WATER POLLUTION CONTROL RESEARCH SERIES
17040 EEE 12/71
Wastewater Demineralization
by Ion Exchange
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Environmental
Protection Agency, through inhouse research and grants
and contracts with Federal, State, and local agencies,
research institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications
Branch, Research Information Division, Research and
Monitoring, Environmental Protection Agency, Washington,
D. C. 20460.
-------
WASTEWATER DEMIIMERALIZATION BY ION EXCHANGE
by
Ed Kreusch
and
Ken Schmidt
CULLIGAN INTERNATIONAL COMPANY
NORTHBROOK, ILLINOIS 60062
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
PROJECT #17040 EEE
CONTRACT #14-12-599
December, 1971
For sale by the Superintendent or Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1 25
-------
Environmental Protection Agency
Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents
necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
Pilot plant studies were conducted on sewage effluent from an activated
sludge treatment plant. These studies used a pretreatment system and
an ion exchange system operating in series. The pretreatment system
consisted of lime clarification, dual media filtration, and granular
activated carbon filtration. This system reduced the total phosphate,
suspended solids (turbidity), and total organic carbon content of the
wastewater as pretreatment for the ion exchange system.
The ion exchange system was the main interest of the project—reduction
of the content of dissolved ionic inorganic salts in the wastewater.
Conventional wastewater treatment plants do not affect the concentration
of these salts. Reuse of wastewater without a reduction of these salts
will cause the recycled wastewater to become unacceptably brackish.
Lime clarification will reduce phosphate concentrations; however, when
ion exchange procedures are applied for wastewater demineralization,
the lime clarification is unnecessary. This is because phosphate removal
will occur in the ion exchange system.
Dual media filtration for removal of suspended solids followed by
granular activated carbon filtration for reduction of total organic
carbon is desirable.
Wastewater demineralization by ion exchange procedures can be success-
fully applied. Conventional ion exchange procedures are recommended.
Partial demineralization can be obtained simply by applying only a weak
acid cation exchange resin. This resin will reduce the total ionized
solids by an amount which is equivalent to the alkalinity present in the
wastewater. The reduction was approximately 50% in these studies. De-
mineralization of wastewater with a high concentration of nonalkaline
inorganic salts requires application of two ion exchange resins: a
strong acid cation exchange resin followed by weak base anion exchange
resin. Operating costs can be reduced by using a third resin: a weak
acid cation resin preceding (in service) the strong acid resin.
Waste regenerants from the ion exchange system must be reused, or
neutralized and disposed of by locally acceptable methods. Disposal
of sludges and saline liquid must be considered.
This report was submitted in fulfillment of Project 17040 EEE, contract
14-12-599, under the sponsorship of the Environmental Protection Agency.
Key words:
Wastewater demineralization, ion exchange, acid neutralization,
regenerant disposal, lime clarification, activated carbon.
111
-------
CONTENTS
Section
1 Conclusions 1
2 Recommendations 3
3 Introduction 5
4 Objectives 7
5 Pilot Plant Description 9
6 Pretreatment Operation 21
7 Weak Base Anion Exchange Resin Performance -
Bicarbonate Form 29
8 Strong Acid Cation Exchange Resin Performance -
Hydrogen Form 37
9 Weak Base Anion Exchange Resin Performance
Free Base Form 53
10 Weak Acid Cation Exchange Resin Performance -
Hydrogen Form 69
11 Weak Acid: Strong Acid Cation Exchange Resin
Performance - Hydrogen Forms 77
12 Waste Regenerant Disposition 97
13 Operating Material Requirements For Ion Exchange
Process 115
14 Acknowledgements 123
15 Definitions 125
y
-------
FIGURES
Figure Page
No. No.
1. Photograph of Pilot Plant Front View 10
2. Photograph of Pilot Plant Rear View 11
3. Photograph of Pilot Plant Wet Analysis Laboratory 12
4. Pilot Plant Pretreatment Flow Sheet 13
5. Photograph of Pilot Plant Pretreatment Systems 14
6. Pilot Plant Ion Exchange Flow Sheet 17
7. Photograph of Pilot Plant Ion Exchange Systems 18
8. Photograph of Pilot Plant Ion Exchange Valve
Control Panel 19
9. pH of Secondary Treated Sewage as Affected by Lime 23
10. Turbidity of Secondary Treated Sewage as Affected
by Lime 24
11. Phosphate Concentration in Secondary Treated
Sewage as Affected by Lime 25
12. Typical effluent conductivity from cation (IRC-120)
resin 40
13. Cation resin (IRC-120) regeneration efficiency,
exhaustion at 6 gpm/cu ft 43
14. Cation resin (IRC-120) regeneration efficiency,
exhaustion at 3 gpm/cu ft 44
15. Costs to produce 1000 gallons of water treated by
cation exchange resins 92
16. Costs to produce 1000 gallons of water treated by
cation exchange resins 93
17. Static neutralization of waste acid (FMA 18900 mg/l)
with 100, 105 and 110% stoichiometric lime dosage at
72o p 99
13. Static neutralization of diluted waste acid (FMA 3210
mg/l) with 100, 120 and 150%stoichiometric limestone
dosages at 72° F 100
vn
-------
FIGURES
Page
Fi9ure No.
19. Dynamic neutralization of waste acid 15
20. Effect of Lime Dosage on Ammonia Recovery 109
21. Effect of Temperature on Ammonia Recovery 110
22. Effect of Vacuum on Ammonia Recovery 111
23. Effect of Air Sweep on Ammonia Recovery 112
24. Ammonia Recovery by Steam Distillation 114
VI11
-------
TABLES
Table Page
No. No.
1. Variation of Influent Wastewater 22
2. Clarifier operation - typical effluent analyses 26
3. Performance of activated carbon filter 27
4. Anion Performance Recap Sheet (IRA-68) 32
5. Anion Resin (IRA-68) Water Sample Analyses'
Abstracts 33
6. Anion Resin (IRA-68) Wash Sample Analyses 34
7. Comparison of estimated performances of strong acid
cation exchange resin (IRC-120) 38
8. Strong acid cation exchange resin (IRC-120) per-
formance summary 42
9. Cation resin (IRC-120) performance recap sheet
(Runs 1-21) 45
10. Cation resin (IRC-120) performance recap sheet
(Runs 22-39) 46
11. Cation resin (IRC-120) performance recap sheet
(Runs 40-67) 47
12. Cation resin (IRC-120) performance, comparison of
estimates and actual 49
13. Water analysis summary of typical exhaustion of
cation exchange resin (IRC-120) 50
14. Regeneration effluent analyses from strong acid
cation resin (IRC-120) 52
15. Weak base anion exchange resin (IRA-93) performance
summary 56
16. Phosphate break-thru from weak base anion resin
(IRA-93) during service exhaustion 57
17. Anion resin (IRA-93) performance recap 58
IX
-------
TABLES
Table Pa9e
No. No-
18. Ionic break-thru from weak base anion resin
(IRA-93) during service exhaustion 59
19. Regeneration effluent analyses from weak base
anion resin (IRA-93) 61
20. Weak base anion exchange resin (IRA-68) per-
formance summary 62
21. Phosphate break-thru from weak base anion resin
(IRA-68) during service exhaustion 64
22. Anion resin (IRA-68) performance recap 65
23. Ionic break-thru from weak base anion resin (IRA-68)
during service exhaustion 66
24. Regenerant effluent analyses from weak base anion
resin (IRA-68) 68
25. Weak acid cation exchange resin (IRC-84) performance
summary 72
26. Weak acid cation exchange resin (IRC-84) performance
recap sheet 73
27. Cation resin (IRC-84) performance - prediction vs.
actual 74
28. Water analysis summary of typical exhaustion of
cation exchange resin (IRC-84) 75
29. Cation Resins Summary 79
30. Carboxylic Cation (IRC-84) Performance Recap 80
31. Sulfonic Cation (IRC-120) Performance Recap; Treating
Carboxylic Effluent 81
32. Sulfonic Cation (IRC-120) Performance Recap; Treating
Filtered Sewage 82
-------
TABLES
Table Page
_No.__ _No._
33. Performance of two beds of cation resin in series,
(IRC-84 preceding IRC-122). 85
34. Performance of two beds of cation resin in series,
(IRC-34 preceding IRC-122). 86
35. Typical effluent quality from system. 88
36. Cation resins summary. 89
37. Costs to produce 1000 gallons of cation resin
treated water. 90
38. Performance of two cation resins in one vessel. 94
39. Comparison of performance; requirements to treat
filtered treated sewage with separate beds or
layered beds of cation resin, 95
40. Analyses of Static Lime - Neutralized Acid Regenerant.101
41. Analyses of Static Lime - Neutralized Acid Regenerant.102
42. Characteristics of Lime Used. 103
43. Characteristics of Limestone Used. 104
44. Analyses of Acid Waste Before Neutralization with 36"
Limestone Bed. 107
45. Chemical requirements for strong acid cation and
weak base anion exchange system. 116
46. Chemical requirements for weak acid cation exchange
system. 118
47. Costs to produce 1000 gallons of demineralized water
with three resins. 120
48. Comparison of systems to produce 1000 gallons of
water by ion exchange. 121
XI
-------
SECTION 1
CONCLUSIONS
Pilot plant studies of several ion exchange systems have been conducted
with a treated sewage effluent from an activated sludge plant. These
studies have led to the following conclusions.
1. The phosphate content of secondary treated sewage is reduced by
addition of hydrated lime (calcium hydroxide). The approximate
reduction is from 18.6 mg/1 (median value) to less than 2 mg/1 at
pH 10.0; or, to a concentration of less than 0.5 mg/1 at pH 10.5.
2. The phosphate concentration can also be reduced from the 18.6 mg/1
median value to less than 2 mg/1 simply by ion exchange.
3. Ion exchange demineralization of this sewage which contains high
alkalinity (median value of 392 mg/1 as calcium carbonate) is more
costly with lime clarification than it is without lime clarification.
The chemical cost for the cation exchange resin partial treatment of
1000 gallons of sewage is 18
-------
8. Organic fouling of the ion exchange resins used did not develop.
9. Waste regenerant acid, from the cation exchange resins, was minimized
by using both weak and strong acid cation exchange resins in series. The
acid is easily neutralized with either limestone or hydrated lime.
10. Waste regenerant ammonium hydroxide from the anion exchange resin
can be treated with hydrated lime for liberation of ammonia, which can
be recovered and reused.
11. Operating costs for chemicals (regenerants and neutralizing agents)
to treat the activated sludge treated sewage effluent at Elgin, Illinois
range from 6.7 to 23.8 cents per 1000 gallons of product water. The low
value is for partial demineralization for removal of alkaline salts by weak
acid cation exchange. The high value is for complete exchange of all ionic
contaminants, without regenerant recovery. A realistic cost for complete
demineralization will be between these two values.
-------
SECTION 2
RECOMMENDATIONS
1. Design of plants for wastewater demineralization by ion exchange
using the demonstrated processes should be initiated. Sizes of
plants should be selected by the Environmental Protection Agency.
Plant design should include cost estimates for capital and operating
expenses.
2. The pilot plant facilities should be maintained operating with the
optimum ion exchange system for this wastewater to demonstrate the
reliability of the resins' operating characteristics. Repetitive
cycling through exhaustion and regeneration to demonstrate reproduc-
ibility of quantity and quality of product should be continued for an
uninterrupted six months test.
3. The pilot plant facilities should be maintained in further demonstra-
tion of other ion exchange processes for wastewater demineralization.
Such other processes should be established by agreement between the
Environmental Protection Agency and the contractor.
-------
SECTION 3
INTRODUCTION
Demands for increased amounts of usable water are prevalent. Satisfy-
ing these demands requires not only prudent use of water supplies, but
its resue wherever practical. Wastewater has, because of its very
nature, always been discarded as unusable, in spite of diminishing
supplies of "fresh water". Removal of objectionable contaminants from
wastewater will permit its reuse. This recycling of wastewater by its
reuse will achieve two primary goals. The first of these is an increase
in the available supply of water at a location of water usage. The
proximity of the wastewater reclamation plant and the need for an in-
creased water supply precludes high distribution costs. The second
achievement is a reduction of pollution downstream from the reclamation
plant.
Wastewater reclamation requires the removal of impurities which may be
broadly classified in three categories: insoluble materials, soluble
organic materials, and soluble inorganic salts. Whereas many operating
plants partially remove the contaminants in the first two categories,
separation of dissolved inorganic salts from wastewater has not been
widely applied. Separation techniques using ion exchange demineraliza-
tion are known, but their application to sewage is not generally prac-
ticed. Nor have these techniques been studied in sufficient detail in
the application to wastewater to permit predictable performance.
