oEPA
Unit*! States
Environmental Protection
Agency
NX*
Laboratory
Cincinnati OH 45268
EPA-600/2-7&-130
July 1978
Research and Development
Aircraft Industry
Wastewater
Recycling
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-78-130
June 1978
AIRCRAFT INDUSTRY WASTEWATER RECYCLING
by
Alan K. Robinson
Donald F. Sekits
Manufacturing Research and Development
Boeing Commercial Airplane Company
Seattle, Washington 98124
Grant: S803073
Project Officer
Hugh B. Durham
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research Laboratory-Cincin-
nati, U.S. Environmental Protection Agency, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and policies of the U.S. Environmental Protec-
tion Agency, nor does mention of trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, converted, and used, the related
pollutional impacts on our environment and even on our health often require that new and increas-
ingly more efficient pollution control methods be used. The Industrial Environmental Research
Laboratory-Cincinnati (IERL-CI) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
This report describes work undertaken to demonstrate the feasibility of recycling certain
categories of water used in the manufacture of airplanes. The results of the work can be used by
those involved in water conservation as a source of technical and cost data that will enable them to
assess the relative merits of recycling compared with other water conservation methods.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
ui
-------
ABSTRACT
This research program was initiated with the objective of demonstrating the feasibility of re-
cycling certain categories of water used in an airplane factory. These categories are: chemical
process rinse water, dye-penetrant crack-detection rinse water, machine shop coolant, and cyanide-
containing rinse water. Water used solely for cooling purposes, and sanitary water, are not included
in the program.
The feasibility of recycling water in each of the four categories was demonstrated in 380-liter
(100-gallon) treatment plants. For each plant, contaminated water was continuously purified, then
recontaminated, in a closed demonstration "loop."
Based on the experiences of constructing and operating the pilot scale treatment plant, an
estimate was developed for the cost of a full-scale water recycling plant. The plant was of a size
suitable for a typical medium-sized airplane factory generating 1.5 Ml/day (0.4 x 10" gal/day). The
estimate was: capital cost including installation, $3.4 million; recycling costs—$0.94/kl (S3.57/
1000 gal) for chemical process rinse water, $1.65/kl ($6.25/1000 gal) for dye penetrant rinse water,
$4.36/kl ($16.50/1000 gal) for cyanide process rinse water, and $12.18/kl ($46.09/1000 gal) for
machine shop coolant.
A color and sound movie, "Closing the Loop," was made that describes the research program.
This IS^-minute movie is suitable for a wide range of audiences from nontechnical ecology groups
to engineers and other specialists in the field.
This report was submitted in fulfillment of Grant Agreement No. S803073-01-1 by the Boeing
Commercial Airplane Company under the sponsorship of the U.S. Environmental Protection
Agency. This report covers the period from August 1, 1974 to August 31, 1976, and work was
completed on October 31, 1976.
IV
-------
CONTENTS
Page
Foreword iii
Abstract iv
Figures vi
Tables vii
Abbreviations and Symbols viii
1. Introduction 1
2. Conclusions ' 2
3. Recommendations 3
4. Pilot-Scale Demonstration of Water Recycling 4
General description 4
Ion-exchange purification of chemical process rinse water 6
Reverse osmosis purification of chemical process rinse water 11
Ultrafiltration and active carbon purification of dye penetrant
inspection rinse water 19
Machine shop coolant waste water purification 26
Cyanide rinse water purification 32
5. Cost of Full-Scale Plant 37
6. Operating Economics of a Full-Scale Plant 41
7. Movie 51
8. Analytical Methods 52
References ,53
Bibliography 53
Appendices
A. Mathematical analysis of three reverse osmosis systems 54
B. Reverse osmosis tests—rejection ratio of nitrates compared with other ions 67
C. Reverse osmosis—experimental verification of mathematical analysis 68
D. Calculator program for computing reverse osmosis unit product and
concentrate concentrations 69
E. Composition of cyanide plating shop solutions 72
F. Equipment list and breakdown diagrams for full-scale water treatment plant 73
G. Operating economics of a full-scale plant—calculations 89
H. Composition of alkaline flushing solution for ultrafiltration membranes 91
-------
FIGURES
Number Page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
General view of demonstration laboratory
Demonstration equipment for ion- exchange purification of chemical
process rinse water
Ion -exchange purification of chemical process rinse water, showing
rapid rise in pH and conductivity at exhaustion
Demonstration equipment for reverse osmosis purification of chemical
process rinse water "
Detail view of reverse osmosis equipment showing reverse osmosis
units
Demonstration equipment for recycling dye penetrant inspection rinse
water
Control panel for demonstration equipment recycling dye penetrant
inspection rinse water
Ultrafiltration unit— permeate flow rate characteristics during flushing
and operation dye penetrant rinse concentrate
Service record of ultrafiltration unit— dye penetrant rinse water puri-
fication
Demonstration equipment for machine shop coolant waste water puri-
fication
Machine shop coolant waste water purification— waste coolant and
recycled water tanks
Demonstration equipment for cyanide rinse water purification
Flow diagram for full-scale cyanide and chemical process rinse water
recycling
Flow diagram for machine shop coolant and dye penetrant rinse water
recycling
5
7
9
12
13
20
21
23
24
27
28
35
38
39
VI
-------
TABLES
Number Page
1 Analysis for Operation of Reverse Osmosis System Counterflow
Arrangement 15
2 Calculated Rejection Ratios 16
3 Comparison of Reverse Osmosis and Ion Exchange Systems 16
4 Comparison of Three Arrangements for Selected Values of R and E. . . . 18
5 Typical Analytical Data for Dye Penetrant Inspection Water Purifica-
tion using Ultrafiltration and Activated Carbon Filtration 22
6 Analyses of Feed and Carbon Filtered Recycled Waters 25
7 Summarized Operating Economics, Recycling System Compared with
Conventional Treatment Plant Discharging to Waste 41
8 Chemical Process Rinse Water, Itemized Operating Economics 44
9 Dye Penetrant Rinse Water, Itemized Operating Ecomonics 47
10 Machine Shop Coolant, Itemized Operating Economics . . 48
11 Cyanide Rinse Water, Itemized Operating Economics 50
Vll
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
SYMBOLS
Amortiz.
ASTM
evap.
ORP
Pa
TDS
UF
uv
//mho/cm
cm
juW/cm
—amortization
—American Society for Testing
and Materials
—evaporator
—oxidation-reduction potential
-Pascal, Newton/meter2
—total dissolved solids
— ultrafiltration
—ultraviolet
— micromhos per centimeter
(same as juS/cm)
-microSiemens per centimeter
(same as ^mho/cm)
—microwatts per square
centimeter
A,Y,D,Z,X —parameters used in math
analysis
Cc —concentration of concentrate
Cf —concentration of feed
—mean concentration
—concentration of permeate
—recovery ratio
cr
cr
E
M
2+
Qc
Qf
QP
R
—(divalent) heavy metal ion
(Cu2+, Cd2+, etc.)
—flow rate of concentration
—flow rate of feed
—flow rate of permeate
—rejection ratio
To Convert
METRIC CONVERSIONS
To
Multiply by
gal 1 3.785 412 E+00
gal/min 1/min 3.785 412 E+00
gal/hr 1/hr 3.785 412 E+00
gal/day I/day 3.785 412 E+00
gal/mo 1/mo 3.785 412 E+00
inches cm 2.540 000 E+00
Ib/hr kg/hr 4.535 924 E-01
Ib/day kg/day 4.535 924 E-01
ppm mg/1 1.000 000 E+00
vm
-------
SECTION 1
INTRODUCTION
To meet the proposed goals of the Federal Water Pollution Control Act for "best available"
technology by 1983, it was felt that a pilot-scale demonstration would be logical before committing
funds for the very high capital and operating costs of a full-scale water recycling plant. Many of the
techniques of water treatment suitable for use in water recycling were well known; some, however,
would require proving tests. A pilot-scale demonstration would provide the opportunity to observe
and test both established and new techniques in a controlled environment.
The water used in an airplane factory falls into three categories: water for sanitary use, water
for cooling air compressors and other machinery, and water for chemical process rinsing and for
making machine shop coolant. This report is concerned only with the third category. Category
two, cooling water, is already recycled by well-known methods, mainly cooling towers, and cate-
gory one, sanitary water, is outside the scope of this work.
The waste water with which this report is concerned originates in four ways in a medium-sized
airplane manufacturing plant. Between 50 and 100 chemical process rinse tanks provide a slightly
acid stream containing small amounts of dissolved chromium, copper, cadmium, zinc, and other
metals. This stream is referred to in the report as the "chemical process rinse water." The second
stream results from 10 to 20 tanks that are used to rinse a crack-detecting oil from the surface of
airplane parts. (Virtually all structural airplane parts are crack tested and rinsed in this way.)
This system is referred to as the "dye penetrant inspection rinse water." A third stream consists of
water-based coolant used in machine shops to assist in the rapid cutting of metals in mills, lathes,
etc. Between 150 and 300 machines contribute to this stream. The last stream consists of rinse
water from electroplating rinse tanks using cyanides. Because of their toxic nature and their ability
to release toxic HCN gas when acidified, these rinse waters require special handling. This stream is
generated by five to ten cyanide rinse tanks. The volumes of each of these streams generated in a
medium-sized plant would be roughly: 1.5 Ml/day (0.4 x 10 gal/day) chemical process rinse water,
0.12 Ml/day (30 x 10 gal/day) dye penetrant inspection rinse water, and 20 kl/day (6x10 gal/
day) each of machine shop coolant and cyanide process rinse water.
The demonstration plant was intended to show the technical feasibility of recycling the water
from each of the above waste streams. A cost analysis, also included in the study, examines the
economic feasibility of this recycling.
-------
SECTION 2
CONCLUSIONS
This pilot scale demonstration of aircraft factory water recycling has demonstrated the tech-
nical feasibility of recycling at least 85% of the water used for chemical process rinsing and machine
shop coolant.
The economic feasibility of recycling depends on the cost of recycled water compared with the
cost of fresh water used in a once-through system, i.e., the initial cost of the fresh water plus the
cost of treating this water to the level required for discharge. As the cost of recycling is almost
twice that of a once-through system in the most favorable case (chemical process rinse water) and
almost five times in the least favorable case (machine shop coolant), the economic feasibility can-
not be said to have been demonstrated.
The above economic feasibility conclusion depends on the assumption that the materials
extracted from the waste have no economic value—which is presently the case.
The overall cost of installing a water-recycling plant in a typical airplane factory treating
approximately 1.5 Ml/day (0,4 x 10° gal/day) is $3.387 million, and the average cost of treatment
in such a plant is $1.21/kl ($4.57/1000 gal).
-------
SECTION 3
RECOMMENDATIONS
To enable recycling to be economically achieved, improved methods of treatment of aircraft
factory process water should be developed. In particular, methods for improving the rejection
of nitrates by reverse osmosis membranes need to be developed. Alternative methods for the treat-
ment of machine shop coolant are also needed to reduce the very high cost of treatment by ultra-
filtration and reverse osmosis.
The development of methods for extracting useful materials from wastes presently trucked
away should also be pursued, and efforts made to develop markets for the materials extracted.
-------
SECTION 4
PILOT-SCALE DEMONSTRATION OF WATER RECYCLING
GENERAL DESCRIPTION
A demonstration laboratory was set up containing working pilot-scale treatment plants. Each
plant took process solutions currently used at Boeing's Plant II facility, Seattle, Washington or at its
Auburn, Washington facility, and diluted them to make simulated rinse water streams for use in the
treatment plants. Each stream was then purified sufficiently to make the water reusable for its
original purpose. The treatment plants were constructed as "loops," i.e., the water—after purifica-
tion—was recontaminated to make more simulated used water which was then repurified and recon-
taminated, etc., in a continuously operating loop.
Five loops were set up:
o Chemical process rinse water purified by ion-exchange resin
o Chemical process rinse water purified by reverse osmosis
o Dye penetrant rinse water purification
o Machine shop coolant water purification
o Cyanide process rinse water purification
The first two loops purified and recycled the same type of waste water but used two different
methods of treatment. The reason for demonstrating two methods of treatment for chemical pro-
cess rinse water was that this type of water forms 80% of the process water from a typical airplane
factory.
Figure 1 shows a general view of the demonstration laboratory.
-------
Figure 1. General view of demonstration laboratory.
-------
ION-EXCHANGE PURIFICATION OF CHEMICAL PROCESS RINSE WATER
General
Figure 2 shows the demonstration equipment for this loop.
Treatment of simulated chemical process rinse water in this loop was comparatively simple,
consisting of filtration (25 jiim) to remove suspended matter, passage through activated carbon to
remove organic matter (particularly traces of oil and surfactants from alkaline detergent cleaner
tanks), and passage through a mixed bed ion-exchange column, to remove ionic contaminants.
Simulation of Rinse Water
Continuous addition of chemical process solutions to the purified recycled water, simulating
the production of chemical process rinse water, was achieved by a steady drip feed via metering
pumps.
— .
ANODIZE 1 DEOXIDIZE
|
1 I 1
r i f i
METERING
PUMPS
\
SIMULATED RINSE
TANK
— -~— —
DETERGENT CHEM-MILL
CLEANER
,J 1 1
1
1
Composition of the metered feed of contaminants was adjusted to the following percentages
to conform to those obtained in a typical airplane manufacturing plant.
Chemical conversion coating (Alodine 1200)* solution 25.5% by volume
Deoxidizer, nitric acid/Amchem 6/16* 22.5%
Chromic acid anodize solution 20.0%
Alkaline detergent cleaner 20.0%
Deoxidizer, sulfuric-chromic acid 7.5%
Aluminum chem-mill (caustic) solution 5.0%
The dilution of this input by the incoming tap water varies in practice over a wide range. For
this demonstration a dilution ratio of 1 in 500 was selected.
A flow diagram for ion-exchange purification of chemical process rinse water follows:
* Amchem Products Inc, Ambler, Pennsylvania, 19002.
6
-------
Figure 2. Demonstration equipment for ion-exchange purification of chemical process rinse water.
-------
METERED FEED OF
CONTAMINANTS
114 l/hr (30 gal/hr)
SIMULATED
RINSE TANK
RECYCLED WATER
SAMPLE TANK
ION-EXCHANGE PURIFICATION SYSTEM
25 MICRON
FILTER
ACTIVE
CARBON
COLUMN
MIXED BED
ION-EXCHANGE
COLUMN
Deletion of Chem-Mill
The loop was operated as planned for the first 80 hours with all six inputs, including the
aluminum chem-mill. However, the presence of sulfides in the chem-mill caused an odor and toxic-
ity problem because small amounts of hydrogen sulfide were liberated in the acid rinse water. Also,
the high aluminum content of the chem-mill solution made it necessary to change filters every 15
hours due to the formation of an aluminum hydroxide precipitate in the rinse water.
It was considered that the cost of modifying the system to remove odor and more efficiently
handle chem-mill rinse was not justified and, accordingly, the chem-mill feed was turned off for the
remainder of the demonstration period.
Quality of Recycled Water
Purification and recycling was accomplished without difficulty in this loop, a high quality
"de-ionized" water being consistently obtained. Quality of the water (less than 0.5 mg/1 total
dissolved solids) far exceeded normal requirements for process rinse water.
