTrtlimil60ma: 1975-657-592/5356 Region No. 5-H
------- |