Reduction of dissolved inorganic salts is readily obtained by the appli-
cation of the ion exchange process. This process has been in wide com-
mercial use for decades, but application to wastewater treatment has been
negligible. A prime deterent has been the affect of high molecular weight
organic compounds present in the wastewater. These compounds have a de-
leterious effect on most anion exchange minerals. Recently, new types
of resins, less affected by organics, have become commercially availabe.
The present investigation includes studies on the performance of these
newer resins on wastewater demineralization.
Ion exchange resins must be periodically regenerated with chemical re-
agents. Regeneration produces a relatively small volume of a highly
mineralized waste. The application of ion exchange systems for waste-
water demineralization must consider reduction of these waste regenerant
volumes together with their disposal. Possible reclamation and reuse
of regenerants has been considered in this investigation.
The Environmental Protection Agency, Water Quality Office and Culligan
International Company have concluded the agency's Contract No. 14-12-599
to study ion exchange processes applied to wastewater reclamation.
-------
Culligan constructed a pilot plant on the properties of the Elgin
Municipal Sanitary Treatment Plant. Daily sewage flow through the
municipal sewage treatment plant is approximately 6 million gallons
per day during the winter months, peaking at 12 million gallons during
the summer. The plant uses both trickling filter and activated sludge
secondary treatment. This investigation used the effluent from the
clarifier of the activated sludge effluent as the feed to the pilot
plant. Most of the work for this investigation has been at the pilot
plant. Detailed chemical analyses in support of the pilot plant were
made at the Analytical Laboratories of Culligan International Company.
Studies on possible regenerant disposal and reuse have been carried
out in the Research Laboratories of the company.
-------
SECTION 4
OBJECTIVES
This project was initiated with two main objectives which were
obtained. The first was to construct a highly flexible pilot
plant to study several ion exchange systems for the treatment
of the effluent from secondary treated sewage.
The second objective was to observe the performance characteristics
of the ion exchange systems by operating the pilot plant.
Operating conditions and processes were varied in order to obtain
performance data for the systems. This performance data is
presented in this report for use in estimating treatment costs
for wastewater demineralization by ion exchange.
Secondary objectives include an evaluation of the pretreatment
processes. Also, neutralization and reuse of regenerants were
studied.
-------
SECTION 5
PILOT PLANT DESCRIPTION
The pilot plant was located on the properties of the Elgin Municipal
Sanitary Treatment Plant. Activated sludge treated sewage clarified
effluent was delivered to the pilot plant. The pilot plant consisted
of two basic systems. The first was a pretreatment system to further
treat and clarify the influent received from the activated sludge
clarifier. The second was the ion exchange system to receive the
effluent from the pretreatment system. Figures 1 and 2 are photo-
graphs showing the exterior of the pilot plant. The pretreatment
system is housed in the frame building, while the ion exchange system
is contained in the adjacent trailer.
Suitable valving and sample points, combined with indicating and re-
cording instruments, were provided in the pilot plant to permit effi-
cient operation and collection of adequate, accurate data.
The pilot plant included facilities, partly shown in Figure 3, for wet
analyses of samples. Immediate chemical analyses and instruments'
readout at the pilot plant permitted rapid decisions concerning the
performance of the ion exchange and pretreatment systems. Verification
of the results with greater analytical detail was obtained by sub-
sequent analyses at the analytical laboratories of Culligan Interna-
tional Company.
Pretreatment equipment description. A brief description of the equip-
ment in the pretreatment system is offered. Figure 4 shows the flow
sheet, while Figure 5 shows a photographic view.
Influent to the pretreatment system was collected from the activated
sludge clarifier effluent by means of a low head centrifugal pump. A
flow restrictor placed in the discharge side of the pump limited the
flow into the pretreatment system to 15.0 gallons per minute. This
flow rate was the maximum design rate to provide 45 minutes detention
in the clarifier in our pretreatment system.
A dry feeder was mounted on top of the clarifier to feed hydrated lime
(calcium hydroxide) for clarification, phosphate precipitation, or other
tests as dictated by the test program. The feed rate was controlled by
a pH indicator - controller which analyzed a small sample stream col-
lected from the bottom of the rapid mixed chamber of the clarifier.
The pH indicator - controller had adjustable set points to permit fine
control of the pH.
The clarifier tank was cylindrical, 6 feet in diameter, 7 feet tall.
Flow was bi-directional: downwardly in the center of the tank, through
-------
o
Figure 1 - Photograph of Pilot Plant front view, showing ion exchange systems trailer against pre-
treatment systems shelter.
-------
Figure 2 - Photograph of Pilot Plant rear view, showing activated sludge clarifiers which supply
influent for tests.
-------
Figure 3 - Photograph of Pilot Plant wet analysis laboratory.
-------
Activated
Sludge
Clarifier
To Ion
Exchange
System
Chemical Feed
Carbonation
Retention
Tank
Activated
Carbon
Filter
Figure 4. Pilot Plant Pretreatment Flow Sheet
13
-------
Figure 5 - Photograph of Pilot Plant pretreatment systems.
14
-------
Effluent
PH
0
90 180 270
Cumulative Time In Minutes
360
Figure 19.
Dynamic neutralization of waste acid. Upflow
through 36 inches of limestone with 45% bed
expansion at 70° F. Influent FMA and sulfate,
in grams per liter as CaC03, shown on curves;
for complete analyses see Table 44.
15
-------
Ion exchange equipment description. The equipment was housed in the
mobile truck trailer shown in Figures 1 and 2. Figure 6 shows a flow
diagram of this equipment. The versatility of the ion exchange system
is illustrated in Figure 7, showing the piping and valve assemblies
located in front of the ion exchange vessels.
Easy manipulation of the valves in the ion exchange system to accom-
plish the desired flow path was readily obtained by the pilot valves
on the control panel illustrated in Figure 8. Operation of a pilot
valve illuminated an indicator light on the graphic panel mounted
above the control valve assembly. Therefore, the status of each
valve was readily known.
The ion exchange system consisted of five pressure vessels, containing
ion exchange resin. These vessels ranged from 10 to 14 inches in
diameter, 48 to 72 inches in height. The smallest vessel accomodated
1-|- cubic feet of resin with a 33 inch bed height, while still allowing
an approximate 30% freeboard. The pressure vessels were operated in-
dividually, or in various combinations. Flow direction during service
or regeneration, through the vessels was upflow or downflow as desired.
Regeneration was accomplished by preparing a solution of the desired
regenerant in a separate open vessel. A pump transferred this solu-
tion through the proper valves operated, at the valve control panel
to the ion exchange vessels. Rinse-out of the regenerant was accom-
plished by pumping rinse water from the regenerant vessel, or by use
of selected water streams from the plumbing network of the ion ex-
change system. The usual procedure was to use both sources of rinse
water - a small quantity of softened potable water following the re-
generant injection so that interconnecting plumbing was flushed;
final regenerant rinse-out was accomplished with "system water".
16
-------
Influent from
Pretreatment
System
5 Ion Exchange Vessels
Regenerant
Vessel
Deqasifier
Product
Water
Figure 6. Pilot Plant Ion Exchange Flow Sheet
17
-------
Figure 7 -
Photograph of Pilot Plant ion exchange systems, showing
piping versatility - vessels located behind piping.
18
-------
Figure 8 - Photograph of Pilot Plant ion exchange valve control panel,
-------
SECTION 6
FRETREATMENT OPERATION
During the first four months of this project, the attention given to
the performance of the pretreatment system (treatment of the influent
treated sewage preparatory for ion exchange treatment) was more detailed
than during the remainder of the project. Later studies did not use the
clarifier and recarbonation processes.
Throughout the project, wide variations were encountered in the raw
water chemical composition. The natural variations of the potable
water supply to the surrounding communities were further varied by
the diverse use of this water before discharge to the sewer. Surface
run-off with its natural contaminants plus street deicing salt during
the winter months were further causes of variation.
Frequent detailed chemical analyses were made of our influent from
the sewage plant during the first four months. Later studies were
to a lesser detail. Table 1 shows the variations which we encountered
in the influent water. The table shows the low, high, and median values
of our analyses for the constituents of our influent received from the
activated sludge clarifier. Table 1 also shows the percent maximum
deviation from the median valve. Largest deviations are to be found
in the nitrogen constituents (ammonia, nitrate, and nitrite) and in
the total organic carbon. These large variations indicate variable
operating efficiency of the activated sludge process at this location.
Repeated variation in our influent water continued throughout the
project.
This influent was pumped to our clarifier as discussed in Section 5.
The clarifier was designed to treat 15 gallons per minute, which provided
45 minutes retention with a rise rate of 0.57 gallons per square foot per
minute. Operational difficulties were encountered with the clarifier due
to non-continuous operation. The clarifier was shut down each night.
Therefore, on the following morning, some time was required to stabilize
the clarifier operation. During this stabilization time, the clarifier
effluent was diverted to drain; while stored, recarbonated, water in the
retention tank was used for ion exchange system tests.
Our clarifier operated with lime (calcium hydroxide) as a precipitant.
A volumetric dry feeder was mounted on top of the clarifier for con-
trolled addition of the lime. Jar tests were first used to establish
the desired dosage rate. Figures 9, 10, 11 show the effects produced
in these jar tests by varied doses of lime. A lime dose of 500-600
mg/1 produced a pH of 10-11, and turbidity of 4-5 JTU, while reducing
the total phosphate (as P04) from 10-12 to less than 1 mg/1.
21
-------
Table 1. Variation of Influent Wastewater,
Which Was Received From Activated
Sludge Clarifier.
Constituent*
Low Value High Value Median Value Deviation
PH
Turbidity, JTU
Total Alkalinity
CaC03
Chlorides,
CaC03
Sulfates
CaC03
Phosphates,
Total P04~~~
Magnesium,
CaC03
Calcium,
CaC03
Ammonia, N
Nitrate, N
Nitrite, N
Sodium,
CaC03
7.3
8
180
128
91
6.2
99
76
1.0
0.1
0.01
198
8.5
37
452
226
144
25.8
222
229
24.0
3.7
1.6
430
7.8
22
392
163
105
18.6
196
120
17.5
1.5
0.04
273
— "•
+68
-49
+39
+37
-67
-50
+91
-94
+147
+Vast
+58
Potassium,
CaC03
Total Organic
Carbon, C
19
25
100
20
39
+25
+157
-^Constituents reported in mg/1, except for pH and turbidity which
are expressed in units.
**Deviation shown in percent maximum from the median value.
22
-------
11
10
200 300 400
Lime Dose, mg/1 as
500
600
Figure 9. pH of secondary treated sewage a.s affected by lime.
Initial pH 8.0.
23
-------
Turbidity
JTU
200 400 60CT
Lime Dose, mg/1 as Ca(OH),
Figure 10i Turbidity of secondary treated sewage as affected
by lime. Settled 80 minutes.
24
-------
12
10
Phosphate
mg/1
as PC)
4
0
200
400
600
Lime Dose, mg/1 as Ca(OH),
Figure 11. Phosphate concentration in secondary treated sewage
as affected by lime.
25
-------
The clarifier was operated in this dosage range (500 - 600 mg/l), con-
trolled by a pH controller recorder. The intermittent operation of the
clarifier resulted in turbidities greater than expected: typically, 6 JTU
turbidity, with normal maximum of about 15 JTU. Similarly, reduction of
phosphate was also less than expected: typically, 1-2 mg/l of phosphate
in the effluent, with normal maximum of about 3 mg/l. Table 2 summarizes
the clarifier operation. Shown are typical results at three pH levels.
It must be remembered that variable raw water composition makes compari-
son of these results difficult.
Table 2. Clarifier operation - typical effluent analyses.
pH, units 10.1 10.3 10.5
Alkalinity, mg/l CaC03:
Hydrate 0 26 3.0
Carbonate 178 125 64
Turbidity, JTU units 864
Phosphate, total mg/l P04 1.9 1.8 0.3
Magnesium, mg/l CaC03 89 35 20
Calcium, mg/l CaC03 69 100 84
Ammonia, mg/l N 13.1 11.2 16.5
Nitrate, mg/l N 2.8 2.1 1.4
Nitrite, mg/l N 0.13 0.04 0.08
Total organic carbon, mg/l C 35 21 16
After clarification, recarbonation with gaseous carbon dioxide was
accomplished in the first compartment of the retention tank. Carbon
dioxide was introduced into the downflow stream of water, while effluent
pH was monitored with a recording pH meter. The pH after recarbonation
was easily maintained at the desired 1 pH unit depression. Some
precipitates settled in the retention tank, but these were inconsequential.