Chemical analysis of the simulated rinse water and the purified recycled water typically gave
the following results:
Simulated rinse
Recycled
IDS
NO3
S04
P04
Cl
F
Na
Total Cr
Surfactant
Conductivity
pH
500 mg/1
300 mg/1
80 mg/1
0.3 mg/1
12 mg/1
Less than 3 mg/1
150 mg/1
40 mg/1
0.8 mg/1
630 ^mho/cm
2.8
Less than
— 0.5 mg/1 total
dissolved solids
5.9
-------
A practical test was made of the acceptability of the recycled water for rinsing chemical treat-
ment solutions from the surfaces of an aluminum alloy. In this test (4), the rinsed surfaces were
bonded together with a high strength aircraft structural adhesive, and the joint tested to failure in a
crack extension test. The rinse water was found to be acceptable in this test.
Ion Exchange Resin Capacity
The mixed-bed ion-exchange column became exhausted after 719 hours of operation during
which time it processed a total of 81,642 liters (21,570 gallons). This volume corresponds to a cal-
culated resin capacity of 1.47 meq/ml for the anions, and 2.26 meq/ml for the cations. These values
are in agreement with the expected values of 1.4 meq/ml for the strongly basic, highly crosslinked
anion resin A-244D*, and 2.1 meq/ml for the strongly acidic, highly crosslinked cation resin
C-361W*.
A graph showing the rapid rise in pH and conductivity as the bed became exhausted is shown
in Figure 3.
NORMAL OPERATION
600
620
640 660 680
HOURS OPERATION
700
720
740
Figure 3. Ion-exchange purification of chemical process rinse water, showing rapid rise in pH and
conductivity at exhaustion.
* The Burhans-Sharpe Company, 2255 Harbor Ave. S.W., Seattle, Washington
9
-------
Activated Carbon Capacity
The small amount of surfactant present in the simulated rinse water, 0.8 mg/1, was readily
absorbed by the 54-liter carbon column which showed no signs of exhaustion during the 488 hours
it was in use. The calculated weight of surfactant absorbed in this period amounted to 0.26 g/100 g
of activated carbon.
An attempt to determine the ultimate capacity of the carbon was defeated by the unexpected
ability of the ion exchange column to absorb at least small amounts of surfactant. (No surfactant
has ever been detected in the recycled water due to this effect. Subsequent discussions with the
resin vendors have confirmed that ion-exchange resins, particularly cation-exchange resins, will
absorb surface active agents in addition to the ions they are intended to absorb.) The attempted
tests indicated only that the carbon capacity was less than 14 g surfactant/100 g of activated carbon.
Regeneration of the Ion-Exchange Resin
The amounts of acid, alkali, and water used in regenerating the ion-exchange column were
as follows:
Commercial sulfuric, 66° Baume 93.2% H2SO4 18.8 kg (41.5 Ib)
Commercial sodium hydroxide 22.7 kg (50.0 Ib)
Water 6,9571 (1,838 gal)
Thus, the ion exchange system may be considered as producing two streams: the "product,"
81,642 liters containing essentially zero dissolved matter, and a waste stream of 6,957 liters con-
taining 11,646 mg/1 dissolved minerals.* The ratio of product water to regenerant water is 11.7
tol.
The water used for regeneration was trucked away for disposal by state-licensed chemical
waste processors.
The volume of water used for regeneration was as recommended by the vendors of the equip-
ment. It is possible that smaller amounts could have been used, particularly in this case where high
quality de-ionized water was not an actual requirement.
Demonstration of Ion-Exchange Purification—Conclusions
This demonstration unit has been operated successfully for over 700 hours, producing very
high quality (de-ionized) water with a minimum of attention.
The ratio of product water to waste regenerant water was 11.7 to 1. Virtually all of the
contaminants in the simulated rinse water were collected and transferred to the regenerant water,
giving it a total solids content of 11,646 mg/1.
* 5778 mg/1 from the regenerating chemicals, 5868 mg/1 from the simulated rinse water.
10
-------
The process has the advantages of being well-established, relatively trouble-free, and it pro-
duces a high quality recyclable water. A waste stream containing approximately equal parts of
regenerating chemicals and contaminants, amounting to approximately one-twelfth the volume of
process water stream, remains for disposal.
REVERSE OSMOSIS PURIFICATION OF CHEMICAL PROCESS RINSE WATER
General
Simulated chemical process rinse water was treated in this loop by a triple RO unit system,
preceded by a clarifier, to produce a high-volume, low-solids recycled water, and a low-volume,
high-solids waste concentrate. As in the case of the ion-exchange unit, rinse water simulation was
accomplished by a continuous addition of chemical process solutions to the recycled water.
Figures 4 and 5 show the demonstration equipment for this loop.
Spiral-wrapped cellulose acetate-type membranes were selected for this work because of their
availability, and relatively lower cost, in a wide range of small and pilot-scale sizes.
Simulation of Rinse Water
Continuous addition of chemical process solutions to the purified recycled water to simulate
the production of chemical process rinse water was achieved by a steady drip feed from the same
type of metering pumps used for the ion-exchange loop.
Clarifier
It is essential that the feed to any RO unit be both filtered to remove all particles that could
clog the membranes, and maintained at such a pH, temperature, composition, etc., that precipita-
tion cannot occur on the membranes. Precipitation resulting from an increase in concentration
must be avoided because of the natural increase in concentration at the membrane surface.
Tests carried out on the simulated rinse water showed that although a pH 7 was necessary
to produce a heavy precipitate (aluminum hydroxide plus heavy metal hydroxides), some precipita-
tion occurred after standing for a few days, even at pH values below 3.0. Since the membrane life
is drastically reduced by chromic acid solutions having pH values of less than 3.0, it was decided to
use a "clarifier" operation on the simulated rinse water before feeding it to the RO units, in order
to avoid all risk of precipitation.
The simulated rinse water was clarified in 379-liter (100-gallon) batches. The pH of the rinse
water was adjusted to 8.5 using (2.5 normal) sodium hydroxide solution. Betz polyelectrolyte no.
1110 was added to aid flocculation. The clear supernatant liquor was drawn off after approxi-
mately 30 minutes of settling and its pH adjusted to 6.0 before treatment by the RO system.
11
-------
Figure 4. Demonstration equipment for reverse osmosis purification of chemical process rinse water.
-------
Figure 5. Detail view of reverse osmosis equipment showing reverse osmosis units.
-------
Reverse Osmosis Purification of Chemical Process Rinse Water
METERED FEED OF
CONTAMINANTS
95 l/hr (25 gal/hr)
SIMULATED
RINSE TANK
RECYCLED WATER
SAMPLE TANK
pH ADJUST
RO SYSTEM
CONCENTRATE
TO WASTE
Special Counterflow Arrangement of RO Units
A special arrangement of three reverse osmosis membranes was developed to give maximum
efficiency of separation of the contaminants from the rinse water. This special arrangement
(Arrangement No. 1) is illustrated below:
SIMULATED
RINSE
ARRANGEMENT NO. 1 OF RO UNITS (COUNTERFLOW ARRANGEMENT)
FINAL
CONCENTRATE
TO WASTE
OUTPUT
PERMEATE
PERMEATE
PERMEATE
Chem-Mill Feed
Hydrogen sulfide generation, from the sulfides present in the chem-mill feed was a problem, as
in the case of the ion exchange purification loop. The problem was overcome by deleting the chem-
mill constituent from the metered drip feeds, and instead adding it as a single shot immediately
before clarification. At the clarification pH of 8.5, rS was not generated.
14
-------
Operation of Reverse Osmosis System
The loop has been operated for a total of 153.5 hours, 104 hours in the "counterflow"
arrangement shown on page 20 and Figure 7, and the remainder of the time in special tests
described in Appendix B, "Reverse Osmosis Tests-Rejection Ratio of Nitrates Compared with
Other Ions," and Appendix C, "Reverse Osmosis-Experimental Verification of Math Analysis."
A typical set of analyses for the system in counterflow arrangement is shown in Table 1.
TABLE 1. ANAL YSIS FOR OPERA TION OF REVERSE OSMOSIS SYSTEM COUNTERFLOW
ARRANGEMENT
Flow
pH
Conductivity
TDS
N03
so4
P04
Cl
Na
Total Cr
Cr+++
Al
Zn
Cu
B
Surfactant
Simulated rinse water
98.4 l/hr
3.0
680 S/cm
534 mg/l
325 mg/l
85 mg/l
0.32 mg/l
12.4 mg/l
210 mg/l
43 mg/l
None detected**
11.4mgA
0.95 mg/l
0.18 mg/l
None detected**
0.77 mg/l
Clean water return
89.9 l/hr
5.8
208 S/cm
172 mg/l
150 mg/l
8.5 mg/l
0.05 mg/l
13.3* mg/l
40 mg/l
6.3 mg/l
None detected**
Removed in clarifier
Removed in clarifier
Removed in clarifier
None detected**
0.23 mg/l
Concentrate to waste
8.5 l/hr
6.2
351 OS/cm
3009 mg/l
560 mg/l
1200 mg/l
0.85 mg/l
460* mg/l
930 mg/l
240 mg/l
None detected**
Removed in clarifier
Removed in clarifier
Removed in clarifier
None detected**
3.47 mg/l
* Higher than simulated rinse input because HCI used internally as pH adjustment.
** Less than 5 mg/l.
The pH values of the clean water return (5.8) and the concentrate to waste (6.2) merely reflect
the value of pH 6.0 chosen to ensure a slightly acid condition to preclude the possibility of precipi-
tation on the membranes.
15
-------
An examination of Table 1 shows that the nitrate concentration in the clean water return is
unexpectedly high, 150 mg/1, for a feed nitrate concentration of 325 mg/1. Measurements of the
rejection ratios* for each RO unit gave the results detailed in Table 2, showing nitrate rejection
ratios of only 0.54 to 0.63 compared with 0.79 to 0.99 for the other ions measured.
* Rejection ratio = (Cm - Cp)/Cm. See equation (4), Appendix A.
TABLE2. CALCULATED REJECTION RATIOS
Ion
N03
Total Cr
Na
Mg
Ca
K
RO-1
0.63
0.89
0.85
0.93
0.98
0.82
RO-2
0.55
0.87
0.79
0.94
0.99
0.79
RO-3
0.54
0.86
0.84
0.98
0.99
0.81
Additional tests, to explore and confirm these low rejection ratios for nitrates are detailed in
Appendix B. These tests showed an unexpected, far lower, rejection ratio of less than 0.2 for the
nitrate ion in acid solution. Tests with the sulfate ion at the same pH showed the expected values
of R = 0.9 or higher.
Comparison of Reverse Osmosis and Ion-Exchange Systems
Comparison of the two systems on the basis of performance shows that they produce roughly
the same volumes of waste water for equal volumes of recycled water. Capture of the solids dis-
solved in the rinse water is, however, 100% for the ion exchange system, but only 71% for the RO
system. The behavior of the two systems, adjusted to a feed of 100 volumes, 500 mg/1 TDS, is
shown in Table 3.
TABLE 3. COMPARISON OF REVERSE OSMOSIS AND ION-EXCHANGE SYSTEMS
Feed
System
Ion exchange
Reverse osmosis
Volume
100
100
TDS
500 mg/1
500 mg/1
Recycle
Volume
92.2
93.3
TDS
NIL
161 mg/1
Dump
Volume
7.8
6.6
TDS
11, 648* mg/1
2.817 mg/l
* Including 5780 mg/1 from the regenerant acid and alkali
16
-------
Alternative RO Arrangements
The arrangement of three RO units into a "counterflow" system, page 14, is only one of a
number of possible arrangements. The more conventional arrangement (No. 2) of three units is
as shown in the following sketch.
SIMULATED RINSE
INPUTQf/Cf
FEED
RO-1
FEED
CONCENTRATE
RO-2
FEED
CONCENTRATE
RO-3
FINAL
CONCENTRATE
TO WASTE Q , C
c c
RECYCLED WATER
PERMEATE
PERMEATE
PERMEATE
OUTPUT Q .C
As in the case of the counterflow system (Arrangement No. 1), the concentrate stream passes
from RO-1 to RO-3, becoming progressively enriched, and finally dumped. Unlike Arrangement
No. 1, however, the three permeate streams are united to form a single output, the recycle stream.
Another arrangement is shown below:
__. RECYCLED WATER
OUTPUT
PERMEATE
CONCENTRATE
SIMULATE RINSE
INPUT
PERMEATE
CONCENTRATE
FEED
CONCENTRATE
I
.FINAL CONCENTRATE
TO WASTE
In this arrangement (No. 3), the concentrate from RO-1 is further concentrated in RO-2, and
the permeate from RO-1 is further purified by permeation through RO-3. The final concentrate has
thus been rejected by two membranes, RO-1 and RO-2, and the final permeate has been passed
through two membranes, RO-1 and RO-3. The concentrate from RO-3 and the permeate from
RO-2 are added to the primary feed to RO-1.
17
-------
Mathematical Analysis of Arrangements Nos. 1, 2, and 3
Mathematical models were constructed of all three arrangements, and calculator programs
written to solve the equations relating input and output flow rates and concentrations for each
arrangement. The equations and their solutions, in the form of operating characteristic curves, are
given in Appendix A.
As examples, the following values were taken from the characteristic curves of Appendix A.
The values selected for recovery ratio R, 0.7 and 0.8, are typical for the mixed ion species present
in the chemical process rinse water, and the values selected for E (recovery ratio)* are commonly
employed in this type of RO unit. A feed rate of 100 1/hr, and a feed concentration of 500 mg/1,
are assumed. (See Table 4).
TABLE 4. COMPARISON
OF RAND E
Item
Arrangement number
Permeate quantity Q l/hr
Concentrate qualtity QC l/hr
Permeate concentration
OF THREE ARRANGEMENTS FOR SELECTED VALUES
Rejection ratio (R) and recovery ratio (E)
R = 0.7 R = 0.7 R = 0.8
E = 0.7 E = 0.8 E = 0.7
1 2 3 12312 3
95.3 97.3 84.5 98.8 99.2 94.1 95.3 97.3 84.5
4.7 2.7 15.5 1.2 0.8 5.9 4.7 2.7 15.5
317 409 195 412 457 314 231 351 105
Cp mg/l
Concentrate concentration 4240 3760 2163 7917 5865 3484 6014 5870 2617
Ccmg/l
Overall performance ratio 13.4 9.2 11.1 19.2 12.8 11.1 25.0 16.7 24.9
Cc/Cp
* Recovery ratio - Ratio of clean water recovered to inpxit = CL/Qf
In all three examples, the overall performance of the system, as measured by the ratio of the
concentration of salts in the final concentrate, divided by the concentration of salts in the permeate,
i.e., Cc/Cp, is the highest for the counterflow Arrangement No. 1.
Examination of the operating characteristic curves, Figures A-l through A-7, Appendix A,
shows clearly that a high rejection ratio R results in the desired low permeate concentrations, and
high concentrate concentrations. The curves also show, however, that the same result cannot be
achieved by adjustments to the recovery ratio E, since increasing E always increases concentrate
concentration (desired) but also always increases permeate concentration (undesired). Consequent-
ly, the choice of recovery ratio becomes a compromise: the highest value possible is selected, con-
sistent with obtaining an acceptable, recyclable quality of permeate. The influence of recovery
ratio E on the volume of concentrate produced by each arrangement is shown in Figure A-7, Appen-
dix A.
* Recovery ratio = Ratio of clean water recovered to input = Qp/Qf
18
-------
Calculation of Permeate Concentration and Concentrate Concentration for a Single RO Unit
Employing Feedback
The full calculation of concentrations in the outputs from a single RO unit employing feed-
back (i.e., returning a portion of the concentrate stream to the intake) is time consuming. A cal-
culator program, for use on a Hewlett-Packard HP-65 programmable calculator for making this cal-
culation, is given in full detail in Appendix D.
Use of this program, which employs a reiterative technique, is useful where feedback rates are
small (less than twice the feed). Where the feedback rates are high (above 10 times the feed),
the approximations used in Appendix A, equations (2) and (4), allow simpler calculations to be
used without introducing significant error.
Demonstration of Reverse Osmosis Purification—Conclusions
Reverse osmosis purification and recycling of simulated chemical process rinse water has been
successfully demonstrated in operation for over 150 hours.
Addition of a clarifier stage was found necessary to avoid the possibility of chemical precipi-
tation on the membranes.
Due to the low rejection rate of the nitrate ion, relatively low quality recycled water was
obtained; with an input of 534 mg/1 total dissolved solids, a recycled water containing 172 mg/1
was obtained.
ULTRAFILTRATION AND ACTIVE CARBON PURIFICATION OF DYE PENETRANT
INSPECTION RINSE WATER
General
Water containing 400 mg/1 of emulsified oil, used to rinse airplane parts after crack detection,
was purified in this loop by passage through an ultrafiltration membrane, followed by "polishing"
in an activated carbon column.
Figures 6 and 7 show the demonstration equipment for this loop. An outline flow diagram of
the equipment is shown below.
CLEAN RECYCLED WATER
DYE
PENETRANT
OIL
DRIP FEED
SIMULATED
RINSE WATER
ULTRAFILTRATION
UNIT
PERMEATE
CONCENTRATE
ACTIVE
CARBON
COLUMN
19
-------
DYE PENETRANTOIL
mfmmaam
EMULSIFIER
•
DEVELOPER
RECYCLED WATER
SAMPLE TANK
ACTIVATED
CARBON
COLUMNS (4)
1ULATED RINSE TANK
ULTRAFI .TRATION UNIT
•ULTRAFILTRATION
i':ONCENTRATE/RECYCLE TANK "-
Figure 6. Demonstration equipment for recycling dye penetrant inspection rinse water.
-------
' )
4_J.
ft ii
ACTIVATED
CARBON
COLUMNS (4)
(PART OF) DEMONSTRATION
EQUIPMENT (SEE Figure 6)
Figure 7. Control panel for demonstration equipment recycling dye penetrant inspection rinse water.
-------
Operation as Designed
The loop was operated as designed, recycling 5.7 to 7.6 1/hr (P/2 to 2 gal/hr) for a total of 627
hours over 98 days. Continuous simulation of rinse water was achieved, as in the previous sections,
by small metering pumps delivering dye penetrant solutions taken from production inspection
tanks. Typical analytical data are shown in Table 5.
TABLE 5. TYPICAL ANALYTICAL DATA FOR DYE PENETRANT INSPECTION WATER
PURIFICATION USING ULTRAFILTRATION AND ACTIVATED CARBON
FILTRATION
Simulated rinse water Recycled water
(mg/l) (mg/l)
Emulsified oil 400-500 1-2
(by freon extraction)
Surfactant (as lauryl 1-2 0.06
alkyl sulfonate)
Flushing of the Ultrafiltration Unit
During the 627 hours of operation, the ultrafiltration unit was flushed only twice, after 41
hours and after 330 hours operation. The flushing after 41 hours of operation was conducted
because two of the four membranes had very low permeability rates, one-fourth that of the other
membranes, and flushing was tried in an attempt to improve their permeability. The attempt was
unsuccessful, and the membranes were replaced.
The high permeate flow rates obtained immediately after flushing, approximately 280 nl/m2-
s-Pa(4 gal/ft2-daylb/m.2) declined rapidly and levelled off at 9.0nl/m2.s-Pa(1.3 gal/ft?day-lb/in-2).
Figure 8 shows the permeate rates achieved during and immediately after detergent flushing, and
much lower rates obtained on a 3000 to 5000 mg/l penetrant oil concentrate.
Maximum Concentration Reached by the UF Unit
The maximum concentration reached during the 627 hours of operation was 5000 mg/l,
representing a 12.5 times concentration of the original 400 mg/l. Fracture of a pump housing,
and consequent loss of concentrate, prevented higher values from being reached during the test
period. There appears to be no reason why much higher concentrations could not be reached,
given sufficient time, since the emulsion was stable, with no signs of "breaking."
Variation in Membrane Permeate Rates
Considerable variation was noted between nominally identical membranes. As an example
of this variation, Figure 9 shows the first 55 hours of operation of the four membranes in the
UF unit.
22
-------
210
CO
a.
CM
140
U-
UJ
cc.
LU
Q.
FLUSHING
CYCLE
^ •»
300 HOURS
OPERATION ON 3000-5000
mg/l EMULSION
3.0
2.0
CM
O)
«
O
cc.
in
O_
TIME (NO SCALE)
Figure 8. Ultrafiltration unit, permeate flow rate characteristics during flushing and operation on
dye penetrant rinse concentrate.
23
-------
K)
350
280-
i 240-
c
O
< 140-
LLJ
5
DC
70-
CONTINUE
FLUSH AT
pH = 11.3
KEY
NO. 1 MEMBRANE
NO.2 MEMBRANE
NO.3 MEMBRANE
NO.4 MEMBRANE
OPERATING PRESSURE =
172-206kPa {25-30 psi)
-400 PPM DYE PENETRANT OIL
4000 PPM DYEPENETRANT OIL
AVERAGE OF ALL
"FOUR MEMBRANES
—r~
4
—1—
6
—i—
8
—T—
10
—I—
12
14
TEST RUN WITH
400 PPM
DYEPENETRANT OIL
START
FLUSH AT
pH= 10.0
OLD NO.2 MEMBRANE
DISCARDED
CONTINUE
i—FLUSH
NEW
NO. 2
MEMBRANE
-5.0
-4.0
CN
-3.0 £
O
NEW NO. 3 MEMBRANE
OLD NO.3 MEMBRANE
DISCARDED
-2.0
-1.0
HOURS IN SERVICE
Figure 9. Service record of ultra filtration unit, dye penetrant rinse water purification.
-------
Simplification of Loop to Operate on Carbon Column Alone
It was reported that one company marketing a dye penetrant crack detection system* recom-
mended recycling its rinse water using only activated carbon filtration for purifying the rinse water.
With this recommendation in mind, a test was made in which simulated dye penetrant rinse water
was passed at rates of between 2.5 and 7.6 1/hr through 6.6 liters of active carbon ("Nuchar") Grade
WV-G 12 x 40) in two columns each 3.7 cm diameter by 45 cm long.
The filtrate from the carbon columns was clear for the first hour of operation, after which it
began to show a white turbidity, i.e., all the yellow dye in the feed appeared to have been removed,
even after 179 hours of operation, by the activated carbon. Examination of the water under UV
light confirmed that the dye was absent (no yellow fluorescence). At this point however, the water
was judged unsuitable for final rinsing of parts undergoing crack detection due to a tendency for
small amounts of free oil to separate from the water, together with the only marginal acceptability
of the water in a laboratory test on residual surface fluorescence. Details of this test are given
below.
Analyses of the feed and carbon-filtered recycled waters gave the results shown in Table 6.
TABLE 6. ANAL YSES OF FEED AND CARBON FILTERED RECYCLED WATERS
Test
Untreated feed
Carbon-filtered
After 13 hours operation
After 179 hours operation
Appearance
UV illumination
Freon extractables
Surfactant as lauryl
alkyl sulfonate
Suitability for reuse,
based on tests of
residual surface
fluorescence
Deep yellow
Vivid yellow
fluorescence
280-515 mg/l
0.04-0.12 mg/l
Marginally
acceptable
White turbidity
Moderate bluish-white
fluorescence
68 mg/l
Less than 0.01 mg/l
Acceptable
White turbidity
Stronger bluish-white
fluorescence
277 mg/l
0.04 mg/l
Marginally acceptable
Test for Acceptability of Rinse Water
Acceptability was determined by comparing fresh and treated water for absence of background
fluorescence and effect on crack sensitivity. Absence of background fluorescence was determined
with a water wash penetrant qualified to Group VI sensitivity with developer, using the water wash-
ability test called out in MIL-I-25135C. Crack sensitivity comparisons were performed on thermally
cracked aluminum blocks, using the same water wash penetrant. The penetrant was rinsed with
* Brent Chemicals, Commerce Road, Brentford, Middlesex, England.
25
-------
either fresh or treated water, using the apparatus and rinse conditions of the water washability test.
Polaroid photographs were taken of the test blocks under a black light intensity of 2200 juW/cm .
The initial exposure was adjusted to give the same intensity of indication in the photo observed on
the test block. All subsequent test blocks were photographed, using the initial exposure. The
results are indicated in Table 6.
Dye Penetrant Inspection Rinse Water Recycling—Conclusions
Dye penetrant rinse water has been successfully recycled in the demonstration unit, using two
methods. A partial purification, using filtration through activated carbon only, has been demon-
strated in operation for 179 hours, and a more complete purification, using ultrafiltration followed
by activated carbon "polishing," has been demonstrated over a period of 627 hours.
Partial purification by filtration through activated carbon yields a dye-free, but not oil-free,
water that is only marginally acceptable for rinsing parts undergoing crack detection.
Purification by ultrafiltration followed by carbon "polishing" yields a higher quality water
(only 1 to 2 mg/1 of oil compared with up to 277 mg/1 oil in the carbon-only filtered water) of more
than adequate purity for its intended purpose.
The simplicity and lack of moving parts in the carbon-only treatment make it an attractive
process. A cost analysis for recycling dye penetrant rinse water, based on a full-scale factory-sized
plant, is given in Section 6.
MACHINE SHOP COOLANT WASTE WATER PURIFICATION
General
In this demonstration loop, clean water was reclaimed from waste machine shop coolant and
reused to make fresh coolant. The total running time was 290 hours over a period of 54 days.
Actual waste machine shop coolant was used for the initial charge in the loop, and metered
additions of new cutting oil concentrates were fed continuously to the recycled water to maintain
coolant strength. Maximum use was made of actual used coolant for replacing evaporation losses,
replacing samples withdrawn, etc., because of the marked differences between used and unused
coolant. Used coolant is vile-smelling, dark, and has a surface layer of "tramp" oil and organic
slime, whereas the fresh product is usually clear to milky white, with an antiseptic odor.
Treatment in this loop consisted of decanting and filtering to remove floating "tramp" oil,
scum, grinding grit and metal particles; ultrafiltration to remove emulsified oil and precipitated
soaps; and reverse osmosis to remove dissolved material. Adjustment of pH from alkaline (pH 7.5
to 9.0) to slightly acid (pH 6.5) was made because the cellulose acetate membranes of the RO unit
could not be used in alkaline conditions.
Figure 10 shows the demonstration equipment and Figure 11 shows the waste coolant, recycle,
and UF concentrate tanks. The floating scum and "tramp" oil on the waste coolant is clearly
shown in Figure 11. Carry over of the scum and oil to the UF concentrate tank, retained by the
divider, was skimmed off daily.
26
-------
•
I
DILUTE
HYDROCHLORIC
FOR pH
ADJUST
DRIP FEED OF CONCENTRATED
COOLANTS AND BIOCIDE
ULTRAVIOLET STERILIZER LAMP
IN RO HEADER TANK
DRIP FEED METERING PUMPS
DISPLAY BOARD
ULTRAFILTRATION FLUSH
SOLUTION TANK
ULTRAFILTRATION
UNIT
WASTE
COOLANT
TANK
ULTRAFILTRATION
CONCENTRATE
TANK
ACTUAL USED WASTE
MACHINE SHOP COOLANT
Figure 10. Demonstration equipment for machine shop coolant waste water purification.
-------
1J
ULTRAFILTRATION
CONCENTRATE TANK
Figure 11. Machine shop coolant waste water purification, waste coolant and recycled-water tanks.
-------
NEW COOLANT
XDNCENTRATES
CLEAN RECYCLED WATER
PERMEATE
PERIODIC DUMP
PERIODIC DUMP
OPERATION
Operation of this loop presented three problems:
o Near-intolerable odor levels unless biocides and sterilizer were used.
o Frequent flushing of the UF membrane was required to maintain adequate permeate
flow rates.
o Organic growths, unless removed, tended to block filters and pipes.
After the above problems were solved (see the following sections for details), the loop was
operated to give the following results.
Total dissolved solids
Suspended solids
Freon extractables
PH
Floating "tramp" oil and slime
Plate count
Waste coolant
1500to2500mg/l
1000 to 2000 mg/1
250 to 1000 mg/1
7.5 to 8.0
s*
1 to 2% by volume
106to 107 micro-
organisms/ml
Reclaimed water
500 to 1500 mg/1
Nil
50 to 100 mg/1
6.5
Nil
Offensive Odor of Waste Coolant
Discarded coolants from a machine shop are normally rancid and offensive in odor. A "plate"
count, i.e., a microbiological count of the number of micro-organisms that can be grown on an agar
plate of over 0.1 million/ml of coolant is sometimes taken as an indication that a coolant has
reached the end of its useful life. At the time of testing, several days after collection from the
machine shops, the 50-gallon samples used to charge the loop contained between one and 10 million
micro-organisms/ml.
29
-------
Due to the heat generated by the UF and RO pumps in the demonstration loop, temperature
of the waste coolant was 21 ° to 29°C (70° to 85°F), in spite of the provision of water cooling. At
these temperatures, conditions for micro-organism growth were highly favorable, and a dark slimy
crust formed on the waste coolant, and a highly offensive odor surrounded the equipment.
Addition of a proprietary biocide, Pace Chemical N-136B, was effective in destroying odors
when added at one percent (10,000 parts per million) by volume to the waste coolant. However,
this is a substantial addition when it is considered that the original coolant contained only 2 to 2¥i%
by volume of added coolant concentrate, i.e., the biocide additive constitutes almost one-third of
the total dissolved load carried by the water.
In spite of the use of biocide in the raw waste coolant, the header tank to the RO unit still
continued to darken, develop a mold-like skin, and emit a decaying odor. As it was undesirable to
add additional biocide to this header tank, (since the whole purpose of the loop was to remove
material from the water) an ultraviolet sterilizer lamp was fitted to the tank. This technique proved
completely successful in eliminating surface growths, and substantially eliminated the odor from
this tank.
Flushing of UF Unit
The ultrafiltration unit required much more frequent flushing—six times in 290 hours opera-
tion—than the identical unit used in the dye penetrant rinse water unit (page 22), which was flushed
only twice in 627 hours.
Immediately after flushing with alkaline detergent solution (see Appendix H), a permeate flow
of over 200 nl/m2
-------
This high velocity was recommended by the manufacturers of the unit, to provide a high
velocity inside the tubes and minimize any tendency to form a coating that would block the
membranes.
Adjustment of pH
Because the cellulose acetate membranes of the RO unit were unsuited to use in alkaline condi-
tions, the pH of the UF permeate, i.e., the RO feed, was experimentally lowered from an original
7.5 to 9.0, to less than 7.0, using 4% hydrochloric acid. However, this procedure resulted in the
formation of a precipitate which rendered the solution unsuitable for introduction to the RO unit
unless completely filtered. After tests, it was found that if the feed to the UF unit was acidified
to pH 6.5, then the permeate from the UF—also at pH 6.5—was suitable for feeding to the RO unit.