The water was then filtered through a dual media filter at a design flow
of 3.1 gpm/ sq ft; thence thru a granular activated carbon filter, which
provided a 40 minute contact time (empty bed basis). The activated carbon
filter reduced the influent total organic carbon approximately 50%. The
removal of total organic carbon during the first 227 bed volumes through
the filter is summarized in Table 3. When the filter was removed from
service after treating 400 approximate bed volumes, reduction of total
organic carbon was still effective - the remaining fraction (C/CO)
ranged 0.50 - 0.60.
26
-------
Table 3.
Performance of activated carbon
for removal of total organic
carbon, mg/1 C.
Total organic carbon
mean mq/1 C
Influent
(C0)
73.0
11.8
20.0
11.4
14.4
Effluent
(C)
21.0
5.1
6.6
6.1
5.8
Remaining
Fraction
(C/Co)
0.34
0.43
0.48
0.54
0.41
Bed Volumes
Treated
0-49
50-99
100-149
150-199
200-227
The effluent of the activated carbon filter was the influent to
the ion exchange treatment system.
27
-------
SECTION 7
WEAK BASE ION EXCHANGE RESIN PERFORMANCE - BICARBONATE FORM
This process employs the unusual charcteristics of a unique weak base
anion exchange resin, Amberlite IRA-68. In the free base form, this
resin (represented as R-N in the first equation below) is capable of
adsorbing carbonic acid to form the bicarbonate salt by exposure to a
solution of carbon dioxide. This bicarbonate form has a bicarbonate -
mineral acidity selectivity such that neutral salts in water can be
converted to alkaline salts. For example, the chlorides and sulfates
of sodium, calcium, and magnesium are converted to the corresponding
bicarbonate salts, as illustrated by the equation employing sodium
chloride:
(R-NH)HCOS + NaCl = (R-NH)Cl + NaHCOs
The effluent water, which has been treated by the bicarbonate form re-
sin, can then be treated with a bed of weak acid cation exchange resin.
This second bed of resin will convert the alkaline salts to water and
carbon dioxide, as illustrated by the equation employing sodium bicar-
bonate:
RCOOH + NaHC03 = RCOONa + H^Os
H2C03 = H20 + C02
A combination of the two ion exchange beds will then remove the ionic
salts from the water, leaving carbon dioxide in solution. The latter
is a gas which is readily eliminated through established technology
with aeration or degasification.
Incorporation of such an ion exchange system after an efficient process
for removal of organic contaminants will result in a reclaimed water
without pollution due to inorganic and organic materials.
Attempts to apply the bicarbonate form of the weak base anion exchange
resin have been only partially successful. The technology for the
application of this process is insufficiently developed to support
pilot plant studies on wastewater with low salinity. Acceptable
techniques for converting the resin to the bicarbonate form were not
established. Several consultations with the manufacturer of the resin
were unrewarding. However, it was mutually agreed that our pilot plant
techniques were proper. It is apparent, therefore, that studies of a
more fundamental nature must be made on the process if it is to be
applied to low salinity wastewaters.
As specified in the research contract, a bed of the weak base anion
exchange resin was employed to treat the effluent from the complete
pretreatment system (lime clarification, recarbonation and settling,
29
-------
filtration, activated carbon adsorption). Two cubic feet of anion
exchange resin (IRA-68) were contained in a 12-inch diameter tank: the
height of mineral was about 31 inches in place. Typically the resin
was regenerated with a 4% solution of ammonium hydroxide at dosages
which exceeded those recommended by the resin manufacturer.
Softened potable water was used for preparation >of regenerant solutions
and for regenerant rinse. Soft water is necessary to prevent the
alkaline regenerant from precipitating the hardness cations of the raw
water.
Conversion of the free base resin to the bicarbonate form was accom-
plished by recirculating softened water saturated with dissolved carbon
dioxide, maintained at a pressure of 75 psi 'in accordance with the re-
sin manufacturer's recommendations. Carbon dioxide was supplied from a
tank of the compressed gas, fitted with standard regulating devices,
pressure gauges, and a flow indicator. The gas was introduced into the
system through an in-line carbon dioxide saturator in the pressurized
recirculated water line flowing through the resin. The conditions
encountered during carbonation were inexplicable. The rate and duration
of carbon dioxide addition were inconsistent. No well defined means were
available to determine when sufficient carbon dioxide had been used. Car-
bonation times varied from 1/2 to two hours to apply the calculated amount
of carbon dioxide required.
Exhaustion of the bicarbonate form resin was with the sewage effluent
from the pretreatment system. The exhaustion flow rate was 2 gpm/cu ft,
in accordance with the resin manufacturer's recommendation and the re-
search contract.
Termination of the resin's exhaustion cycle was determined in the field
by analyses of the alkalinity and chloride in the effluent water. During
service, all non-alkaline anions, such as chloride and sulfate, were
exchanged for alkalinity. As exhaustion of the resin approached, com-
plete removal of non-alkaline anions no longer occurred; as a result,
the alkalinity in the effluent correspondingly decreased. Service ex-
haustion was terminated when the effluent alkalinity decreased about
10% from the typical value obtained on similar samples analyzed through
the service cycle. Confirming chemical analyses were made on selected
samples submitted to the Analytical Laboratory of the parent company.
Similar field and laboratory analyses were made for chloride in the
effluent. During most of the service cycle, the effluent chloride
concentration was low. As the resin approached exhaustion, the effluent
chloride concentration increased while the effluent -alkalinity corres-
pondingly decreased.
To determine the capacity of the resin, the influent loading factor
must be known. Determination of the loading factor was made by using
a small ion exchange column filled with hydrogen form strong acid
30
-------
cation exchange resin. Passage of water through this column removes
all cations by hydrogen exchange: acids result in the effluent. These
acids are of two types: strong acids produced essentially from chlorides
and sulfates; and weak acids produced essentially from the alkaline
salt. The strongly acidic anions are those which must be removed in
the pilot plant by the bicarbonate form anion exchange resin: the
weakly acidic anions are unaffected by the bicarbonate form resin.
The concentration of strong acids produced by the column was easily
measured by titration with a standard solution of sodium hydroxide.
Proper sample preparation through the cation resin column requires
pre-rinsing the column with demineralized water, passage of a measured
sample through the column, and its rinse-out with demineralized water.
The rinse water must be combined with the effluent collected from the
sample passage to insure total recovery of all strong acids produced
in the test column. The titration is reported in terms of calcium
carbonate equivalents. This is the loading factor for calculating the
anion exchange resin's operating capacity. The total operating capacity
of the anion resin was then determined from the loading factor and the
volume of water treated during service. For example, Table 4 attached,
shows (on a per cubic foot basis) that run number one treated 86 bed
volumes of water with a loading factor of 340 mg/1. The 86 bed volumes
is 645 gallons (each cubic foot has a bed volume of 7.5 gallons) per
cubic foot of ion exchange resin. The loading factor of 340 mg/1 is
19.9 grains (340 mg/1 divided by 17.1) of loading per gallon of water
treated. Multiplying these two products results in a capacity of
12,280 grains per cubic foot; or 12.8 kgr/cu ft.
The performance of the resin during exhaustion is summarized in the
attached TE ble 4. The first exhaustion produced a capacity of 12.8
kilograins per cubic foot to the alkalinity - chloride breakthrough.
The next 11 cycles produced results which were too variable to consider.
Regeneration to the free base form for these 11 cycles used dosages of
ammonium hydroxide calculated to 125% of stoichiometric to the capacity
obtained in exhaustion. Because of the poor performance, runs 12, 13,
and 14 were made with the resin regenerated with ammonium hydroxide
equivalent to 100$ of the total capacity of the resin, which is several
times the equivalent capacity realized. A stable average capacity of
10.8 kgr was obtained per cubic foot of resin, employing the excessive
ammonium hydroxide regeneration.
Table 5 shows the typical partial analyses of effluent water samples
collected from the exhaustion runs. The influent samples were a com-
posite (drip sample) collected throughout the exhaustion; while effluent
samples were collected' at the end of the exhaustion to show ionic
breakthrough. The partial analyses are presented to show the effect
of the bicarbonate exchange process on the three ions, as well as to
show the chloride breakthrough.
31
-------
00
ro
RUN NO.
2-11
21
22
23
24
TABLE 4. ANION PERFORMANCE RECAP SHEET, per cubic foot. Regenerated at 1/2 gpm/cu ft
(4 bed volume/hour), exhausted at 2 gpm/cu ft (16 bed volume/hour). Bicarbonate
form of IRA-68 system.
Exhaustion **
Regeneration
NH OH
Used
Kgr*
5-10
12
13
14
15
16
17
33.9
33.9
33.9
13.5
13.5
13.5
13.5
13.5
co2
Used
Lbs
Bed***
Volumes
Load
mg/1*
1-2
3-3/4
40
45
35.0 NaOH
102
34
46
216
218
214
208
Capacity
Kqr*
86
--
92
90
96
54
84
62
340
--
275
264
252
245
220
236
12.8
5-8
11.1
10.4
10.6
5.8
8.1
6.4
18
19
20
13.5
13.5
13.5
1-3/4
1-3/4
2
79
68
62
260
296
294
9.0
8.9
8.0
3.8
4.3
3.2
4.2
As CaCOs equivalent. ** to 10% Chloride break-thru.
gal. per bed volume.
COMMENTS
New resin. Carbonation was difficult
unfamiliar.
Results too variable.
Co-current regeneration for 12, 13,
14. Average capacity 10.8 Kgr.
Co-current regeneration, NH^OH used at
125% of average capacity for 12, 13, 14.
Leakage for 15 is high. Capacities are
unstable.
Countercurrent regeneration, NFLOH
used at the same 125% of 12, 13, 14.
Leakage for 18 & 19 is high, but capacity
is fairly stable at average 8.6 Kgr.
Concern expressed about resin fouling.
Treated with two bed volumes of 3% HC1.
Clarifier, recarbonation by-passed.
Countercurrent regeneration, Drastic
capacity loss questions loss of resin.
Physical inspection shows resin present
in good condition.
Low capacityl
Regenerated high to determine maximum
capacity obtainable.
-**-*• Volume resin occupies; 2 cu ft, or 15
-------
TABLE 5. Anion resin (IRA-68) water sample analyses' abstracts
collected during exhaustion. Results expressed in
mg/1, as calcium carbonate equivalents. Bicarbonate
form resin system.
Run No. 13 17 20 22 23 24
Alkalinity:
Influent 150 102 78 404 330 326
Effluent 368 328 390 564 444 416
Chlorides:
Influent 172 120 152 130 142
Effluent 40 60 20 34 52 58
Sulfate:
Influent 97 116 144 88 100
Effluent 010 200
Note: Influent samples were composited throughout the exhaustion,
effluent samples were spot-collected at the endpoint of
the exhaustion.
An attempt was made to reduce the amount of regenerant required by
returning the dosage to 125% of the stable average 10.8 kilograin
capacity obtained in runs 12 - 14. This was a regenerant dosage of
13.5 kilograins (as calcium carbonate equivalent) for runs 15, 16, and
17. Co-current flows (downflow regeneration - downflow service) were
used. The capacities obtained from these three runs were low and
unstable, while the anion concentration to service was high.
In an attempt to improve the performance, counter-current procedures
(downflow exhaustion - upflow regeneration) were adopted for runs 18,
19, and 20. The same regeneration level, 13.5 kilograins per cubic
foot, was used. It produced a higher, fairly stable average capacity
of 8.6 kilograins per cubic foot; however, anion leakage was of
questionable quality.
At this point, concern was expressed about possible fouling of the
resin. Therefore, the resin was washed with 2 bed volumes of 4%
hydrochloric acid, downflow. The effluent during this acid treatment
was sampled and analyzed for suspected contaminants, including sulfate,
phosphate, iron, total organic carbon. No indication of foulants was
found in the analyses. Table 6 shows the analyses of selected samples
of the wash effluent where ionic concentrations peaked. The table also
shows similar effluent analyses for samples collected during a later
conducted sodium hydroxide wash. This latter wash is discussed below.
33
-------
TABIjE ,6. Anion resin (IRA-68) wash sample analyses. Collected
during--'clean-up procedures after separate exhaustions.
' AlT'results in mg/1. Bicarbonate form resin system.
Hydrochloric Acid Wash* Sodium Hydroxide Wash*
Bed volume thru 1.8 2.5 ' 3.2 1.8 2.5 3.2
Sulfate,, CaC03 2600 220 8- 2600 1800 600
Phosphate, P04"T> 28.5 9.4 1.5 340 183 145
Calcium, QaC03 15 6" 3 2 0.4 0
Iron., Fe '.. 36.6 4.8 0.6 0.8 1.3 0.8
Aluminum, Al 150 40 10 6.0 6.5 4.5
Organic Carbon-, C 16 84 4625 2000 1750
* Two bed volumes of 4%' strength, downflow at 0.5 gpm/cu ft.