The unit was accordingly operated in this manner.
Reverse Osmosis Unit
This unit has operated with a minimum of attention, except for weekly operation for 10 min-
utes at increased recycle rates, to "wash" the membranes, plus mild chlorination (5 ppm) during
^
periods of shutdown of more than two weeks. At an operating pressure of 1345 kPa (195 Ib/in. ),
the permeate flow was between 50 and 100 ml/min (0.79 to 1.59 gal/hr). Membrane area on this
unit was 0.46 m2/(5 ft2).
A rejection ratio of approximately 0.77, based on conductivity, was obtained for the unit
when the demonstration loop was operated on actual machine shop waste coolant. At the upper
limit of 1500 mg/1 total dissolved solids in the recycled water, with this rejection ratio, the waste
from the RO unit, concentrated in the header tank, contains slightly more than 6520 mg/1 total
dissolved solids.
Factory Test of Recycled Water
A sample of 76 liters (20 gallons) of reclaimed water from the demonstration unit was used in
a production turret type drilling, tapping and honing machine (a Burgermaster Model 2B). Used to
prepare a proprietary coolant (CIMCOOL "Five Star") at a dilution of 40:1, the coolant was indis-
tinguishable from coolant made from the city water supply, over a period of at least 2Vi months.
The water had the following characteristics:
pH 6.75
Freon extractables 64 mg/1
TDS(50°C) 1413 mg/1
Ash on TDS 437 mg/1
Appearance Faintly cloudy
Odor Musty
31
-------
Machine Shop Coolant Waste Water Purification—Conclusions
The demonstration unit has successfully reclaimed water from used machine-shop coolant.
However, more problems were encountered in operating this loop than were originally anticipated.
These included:
o The development of a strongly offensive odor unless relatively large amounts of biocide
were added to the waste coolant tank
o A considerable organic growth in the RO header tank unless UV sterilizer used
o Only 50 hours average operating time before the UF unit required flushing to restore its
permeate flow rate
o A relatively low concentration of 6500 mg/1 total dissolved solids in the concentrated
waste from the RO unit before dumping becomes necessary
CYANIDE RINSE.WATER PURIFICATION
Electrolytic Destruction of Cyanide
The original plan for this loop was to destroy the cyanide in the rinse water from plating shop
waste, by electrolysis. Experiments with a prototype unit had demonstrated the feasibility of this
plan, which converts cyanide to carbon dioxide and nitrogen, and at the same time removes heavy
metals from the rinse water by plating them out onto the electrodes.
The prototype unit had a unique design in which electrode area and electrode velocity were
maximized by passing the electrode successfully through a series of concentric stainless steel cylinder
electrodes. In the prototype development runs, d.c. current reversal was employed to prevent the
buildup of copper, etc. on the electrodes.
The simplified cyanide destruction equations are as follows:
ANODE : 2CN""~ + 4H2O -*- 2CO2 1 + N2 \ + 8H+ + 1 0 e
CATHODE:
8e
Me
4H2 f
2Me |
The flow diagram of the original plan (not used) is shown below:
CYANIDE
PLATING
SOLUTIONS
DRIP FEED
1
—>»s— ^N.
CY/
RECYCLED WATER
7
\
i
^. — ^ — ^ — ^^
kNIDE RINSE
WATER
A^.
1
c!
C
DC POWER SUPPLY
AND REVERSER
•
\
—->
AC
CARE
FOR
BRK
RE
VIVA
ON F
ORC
3HTE
MOV
TED
•ILTER
NER
AL
^T
"* 1
GAS
SEPARATO
R
CYANIDE DESTRUCT UNIT
32
-------
Change to Nonelectrolytic Cyanide Destruction
During the equipment design phase of this program, it became evident that the generation of
hydrogen and oxygen within the electrolytic cyanide destruction unit posed a safety problem, par-
ticularly if the equipment was to operate semicontinuously with the minimum of attention. The
safety concern was that metals such as copper, loosely deposited on the electrodes, could possibly
form flakes that might internally short the electrodes and ignite the explosive gas mixture. In brief,
the unit was a potential "bomb."
Therefore, in this demonstration program, it was decided to use a well-established chemical
destruction method for removing cyanide.
Outline of Cyanide Destruct Unit
The purification and recycling of rinse water containing 100 to 200 mg/1 of cyanides was dem-
onstrated in a four-step operation:
o Accumulation of a batch of cyanide-containing rinse water
o Destruction of the cyanide in the batch by addition of sodium hypochlorite
o Removal of solids and excess chlorine
o Removal of all anions and cations by ion-exchange resin*
The last operation was accomplished by feeding the treated, cyanide-free water to the ion-
exchange loop to avoid duplication of equipment. The flow diagram of the plan actually used is
shown below.
CYANIDE PLATING
SOLUTIONS
LW-V^--M^_-*-*"J
RECYCLED WATER
DRIP FEED
SODIUM HYPOCHLORITE.SULFURIC ACID
AND SODIUM HYDROXIDE
L^ (J k— *~<^J t~_~- 1
TO ION-
EXCHANGE
COLUMN
(FIGURES2&3)
ACTIVATED CARBON
COLUMN TO FILTER AND
REMOVE FREE CHLORINE
REACTOR TANK FOR BATCH
DESTRUCTION OF CYANIDES
*Cyanide must be removed (except in certain special cases) from the feed to an ion-exchange sys-
tem because of the risk of toxic HCN production during regeneration with acid.
33
-------
Batch Treatment of Cyanide Rinse Water
Simulated cyanide rinse water was fed to the reactor tank at a rate of 50 ml/min,accumulating
a batch of 30 liters (8 gallons) every 10 hours of operation. A series of valves, timer-controlled,
together with pH and ORP (oxidation-reduction potential) probes governed the addition of oxidant
(5% sodium hypochlorite solution), and pH adjusting solutions, over a pre-set time schedule as
follows:
0 — 30 minutes Adjust pH to 9.0 and add excess sodium
hypochlorite to ORP 400 mV minimum
30 - 60 minutes Lower pH to 8.0 and maintain ORP 400 mV
minimum
60-75 minutes Dwell
75 — 90 minutes Empty reactor tank
Cyanide Rinse Water Simulation
Rinse water from cyanide plating operations was simulated by metering actual plating shop
solutions into the recycled water stream at the following rates:
Copper Plating Bath 0.057 ml/min
Cadmium Plating Bath 0.057 ml/min
Titanium-Cadmium Plating Bath 0.057 ml/min
Enstrip * 0.029 ml/min
Recycled Water 50.0 ml/min
Composition of each of the above plating and stripping solutions is detailed in Appendix E.
Analysis of Simulated Rinse Water, Detoxified Rinse Water and Recycled Water
Typical analysis of the input, output and recyled waters are as follows:
Simulated rinse Detoxified De-ionized
Nil Nil
Nil
Nil
Nil
0.05 mg/1 0.05 mg/1
8.0 5.9
Figure 12 shows the demonstration equipment for the cyanide destruct system.
* Enthone Inc., 2751 El Presidio St., Long Beach, California 90810
34
Free CN
Cu
Cd
NaOH
Freed 2
pH
186 mg/1
30 mg/1
63 mg/1
45 mg/1
—
9.5
-------
-------
Cyanide Rinse Water Purification—Conclusions
The recycling of rinse water containing approximately 200 mg/1 cyanide was demonstrated in
a conventional chlorine oxidation operation to destroy cyanides followed by ion-exchange treat-
ment to remove the remaining dissolved minerals.
Original plans to use an electrolytic cell to destroy cyanide were abandoned because of the
explosion risk from the hydrogen-oxygen mixture generated as a byproduct of the electrolysis.
36
-------
SECTION 5
COST OF FULL-SCALE PLANT
GENERAL
The estimated cost of installing a full-scale facility, excluding land cost, is detailed in this
section.
The facility would receive an average flow of 1.68 million I/day (444,000 gal/day) made up
as follows: .
I/day gal/day
Chemical process rinse water 1,514,000 400,000
Dye penetrant rinse water 121,120 32,000
Machine shop coolant 24,224 6,400
Cyanide process rinse water 22,710 6,000
Figures 13 and 14 show overall flow diagrams for the full-scale facilities of this cost estimate.
Water used for cooling compressors, air-conditioning units and other machines, and sanitary
waste water would not be treated by the facility.
ESTIMATE
Cost Summary 1—By Process
Site preparation,
Equipment building, wiring
Process purchase cost installation Total
Chemical process rinse water $1,050,173 $1,695,668 $2,745,841
Dye penetrant rinse water 100,253 161,901 262,154
Machine shop coolant 79,927 129,051 208,978
Cyanide rinse water 64,980 105,047 170,027
Totals $1,295,333 $2,091,667 $3,387,000
37
-------
oo
22,710 I/day
(6000 gal/day)
15141/hr
(400 gal/hr)
RINSE TANKS
IN CYAN IDE-
CONTAINING
CHEMICAL
PROCESSES
200 mg/l CN
1200mg/l
1.514,000 I/day
(400,000 gal/day)
RINSE TANKS
IN NONCYANIDE
CHEMICAL
PROCESSES
600 mg/l
LEGEND
gal/day :
I/day
gal/hr
l/hr
Ib/day
kg/day
mg/l
CYANIDE
4.5 kg/day
(10 Ib/day)
113,550 l/hr
(30,000 gal/hr)
GALLONS PER DAY
LITERS PER DAY
GALLONS PER HOUR
LITERS PER HOUR
POUNDS PER DAY
KILOGRAMS PER DAY
MILLIGRAMS PER LITER
(TOTAL SOLIDS)
Figure 13. Flow diagram for cyanide and noncyanide chemical process rinse water.
-------
TO MACHINE
SHOP COOLANT
MAKE-UP TANK
WASTE MACHINE
SHOP COOLANT
1514 l/hr
(400 gal/hr)
RO
106OI/hr
(280 gal/hr)
RECYCLE
STORAGE
INSOLUBLE
OILS
BATCH REMOVAL
OF OIL/WATER
CONCENTRATE
18,925 I/mo
(5000 gal/mo)
I 379 l/hr
(100 gal/hr)
OIL FREE
WASTE TO
CLARIFIER
MACHINE SHOP COOLANT
TO DYE
PENETRANT
RINSE TANK
7570 l/hr
(2000 gal/hr)
7570 l/hr
(2000 gal/hr)
7570 l/hr
(2000 gal/hr)
BATCH REMOVAL
OF OIL/WATER
CONCENTRATE
947 I/mo
(250 gal/mo)
BATCH REMOVAL OF
SPENT CARBON
DYE PENETRANT INSPECTION RINSE WATER
LEGEND
gal/hr = GALLONS PER HOUR
l/hr = LITERS PER HOUR
gal/mo» GALLONS PER MONTH
I/mo = LITERS PER MONTH
Figure 14. Flow diagram for machine shop coolant and dye penetrant rinse water recycling.
-------
PROCESS OPTION
The above cost summary assumes that the chemical process rinse water is purified by the
reverse osmosis process. If the reverse osmosis option were replaced by ion exchange, there would
be a saving of $40,800 (1.2% of total).
Cost Summary 2—By Overall Items
Item Cost
Building, approach road, drains, fence, $ 740,000
sumps, (capital cost plus installation)
Tanks, equipment, plumbing, controls 1,976,000
(capital cost plus installation)
Electrical (capital cost plus installation) 308,000
Construction management (4%), architec- 363,000
tural and engineering fee (8%)
Total $3,387,000
Equipment List
An equipment list of all pumps, tanks, RO units, UF units, separators, centrifuges, driers,
evaporators, filters, etc., is included in Appendix F, together with a complete schematic for the
full-scale recycling plant.
40
-------
SECTION 6
OPERATING ECONOMICS OF A FULL-SCALE PLANT
GENERAL
The operating cost of a full-scale recycling plant, based on the principles of the pilot scale
demonstration units, is developed in this section.
A summary of the operating costs, and conclusions, are presented first. Details of the costs of
each process, and the calculations on which these costs are based, can be found in later pages in this
section.
For comparison purposes, the cost of operating a full-scale conventional treatment plant is also
included.
SUMMARY
Table 7 summarizes the operating economics of a recycling system compared with the econ-
omics of a conventional treatment plant discharging to waste.
TABLE 7. SUMMARIZED OPERATING ECONOMICS, RECYCLING SYSTEM COMPARED
WITH CONVENTIONAL TREATMENT PLANT DISCHARGING TO WASTE
Cost of recycled water
$ per 10001 $ per 1000 gal
Chemical process 0.94
rinse water
Dye penetrant 1 .65
rinse water
Machine shop 12.18
coolant
Cyanide process 4.36
rinse water
3.57
6.25
46.09
16.50
Cost of once-through water
$ per 10001 $ per 1000 gal
0.52 1.97
0.54 2.06
2.56 9.69
3.21 12.17
TOTAL 1.21 4.57 0.59 2.24
41
-------
DISCUSSION AND CONCLUSIONS
The calculations show that, in all four cases, the cost of recycling is higher than the cost of
single-use. Recycled chemical process water costs 80% more, dye penetrant rinse water 203% more,
machine shop coolant water 376% more, and cyanide rinse water 36% more. Except in the case of
machine shop coolant water recycling, higher equipment cost is the major expense factor. (The
major cost item for machine shop cooling recycling is labor.)
The minimum cost, $0.94/kl of recycled water applies to water that can be purified with rela-
tively little difficulty, and which is used in large volumes (the chemical process rinse water). The
cost of recycling rises steeply with the technical difficulty of purification, and with decreasing
volumes to be treated (e.g., the cyanide rinse water and the machine shop coolant).
It should be noted that the cost of fresh water, taken as $0.0794/kl or $0.30/1 OQO gal, is the
average cost for water in large airplane factories in the States of Washington, California, and New
York. This is only 3 to 15% of the costs of treating that water after use, before it can be discharged
to waste.
The high cost of extracting clean water from used machine shop coolant by ultrafiltration and
reverse osmosis, $12/kl or $46/1000 gal, compares very unfavorably with the $2.56/kl or $9.69/
1000 gal to purify the same coolant by chemical methods. However, the chemical method cannot
be used by itself in a closed loop system because it involves the addition of soluble sulfates to the
water. These sulfates are permitted additives to water discharged to a natural waterway or sewer,
but are forbidden to any closed system where they could accumulate without limit. Further research
is needed to reduce the high cost of treatment of machine shop waste coolants for recycling.
Since equipment is the major factor governing treatment cost, it is evident that reduction of
rinse water volumes will effect the greatest overall economies in the cost of water treatment. Stand-
ard procedures for water volume reduction, such as provision of multiple counter-current rinses,
"on-demand" provision of rinse water (e.g., by conductivity-controlled supply valves), and vigilance
by factory operators, supervision and maintenance in shutting off rinse tank feeds during inactive
shifts and weekends, will become of increasing importance when water recycling is employed.
OPERATING ECONOMICS-CALCULATIONS
Assumptions Used
1. All costs are in 1976 dollars.
2. The following volumes of water are received daily:
Chemical process rinse water 1.54 Ml/day (400,000 gal/day)
Dye penetrant rinse water 0.121 Ml/day (32,000 gal/day)
Machine shop coolant 24.2 kl/day (6,400 gal/day)
Cyanide process rinse water 22.7 kl/day (6,000 gal/day)
42
-------
3. Amortization of chemical process equipment cost (purchase price plus installation cost) is
calculated at 8% over a 12-year life.
4. Amortization of building and utilities (materials, construction and installation cost) is calcu-
lated at 8% over a 15-year life.
5. Operating labor costs are $30/hr total, including basic wage, overtime, fringe benefits and
management and overhead burden.
6. Cost of fresh water is $0.0794/kl ($0.30/1000 gal).
7. Other assumptions, such as cost of disposal of sludge, disposition of operator time, etc. are
treated in the detailed calculations, pages 43 through50.
8. The recycling plant has been designed for operation by one operator on each of two shifts.
The nonrecycling plant requires one operator on only one shift. The reason for this is to
reduce overall costs. Operating on two shifts obviously doubles the total operating labor costs,
but allows a reduction in equipment costs (the major cost item) since the equipment can be
sized to handle one half of the load of a single shift operation.
9. The following items are not taken into account in this study: startup costs, operating profit,
site purchase cost, sewer disposal charges, tax advantages.
Operating Cost Details
Tables 8 through 11 itemize the operating costs on which the summary table, page 41, is based.
Costs are identified by a letter: this letter refers to a calculation given on pages immediately follow-
ing the table.
BASES AND CALCULATIONS FOR TABLE 8
(a) From Appendix F, summarized on page 37.
(b) From unpublished Boeing data for its Auburn waste treatment plant. Dollar costs for 1967
have been converted to 1976 dollars, using an inflation factor of 1.739 (average of Chemical
Engineering and Marshall & Swift indices for 1967 to 1976).
Equipment plus installation cost Equivalent 1976 dollars
Chemical process rinse water $ 809,499
Dye penetrant rinse water 80,950
Machine shop coolant 52,800
Cyanide rinse water 112,7 51
Total $1,056,000
43
-------
TABLE & CHEMICAL PROCESS RINSE WA TER, ITEMIZED OPERA TING ECONOMICS
Total equipment cost including installation
Equipment amortization cost per day
Operating labor cost per day
Chemicals— cost per day
Electric power cost per day
Maintenance cost per day
Sludge disposal cost per day
Fresh water purchase cost per day
Membrane replacements
Total
Total volume processed per day
Cost per 1000 liters
Cost per 1000 gallons
Recycling Plant
$2,745,841 (a)
1,014 (c)
192 (e)
52 (g)
12 (i)
14 (k)
56 (m)
12 (o)
76 (g)
$1,428
1,514,000 liters
(400,000 gallons)
$0.94
$3.57
Nonrecycling Plant
$809,499 (b)
299 (d)
180 (f)
114 (h)
3 (j)
15 (I)
56 (n)
120 (p)
$787
1,514,000 liters
(400,000 gallons)
$0.52
$1.97
(c) $2,145,923 (equipment) x 0.132695*/350 days/year =$ 813.58/day
$ 599,918 (buildings) x 0.116830**/350 days/year = $ 200.25/day
$2,745,841 (total) x (0.0003692)***days/year = $l,013,83/day
(d) $809,499 (total) x (0.0003692)*** t = $298.89
* 0.132695 = Periodic payment for annuity, 8%, 12 years (2)
** 0.116830 = Periodic payment for annuity, 8%, 15 years (2)
*** 0.000369 = Combined periodic payment for annuity calculated from $1,013.834- $2,745,841
t It is assumed that the same relative percents of costs between equipment and buildings exist for
the nonrecycle plant as in the recycle plant.
44
-------
(e) Labor distribution, recycle plant:
Chemical process rinse water
Dye penetrant rinse water
Machine coolant
Cyanide rinse water
Labor cost:
Chemical process rinse =
Dye penetrant rinse
Machine coolant =
Cyanide rinse
40%
15%
40%
16hr/dayx$30/hrx40%
16hr/dayx$30/hrx 15%
16hr/dayx$30/hrx40%
16hr/day x$30/hrx 5%
$192/day
$ 72/day
$192/day
$ 24/day
(f) Labor distribution, nonrecycling plant:
Chemical process rinse water =
Dye penetrant rinse water =
Machine coolant =
Cyanide rinse water =
Labor cost:
Chemical process rinse water =
Dye penetrant rinse water =
Machine coolant =
Cyanide rinse water =
75.0%
7.5%
8.75%
8.75%
8 hr/day x $30/hr x 75%
8 hr/day x $30/hr x 7.5%
8 hr/day x$30/hrx 8.75%
8hr/dayx$30/hrx8.75%
$180/day
$ 18/day
$ 21/day
$ 21/day
(g) Cost of chemicals, chemical process rinse water, recycling plant:
Lime for clarifier = $5976/hr 350 days/yr x 90% = $14.90/day
Flocculant clarifier = $1944/hr 350 days/yr x 90% = $ 5.00/day
Antifoam clarifier = $12,600/hr 350 days/yr x 90% = $32.40/day
Total
$52.30/day
(h) Cost of chemicals, chemical process rinse water, nonrecycling plant:
Lime for clarifier =
Flocculant clarifier =
Antifoam clarifier =
Sulfuric acid =
(Cr reduction)
Sulfur dioxide
Total
$ 5,796/yr
$ 1,944/yr
$12,600/yr
10,552/yr
$11,205
350 days/yr x 90% =
350 days/yr x 90% =
350 days/yr x 90% =
3 50 days/yr x 100% =
$14.90/day
$ 5.90/day
. $32.40/day
$30.15/day
xlOO%= $32.01/day
$114.46/day
45
-------
(i) Electric power, chemical process rinse water recycling:
RO pumps:
210 hp x 0.746 kW/hp x 0.0035 $/kWh x 16 hr/day = $8.77/day
Other pumps: (160 hp total, from Appendix F)
160 hp x 50% duty cycle x 0.746 kW/hp x 0.0035 $/kWh x 16 hr/day = $3.34/day
Total $12.11/day
(j) Same as (i), "Other pumps"
(k) 10% of operating labor costs
(1) Same as (k)
(m) From 49.7 m3/mo (65 yd3/mo) at $32.70/m3 ($25/yd3)
49.7 nr/mo x $32.70/m3 * 29.25 days/mo = $55.56/day
(n) Same as (m)
(o) 151,400 liters (40,000 gallons) @$0.0793/kl ($0.030/1000 gal) = $12/day
(p) 1,514,000 liters (400,000 gallons) @$0.0793/kl ($0.30/1000 gal) = $120.06/day
(q) Membrane replacement cost, 20% x $400,000 (Appendix F, page 78 , item 70) every 3 years
= $80,000 •=• (3 yr x 350 days/yr) = $76.19/day
BASES AND CALCULATIONS FOR TABLE 9
(a) From Section 5, page 37.
(b) From (b), page 43.
(c) $ 100,252 (equipment) x 0.132695*/350 days/yr = $38.01/day
$161,901 (buildings) x 0.116830**/350 days/yr = $54.04/day
$262,154 (total) x (0.0003511)/350 days/yr = $92.05/day v
(d) $80,950x0.0003511 [from (c) above] = $28.42
(e) See (e), page 44 = $72.00
(f) See (f), page45 = $18.00
(g) Chemicals for recycling dye penetrant rinse water:
Activated carbon 35.3 kg/day (77.7 Ib/day)
77.7 Ib/day x $0,50/lb x 50% (saved by pyrolytic regeneration) = $19.4/day
*0.132695 = Periodic payment for annuity, 8%, 12 years (2).
**0.116830 = Periodic payment for annuity, 8%, 15 years (2).
46
-------
TABLE 9. DYE PENETRANT RINSE WA TER, ITEMIZED OPERA TING ECONOMICS
Recycling Plant
Total equipment cost including installation
Equipment amortization cost per day
Operating labor cost per day
Chemicals— cost per day
Electric power cost per day
Maintenance cost per day
Sludge disposal cost per day
Fresh water purchase cost per day
Membrane replacements
Total
Total volume processed per day
Cost per 1000 liters
Cost per 1000 gallons
$262,154
92
72
19
1
7
0
1
9
$200
(a)
(c)
(e)
(g)
(i)
(j)
(I)
(o)
(P)
121,120 liters
(32,000 gallons)
$1.65
$6.25
Nonrecycling Plant
$80,950
28
18
4
2
4
10
$66
(b)
id)
(f)
(h)
(k)
(m>
(n)
121,120 liters
(32,000 gallons)
$0.54 ^
$2.06
(h) 8% of chemical process rinse water chemical costs [(g), page 451
(i) 10% of power for RO [(i), page 46 ]
$12/dayx 10% = $1.20/day
(j) 10% of operating labor cost
(k) Same as (j)
(1) 947 1/mo (250 gal/mo) of water/oil mixture
250 gal/mo 29.25 days/mo x $0.04/gal = $0.34/day
(m) 8% of sludge disposal costs for chemical process rinse water,
[(m), page 63) $55.56x8% = $4.44/day
47
-------
(nj 121,120 liters (32,000 gallons)/day @ $7.93/kl ($0.30/100 gal)
(o) 10%of(n)
(p) Membrane replacement cost, 20% x $45,000 (Appendix F, page 79, item 83) every 3 years
= $9,000/3yr/350day/yr = $8.57/day
TABLE 10. MACHINE SHOP COOLANT, ITEMIZED OPERA TING ECONOMICS
Total equipment cost including installation
Equipment amortization cost per day
Operating labor cost per day
Chemicals— cost per day
Electric power cost per day
Maintenance cost per day
Sludge disposal cost per day
Fresh water purchase cost per day
Membrane replacements
Total
Total volume processed per day
Cost per 1000 liters
Cost per 1000 gallons
Recycling Plant
$208,978 (a/
73 (c)
192 (e)
1 (h)
19 (i)
7 (k)
•
3 (n)
$295
24,224 liters
(6,400 gallons)
$12.18
$46.09
Nonrecycling Plant
$52,800 (b)
19 (d)
21 (f)
11 (g)
2 (j)
7 (I)
2 (m)
,
$62
24,224 liters
(6,400 gallons)
$2.56
$9.96
BASES AND CALCULATIONS FOR TABLE 10
(a) From Section 5, page 3 7 .
(b) From (b), page 43.
(c) $79,927 (equipment) x 0.132695*/350 days/yr = $30.30
$129,927 (buildings) x 0.116830**/350 days/yr = $43.08
$208,978 (total) x (0.0003511) = $73.78
*0.132695 = Periodic payment for annuity, 8%, 12 years (2).
**0.116830 = Periodic payment for annuity, 8%, 15 years (2).
48
-------
(d) $52,800 x 0.0003511 [from (c) above].
(e) See (e), page 45.
(f) See (f), page 45.
(g) Hydratedlime $1320/yr
Ferric sulfate $2520/yr
Total $3840/yr
$3840/350 days/yr = $10.97/day
(h) 10% of power for RO [(i), page 46].
(i) 10% of operating labor cost.
G) Same as (i) above.
(k) 18,925 1/mo (5000 gal/mo) at $0.04/gal disposal cost.
5000 gal/29.25 days/mo x $0.04 = $6.84/day
(1) Same volume and cost as (k) above.
(m) 24,224 I/day (6,400 gal/day) @ $0.0793/kl ($0.30/1000 gal) = $1.92/day
(n) Membrane replacement cost, 20% x $15,000 every 3 years
= 20% x$ 15,000/3 yr/350day/yr = $2.86
BASES AND CALCULATIONS FOR TABLE 11
(a) From Section 5, page 37.
(b) From (b), page 43.
(c) $ 64,980 (equipment) x 0.132695*/350 days/yr = $24.64/day
$105,047 (buildings) x 0.116830**/350 days/yr = $35.06/day
$ 170,027 (total) x (0.0003511)
(d) $112,751 x 0.0003511 [from (c) above]
(e) See (e), page 45.
(f) See (f), page 45-
*0.132695 = Periodic payment for annuity, 8%, 12 years (2).
**0.116830 = Periodic payment for annuity, 8%, 15 years (2).
$59.70/day
$39.59/day
49
-------
TABLE 11. CYANIDE RINSE WA TER, ITEMIZED OPERA TING ECONOMICS
(g)
(h)
(i)
0)
(k)
(1)
Total equipment cost including installation
Equipment amortization cost per day
Operating labor cost per day
Chemicals-cost per day
Electric power cost per day
Maintenance cost per day
Sludge disposal cost per day
Fresh water purchase cost per day
Total
Total volume processed per day
Cost per 1000 liters
Cost per 1000 gal Ions
Chlorine = $2430/yr/350 days/yr
Same as (g) above.
1 .4% of power for RO unit [(i), page 46]
1 0% of operating cost
1 0% of operating labor cost
1 .4% of sludge disposal cost for chemical
Recycling Plant
$170,027 (a)
\
60 (c)
24 (e)
7 (g)
fit
\>>
7 (j)
1 (I)
0 (o)
$99
22,710 liters
(6,000 gallons)
$ 4.36
$16.50
= $ 6.94
less than $1.00
= $ 7.20
= $ 2.10
$ 0.78
Nonrecycling Plant
$112,751 (b)
40 (d)
21 (f)
7 (h)
2 (k)
1 (m)
2 (n)
$73
22,710 liters
(6,000 gallons)
$ 3.21
$12.17
-
process rinse water [(m), page 4 6 ]
(m) Same as (1) above ,;
(n) 22,710 1/day (6,000 gal) at $0.0793/kl = $1.80
($0.30/1000 gal)
(o) 10%of(n)
* 0.132695 = Periodic payment for annuity, 8%, 12 years (2).
** 0.116830 = Periodic payment for annuity, 8%, 15 years (2).
50
-------
SECTION 7
MOVIE
SHORT TITLE: "Closing the Loop"
DESCRIPTIVE TITLE: "Water Recycling in the Airplane Manufacturing Industry"
GENERAL
A 16mm color and sound movie dealing with the subject of water usage and water recycling
in the airplane industry has been made. The movie explains where, how and why water is used in
the manufacture of airplanes, emphasizes the importance of protecting the environment, and shows
how the airplane industry and the Environmental Protection Agency are working together to develop
water recycling techniques. The water treatment loops described in this report are shown in the
movie.
Intended Audience
The movie is suitable for showing to a broad range of audiences including those in:
o Technical societies
o High schools
o Colleges
o Companies employing electropolating and
metal finishing processes
o Ecology groups
o Government agencies concerned with environmental protection
o Aircraft manufacturing companies
Equipment costs and operating costs of water recycling equipment are not discussed in the
movie.