It was 'postulated ,that high pH in the influent to the anion resin could be
a cause:for the irregylar results. Therefore, the operation of the lime
clarifier and reearbonation was'" discontinued for the next four runs.
Pretreatment merely consisted of filtration through the dual media filter
and activated carbon. Runs 21 and 22 were duplicate attempts made with
counter-current regeneration using the 13.5 kilograins regeneration
level. The average capacity obtained for the two exhaustions was 4.0
kilograins per cubic foot. The resin column was then opened for internal
inspection, which showed no physical damage to the internals or loss of
resin. A representative'sample of the resin in the exhausted state was
removed: for laboratory analysis by Culligan and the resin manufacturer.
Resin analyses from both laboratories, showed that the resin still possessed
good chemical, and physical properties. In addition, the resin manufacturer
ran additional laboratory evaluation tests on the resin and concluded that
the resin' had been apparently properly regenerated and carbonated by our
procedures. ? -
;Run #23 employed sodium hydroxide (NaOH) in an attempt to achieve better
capacity, The sodiuip hjydroxide is a stronger base than is the ammonium
hydroxide previously used for regeneration. It should regenerate the
resin more efficiently', and also better elute organics from the resin.
Samples-of the regenerant effluent were collected during the regeneration
with spdium hydroxide. Table 6;shows-the analyses of effluent samples.
These samples .did,-indeed, show high concentrations of total organic
carbon dufe to elyted organic material. The total organic carbon concen-
tration in the regenerant effluent samples peaked at a value greater than
4,000 mg/1 as 'carbon; This high organic level indicates that previous
regenerations with ammonium'hydroxide were insufficient to remove most
of the organics from the' resin. In spite of this organic elution and
34
-------
high regenerant dosage, a low capacity (3.2 kilograins per cubic foot)
was obtained.
A final attempt was made using a massive regeneration dosage (102
kilograins as calcium carbonate equivalent) of ammonium hydroxide.
The low capacity (4.2 kilograins per cubic foot) obtained confirmed
the agreement between the Project Officer of the Water Quality Office
and the Project Manager at Culligan that the process was not suitable
for demineralization of the secondary sewage effluent at Elgin, 111.
The recommendation is therefore made that this process should not be
applied to wastewater of salinities less than 600 mg/1. The following
observations were offered as a basis for discontinuing studies on this
process.
A. Operating procedures had not been fully established.
B. Recarbonation procedures, in particular, had not been
detailed.
C. An unacceptably low volume of product water is produced
to a typical operating endpoint.
D. Excessive anionic concentration in the product water occurs
when the resin is exhausted to a high operating capacity.
E. High regenerant dosages, particularly for ammonium hydroxide,
are necessary to obtain realistic capacities.
F. The resin manufacturer's published data indicates that the
capacity of the resin decreases very rapidly as the influent
salt content decreases below 500 mg/1, which is the approxi-
mate salt concentration of our influent water.
35
-------
SECTION 8
STRONG ACID ION EXCHANGE RESIN PERFORMANCE-HYDROGEN FORM
This process employs a strong acid cation exchange resin. In the
hydrogen form, this resin is capable of "splitting" ionized salts
in water to remove cations by replacing them with an equivalent
hydrogen ion. This resin has a cation - hydrogen selectivity such
that neutral mineral salts in water are converted to their correspond-
ing mineral acids. For example, the chlorides and sulfates of sodium,
calcium, and magnesium are converted to the corresponding acids, as
illustrated oy the equation employing sodium chloride:
R-H + NaCl = R-Na + BC1
The effluent water, which has been treated by the hydrogen form resin,
can then be treated with a bed of anion exchange resin (weak base) in
the free base form. The function of this second bed of resin is to
adsorb the acids produced by the cation exchanger. This is illustrated
by the following equation employing the hydrochloric acid produced
above:
RNH2 + HC1 = (R-NH2)HC1
A combination of the two ion exchange beds will remove the mineral salts
from the water. Incorporation of such a two bed ion exchange system
after an efficient process for removal of organic contaminants will
result in a reclaimed water without pollution due to inorganic and
organic materials. This section of the report, however, discusses only
the performance of the hydrogen form resin.
Our attempts to apply the hydrogen form of the strong acid cation ex-
change resin have been successful. We have used accepted techniques for
the ion exchange industry in applying this resin to sewage treated by
the activated sludge process and our pretreatment system. It is demon-
strated, therefore, that the process can be used in the design of
treatment plants.
The strong acid cation exchange resin was studied for treating the
effluent from the pretreatment system, (settling, filtration, activated
carbon adsorption). The clarifier and recarbonation components of
pretreatment were not used. These latter components will remove phos-
phate, but this can be done by the anion exchange process as well.
Turbidity removal will be accomplished easier by mechanical filtration
than with clarification. Cost estimates based on influent water compo-
sition and the resin manufacturer's procedures for performance estima-
tion show lime clarification to be costly. Table 7, "Comparison of
estimated performances of strong acid cation exchange resin", details
37
-------
Table 7. Comparison of estimated performances of strong
acid cation exchange resin (IRC-120). Resins
treating activated sludge treated sewage, with
and without lime clarification.
With Without
Lime Lime
Composition of resin influent:
Sodium, mg/1 CaC03 298 205
Total hardness, mg/1 CaCOs 132 248
Chloride and Sulfate, mg/1 CaC03 301 222
Alkalinity, mg/1 CaC03 129 331
Total Cations, mg/1 CaCOg 430 553
Percent ratio of:
Sodium 69 37
Calcium 15 28
Alkalinity 30 60
Performance prediction:
Iso - capacity, Kgr/cu ft 10.4 10.3
Correction for alkalinity 1.06 1.12
Adjusted gross capacity,
Kgr/cu ft 11.0 11.5
Deduct for rinse requirement,
Kgr/cu ft 1.2 1.3
Net predicted capacity,
Kgr/cu ft 9.8 10.2
Leakage, % of influent cations 12.0 4.0
Requirements for 1000 gallons:
Resin, cu ft 2.57 2.83
Acid (66° Be sulfuric) :
pounds (3/cu ft) 7.7 8.5
Cost (1.8 $/lb), $ 13.9 15.3
Lime:
Pounds 4.6 0
Cost (1.0
-------
the cost estimate for producing 1000 gallons of resin treated water
with and without lime clarification. As shown, lime clarification
increases the operating chemical costs from 15.3 to 18.5 cents per 1000
gallons of water from the cation resin. This cost does not include
costs due to recarbonation which would apply only to the clarifier
operation, although these costs should be a comparatively small factor.
Studies of this strong acid cation exchange resin used 2 cu ft of resin
in a 12" diameter tank, so that the height of resin was about 30" in
place. Regeneration of the resin was accomplished with sulfuric acid,
ranging from 2% to 5%. The use of sulfuric acid for regeneration can
cause resin fouling by precipitation of calcium sulfate during regener-
ation. This precipitation in the resin is avoided by using an acid
strength no greater than 2% when large amounts of calcium are being
eluted. Increased capacity and reduced cation leakage to service can
be obtained by increasing the strength and dosage of acid after eluting
most of the calcium with an initial dosage of 2 pounds of acid per cubic
foot of resin. This is sufficient to reduce the amount of calcium re-
maining on the resin so that the calcium sulfate solubility will not be
exceeded when the stronger sulfuric acid solution is used. This proce-
dure is refered to as "step-wise regeneration". Therefore, during
regeneration, the first 2 pounds of sulfuric acid regenerant were
prepared as a 2% solution while the remaining regenerant dosage beyond
2 pounds per cubic foot was fed at higher concentrations. We used
regenerant dosages of 1 - 5 pounds of 66° Baume sulfuric acid per
cubic foot of resin. Soft potable water was used for regenerant solu-
tion preparation and for the first portion of rinse. Final rinse was
made with sewage water from the pretreatment system.
Performance of ion exchange resins during exhaustion was monitored by
indicating and recording instruments. Figure 12, "Typical Effluent
Conductivity From Cation (IRC-120) Resin", is a representative exhaustion
effluent conductivity pattern, which shows three distinct zones as
follows:
High, but rapidly decreasing conductivity during regenerant
acid rinse out.
Moderate and stable conductivity during exhaustion. The
conductivity is caused by the free mineral acidity formed
by the hydrogen ion exchange for the cations.
Rapidly decreasing conductivity, signaling increasing break-
through of cations as the ion exchanger becomes exhausted.
The exhaustion cycles for the cation exchange resin were based on actual
curves of this nature which were obtained for each run. The exhaustion
started and ended when the effluent conductivity was 500 micromhos higher
39
-------
Figure 12. Typical Effluent Conductivity
From Cation (IRC-120) Resin
End of Exhaustion
10
Middle of Exhaustion
456
MICROMHOS
Acid Rinse-Out
15
20
15
20
40
-------
and lower respectively than the average conductivity during service.
The total cationic loading to the cation exchanger was immediately
determined by field analysis of a representative sample of the average
influent water to the resin. This sample was collected at a steady,
slow rate (drip sample). Two determinations were made on this sample.
The first determination was for total alkalinity as collected (but
after laboratory filtration). This determination equaled the total
cations associated with alkalinity. The remaining cations were deter-
mined by passing the laboratory filtered sample through an analytical
column packed with hydrogen form strong acid cation resin which was
totally regenerated. The effluent from this laboratory column was
analyzed for free mineral acidity (FMA), equal to the total cations
except those associated with alkalinity. The latter cations are
neutralized by the alkalinity present, and are therefore not determined
in the FMA titration. The sum of these two results then equaled the
total cation load in service to the cation exchange resin bed. The
analytical laboratory at Culligan International Company was used to
corroborate the field analyses as well as to obtain analyses in greater
detail.
Table 8, "Strong Acid Cation Exchange Resin (IRC-120) Performance Re-
cap", summarizes the performance of this resin. The top half of the
table pertains to counter-current flows, while the bottom half pertains
to co-current operation. A comparison of these two areas shows that
the co-current flow gave better performance than did the counter-
current. More capacity, less cation leakage in service, and better
acid utilization resulted. Figures 13 and 14 attached are graphic
representations of the regeneration efficiency.
Table 8 also shows that there is slightly better co-current performance
at 3 gpm/cu ft than at 6 gpm/cu ft. Equipment designs can use the
lower flow rate with the slightly better performance while considering
the higher flow rate as a reserve for future expansion.
The requirements (of resin, regenerant acid, and rinse) to produce
1000 gallons of treated water are also shown on Table 8. Low regener-
ation levels decrease the operating costs (less pounds of regenerant)
but increase the capital cost (more cubic feet of resin). Amortization
favors use of low regenerant levels.
Tables 9, 10, and 11 attached, recap the individual exhaustion runs
used in preparing the summary of Table 8. The data has been calculated
to a "per cubic foot" base although more resin was actually used. The
second column of these tables shows the weight of 66° Baume sulfuric
acid used for regeneration. The third column shows the exhaustion
flow rate in gallons per minute, which ranged from 3-6 gpm/cu ft.
The mineral loading during exhaustion is also shown in tables 9, 10,
and 11.
41
-------
Table 8'. Strong acid' cation exchange resin (IRC-120)
performance summary. Per cubic foot base.
Co-current flows: downflow regeneration, downflow exhaustion.
Test Series
D
Regenerant, Ibs/cu ft
Exhaust, gpm/cu ft
Capacity, kgr/cu ft
Acid Utilization %
Leakage, Average:
mg/1
% of influent
Requirements per 1000
Gallons of water
produced:
Cation resin, cu ft
Acid, Ibs 66° Be
Rinse water,
at 35 gal/cu ft, gals.
5
3
14.7
44.0
16.2
2.7
2.38
11.9
83:3
5
6
14.6
43.7
20.7
3.6
2.31
11.6
80.9
3
6
10.4
52.0
32.0
5.6
3.22
9.7
113
3
3
10.6
53.0
61.6
10.4
3.13
9.4
110
1
3
6.0
90.0
94.3
16.4
5.56
5.6
195
1
6
5.0
74.9
96.5
17.6
6.25
6.3
219
.Counter-current flows: upflow regeneration, downflow exhaustion.
Test Series
Regenerant Ibs/cu ft
Exhaust, gpm/cu ft
Capacity, kgr/cu ft
Acid Utilization %
Leakage, Average:
rag/1
% of influent
Requirements per 1000
Gallons of water
produced:
Cation resin, cu ft
Acid, Ibs 66° Be
Rinse Water,
at 35 gal/cu ft, gals.