Duration of movie: 13 minutes - 30 seconds
Availability
The movie is available from Boeing. Requests should be addressed to, and should refer to
"Closing the Loop," Boeing movie reference no. 4249. > Boeing Commercial Airplane Company
Public Relations, Organization 6-1051
Mail Stop 65-47
P. O. Box 3707
Seattle, WA 98124
51
-------
SECTION 8
ANALYTICAL METHODS
The chemical analyses were made using the following methods:
Nitrate
Sulfate
Orthophosphate
Chloride
Surfactant
Freon extractables
Total cyanide
Free Chlorine
Floating oil and slime
Plate count
Sodium, magnesium,
potassium, aluminum,
zinc, total chrome,
copper, cadmium
Trivalent chrome
Specific ion electrode, Orion Research Analytical Methods Guide
7th Edition, May 1975
ASTMD516MethodA
ASTMD 151 Method A
Silver nitrate titration using chloride specific ion electrode, Orion
Research Analytical Methods Guide, 7th Edition, May 1975
Methylene blue method, Standard Methods of Water Analysis, 13th
Edition, Section 159A
Method No. 0056, Methods for Chemical Analysis of Water and
Waste, U.S. Environmental Protection Agency
Liberation by Method No. 001'22, Methods for Chemical Analysis of
Water and Waste, U.S. Environmental Protection Agency; Measure-
ment by specific ion electrode, Orion Research Analytical Methods
Guide, 7th Edition, May 1975
Colormetric Test per ASTM Dl 253
Visual measurement of sample in graduated cylinder
Tryptone agar incubation, Standard Methods of Water A nalysis, 13th
Edition, Section 400
Atomic absorption spectrophotometry, Methods for Chemical Anal-
ysis of Water and Waste, U.S. Environmental Protection Agency
Cerimetric titration, Treatise on Analytical Chemistry, Part II,
Analytical Chemistry, Volume 8
Total dissolved solids
Evaporation at 50°C per ASTM D 1888
-------
REFERENCES
1. Anon., Standard Methods for the Examination of Water and Waste-water, American Public
Health Association, Washington, D. C. 1193 pp.
2. Gushee, C. H., Financial Compound Interest and Annuity Tables, Boston Financial Publishing
Company. 884 pp.
3. ASTM Metric Practice Guide, E-380-74, American Society for Testing and Materials. 34 pp.
4. "Crack Extension Test," para. 4.4.2, Qualification of Subcontractors to Perform Structural
Bonding, Boeing Document No. D-16925.
BIBLIOGRAPHY
Perry, J. H. Chemical Engineers' Handbook. McGraw-Hill.
Sourirajan, S. Reverse Osmosis, Academic Press Inc., New York. 580 pp.
Webber, Jr., W. J. Physicochemical Processes. Wiley-Interscience, New York. 640 pp.
Wixon, R., W. G. Kell, and N. M. Bedford. Accountants' Handbook. The Ronald Press Company,
New York. 800 pp.
53
-------
Abbreviations
c
Qf
'm
R
APPENDIX A
MATHEMATICAL ANALYSIS OF THREE RO SYSTEMS
Concentration of feed
Concentration of permeate
Concentration of concentrate
Flow rate (quantity) of feed
Flow rate (quantity) of permeate
Flow rate (quantity) of concentration
Mean concentration, pressure side of membrane
Rejection ratio of membrane
= Concentration difference across membrane
Concentration on pressure side of membrane
= (cm-cp)/cm= i-cp/cm
E = Recovery of RO unit = Qp/Qf
Basic Formula for a Single RO Unit
A single RO unit is represented diagrammatically as follows:
«vv HS*
MEMBRANE
RO UNIT
PERMEATE
CONCENTRATE
Qc'Cc
54
-------
Derivation of Output of a Single RO Unit in Terms of Input E, and R
By definition, E = Qp/Qf
and, R = 1 - C /Cm
By volume balance, Qf = Qn + QP
CD
(2)
(3)
The RO units used in the experimental work all employed feedback of concentrate in a vendor-
installed loop:
FEED.
LOOP
I
PERMEATE
CONCENTRATE
Since the loop flow was at least ten times the feed flow, the mean concentration at the membrane
differs negligibly from the concentrate concentration.
In this study, therefore, Cm and Cc are assumed to be the same, so that
From (l)and(3)
R = l-Cp/Cc
Qp = EQf
Qc = QrQp = Qf 0-
By material balance, and (5) and (6):
QfCf = Qf(l-E)Cc + QfECp
which can be simplified, using (4), to:
Cp = Cf(l-R)/(l-ER)
Cc = Cf/d-ER)
(4)
(5)
(6)
(7)
(8)
(9)
The basic single RO unit thus has the following outputs for unit inputs, i.e., for Qf - 1, Cf - 1.
55
-------
Qf=1
Cf-1
Derivation of Formula for Arrangement No, 1
FEED
(SIMULATED
RINSE INPUT)
VE
C = <1
Qc=1-E
RECYCLED WATER OUTPUT
FINAL
CONCENTRATE
TO WASTE
By starting with RO-2, assigning values of flow and concentration to each stream in accordance with
(5), (6), (8) and (9), using flow balance and material balances at junctions A and B, the following
relations were arrived at:
Qp = Flow of permeate when feed flow (Qf) = 1
= E - E2(1-E)/A(1-E)
Qc = Flow of concentrate when feed flow (Qf) = 1
= (1-E)2/A
Cp = Concentration of permeate when Cf = 1
= YD (1-R)
Cc = Concentration of concentrate when Cf = 1
= Y/(1-ER)
r
The parameters A, Y, and D are:
A = [(1-(1-E)E)/(1-E)] -E
Y = A/[(A + E)D(1-ER)-E(1-R)]
D = [d-ER)2-(l-E)E(l-R)]/[l-(l-E)E] (1-ER)
56
(10)
(11)
(12)
(13)
(14)
(15)
(16)
-------
Numerical values derived from formulae (10), (11), (12) and (13) are shown graphically in Figures
A-l and A-2 and tabulated in Table A-l.
1.000
FEED CONCENTRATION = 1.000
06 07
RECOVERY, E = Qp/Qf
08
0.9
Figure A-1. Permeate concentration plotted as a function of recovery (E) and rejection ratio (R)
for RO Arrangement No. 1.
57
-------
100
0.6
0.7
RECOVERY, E = Q
0.8
o.g
Figure A-2. Concentrate concentration plotted as a function of recovery (E) and rejection ratio
(R) for RO Arrangement No. 1.
58
-------
TABLE A-1. CALCULATED FLOWS AND CONCENTRATIONS FOR RO ARRANGEMENT
NO. 1, WITH UNIT INPUT FLOW AND CONCENTRA TION
E R Qp
0.5
0.6
0.5 0.7 0.750
0.8
0.9
0.5
0.6
0.6 0.7 0.877
0.8
0.9
0.5
0.6
0.7 0.7 0.953
0.8
0.9
0.5
0.6
0.8 0.7 0.988
0.8
0.9
0.5
0.6
0.85 0.7 0.996
0.8
0.9
0.5
0.6
0.9 0.7 0.999
0.8
0.9
Qc CP
0.62
0.51
0.250 0.39
0.26
0.13
0.72
0.62
0.123 0.49
0.34
0.17
0.84
0.76
0.047 0.63
0.46
0.24
0.94
0.90
0.012 0.82
0.67
0.39
0.98
0.95
0.004 0.91
0.80
0.52
0.99
0.99
0.001 0.97
0.92
0.71
Cc
2.1
2.5
2.8
3.2
3.6
3.0
3.7
4.7
5.7
6.9
4.2
5.9
8.5
12.0
16.5
5.7
9.2
15.8
28.8
52.3
6.4
11.0
21.0
45.3
106.7
7.0
12.8
26.7
69.1
238.4
59
-------
Derivation of Formula for Arrangement No. 2
Arrangement No. 2, the conventional arrangement in which all permeate flows are united, is
as follows:
SIMULATED
RINSE INPUT
Qf.Cf
RECYCLED^
WATER OUTPUT
QP'CP
• POINT "A"
"[FEED
RO-1
CONCENTRATE
FEED
RO-2
PERMEATE
I
CONCENTRATE
PERMEATE
FEED
RO-3
J
FINAL
CONCENTRATE
TO WASTE QC, Cc
PERMEATE
Starting at point A, and assigning the values Qf = 1 and Cf = 1, applying equations (5), (6), (8)
and (9) leads to the following:
Qp = Flow of permeate
Qp = E3 - 3E2 + 3E
Qc = Flow of concentrate
(17)
(18)
Cp = Concentration of permeate
[1-((1-E)/(1-ER))3]/(E3 - 3E2 + 3E)
Cc = Concentration of concentrate
= (1-ER)~3
(19)
(20)
Numerical values derived from equations (17), (18), (19) and (20) are shown graphically in
Figures A-3 and A-4 and tabulated in Table A-2.
60
-------
1.000
0.5 0.6 0.7
RECOVERY, E = Q /Qf
Figure A-3., Permeate concentration plotted as a function of recovery (E) and rejection ratio (Ft)
for RO Arrangement No. 2.
80
0.5
0.7
RECOVERY,E=(
Figure A-4. Concentrate concentration plotted as a function of recovery (E) and rejection ratio
(R) for RO Arrangement No. 2.
61
-------
TABLE A-2. CALCULA TED FLOWS AND CONCENTRA TIONS FOR RO ARRANGEMENT
NO. 2, WITH UNIT INPUT FLOW AND CONCENTRA TION
E R Qp
••••••^ ^ ^^^^^^^^^^^HH^^^H^^^^^^^^^^^^^^^H
0.5
0.6
0.50 0.7 0.875
0.8
0.9
0.5
0.6
0.60 0.7 0.936
0.8
0.9
0.5
0.6
0.70 0.7 0.973
0.8
0.9
0.5
0.6
0.75 0.7 0.983
0.8
0.9
0.5
0.6
0.80 0.7 0.992
0.8
0.9
0.5
0.6
0.85 0.7 0.997
0.8
0.9
0.5
0.6
0.90 0.7 0.999
0.8
0.9
Qc CP
^^•^^^^^^•^^^^^^•••••^^^^^^^^•••v^^^^^^^^^— __*M^M-*^^^M^MBI*l*v«v^^ta
0.80
0.73
0.125 0.62
0.48
0.28
0.87
0.81
0.064 0.72
0.58
0.37
0.93
0.89
0.027 0.82
0.70
0.48
0.95
0.92
0.016 0.87
0.77
0.55
0.97
0.95
0.008 0.91
0.84
0.64
0.99
0.97
0.003 0.95
0.90
0.74
1.00
0.99
0.001 0.98
0.96
0.86
Cc
— — . _ ^^_ ^^^^^^^
2.4
2.9
3.6
4.6
6.0
2.9
3.8
5.1
7.1
10.3
3.6
5.1
7.5
11.7
19.7
4.1
6.0
9.3
15.6
29.1
4.6
7.1
21.4
45.6
5.3
8.5
15.0
30.5
77.1
6.0
10.3
19.7
45.6
145.8
62
-------
Derivation of Formula for Arrangement No. 3
Arrangement No. 3 is as below:
Applying equations (5), (6), (8) and (9) to the arrangement leads to the following values of
output for the unit input of Qf = 1 , Cf = 1 :
QP =
Flow of permeate
E2/[1-2E(10E)]
= Flow of concentrate
Concentration of permeate
Cc =
Concentration of concentrate
(l-REr2/X
The parameter X is:
X = (1+RE-2E-2E2R + 2E2)/(1-2E + 2E2 - ER + 2 E2R -2E3R)
(21)
(22)
(23)
(24)
(25)
Numerical values calculated from (21), (22), (23) and (24) are shown graphically in Figures A-5
and A-6 and tabulated in Table A-3.
63
-------
1 000
a 0.750
o 0.500
o
o
ID
0.250
R =0.5
R -0.6
R = 0.7
0.5
0.6
0.7
0.8
RECOVERY, E = Qp/Qf
0.9
Figure A-5. Permeate concentration plotted as a function of recovery (E) and rejection ratio (R)
for RO Arrangement No. 3.
0.7
RECOVERY, E
Figure A-6. Concentrate concentration plotted as a function of recovery (E) and rejection ratio
(R) for RO Arrangement No. 3.
-------
TABLE A-3. CALCULATED FLOWS AND CONCENTRATIONS FOR RO ARRANGEMENT
NO. 3, WITH UNIT INPUT FLOW AND CONCENTRA TION
E R Qp
0.5
0.6
0.5 0.7 0.500
0.8
0.9
0.5
0.6
0.6 " 0.7 0.692
0.8
0.9
0.5
0.6
0.7 0.7 0.845
0.8
0.9
0.5
0.6
0.8 0.7 0.941
0.8
0.9
0.5
0.6
0.9 0.7 0.988
0.8
0.9
Qc CP
0.40
0.28
0.500 0.17
0.08
0.02
0.52
0.38
0.308 0.24
0.12
0.03
0.68
0.55
0.155 0.39
0.21
0.07
0.85
0.76
0.059 0.63
0.42
0.14
0.96
0.94
0.012 0.89
0.77
0.45
Cc
1.6
1.7
1.8
1.9
2.0
2.1
2.4
2.7
3.0
3.2
2.7
3.4
4.3
5.2
7.0
3.4
4.8
7.0
10.4
13.5
3.9
5.9
9.9
19.3
45.3
65
-------
Volume of Permeate and Concentrate
The volumes of permeate and concentrate for each of the RO arrangements are shown graph-
ically in Figure A-7.
SINGLE RO UNIT
0.20
RO
ARRANGEMENT
NO. 3
RO RO
ARRANGEMENT ARRANGEMENT
NO. 2 NO. 1
0.80
0.85
0.90
0.95
D
_J
O
>
LU
<
LU
s
cc
LU
O.
1.0
Figure A-7. Volume yields for Arrangements Nos. 1, 2 and 3.
66
-------
APPENDIX B
REVERSE OSMOSIS TESTS-REJECTION RATIO OF NITRATES
COMPARED WITH OTHER IONS
The very low rejection ratios obtained on NO3 by chemical analysis, on solutions of mixed
ions (simulated rinse water from anodizing, deoxidizing conversion coating, alkaline cleaning and
chem-milling), suggested the need for confirmatory tests. Accordingly, the series of tests shown in
Table B-l was run, all with conductivity measurements as a measure of concentration. All conduc-
tivities were corrected to a standard temperature of 20°C.
TABLE B-1. REVERSE OSMOSIS TESTS, REJECTION RA TIO OF NITRA TES COMPARED
WITH OTHER IONS
Test no. Electrolyte
1 NaCI
2 Tap water
3 HN03,
approxi-
mately
1500mg/l
pH = 2.0
Purpose of test
Confirm overall
functioning
Clean out equipment
Measure rejection
ratio
RO units
RO-1
RO-2
"RO-3
RO-1
RO-2
RO-3
RO-1
RO-2
RO-3
Conductivities
P/C, S/cm
370/2909
395/2645
187/2310
10/81
10/89
10/110
2932/3788
2500/3060
2339/3069
Rejection ratio
R
0.87
0.85
0.92
0.88
0.89
0.91
0.23
0.18
0.24
pH = 2.0
MgS04
pH = 7,0
Measure rejection
ratio of SO^—at
lowpH
Reconfirm overall
functioning
RO-1
RO-2
RO-3
RO-1
RO-2
RO-3
Not tested
28/985
35/1532
243/5523
0.91
0.97
0.96
67
-------
APPENDIX C
REVERSE OSMOSIS-EXPERIMENTAL VERIFICATION OF MATH ANALYSIS
Purpose of Tests
Practical tests were made using the (modified) demonstration equipment, to check the overall
accuracy of the mathematical models.
Test and Results
The demonstration equipment was modified to operate in Arrangement Nos. 1 and 2. Recovery
ratios for each RO unit were adjusted by reducing pressure on the membranes.
Arrangement No. 1
Feed concentration Cf
Permeate concentration Cp
Concentrate concentration Cc
Arrangement No. 2
Feed concentration Cf
Permeate concentration Cp
Concentrate Concentration Cr
Recovery ratio E = 0.6
Actual
563
236
2967
Recovery ratio E = 0.6
Actual
638
401
3622
Rejection ratio R = 0.70 (av.)