103.3
18.3
7.04
7.04
H
63.7
12.8
3.64
10.92
246 ' 127 '
64.7
11.5
4.00
12.00
140'
J
1
3
4.7
70.5
3
3
8.0
40.0
3
6
8.1
40.5
5
6
10.1
30.3
68.1
11.2
3.51
17.55
123
5
3
10.2
30.6
64.8
11.0
3.38
16.90
118
42
-------
Figure 13. CATION RESIN (IR-120) REGENERATION EFFICIENCY
Exhaustion Flow Rate, 6 gpm/cu ft
105
CO
OJ
o>
DH
05
c
•H
03
f-t
C
•H
•H
O
(0
a
03
O
0
123
Regenerant Level, Pounds H2S04 Per Cubic Foot
43
-------
Figure 14. CATION RESIN (IR-120) REGENERATION EFFICIENCY
Exhaustion flow rate, 3 gpm/cu ft
o>
a
01 3
c
•H
CO
h
CT>
O
• H
O
CO
a
as
O
01234
Regenerant level, pounds ^804 per cubic foot
105
90
75
60
45
0)
O
h
-------
Table 9. CATION RESIN (IR-120) PERFORMANCE
RECAP SHEET
(per cubic foot)
Counter-current flows
Run
#
1
2
3
4
5
6
7
Series
8
9
10
11
12
13
14
Hoso,,
Ibs?
30
5
it
ti
11
ii
it
A
5
it
it
ti
ii
H
ii
Series B
15
16
17
18
19
20
21
Series
3
ii
ii
it
ti
ii
n
C
Exhaust
GPM
3
3
ti
IT
tt
It
tt
Average
6
H
ti
H
H
H
H
Average
6
ii
ii
H
ii
H
H
Average
Loading
mg/1**
588
592
552
566
592*
600*
600*
597
568
576
596
576
564*
584*
580*
576
592
592*
592*
592*
572*
556*
550*
576
Capacity
qals.
1070
523
436
350
430*
425*
404*
420
449
413
263
365
435*
425*
435*
432
319
307*
296*
300*
312*
329*
325*
311
Kqr**
36.8
18.1
14.1
11.6
14.9*
14.9*
14.2*
14.7 (*based)
14.9
13.9
9.2
12.4
14.4*
14.5*
14.8*
14.6 (*based)
11.0
10.6*
10.2*
10.4*
10.4*
10.7*
10.5*
10.4 (*based)
** As calcium carbonate equivalent.
45
-------
Table 10. CATION RESIN (IR-120) PERFORMANCE
RECAP SHEET
(Per Cubic Foot)
Counter-current flows
Run ,
#
22
23
24
25
26
27
28
29
30
Series
31
32
33
34
35
Series
36
37
38
39
Series
H2S04 Exhaust
Ibs GPM
3 3
n n
M n
n n
M n
n n
n ti
n n
11 n
D Average
1 3
ii n
11 n
11 n
it n
E Average
1 6
n n
ii ii
" "
F Average
Loading
572
584
588
608
584
568*
568*
592*
564*
573
524*
572*
600*
600*
580*
575
528*
532*
548*
580*
547
Capacity
gal. Kgr**
348
326
308
307
224
322*
319*
315*
319*
319
203*
179*
171*
168*
178*
180
168*
167*
153*
150*
160
11.6
11.1
10.5
10.9
7.6
' 10.7*
10.6*
10.8*
10.5*
10.6 (*Based)
6,1*
6.0*
6.0*
5.9*
6.0*
6.0 (Based)
5.0*
5.2*
4.9*
5.1*
5.0 (Based)
** As calcium carbonate equivalent.
46
-------
Table 11.
CATION RESIN (IR 120) PERFORMANCE
RECAP SHEET
(Per Cubic Foot)
Co-current Flows
Run
#
H2S04
Ibs
Exhaust'
qpm
Loading
mq/1**
Capacity
qal. Kqr.**
40-44
45
46
47
48
49
Series
50
51
52
53
Series
54
55
56
57
58
59
Series
60
61
62
63
Series
64
65
66
67
Series
l(upflow)
G Average
3(upflow)
H Average
3(upflow)
I Average
5(upflow)
J Average
5(upflow)
n
it
K Average
6
n
Conditioning runs on resin transferred
to smaller tank for upflow studies.
562* 133* 4.4*
556* 160* 5.2*
Invalid results due to mechanical
difficulties
576* 133* 4.5*
565 142 4.7 (*Based)
488
484*
490*
516*
497
546
564*
464*
596
592*
598*
561
608*
608*
612*
608*
609
616
584*
588*
592*
588
198 5.7
291* 8.2*
264* 7.6*
271* 8.1*
275 8.0 (*Based)
261 8.3
240* 7.9*
301* 8.2*
246* , ,8.6*
233* 8.1*
229* 7.9*
250 8.1 (*Based)
293* 10.4*
260* 9.2*,
285* 10.2*'
300* 10.6*
284 10.1 (*Based)
301 10.9
307* 10.5*
291* 10.0*
291* lO^l**
297 10.2 (Based)
** As calcium carbonate equivalent.
47.
-------
The last two columes of the three tables show the capacity obtained
during exhaustion. The gallons shown are those number of gallons delivered
between the two selected start and endpoint conductivity values for the
effluent as previously discussed.
The capacity has been calculated from the gallons produced and the load-
ing factor to obtain the values shown as kilograins (kgr). The loading
factor is shown as mg/1, as calcium carbonate equivalent: this is
changed to grains per gallon (gpg) by dividing by 17.1. The capacity,
in kgr/cu ft is then calculated from the following formula:
kgr/cu ft = VL
17.1 x 1000
Where: kgr/cu ft is the ion exchange capacity per cubic foot of
resin.
V is the number of gallons treated by one cubic foot of resin.
L is the loading factor, in mg/1 as calcium carbonate equivalent.
The performance obtained with this resin is encouraging. The results
do not differ greatly from those which would be predicted on a potable
water. The published procedures of Rohm & Haas were used to predict
the capacity and leakage from this cation exchange resin on potable
water. The capacity and leakage predicted by calculation, as compared
with the values obtained for the first four test series, are shown on
the attached Table 12, "Cation Resin (IRC-120) Performance, Comparison
of Estimates and Actual". This comparison shows that the predicted
capacities are comparable to those which are actually obtained. Cation
leakage from the exchanger, however, has been somewhat higher than
predicted values.
The quality of water produced from the cation exchange resin during
service is shown in the attached Table 13. Each test series consisted
of several regeneration and exhaustion cycles made until stable perfor-
mance was obtained. Table 13 contains typical results of the analyses
made of composite samples of influent and effluent during the stable
performance runs. As shown, the influent contains a typical turbidity
of 2.0 - 3.4 JTU; the mechanical filtration of the pretreatment system
was performing adequately. The table also shows that the influent total
organic carbon ranged from 5.4 - 10.2 mg/1. (The previous section of
this report which discussed the pretreatment equipment stated that the
activated carbon filter of the pretreatment system removed about one-
half of this organic carbon from the sewage). This Table 13 shows that
the total organic carbon was unaffected by the cation exchange resin.
48
-------
Table 12. Cation Resin (lR-120) Performance
Comparison - Estimates and Actual
Test Series ABC D
Regenerant, Ibs/cu ft 5533
Exhaust, gpm/cu ft 3663
Loading, mg/1 as CaCOo:
Ca 166 144 156 145
Mg 188 198 202 187
Monovalents 246 234 214 260
Alkalinity 342 312 344 374
Loading %:
Ca 28 25 27 25
Mg 31 35 35 31
Monovalent 41 40 38 44
Alkalinity 57 54 60 63
Capacity, kgr/cu ft:
Predicted:
Unadjusted 14.4 14.6 10.4 10.4
Adjusted 1.12 1.12 1.13 1.13
Adjusted Gross 16.1 16.4 11.8 11.8
Rinse Deduct 1.3 1.2 1.2 1.2
Net Prediction 14.8 15.2 10.6 10.6
Obtained 14.7 14.6 10.4 10.6
Leakage %:
Predicted 224 4
Obtained 2.7 3.6 5.6 10.4
49
-------
Table 13. Water analyses summary of typical exhaustion
of cation exchange resin (IRC-120)
Test Series
Regenerant, Ib/cu ft
Exhaust, gpm/cu ft
Influent:
pH, units
Turbidity, JTU
Total organic carbon
mg/1 C
Magnesium, mg/1 CaCOg
Calcium, mg/1 CaCO
O
Sodium, mg/1 CaC03
Potassium, mg/1 CaCOg
Ammonia, mg/1 N
Effluent:
Total organic carbon
mg/1 C
Acidity, mg/1 CaCOg
Magnesium, mg/1 CaCOg
Calcium, mg/1 CaCOg
Sodium, mg/1 CaCOg
Potassium, mg/1 CaCOg
Ammonia, mg/1 N
A
5
3
7.7
3.4
5.4
192
164
207
10.0
2.0
5.3
217
Trace
Trace
12.8
1.8
0.0
B
5
6
8.0
2.4
5.5
193
159
178
11.4
2.6
5.4
206
1.7
5.3
12.8
1.6
0.0
C
3
6
8.0
2.0
7.2
195
169
163
11.0
2.0
6.6
191
0.9
8.2
19.7
1.9
0.0
D
3
3
8.1
3.0
8.6
193
155
164
11.2
6.7
9.5
172
9.4
8.4
25.8
2.3
1.3
E
1
3
7.8
2.6
10.2
177
146
223
11.7
3.3
9.0
133
4.3
6.8
71.1
4.4
2.7
F
1
6
8.0
2.8
9.9
174
161
175
10.5
5.8
4.6
144
6.0
10.0
47.9
4.6
2.9
50
-------
The effluent analyses also show the production of acidity (which would
result in a low pH) due to the hydrogen ion exchange for cations. This
acidity can be easily removed by subsequent ion exchange treatment
with a second ion exchange resin as discussed in later sections of
this report. The analyses show great reduction in the cationic compo-
sition of the influent water. Ammonia is significantly reduced; however,
total avoidance of this cation in the effluent water will require sub-
sequent break-point chlorination techniques.
Regeneration of this cation exchange resin with sulfuric acid has been
accomplished without difficulty. Precipitation of calcium sulfate in
the resin bed has been avoided; however, such precipitation occurs in
the regeneration effluent about 10 minutes after leaving the column.
The composition of regeneration effluents, as shown in Table 14,
duplicate those from conventional procedures. Specific analyses for
total organic carbon elution show only analytical and sampling varia-
tions throughout the regeneration. Excess acid, which requires dis-
posal, is present in the regeneration effluent. Neutralization of this
acid is discussed in a later separate section of this report. An
alternate means of acid disposal is by its reuse for the regeneration
of a weak acid cation exchange resin, which is a separate ion exchange
system, also discussed in a later section.
The recommendation is made that strong acid cation exchange resin can
be applied to treat the effluent from secondary sewage treatment plants.
Use of the strong acid cation exchange resin, of course, is only a
partial means of wastewater demineralization. The effluent is un-
acceptable and must be post treated. Subsequent sections in this
report will consider the application of ion exchange treatment for
this post treatment to produce a demineralized wastewater with
acceptable mineral concentration. The operating material requirements
for resins, regenerants, rinse water, etc. is developed for each total
ion exchange system in portions of the section titled, "Operating
Material Requirements for Ion Exchange Processes".
51
-------
Table 14. Regeneration effluent analyses from strong acid cation resin IRC-120,
WATER ANALYSIS
FWPCA Contract No. 14-12-599 DATE July 1970
TEST (*AS CoC03)
pH
ACIDITY, MINERAL '
ALKALINITY, OH'
ALKALINITY, C03-
ALKALINITY, HCO3"
CHLORIDES-
TURBIDITY
IN J.T.U.
SULFATES-
TOTAL PHOSPHATE
AS P04
TOTAL HARDNESS-
MAGNESIUM'
CALCIUM*
AMMONIA
NITROGEN AS N
NITRATE
NITROGEN AS N
NITRITE
NITROGEN AS N
SODIUM-
POTASSIUM'
TOTAL
ORGANIC CARBON
TOTAL
DISSOLVED SOLIDS
Bed Volume
1000
40
7200
5400
1800
10
5100
250
10.5
1
i
5300
38
7700
4400
3300
140
3000
100
11.0
1-2/3
6600
40
6300
3400
2900
220
2600
100
10,, .L
2
7600
40
2900
1700
1100
230
]800
120
9.8
2-1/3
8100
38
2100
1300
800
110
1600
100
10.6
2-2/3
3100
40
340 ,
200
140
40
260
15
9.3
3
Regeneration Effluent IRC- 120
-------
SECTION 9
WEAK BASE ANION EXCHANGE RESIN PERFORMANCE-FREE BASE FORM
This process employs a weak base anion exchange resin. In the free
base form, this resin is capable of adsorbing strong acids. In
multi-column demineralization processes, such acids are produced by
the strong acid cation exchange resin.
Whereas the strong acid exchange resin removes cations to form both
strong acids and weak acids, the weak base anion exchange resin can
only adsorb strong acids. Weak acids are not removed by the weak
base anion exchange resin. One weak acid not removed is carbonic
acid, produced by the cation exchange of an alkaline salt, as illus-
trated by the following equation employing sodium bicarbonate:
RH + NaHC03 = RNa + H2C03
H2C03 = H20 + C02
The carbon dioxide produced is unaffected by the weak base anion
exchange resin: the gas is readily eliminated through established
technology with aeration or degasification.
The performance characteristics of the weak base anion exchange resii
have been encouraging. The technology for their application is
sufficiently developed for application to wastewater demineralizatio
Standard techniques of the ion exchange industry for application of
this resin have been employed. This resin successfully treated the
effluent from a strong acid cation exchange resin (hydrogen form).
It is demonstrated, therefore, that a combination of the two resins
can be used in the design of equipment for demineralization of
secondary sewage effluents.
The product water produced by the above system is easily degased to
remove carbon dioxide. The resulting product will have a low conter
of dissolved solids. Total phosphate will be reduced to a typical
average concentration of about one mg/1; further reduction is possi):
by treating less water per cycle with a corresponding increase in
chemical operating costs per unit volume of water produced.
Ammonia leakage to service does occur from the cation exchange resii
and this ammonia is unaffected by anion exchange. The ammonia can 1
eliminated from the product water by break-point chlorination. Nit:
nitrogen was a constant 0.1 mg/1 as nitrogen in the effluent. Nitr
nitrogen was present to a much lesser degree.
53
-------
Algae regrowth did not occur during an approximate one month's storage
in vented clear glass bottles which were half filled. Product water
was collected directly from the anion exchanger three times during the
exhaustion cycle; these were at the start, middle, and at the end of
exhaustion. These samples were exposed to direct sunlight on the south-
end of the pilot plant during early fall. All samples remain clear, with
no noticeable growth appearing.
Degasification of the product water was also studied. Product water from
the weak base anion exchange resin contains dissolved carbon dioxide
equivalent to the bicarbonate alkalinity of the influent water. This
carbon dioxide should be removed from the product water if this water is
to be transported in pipes. Mechanical degasification was readily
obtained without difficulty. Foaming in the degasifier was not observed;
nor were growths or precipitate formation in the system observed.
The performance characteristics of two different weak base anion resins
were studied. A macroporous resin (IRA-93) has been used. This resin
has large pores, thereby minimizing the tendency of organic fouling.
Sewage effluent normally contains ionic organic chemicals of compara-
tively large size. The organic molecules can be partially removed by
anion exchange resin; however, they usually are difficult to elute during
regeneration. This difficulty is minimized with macroporous resin.
A second weak base resin (IRA-68) has been similarily evaluated. This
resin also possesses a high degress of porosity. It has been used
efficiently in industry as a decolorizing agent in a variety of situations.
Because of its resistance to organic fouling, it was established that both
types of weak base resin should be evaluated.
Both weak base anion exchange resins were individually studied in 10"
diameter tanks containing 1-1/2 cu ft of resin. This amount of resin
resulted in a bed height of about 31".
Regeneration was with a 4% solution of ammonium hydroxide made in soft
potable water. This water was used to avoid precipitation of polyvalent
cations in the regenerant solution. The solution was prepared in the
open regenerant solution vessel, and pumped through the anion resin bed.
Rinse was in two stages: first with approximately 2-1/3 bed volumes of
softened water at the regeneration flow rate; second, with water from the
cation exchange resin (as in service) at the service flow rate.
Observation of the recorded effluent conductivity indicated when the
fast rinse was finished and the service exhaustion was initiated. Prior
to this point, the conductivity decreased rapidly; during service it was
stablized; at exhaustion, the conductivity increased rapidly. Capacity
54
-------
was calculated based on a conductivity endpoint fifty micromhos higher
than the stable minimum conductivity obtained during service.
Capacity calculations for the resin were based upon the above determined
volume of water and the average loading factor of the water from the
cation exchange resin during the exhaustion. The loading factor was
determined from the analysis of the composite sample of the influent.
The sample was titrated for its content of free mineral acidity (FMA).
FMA is what must be removed by the weak base resin.
The performance of each of the two resins is discussed separately below.
Weak base resin IRA-93 performance.
The performance of the IRA-93 weak base anion exchange resin is summa-
rized in the attached Table 15. As shown, capacities of about 17
kilograins per cubic foot were obtained under the conditions investi-
gated. The capacity obtained is compatible with the resin manufacturer's
(Rohm & Haas) literature, which shows 21.0 kilograins. The slight
capacity difference is probably due to the effect of unknown organic
loading. The only significant difference in capacity under varied
operating conditions was obtained by reducing the exhaustion flow rate
from 6 gpm to 2 gpm per cubic foot. This flow reduction increased
the capacity obtained by about 10%. This capacity increase is in-
cufficient to justify the increased plant size which would be necessary
to accommodate the slower flow rate.
Table 15 shows the average leakage of phosphate to be about 1 mg/1. A
lower level could be obtained by reducing the operating capacity. Table
16 shows the phosphate break-through from this weak base anion resin
during a typical exhaustion cycle. As shown, after treating 1,380
gallons of water, the effluent phosphate concentration was 17.6 mg/1
phosphate as PC>4 . The average concentration in the effluent was
1.2 mg/1 P04. Obviously, avoiding the higher phosphate concentra-
tion present at the end of the exhaustion cycle would reduce the average
phosphate concentration in the treated water.
Individual results for the test on the IRA-93 resin are summarized in
Table 17, "Anion Resin (IRA-93) Performance Recap Sheet". As shown,
stable capacities were quickly obtained. The regenerant dosage used
ranged from 75% to 117% stoichiometric to the maximum capacity of the
resin.
The ionic concentration of the effluent from the IRA-93 resin (which
has previously been treated with cation exchange resin in hydrogen
form) is shown in Table 18 for selected samples during a typical ex-
haustion cycle. The table shows low concentrations for both cations
and anions. Phosphates, however, increase rapidly as the endpoint is
approached. As previously discussed, these higher concentrations can
be easily avoided.
55
-------
TABLE 15. Weak base anion exchange resin (IRA 93) performance
summary. Treating effluent from cation exchange
(per cubic foot)
Test Series
Regeneration:
A.
B.
C.
D.
E.
NH3 (28%) Ibs
NH4OH (100%) Ibs.
Direction
Flow, gpm
Exhaustion:
Direction
Flow, gpm
Capacity, Kgr
Ammonia utilization
percent
Leakage, average
mg/1 P04
Requirements per 1000
Gallons produced:
Anion resin,
cu ft
Ammonia, 28%,
Ibs
Rinse water,
gal.
7.3
4.2
Down
0.3
Down
6
17.1
5
42
2.5
0.66
4,75
75
7.3
4.2
Down
0.3
Down
2
18.8
46
0.5
0.58
4.19
75
6.6
3.8
Down
0.3
Down
6
16.9
46
1.3
0.71
4.70
69
6.6
3.8
Down
0.9
D own
6
17.4
47
1.4
0.67
4.48
70
4.8
2.8
Down
0.9
Down
6
17.0
62
1.2
0.75
3.58
65
56
-------
TABLE 16. Phosphate break-thru from weak base anion resin
(IRA-93) during service exhaustion on cation
resin effluent, which contains 19.2 mg/1 phosphate.
Per cubic foot basis.
Gallons to
service
0
667
1270
1310
1330
1350
1360
1370
1380*
1385
Composite Average
Phosphate,
mg/1, PO}"
0.5
0.2
1.5
3.2
6.4
12.0
10.4
14.8
17.6
17.6
1.2
Effluei
Conduc-
microml
40
50
60
70
80
90
100
Endpoint of service exhaustion.
57
-------
TABLE 17. AN ION RESIN (IRA-93) PERFORMANCE
RECAP SHEET
Treating Cation Resin Effluent
(per cubic foot) '
Run NH4OH
# ( 100$)
Lbs **
1 New
2 2.8
3
4
5 4.2
6
7
8
9
10
11
12
13
Series A average
14 4.2
15
16
Series B average
17 3.8
18
19
20
Series C average
21 3.3
22
23 "
24
25
26
27 "
Series D average
28 2.7
29
30
Series E average
Exhaust
GPM
6
!f
tl
II
6
tt
tt
M
tf
II
II
If
tl
(* based)
2
II
IT
(* based)
6
(1
tl
If
(* based)
6
Loading
mg/1
.._
200
220
220
176
208
Capacity
gals. kgr
...
1270
1277
1193
1314
1247
...
14.8
16.4
15.3
13.5
15.2
incompleted run
204
212
200
134*
200*
192*
192
172*
200*
192*
188
112
196*
212*
212*
207
210
1361
1167
1227
1584*
1454*
1524*
1521
1761*
1700*
1684*
1715
2094
1487*
1354*
1369*
1403
1383
16.2
14.5
14.4
17.0*
17.1*
17.1*
17.1
17.6*
19.8*
18.9*
18.8
13.7
17.0*
16.8*
17.0*
16.9
17.0
incomplete run
it
"
I!
II
"
(* based)
6
II
11
(* based)
212
216
220
*204
*200
202
*216
*220
*220
219
1253
1384
1035
*1421
*1525
1473
*1360
*1367
*1301
1343
15.5
17.5
13.3
*16.9
*17.8
17.4
*17.2
*17.2
*16.7
17.0
* 3.6 Ibs is stoichiometric to maximum capacity of resin.
58
-------
Table la. tonic break-thru from weak bjso onion rosin (IRC-IM) during service exhaustion.
WATER ANALYSIS
FWPCA Contract No. 14-12 599
TEST CASCoCOj)
»H
ALKALINITY. OH1
ALKALINITY, C03*
CHLORIDES-
TURBIDITY
IN J.T.U.
SULFATES-
AS POj
MAGNESIUM*
CALCIUM-
AMMONIA
NITRATE
NITRITE
NITROGEN AS N
SODIUM-
POTASSIUM-
TOTAL
TOTAL
Number
_Raw
8.0
8.0
16.0
90.4
956
Carbon
Filtered
8.1
1.8
17.6
8.4
955
Cation
Effluent
220.
140.
56.
19.2
1.4
7.2
953
Anlon E
12.
6.
0.
0.5
0.
6.5
950
fluent
~
10.
0.
0.
0.2
0.
7.5
951
.
2.
4.
0.
27.2
0.
9.1
952
12.
0.
0.
1.2
0.
7.7
954
Gallons For
cu ft
Drip
Drip
Drip
1000 2078 Drip
IRA 93 Run No. 24
WATER ANALYSIS
FWPCA Contract No. 14-12-599 DATE October, 1970
TEST CAS C-COj)
•
ALKALINITY, OH'
ALKALINITY, CO-j'
CHLORIDES-
TURBIDITY
IN J.T.U,
SULFATES'
TOTAL PHOSPHATE
AS P0t
MAGNESIUM-
CALCIUM'
AMMONIA
NITROGEN AS N
NITRATE
NITROGEN AS H
NITRITE
NITROGEN AS N
SODIUM-
POTASSIUM-
TOTAL
TOTAL
DISSOLVED SOLIDS
Number
30
f,.
n.
0.
1.5
0.
0.
0.
6.8
0.2
9.B
957
40
fi.
n.
0.
3.2
0.
0.
0.
7.3
0.2
8.0
958
50
H.
n.
0.
6.4
0.
0.3
0.
7.3
0.3
10.8
959
60
6.
0.
0.
12.0
0.
0.3
0.
7.5
0.2
15.1
960
70
4.
0.
0.
10.4
0.
0.3
0.
7.3
0.2
13.3
961
80
6.
2.
0.
14.8
0.
0.3
0.
7.9
0.1
13.6
962
90
H.
2.
0. ,
17.6
0.
0.3
0.
8.2
0.1
16.1
963
100
2.
6.
0.
17.6
0.
0.
0.
7.8
0.1
14.1
964
—
12
0
0
1.2
—
0
—
7.7
954
Gallons For 1908 1965 1993 2020 2038 2053 2065 2078 Drip
Break-thru samples, IRA-93, Run No. 24
59
-------
The influent to this anion exchange system has been treated with
activated carbon; consequently, the influent total organic carbon
was typically 6-8 mg/1 as carbon, and the composite (drip) shows
the treated effluent to contain an average of 7.7 mg/1. However, the
analyses of samples collected as exhaustion approach showed breakthrough
of organic materials from this resin. Therefore, organics are removed
by IRA-93 to a small extent.