Predicted by math analysis
275
2646
Rejection ratio R = 0.72 (av.)
Predicted by math analysis
440
3510
Discussion of Test Results
Agreement between theoretical and actual concentrations was excellent for both arrangements
at the value of recovery ratio E, (0.6) used. Equipment limitations prevented testing at other values
of E. (If RO-1 is adjusted to a recovery of, e.g., 0.8, then the feed to RO-2 becomes inadequate to
maintain it in operation. Similar considerations apply to RO-2 and RO-3.)
68
-------
APPENDIX D
CALCULATOR PROGRAM FOR COMPUTING REVERSE OSMOSIS UNIT PRODUCT AND
CONCENTRATE CONCENTRATIONS
The general problem of calculating concentrate concentration and permeate concentration in
an RO unit employing feedback (i.e., having a proportion of the concentrate stream fed to the in-
take) is best solved by a reiterative technique. Such calculation is time consuming unless a computer
or programmable calculator is available. This appendix contains a 99-step program devised for use
on a Hewlett-Packard HP-65 calculator, that reiterates the necessary calculations until two succes-
sive results differ by less than one part in 1000.
REVERSE OSMOSIS PROGRAM-CALCULATION OF PRODUCT AND
CONCENTRATE CONCENTRATIONS HP PROG. NO. 04198A
Program Description, Equations, Variables
This program computes the concentrations of permeate and concentrate in an idealized reverse
. osmosis system employing feedback.
Qf = Quantity of feed
Q = Quantity of permeate
Input data required: Qc = Quantity of concentrate
Qj = Quantity of loop (feedback)
Cf = Concentration of feed
R = Rejection of membrane
The program computes Cp and Cc (concentrations of permeate and concentrate)
CALCULATION STEPS
1. Calculates a first rough approximation for Cp from Cp = Cf/3
2. Uses this value of Cp to compute Cc from Cc = (Qf Cf - Qp Cp)/Qc
3. Cm= [QfCfH-Cc(2Q1+Qc)]/(Qf+2Q1+Qc)
69
-------
4. Uses this value of Cm to compute Cp from Cp - (1 -R) x Cm
5. Inserts this value of Cp into step 2 and recalculates Cc
6. Repeats steps 3, 4, and 5 until two successive values of Cc differ by less than one part in a
thousand.
Operating Limits and Warnings
1. Any units may be used for Qf, Qp> etc., such as gallons per hour, liters per hour, etc., but must
be consistent.
2. Similarly, Cf may be in any units. Cp and Cc will be in the same units as Cf.
3. Rejection, R, must be in the form of a decimal fraction, e.g., 0.8, not 80%.
i
Sample Problem
Given a reverse osmosis unit with rejection R = 0.8 for a given ionic species, a feed concen-
tration of 200 mg/1 in the same ionic species, and the following flow rates, Qf = 3.0, Qp = 2.0,
Qc = 1.0, and Qj = 5.0 (all in gallons per hour), calculate the concentration of the ionic species in
the permeate (Cp) and in the concentrate (Cc).
Solution
INPUTS: A 3 R/S 2 R/S 1 R/S 5 R/S 200 R/S 0.8 R/S
ANSWERS: B 78.25 (i.e., Cp = 78.25 mg/1)
C 443.41 (i.e., Cc = 443.41 mg/1)
Reference
Weber, Formulae 7-14 through 7-21, Physicochemical Processes for Water Quality Control,
Wiley Interscience, 1972, pages 326, 327.
70
-------
PROGRAM
KEY
ENTRY
LBL
A
f
SF 1
0
STO 7
STO 8
STO
g
R/S
STO 1
R/S
STO 2
R/S
STO 3
R/S
STO 4
R/S
STO 5
R/S
STO 6
RTN
LBL
B
D
RCL 7
RTN
LBL
C
D
RCL 8
RTN
LBL
D
f""1
TF1
CLX
RTN
RCL 5
3
^
STO 7
LBL
3
RCL 2
X
CHS
RCL 5
RCL ?
X
CODE
SHOWN
23
11
31
51
00
3307
3308
33
09
84
3301
84
3302
84
3303
84
3304
84
3305
84
3306
24
23
12
14
3407
24
23
13
14
3408
24
23
14
32
61
44
24
3405
03
81
3307
23
03
3402
71
42
3405
3401
71
COMMENTS
INITIALIZE
STORE o IN REG 7
STOREo IN REGS
STOREo IN REG 9
STOP FOR ENTRIES
STORE Of IN REG 1
STORE Op IN REG 2
STORE Qc IN REG 3
STORE QI IN REG 4
STORE Cf IN REG 5
STORE R IN REG 6
-END OF PROG. "A"-
RUN PROG. "D"
RECALL Cp
-END OF PROG. "B"-
RUNPROG. "D"
RECALL Cc
-END OF PROG. "C"-
(MAIN PROGRAM)
IS FLAG 1 LOWERED?
IE HAS CALC. BEEN RUN?
IF SO, CLEAR DISPLAY
AND RETURN
OTHERWISE, PROCEED
WITH CALC. CALC.
APPROX.Cp
STORE APPROXCp IN 7
IDENTIFY BEGINNING
OF MAIN CALCULATION
CALCULATE Cc
FROM
= QfCf-QpCp
0.
KEY
ENTRY
+
RCL 3
-=•
RCL 8
g x 2 y
STO8
-
g
ABS
RCL 8
-=-
EEX
CHS
3
gxzy
f-1
SF 1
RCL 1
RCL 5
X
RCL 8
RCL 4
2
X
RCL 3
+
X
+
RCL 1
RCL 4
2
X
+
RCL 3
+
-=-
RCL 6
CHS
ENTER
1
+
X
STO 7
f
TF 1
GTO
3
CLX
RTN
CODE
SHOWN
61
3403
81
3408
3507
3308
51
35
06
3408
81
43
42
03
3524
32
51
3401
3405
71
3408
3404
02
71
3403
61
71
61
3401
3404
02
71
61
3403
61
81
3406
42
41
01
61
71
3307
31
61
22
03
44
24
COMMENTS
CALCULATE
FRACTIONAL CHANGE
IN Cc SINCE LAST
IS FRAC. CHANGE
LESS THAN TO"3?
IF SO, LOWER FLAG 1
"CALCULATION OVER'
OTHERWISE, GO
ON WITH CALC.
CALCULATE Cm
FROMCm =
QfCf + Cc(2Qi+Qc)
Of + 2Qi + Qc
CALCULATECp
= Cm(1-R)
IS FLAG 1 STILL
FLYING?
IF SO, REITERATE
CALCULATION
OTHERWISE, CLEAR
DISPLAY AND STOP.
71
-------
APPENDIX E
COMPOSITION OF CYANIDE PLATING SHOP SOLUTIONS
The cyanide solutions used to simulate plating shop rinse water were production plating shop
solutions of the following compositions:
Copper plate Cadmium plate Ti-cadmium plate Enstrip S
g/1 oz/gal g/1 oz/gal g/1 oz/gal g/1 oz/gal
Cu 23.9 3.19 - -
NaOH - - 18.95 2.53 17.15 2.29
144.5 19.3
v2
Rochelle
salt
NaCN
Cd
Ti
46.0
to
43.8
9.2
—
Nil
6.15
to
5.85
1.23
—
Nil
103.3
24.6
Nil
13.8
3.24
Nil
107.1
26.28
0.100
14.3
3.51
1.33x
72
-------
APPENDIX F
EQUIPMENT LIST AND BREAKDOWN DIAGRAMS FOR
FULL-SCALE WATER TREATMENT PLANT
The equipment list for the water recycle system (Table F-l) precedes Figures F-l through F-8,
breakdown diagrams of the overall schematic for the full-scale water recycling plant.
The circled numbers on these diagrams also appear in the left-hand columns in Table F-l,
Equipment List, where they are listed in numerical order.
73
-------
TABLE F-1. EQUIPMENT LIST, WATER RECYCLE SYSTEM
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Name
Pumps
Acid operating tank to clarifier, pump
Acid upset tank to clarifier pump
Cyanide operating tank to cyanidestruct
unit, pump
Cyanide upset tank to cyan destruct
unit, pump
Treated cyanide - cyan destruct unit to
clarifier pump
Centrifuge surge tank to clarifier pump
Acid underground tank 50,000 gal, to
upset and opertaing tanks pump
Cyanide underground tank, 10,000 gal,
to upset and operating tanks pump
Caustic metering pump NaOH to clarifier
and cyanide destruct unit
Pump polyelectrolyte tank to clarifier
Filter feed tank to sand filters pump
RO feed tank to sand filters pump
RO feed tank to reverse osmosis pump
Reverse osmosis to evaporator pump
Reverse osmosis to water recycle storage
pump
Size
6 in.
6 in.
1 in.
1 in.
1 in.
1 in.
4 in.
1 in.
14 in.
1/2 in.
6 in.
4 in.
6 in.
2 in.
6 in.
Hp Capacity
20 500 gal/min
20 500 gal/min
1/2 7 gal/min
)4 7 gal/min
Vx 7 gal/min
Vz 5 gal/min
10 200 gal/min
2 10 gal/min
1 20 gal/hr
1 30 gal/hr
20 500 gal/min
10 300 gal/min
15 500 gal/min
5 100 gal/min
20 500 gal/min
Material
SS Pacific pump
model 4070-5
SS 4070-5
Iron 1250-1 LN
Iron 1250-1 LN
Penton
Iron 1250-1 LN
SS
Iron
Iron
SS
Iron 4070-5
Iron 3070-5
Iron 4070-5
Iron 1570-5
Iron 4070-5
Remarks
In-line easy removal
In-line easy removal
Duplex sump pump with float
switch and auto alternator
Duplex sump pump with floor
switch and auto alternator
Variable meter
Variable meter
In-line easy removal
In-line easy removal
In-line easy removal
In-line easy removal
In-line easy removal
$5,850
5,850
400
400
800
400
10,000
1,200
1,000
1,400
1,125
780
950
520
1,150
t A conversion table for the engineering units used in Table F-1 and Figures F1 thru F8 is provided on page viii.
-------
Item
16
17
18
20
21
22
23
24
25
26
27
28
29
30
Name
Evaporator to water recycle storage pump
Evaporator to dryer pump
Water recycle storage to process rinse
tank pump
Coolant upset and surge tank to oil
separator tank pump
98% H2SO4 drum to 5% H2SO4 tank
pump
5% H2SO4 tank to pH adjust and 2nd
oil separator pump
pH adjust and 2nd oil separator to
ultrafilter pump
RO feed tank to RO pump
Coolant recycle tank to coolant make
up tank pump
Dye upset and surge tanks to ultrafiltration
pump
Ultrafiltration to carbon filters pump
Recycle tank to dye penetrant rinse
pump
50% H2SO4 tank to cyanide destruct unit
50% H2S04 tank to RO feed tank
Size Hp
2 in. 5
1 in. 1/2
6 in. 25
1 in. 1/2
1/2 in. 1/3
1/2 in. 1/3
1 in. 1
1 in. %
1 in. '/2
2 in. 3
2 in. 3
2in. 3
1/2 in. 1
1/2 in. 1
Capacity
100 gal/min
10 gal/min
500 gal/min
7 gal/min
5 gal/min
4 gal/min
7 gal/min
7 gal/min
6 gaf/min
70 gal/min
70 gal/min
70 gal/min
lOgal/hr
1 gal/hr
Material Remarks
Iron 1570-5 In line easy removal
Iron 1250-1 LN
Iron 4070-5 In-line
Iron 1250-1 LN
Polypropylene Drum pump, metering
Polypropylene or Metering pump
H2S04 acid res.
Iron 1250-1 LN
Iron
1250-1 LN
Iron 1270-5 In-line easy removal
Iron 1270-5 In-line easy removal
Iron 1270-5 In-line easy removal
Polypropylene Metering pump
Polypropylene Metering pump
$520
400
1,200
400
1,200
1,200
400
400
400
500
500
500
1,400
1,400
TABLE F-1. (continued)
-------
Item
Name
Size Hp Capacity
Material
Remarks
33 Coolant RO unit sump pump to sewer 1 in.
34 Cyanide spillage sump pump
35 Acid spillage sump pump
36 Yard sump pump
Tanks
39 50% H2SO4 tank
40 Acid upset tank No. 1
41 Acid upset tank No. 2
42 Acid operating tank
43 Acid spillage underground tank
44 Cyanide upset tank
45 Cyanide operating tank
46 Cyanide spillage underground tank
2 in.
2 in.
2 in.