The regeneration of the IRA-93 has been characterized when 4.2 Ibs of
ammonium hydroxide were used; this was equal to 115$ of the stoichio-
metric requirement for the total capacity of the resin. The regenerant
effluent was sampled in 1/4 or 1/2 bed volume increments. The analyses
for the samples collected within the range of 1 to 3-3/4 bed volumes
are shown in Table 19, "Regenerant Effluent Analyses From Weak Base
Anion Resin (IRA-93)". As shown, regenerant ammonium hydroxide was
initially neutralized by the adsorbed acids (chloride, sulfate, and
phosphate) so that low alkalinity resulted. However, by the third
bed volume, most of the ammonium hydroxide was not being used. The
concentration of ammonium nitrogen shows that the regenerant concen-
tration reached a maximum at about 2 bed volumes and started to decrease
at around 3 bed volumes. Elution of chlorides is slower than for the
other anions. However, this will not be of significance during the
service cycle.
Total organic carbon elution from this resin is difficult, as seen by
Table 19. The consequence of this is that organic fouling of the resin
may result. This in turn, would result in reduced capacity and perfor-
mance during service. Remedial maintenance procedures may be required
to remove the organics. However, stable capacities were observed during
the studies.
Pilot plant studies have shown that weak base resin IRA-93 is easily
applied in demineralization of secondary sewage. Stable capacities were
readily achieved; a reasonable quantity of water was treated; regenera-
tion with ammonium hydroxide was without difficulty. Although the
possibility of organic fouling of this resin cannot be ignored, it was
not encountered. Excess ammonium hydroxide is required to regenerate
the resin. This excess can be neutralized or recovered for reuse.
Additional treatment beyond demineralization is recommended: first,
degasification for removal of carbon dioxide; second, break-point
chlorination for disinfection and removal of ammonium ion leakage.
Weak base resin IRA-68 performance.
The^perfprmance of the IRA-68 weak base anion exchange resin is sum-
marized in Table 20. As shown, operating capacities of about 24 kilo-
grains per cubic foot were obtained, representing 70% of the total
capacity of the resin.
60
-------
TABLE 19. Regenerant effluent analyses from weak base anion resin (IRA-93)
Regenerated with 4.2 Ibs of 100% ammonium hydroxide.
WATER ANALYSIS
FWPCA Contract No. 14-12-599 D.TE September, 1970
TEST CAS CoCOj)
,H
ALKALINITY, OH'
CHLORIDES'
TURBIDITY
IN J-T.U.
SULFATES'
TOTAL RD E
MAGNESIUM-
CALCIUM-
AMMONIA
NITROGEN AS N
NITRATE
NITROGEN AS N
NITRITE
NITROGEN AS N
SODIUM'
POTASSIUM-
TOTAL
ORGANIC CARBON
TOTAL
DISSOLVED SOLIDS
60
40
0
3.8
30.0
0,3
0. 1
10.1
860
5HO
Rfin
-56.
1930
^
515
n.fi
22.6
769-
99SO.
41400.
5430
35 0
9 5
33.8
1
R61
862
3900.
9170.
IfiROO.
77000.
191HO
107.5
15-0
35.0
863
14700
96500.
90000.
690.
14000
107.5
6-4
37.4
864
'7500.
22000.
1200.
90.
8.8
14000
10.0
0.9
38.2
865
24800.
10400.
800.
33.
3.0
12180
9.0
0.3
29.2
866
13900.
10200.
22.
2.0
7660
8.7
0.3
27.4
867
13300.
3600.
600.
11.
1.3
5190
8.2
0.1
21.8
868
9600.
4000.
580.
11.
3.0
407
7.6
0.1
22.8
869
Bed Volume 1 It It 1-3/4 2-1/4 i-3/4 33* 3£ 3.3/4
IRA 93 Regeneration Effluent
-------
TABLE 20. Weak base anion exchange resin (IRA-68) performance
summary, (per cubic foot basis). Treating cation
resin effluent.
Test Series A.
Regeneration:
NH3 (28%) Ibs
NH4OH (100%) Ibs 3.2
Direction
Flow, gpm
Exhaustion:
Direction
Flow, gpm
Capacity, Kgr
Ammonia utili-
zation percent
68
Leakage, average
mg/1 PO^"" 0.2
Requirements per
1000 Gallons
produced:
B.
77
1.5
C.
D.
76
0.4
76
0.8
E.
86
1.2
F.
5.5
3.2
Down
2/3
Down
6
21.7
5.5
3.2
Down
1/2
Down
4
24.8
5.5
3.2
Down
1/2
Down
2-1/3
24.4
5.5
3.2
Down
1/2
Down
6
24.5
4.6
2.8
Down
1/2
Down
6
23.6
5.5
3.2
Down
1/3
Down
6
25.0
78
0.6
Anion resin
Ammonia, 28%
Ibs
Rinse water,
gal.
0.54
2.96
25
0.53
2.89
27
0.53
2.89
29
0.55
3.02
27
0.66
3.03
34
0.55
3.03
27
62
-------
4% ammonium hydroxide was employed for downflow regeneration of the
resin. In five of the six test series, the regenerant dosage has been
3.2 pounds of ammonium hydroxide per cubic foot - as specified by the
resin manufacturer. In test series E, however, dosage was reduced to
2.8 pounds per cubic foot in an attempt to achieve better regenerant
utilization. This lower regeneration level increased the ammonia
utilization from about 75 to 86% of the applied dosage. The reduced
dosage did not significantly affect capacity.
Further inspection of the summary Table 20 (test series A, B, and F)
shows that regeneration flow rates of 1/2 and 1/3 gpm per cubic foot
produced equivalent capacities (24.5 and 25.0 kilograins). Increasing
the regeneration flow rate to 2/3 gpm per cubic foot, however, reduced
the capacity to 21.7 kilograins. Varying the exhaustion flow rate
between 2 and 6 gpm per cubic foot (test series B, C and D) did not
affect the performance significantly.
The table also shows the average leakage of phosphate (as P04) is
no greater than 1.5 mg/1. This phosphate concentration is similar
to that obtained with IRA-93 weak base anion resin previously dis-
cussed. Also, a lower phosphate level can be obtained by reducing
the operating capacity.
Table 21 shows the phosphate break-through from the IRA-68 resin
during a typical service exhaustion. After treating 2,000 gallons of
water, the effluent phosphate concentration was 30.0 mg/1 phosphate
as P04. The phosphate concentration in the composite effluent was
0.6 mg/1. Obviously, as in the case with the IRA-93 resin, avoiding
the higher phosphate concentration present at the end of the exhaus-
tion cycle would reduce the average phosphate concentration in the
treated water.
Individual results for the tests on the IRA-68 resin have been
summarized in Table 22, "Anion Resin (IRA-68) Performance Recap
Sheet". This sheet shows that stable capacities were quickly obtained.
The ionic concentration of the effluent from the IRA-68 resin is
shown in Table 23 for selected samples during a typical service ex-
haustion. This table is similar to Table 18 presented for IRA-93
resin. A comparison of these two tables shows that the IRA-93 resin
has less chloride leakage than does the IRA-68. The phosphate con-
centration is also lower during break-through with the IRA-93. An
important feature in the comparison is the concentration of total
organic carbon: during exhaustion, IRA-93 shows (Table 18) some
break-through of organic carbon, while IRA-68 does not. Organic
compounds are apparently not accumulating on the resin. The perfor-
mance of the resin in service does not indicate fouling tendencies.
63
-------
TABLE 21. Phosphate break-thru from weak base anion resin (IRA-68)
during exhaustion with cation effluent.
Influent contains 20.8 rag/1 phosphate. Per cubic
foot basis.
Gallons to
service
0
1000
1900
1950
1965
1975
1985
1995
2000*
2005
Composite Average
Phosphate,
mg/1, PO^—
0.2
0.2
6.0
18.0
22.0
21.6
28.0
28.8
30.0
32.0
0.6
Efflu
Condu
micro
--
__
30
40
50
60
70
80
90
100
— — —
*Endpoint of service exhaustion.
64
-------
TABLE 22.
ANION RESIN ( IRA 68) PERFORMANCE
RECAP SHEET
Exhausted with cation effluent
(per cubic foot)
Run
#
NH4OH
( 100%)
Lbs.**
Exhaust
GPM
Loading
mg/1
Capacity
gals.
Kgr
1 New
2 3.2
3 3.2
4
5
Series A average (* based)
6 3.2
7 "
8
9
10
11
Series B average (* based)
12 3.2
13
14
15
16
Series C average (*based)
17 3.2
18
Series D average
19 2.8
20
Series E average
21 3.2
22
Series F average
220
1878
6
6
2-1/3
24.1
212
192
188*
212*
200
228*
212*
206*
240*
216*
240*
224
228*
228*
206*
Incompleted
216*
220
240*
220*
230
216*
340*
278
236*
232*
234
1868
2134
1921*
1794*
1858
1878*
2008*
1994*
1731*
1994*
1808*
1902
1767
1907
1897
2040
1904
1763*
1873*
1819
1840*
1200*
1520
1813*
1840*
1827
23.1
23.9
21.1*
22.2*
21.7
25.0*
24.9*
24.0*
24.3*
25.2*
25.3*
24.8
23.5*
25.4*
22.9*
25.7*
24.4
24.8*
24.1*
24.5
23.3*
23.9*
23.6
25.0*
24.9*
25.0
65
-------
Table 23. Ionic br,ok-thru from w,,,k K,,, dnlo,, resin ( UfA-DO) during service exhaustion.
WATER ANALYSIS
FWPCA Contract NO. 14-12-599 DATE .._.._. November _1970
TEST {'AS CoC03)
PH
'
CHLORIDES'
TURBIDITY
IH J.T.U.
SULFATES'
AS P04
TOTAL HARDNESS-
MAGNESIUM'
CALCIUM-
AMMONIA
HITRATE
HITROGEN AS N
NITRITE
NITROGEN AS N
SODIUM-
POTASSIUM'
TOTAL
ORGANIC CARBON
TOTAL
DISSOLVED SOLIDS
-Wymbep
9.0
15.6
10.8
1053
Carbon
2.5
13.6
7.8
1052
Cation
904.
130.
72.
20.8
5.8
9.1
1050
4.
10.
0.
0.2
1.4
9.3
1047
3.
8.
0.
.0.2
0.1
9.3
1048
2.
18.
0.
30.4
0.1
7.2
1049
12.
6.
0.
0.6
0.5
8.0
1051
Gallons For
It cu ft
Drip Drip
Drip
3010 Drip
IRA-68 Run #7
WATER ANALYSIS
FWPCA Contract No. 14-12-599 DATE November. 1970
TEST CAS CoCOj)
88 Mi.-T.omh,.;
ALKALINITY, OH'
ALKALINITY, C03'
ALKALINITY, HC03*
CHLORIDES-
TUR8IDITY
IN J.T.U.
SULFATES-
TOTAL PHOSPHATE
AS P04
TOTAL HARDNESS-
MAGNESIUM-
CALCIUM'
AMMONIA
NITROGEN AS N
NITRATE
NITROGEN AS N
NITRITE
NITROGEN AS N
SODIUM-
POTASSIUM-
TOTAL
ORGANIC CARQON
TOTAL
DISSOLVED SOLIDS
., Numbei -. ,
Gallons For
15 cu ft
no
IB.
R.
0.
fi.n
n.
0.
0.
0.2
n.i
9.6
0.07
7.1
1054
2860
40
IIS.
R.
0.
1R.
n.
0.
0.
0.3
0.1
12.
0.12
6^4 .
1055
50
17.
14.
0.
99
0.
n.
0.
0.4
14.
0.12
6.1.....
1Q')6
2917 2955
60
R.
10.
0.
91.fi
0.
n.
0,
0.3
0.1
15.
0,14
6.2..
10r,7
2965
70
H.
If,.
0.
9B.S
0.
0.
o
0 0
0.1
16.
0.15
_ 5.9
105F1
RO
fi.
Id.
0.
9R.O
0.
0.
0.
0.3
0.1
17.
0.11
6,1
1059
90
6.
In.
0.
30.
0.
0.
0.
0.3
0.1
19.
0.16
6.6 .
1060
100
4.
90.
0
39.
0.
0.
0.
0.3
0.1
19.