1 10gal/hr
2 50 gal/min
T/2 200 gal/min
1 30 gal/min
5ftdia
7fthi
35 ft dia
35 ft hi
35 ft dia
35 ft hi
15 ft dia
15 ft hi
12 ft dia
60ftlg
15 ft dia
15ft hi
7 ft dia
7 ft hi
8 ft dia
27ftlg
Iron Duplex sump pump with float
switch and auto alternator
Iron 2095-0 Duplex sump pump with float
switch and auto alternator
SS or acid reservoir Duplex sump pump with float
2011-0 % switch and auto alternator
SS or acid reservoir Duplex sump pump with float
2095-0 switch and auto alternator
1,000 gal Steel lined with
polyethylene or
PVC
250,000 gal Steel w/epoxy paint
250,000 gal Steel w/epoxy paint
20,000 gal Steel w/epoxy paint
50,000 gal Steel w/epoxy paint
20,000 gal Steel
2,000 gal Steel
10,000 gal Steel
$1,000
2,000
8.500
8,000
330
24,875
24,875
9,370
19,611
9,600
465
2,900
TABLE F-1. (continued)
-------
Item
47
48
49
50
51
52
i 53
i
54
55
56
57
58
59
Name
50% NaOH tank
Poly electrolyte tank
Centrifuge surge tank
Filter feed tank
RO feed tank
Water recycle storage tank No. 1
Water recycle storage tank No. 2
Coolant upset tank
Coolant surge tank
5% H2SO4 tank
Tanks and equipment
pH adjust and 2nd oil "separator
Insoluble oil tank
Oil/water concentrate
batch removal tank
Size Hp
9 ft dia
1 1 ft hi
9 ft dia
1 1 ft hi
5 ft dia
7 ft hi
15 ft dia
15 ft hi
15 ft dia
15 ft hi
35 ft dia
35 ft hi
35 ft dia
35 ft hi
10 ft dia
10 ft hi
5 ft dia
5 ft hi
31/2 ft dia
4 ft hi
3% ft dia
4 ft hi
8 ft dia
8 ft hi
Capacity
5,000 gal
5,000 gal
1,000 gal
20,000 gal
20,000 gal
250,000 gal
250,000 gal
6.000 gal
600 gal
300 gal
300 gal
300 gal
3,000 gal
Material Remarks
Steel
Steel w/epoxy paint
Steel
Steel
Steel w/epoxy paint
Steel w/epoxy paint
Steel w/epoxy paint
Steel
Steel
Steel lined with
polyethylene or PVC
Steel lined w/
polyethylene or PVC
Steel
Steel lined w/PVC
$1,500
1,937
300
10,289
10,389
29,956
30,590
6,640
235
140
206
206
810
TABLE F-1. (continued)
-------
Item
60
61
62
63
64
65
66
-j
00
67
68
69
70
71
72
73
74
Name
RO feed and ultraviolet sterilizer
Coolant recycle storage tank
Dye penetrant rinse water upset tank
Dye penetrant rinse water-surge tank
Dye penetrant oil/water concentrate
batch removal tank
Dye penetrant water recycle tank
Centrifuge to clarif ier surge tank
Cyanide destruction unit
Chlorine, supply
Sludge dryer
RO unit (2) 500 gal/min units
Evaporator 10:1
Clarifier
Centrifuge
Dry solids hopper
— — • ^^===^==
Size
2 ft dia
3 ft hi
10 ft dia
10 ft hi
17 ft dia
18 ft hi
8 ft dia
8 ft hi
31/2 ft dia
4 ft hi
17 ft dia
17 ft hi
2 ft dia
3 ft hi
=^=^=^^=
HP Capacity
75 gal
6,000 gal
30,000 gal
3,000 gal
250 gal
30,000 gal
75 gal
10 Ib/day
at 400 gal/hr
Four 400 Ib
cylinders
520 gal/hr
Two 500 gal/hr
5000 gal/hr
30,000 'gal/hr
500 gal/hr
5 tons
=^=^==
Material Remarks
Steel w/epoxy paint $1,000 for sterilizer
Steel w/epoxy paint
Steel
Steel
Steel
Steel w/epoxy paint
Steel
Steel, rubber lined
or equivalent
Steel Tanks by vendor
Stainless steel
By supplier Two units =
Steel, brass SS
Steel
Stainless steel
Steel
•••^^^••^•^^^•"•••^^^•^•^^^^^^^^^^•^^^^^^^^^^^^^^^^™ iimimmiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiimM^i^^^^^^^^
$2,118
11,400
1,118
206
12,847
200
29,000
50,000
400,000
200,000
45,000
25,000
2,000
TABLE F-1. (continued)
-------
Item
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Name
Sand filter unit
Mixer, surge tank No. 49
Motor, surge tank No. 49
Mixer, H2S04 tank No. 56
Motor, H2SO4 tank No. 56
Coolant oil separator
Coolant ultrafiltration unit
Coolant reverse osmosis unit
Dye penetrant rinse water
ultrifiltration unit
Cargon filters unit
Chlorine ORP controller unit complete
NaOH pH controller unit complete,
cyanide destruct unit
NaOH pH controller unit complete,
clarifier unit
Conveyor system centrifuge solids
to dryer
Conveyor system dryer to hopper
50% H2SO pH controller unit complete,
cyanide destruct unit
Size HP Capacity Material Remarks
30,000 gal/hr Steel
2 SS shaft and propeller
2
2 Corrosion reservoir
elastomer coated
shaft propeller
2
200 gal Steel
400 gal/hr By supplier
400 gal/hr By supplier
2000 gal/hr By supplier
2000 gal/hr Polypropylene
Included in item 67
200 Ib/hr Steel
100 Ib/hr Steel
Included in item 67
$55,000
2,500
3,000
21,300
1 5,000
15,000
45,000
20,000
2,000
1,500
TABLE F-1. (continued)
-------
Item
91
92
93
94
95
96
97
98
00 "
o
100
101
102
103
104
105
106
107
Name
Size HP
Capacity
Material
Remarks
50% H2SC>4 pH controller unit complete,
RO feed tank
5% H2SO4 pH controller unit
complete, pH adjust and 2nd oil separator
Control room monitoring instrumentation
Level control
Level control
Level control
Gate
Ball
Ball
Check, swing
Ball
Gate
Gate
Gate
Ball
Check, swing
i
Check, swing
% in. sere wed
I/a in. screwed
1 in.screwed
1 in.screwed
2 in.flg
4 in.flg
6 in.flg
4 in.flg
6 in.flg
6 in.flg
4 in.flg
125 Ib
125 Ib
125 Ib
125 Ib
125 Ib
125 Ib
150 Ib
150 Ib
150 Ib
150 Ib
Steel
31 6 SS crane
No. 950 TF
316SScrane
No. 950 TF
All iron No. 36
316SS
Cast iron No. 4651/2
Cast iron No. 4651/4
316SS 61176
316SS 61176
316 SS 61676
316 SS 61676
Required
30 required x $24 =
48 required x $38
6 required x $24
28 required x $87
18 required x $190
24 required x $295
5 required x $620
12 required x$1100
4 required x$1243
1 required x $710
$5,000
50,665
720
1,824
144
2,436
3,420
7,080
3,100
13,200
4,972
710
TABLE F-J (continued)
-------
6"
RECYCLED WATER
J
"F"( Figure F-4)
* t-
RINSE TANKS
IN CHEMICAL
PROCESS LINE
(EXCEPT 1
CYANIDE) 1
I PH 2-3 1
1
6"
000 gal/day
ii
* O
r ?
SUMP | <°
i_i
U
o.
5*
TO SEWER ^ "
RINSE TANKS
IN CYANIDE
PROCESS
LINES i
1
I
UPSET UPSET
TANK NO. 1 TANK NO. 2
250,000 gal 250,000 gal
_£. « JL ± n JL
••u J »<7i ,4. © MO*!
"^ 6" PVC *£ J ' •
l_ 6" PVC
OPERATING ^ fa
TANK
20,000 gal (T)
6" pvc~T (a} -L ^j^ «*
600 ppm
30,000 gal/hr
ACID * 1
V UPSET x-v
J 1 TANK 1©
0
— — J*— REMOTE
L LEVEL TO CL
(Figu
x ^^- CuNUUCI'vUY
/^
— ^ FLOW
1
^
D
"D"
ARIFIER
re F-2)
"A" (Figure F-3)
NONRETURN VALVE
Figure F-1. Collection and return system for chemical process and cyanide process rinse waters.
81
-------
FROM COLLECTION SYSTEM
TO H2SO4
TANK
"C"
FROM CYANIDE
DESTRUCT UNIT
oo
10
POLY-
ELECTRO-
LYTE
TANK
1000 PPM
30 gal/hr
G" (Figure F-4]
OVERFLOW TO FILTER FEED
4
700 PPM
30.000 gal/hr
•B" (Figure F-1)
FILTER FEED
SAND FILTERS
RO FEEDTANK
30,000 gal/hr
750 PPM pH8.5
CLARIFIER
FROM YARD DRAIN SUMPS AND
ACID UPSET AND OPERATING
TANKS SPILLAGE SUMP
UNDERGROUND PV
TANK i
50.000 gal I
"D" (Figure F-1)
TO ACID UPSET AND
OPERATING TANKS
CENTRIFUGE
CHUTE OR CONVEYOR
8% SOL I PS 20 gal/hr
TO REVERSE
OSMOSIS UNIT
E" (Figure F-4)
TO DRYER
Figure F-2. Clarifier and filters, chemical process rinse water purification.
-------
50%
oo
OJ
FROM CYANIDE
RINSE TANKS
J"A" (Figure F-1)
> m-
-8 J»Ki
-5 ' *
a
~u — r *~
T f
SUMP LJ
t
TANK
1000
n *•'
CHLORINE *
70 Ib/dav '
~LJ - ® ®
UPSET TANK : (»1 ;;
20.000 gal S r"^ 3
<3day) ^ ^^r^r^ \ °:
CO ' 4£---4ORp|\55) >
^? ^v Jv T^LJ V^ p^l o.
vJ **^ r (g) i '"l
1 1 i
1 i [j^^
200 PPM CYANIDE
400 gal/hr . DESTRUCT
5- 10lb/day UNIT
CYANIDE
/N ... & I
— -, V/ - w
m>CD ATlMfi » 1
TANK .I
2000 gal '": L L
X*™\ «™_
1 » 1 J •"•* *- —
1 1 ^*V
?•
[10gal/rnin ^_
J _ __— FROM CYAN IDE UPSET
FROM NaOH TANK («)
' I
"C" TO RO
^^ 1-btU I ANK.
^ 1 gal/hr 1/2"
2500 PPM 7f
400 gal/min r^ /*s
OH
^ «
*—
TOCLARIFIER
"B" (Figure F-2)
[UNDERGROUND "^ AND OPERATING SPILLAGE
I TANK
i 10,000 gal fa
Figure F-3. Cyanide destruct unit
-------
700 PPM
30,000 gal/hr
REVERSE
OSMOSIS
TWO (2)
TO RINSE TANKS
"F" (Figure F-1)
CITY WATER
MAKEUP
2"
2" 5000 gal/hr
"G" (Figure F-2)
FROMCLARIFIER
AND FILTERS
LJL
WATER
RECYCLE
STORAGE
250,000 gal
NO. 1
2"
WATER
RECYCLE
STORAGE
250,000 gal
NO. 2
6"
EMERGENCY OVERFLOW
TO SEWER (OPTIONAL)
CHUTE OR CONVEYOR
100lb/hr
D
10 gal/hr
200 PPM
(I*)
FROM CENTRIFUGE
F-4. Reverse osmosis unit, chemical process rinse water purification.
84
-------
OUTPUT
RECYCLED WATER
INPUT
2,000 gal/hr
UPSET TANK
30,000 gal
(1 day)
-03-
SURGE
3000 gal
-co-
—00-
BATCH REMOVAL OF
OIL-WATER CONCENTRATE
250 gal/mo
SOOOgal/hr
2"
RECYCLE STORAGE
30,000 gal
BATCH REMOVAL
OF SPENT CARBON
Figure F-5. Dye penetrant inspection rinse water purification.
85
-------
WASTE MACHINE
SHOP COOLANT -
400 gal/hr
pH = 8-9
UPSET TANK
6000 gal
(1 day)
+4-
SURGE AND II
SETTLING
600 gal
OIL
SEPARATOR
1/2"
i -
) 396 gal/hr pH ADJUST
AND 2ND
OIL SEPARATOR
_... CDCC
WASTES
BATCH REMOVAL OF OIL-WATER
CONCENTRATE (5000 gal/mo)
;
RO
®
.
280 gal/hr _ __
RECYCLE STORAGE
6000 gal
©
V< "C
/7p-*«-«-TOCLARIFIER
IT® 100 gal/hr
u
TO MACHINE SHOP
COOLANT MAKEUP TANK
280 gal/hr
Figure F-6. Machine shop coolant water purification.
86
-------
FROM ACID RINSE
TANKS
FROM CYANIDE RINSE
TANKS
4 IN. TO SEWER •
WASTE MACHINE SHOP-
COOLANT
TO MACHINE SHOP
COOLANT MAKEUP
TANK
CYANIDE
DESTRUCT
UNIT
j
©/•N. SURGE
VWASTEN x^x
1 r*r\m AMT 1(5511
INSOLUBLE OILS
98% H
«QM<(§> 25 gal CARBOY
(Figure F-8)
Figure F-7. Plant layout, water recycle system.
87
-------
TO RINSE TANKS IN PROCESS LINE t
FROM CITY WATER I •»
FROM DYE
PENETRANT RINSE TANKS
TO DYE
PENETRANT RINSE TANKS
OIL/WATER
CONCENTRATE
WATER
RECYCLE
STORAGE
WATER
RECYCLE
STORAGE
Ci-
YARD DRAIN
I
4",
4" 4."
-------
APPENDIX G
OPERATING ECONOMICS OF A FULL-SCALE PLANT-CALCULATIONS
Equipment and Installation Amortization Costs
Interest rate =
Amortization period =
Periodic payment, from annuity =
tables, reference (2) p. 53
Payment per day =
Hence:
1. Chemical process rinse water =
equipment =
2. Dye penetrant rinse water =
equipment =
3. Machine shop coolant =
equipment =
4. Cyanide rinse water equipment =
Building and Utilities Amortization Costs
Interest rate
Amortization period =
Periodic payment, from annuity =
tables, ref (2)
Payment per day =
Hence:
1. Chemical process rinse water =
building and utilities =
12 years
0.132695
0.13295/350
3.79129E-04
$2,145,923x3.79129E-04
$813.58/day
$225,006x3.79129E-04
$85.30/day
$163,320 x 3.79129 E-04
$61.92/day
$112,751 x 3.79129 E-04
$42.75/day
8%
15 years
0.116830
0.116830/350
3.33799 E-04
$599,918x3.33799 E-04
$200.25
89
-------
2. Dye penetrant rinse water
building and utilities
3. Machine shop coolant
building and utilities
4. Cyanide rinse water
building and utilities
Operating Costs
Man-hours per day
Cost per day
Hence:
$37,148 x 3.33799 E-04
$12.40
$45,658x3.33799 E-04
$15.24
$57,276x3.33799 E-04
$19.12
16
16x$30
$480
1 . Chemical process rinse water
plant labor
2. Dye penetrant rinse water
plant labor
3 . Machine shop
plant labor
coolant
4. Cyanide process rinse
plant labor
Total costs
Chemical process
rinse water
Dye penetrant
rinse water
Machine shop
coolant
Cyanide rinse
water
Equipment plus
installation
cost amortiz.
per day
$813.58
$ 85.30
$ 61.92
$ 42.75
$480 x 30%
$144/day
= $480x15%
= $77/day
$480 x 40%
= $192/day
= $480x15%
= $77/day
Building Total
and util. facility
cost amortiz. cost amortiz.
per day per day
$200.25 $1,013.83
$ 12.40 $ 97.70
$ 15.24 $ 77.16
$ 19.12 $ 61.87
Operating
cost
per day
$144.00
$ 77.00
$192.00
$ 77.00
90
-------
APPENDIX H
COMPOSITION OF ALKALINE FLUSHING SOLUTION FOR
ULTRAFILTRATION MEMBRANES
NaOH 70 grams
Proctor & Gamble's ERA* 200 milliliters
Formalin * * 100 milliliters
Water 100 liters
The above solution, at 21°-38°C (70°-100°F), was circulated under approximately 200 kPa
(30 lb/in.2) pressure until an acceptable permeate flow rate was restored—normally between 1 and
6 hours.
* Heavy duty anionic and nonionic laundry surfactant
** 34% formaldehyde
91
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-78-130
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
AIRCRAFT INDUSTRY WASTEWATER RECYCLING
. REPORT DATE
June 1978 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robinson, A. K. and Sekits, D. F.
8. PERFORMING ORGANIZATION REPORT NO.
D6422054
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The Boeing Commercial Airplane Company
P.O. Box 3707
Seattle, Washington 98124
10. PROGRAM ELEMENT NO.
1BB610
11. CONTRACT/GRANT NO.
S 803073
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory—Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati. Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The feasibility of recycling certain categories of water used in the manufacture of airplanes was demon-
strated. Water in four categories was continuously recycled in 380-liter (100-gallon) treatment plants;
chemical process rinse water, dye-penetrant crack-detection rinse water, electroplating process rinse water
containing cyanides, and machine shop water-based coolant.
The estimated capital cost was $3.4 million for equipment to recycle the above categories of water in a
typical, medium-sized airplane factory generating 1.5 Ml (0.4 x 10" gal)/day. Recycling costs were esti-
mated to be: $0.94/kl ($3.57/1000 gal) for chemical process rinse water; $1.65/kl ($6.25/1000 gal) for
dye penetrant rinse water; $4.36/kl ($16.50/1000 gal) for cyanide process rinse water; and $12.18/kl
($46.09/1000 gal) for machine shop coolant.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Circulation
Water Recovery
Water Reuse
68D
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
TInclassifip.fi
21. NO. OF PAGES
100
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
92
* us. gwDMMiTnmiMG OFFICE. mt-w - uo /1383
------- |