0.15
1061
2975 7990 3000 3010
Break thru Samples IRA-IJfl Run »7
66
-------
The regeneration of the IRA-68 has been characterized when it was
regenerated with the recommended 3.2 pounds of ammonium hydroxide per
cubic foot. The regenerant effluent was sampled in fractional bed
volume increments. The analyses for the samples collected within the
range of 2/3 to 4 bed volumes are shown on Table 24. As shown, re-
generant ammonium hydroxide was initially neutralized by the adsorbed
acids so that low effluent alkalinity resulted. This characteristic
as well as others are much the same as they were for IRA-93 resin as
shown in Table 19. The main differences between analyses for the two
elutions are that phosphate elution is better with IRA-93 than it is
with IRA-68; nitrate elution is also better with the IRA-93; total
organic carbon is better from the IRA-93 in view of the continued
elution for a longer period of time than occurs for the IRA-68.
The pilot plant studies have shown that this weak base resin IRA-68 is
easily applied for demineralization of secondary treated sewage effluent,
Stable operating capacities have been obtained without operational dif-
ficulties. Performance slightly favors the choice of IRA-93 over that
of IRA-68 with the exception of the slightly increased capacity from
IRA-68. More importantly, the capital cost for IRA-93 is about 2/3
that of the IRA-68 resin.
The recommendation is made that IRA-93 weak base resin is the choice
in the selection of such a resin in combination with (and preceded by)
a strong acid cation exchange resin in the hydrogen form.
67
-------
00
TABLE 24. Regenerant effluent analyses from weak base anion resin (IRA-68)
Regenerated with 3.2 Ibs of 100% ammonium hydroxide.
WATER ANALYSIS
FWPCA Contract No. 14-12-599
PATE December. 1970
TEST CASCaCO,)
PH
ACIDITY, MINERAL *
ALKALINITY, OH*
ALKALINITY, CO}*
CHLORIDES'
TURBIDITY
IN J.T.U.
SULFATES*
TOTAL PHOSPHATE
AS P04
TOTAL HARDNESS*
MAGNESIUM*
CALCIUM*
AMMONIA
NITROGEN AS N
NITRATE
NITROGEN AS N
NITRITE
NITROGEN AS N
SODIUM*
POTASSIUM*
TOTAL
ORGANIC CARBON
TOTAL
DISSOLVED SOLIDS
Number
140.
660.
40.
250.
344.
2.0
0.009
16.7
1139
200
7300.
13600.
3760.
9659.
30.
0.034
27.4
1140
575
14500.
18800.
3200.
11700.
60.
0.042
21.8
1141
780.
27100.
11200.
1330.
loasn.
i*-,n.
0.029
20.4
1142 .
2440
2720.
15600.
1400.
285.
r-,-i4n
ion.
0.003
10.0
1143
1660.
1160.
3600.
540.
85.
1RQD
4<">
0.003
8.0
1144
840.
7JO.
mon.
.1HO-
BH.S
]0?2
30
0.002
9.2
1145
520.
SfiO.
7*1(1-
->hri.
7-S.9
550
°5
0.003
6.?
1146
2?0.
ddfl.
tan
09n_
•iR.n
3 10
19.
0.004
fi.5
1147
Bed Volume "2/3 1 1-1/3 1-2/il 2-1/3 3 3-1/3 3-2/3 4
Regeneration Effluent IRA-68
-------
SECTION 10
WEAK ACID CATION EXCHANGE RESIN PERFORMANCE-HYDROGEN FORM
This process employs a weak acid cation exchange resin. Rohm & Haas'
resin IRC-84 was used. This resin contains carboxylic acid functional
groups. Whereas strong acid cation exchange resin is capable of
splitting all ionized salts in water to remove cations by an equivalent
exchange for hydrogen ions, weak acid cation exchange resin can only
neutralize alkaline salts. Calcium and magnesium associated with the
alkalinity present in the water, are readily removed by weakly acidic
resin; the capacity for sodium removal is quite low.
An example of calcium removal with this resin, represented as RCOOH,
is shown in the following chemical equation employing calcium bi-
carbonate, one of the most typical inorganic chemicals found in water.
2R-COOH + Ca(HC03)2 = (R-COO^a + H2C03
Calcium is removed by an equivalent exchange for hydrogen ions from the
ion exchange resin. The hydrogen ions combine with the bicarbonate ions
already present in the water to form carbonic acid - H^Os. The carbonic
acid so formed can disassociate to form water and dissolved carbon dioxide
according to the following equation.
H2C03 = H20 + C02
The carbon dioxide can be removed by degasification or aeration
techniques.
The resin which has been exhausted through the above process can be
regenerated with strong acids. The following equation illustrates the
regeneration with sulfuric acid.
(R-COO)2Ca + H2S04 = 2R-COOH + CaS04
Sulfuric acid is normally used for regeneration because of its availa-
bility and low cost. Sulfuric acid can cause precipitation of calcium
sulfate during the regeneration; therefore, care must be taken to use
a low acid concentration to prevent precipitation within the ion ex-
change resin. If not prevented, such precipitation will result in
poor performance. The precipitate will physically plug resin pores,
which in turn results in reduced capacity. The precipitate will also
slowly dissolve during the service cycle so that increased calcium
will appear in the treated water.
69
-------
Weak acid cation exchange resin will only partly demineralize waste
water. The extent of demineralization will not exceed the content of
alkalinity present in the water. The effluent from this resin will
contain all cations associated with non-alkaline cations. Complete
cation removal requires post-treatment with a strong acid cation ex-
change resin. If complete demineralization is not required, the weak
acid cation exchange resin is the only resin process which need be
applied. For example, with the sewage at Elgin, Illinois, this resin
by itself reduced the inorganic salts from about 600 mg/1 to about
250 mg/1. Application of the process would dictate what post-treat-
ment methods would be required for removal of phosphate or ammonia.
Attempts to apply the hydrogen form of the weak acid cation exchange
resin have been encouraging and successful. Conventional techniques
of applying this resin were used. This process can be used in the de-
sign of treatment plants where only partial demineralization of the
waste water is desired.
The influent water employed in the IRC-84 study was the sewage treated
in the pre-treatment section of the pilot plant. This treatment con-
sisted of dual media filtration and activated carbon adsorption.
Weak acid cation exchange resin studies were conducted in a 14" diameter
tank containing 2-g- cubic feet of mineral. The resin bed height was
about 29" in the hydrogen form.
Regeneration was accomplished with a 0.7% sulfuric acid solution. This
strength is sufficient to regenerate the resin, while avoiding precipi-
tation of calcium sulfate. The regenerant solution was made with soft
potable water as a matter of convenience. The solution was prepared in
the open regenerant solution vessel, then pumped through the cation ex-
change resin bed. Rinse was in two stages: first with approximately
1-1/3 bed volumes of softened water at the regeneration flow rate;
second, with influent service water at the service flow rate.
Observation of the recorded effluent conductivity indicated when the
fast rinse was completed. Service exhaustion was then immediately
initiated. Prior to this point, the conductivity decreased rapidly;
during service it was stable; at exhaustion the conductivity increased.
Resin capacity was calculated from the volume throughput to an effluent
conductivity rise of 50 micromhos over the average (stable) conductivity
during exhaustion. This exhaustion endpoint corresponds to a 10% alka-
linity leakage; i.e., the effluent alkalinity was about 10% of that in
the influent.
Capacity calculations were based on the above measured volume of water
and its average total alkalinity. Alkalinity was determined from an
analysis of the composite sample of the influent.
70
-------
Table 25, "Weak Acid Cation Exchange Resin (lRC-84) Performance
Summary", summarizes the performance of this resin. Four test series
were run. A comparison of series A and B shows capacity is greatly
affected by flow rate, which is typical for the weak acid exchange
resin. At exhaustion flow rates of 6 and 3 gpm/cu ft, capacities
of 17.2 and 21.5 kilograins were obtained. Effluent quality is about
the same at the two flow rates. Test series C shows that greater acid
utilization is obtained at a lower regenerant dosage. Test series D
shows that increasing the regenerant strength and flow rate reduced
the acid utilization. Series B appears to be the best system because
of the lesser requirements of resin, acid, and rinse water to produce
1,000 gallons of treated water.
Table 26 recaps the individual exhaustion runs summarized in Table
25. The data has been calculated to a "per cubic foot" base, al-
though more resin was actually used. The second column of this table
shows the weight of 66° Baume sulfuric acid used for regeneration.
The third column shows the exhaustion flow rate, in gallons per
minute.
The loading during exhaustion, shown in the fourth column, is expressed
in mg/1 as calcium carbonate equivalents. The determination of this
loading factor has been previously described.
The last two columns of Table 26 show the capacities obtained. The
gallons shown are those delivered between the two selected start and
endpoint conductivity values for the effluent as previously discussed.
The capacity has been calculated, as shown in earlier sections, from
the gallons produced and the loading factor to obtain the values shown
in kgrs/cu ft.
The performance obtained with this resin is encouraging. Stable
capacities have been fairly easily obtained, with no indication of
organic fouling. Unfortunately, the methods of Rohm & Haas, the resin
supplier cannot be used to predict the performance of this resin employed
for sewage demineralization. Predicted capacities, as compared with the
values obtained for these four test series are shown in Table 27, "Cation
Resin (IRC-84) Performance-Prediction Vs Actual". The actual capacity
is significantly less than that predicted.
The quality of water produced from the cation exchange resin is shown in
Table 28. The table shows the typical analysis of the influent and
effluent for the cation exchange resin for the four test series. As
shown, the resin reduced the influent total organic carbon from a range
of 6.2 - 17.6 mg/1 to 6.9 - 8.9 mg/1, as measured by a Beckman Carbon
Analyzer. However, analysis of samples of regeneration effluent, and
resin samples removed from the test vessel, showed no evidence to sub-
stantiate this observation. The regeneration effluent does not show
any elution of total organic carbon. Resin analysis does not indicate
accumulation of organics.
71
-------
Table 25. Weak Acid Cation Exchange Resin
(IRC-84) Performance Summary,
Per cubic foot base.
Test Series
Regenerant: Ibs/cu ft
Strength %
gpm/cu ft
Exhaust, gpm/cu ft
Capacity, Kgr/cu ft
Acid Utilization %
Leakage, Average mg/1:
Total hardness
Total cations
Requirements per 1000
Gallons of Water
Produced:
Cation resin,
cu ft
Acid, Ibs 66° Be
Rinse Water, Gals.
A.
3.9
0.7
1.0
6
17.2
66.6
23
236
1.22
4.8
66
B.
3.9
0.7
1.0
3
21.5
83.5
30
229
1.00
3.9
54
C.
3.4
0.7
1.0
3
18.5
93.5
54
258
1.15
3.9
62
D.
3.9
1.0
1.4
3
18.6
71.5
49
245
1.19
4.6
64
-------
TABLE 26
WEAK ACID CATION EXCHANGE RESIN (IRC-84)
PERFORMANCE RECAF SHEET
PER CUBIC FOOT BASE
Run
#
1
2
3
4
5
6
7
8
9
10
Series
11
12
13
14
15
Series
16
17
18
Series
19
20
21
22
QOT T PQ
H2S04
Ib.
New
3.9
It
1T
IT
fl
11
tl
It
11
A Average
3.9
11
tt
11
!l
B Average
3.4
II
1!
C Average
q a
o . ™
ii
M
ii
n Averaae
Exhaust
qpm
6
6
It
M
11
tt
11
If
11
tl
(* based)
3
It
ft
tl
11
(* based)
3
ti
M
(* based)
3
It
M
11
(* based)
Loading
mq/1**
360
360
380
352
348
368
360*
350*
362
368*
359
379
368*
372*
366*
372*
369
352*
368*
368*
363
Mechanical
372*
380*
412
376
Capacity
qal. Kqr.
1066
1210
910
558
640
836
827*
840*
927
790*
819
1124
1000*
978*
1026*
984*
997
881*
847*
885*
871
Difficulties
843*
847*
843
845
22.4
25.5
20.2
11.5
13.0
18.0
17.4*
17.2*
19.6
17.0*
17.2
24.9
21.5*
21.3*
21.9*
21.4*
21.5
18.1*
18.3*
19.1*
18.5
18.4*
18.8*
20.3
18.6
** as calcium carbonate equivalent
73
-------
A. B.
6 3
639 551
C.
3
594
D.
3
603
TABLE 27
CATION RESIN (IRC-84) PERFORMANCE
PREDICTION VS ACTUAL
Test Series
Exhaust, gpm (cu ft)
Total cation influent
mg/1
Hardness/ alkalinity
influent 1.03 0.79 0.89 0.88
Capacity predict:
Gross Kgr/cu ft 54.0 26.0 31.0 3.1.0
Correction for gpm 32/58 51/58 51/58 51/58
Net Kgr/cu ft 29.8 22.9 27.2 27.3
Actual Kgr/cu ft 17.2 21.5 18.5 18.6
Actual/predict capacity 58% 94%
74
-------
Table 28 Water analyses summary of typical ex-
haustion of cation exchange resin (IRC-84).
All influent and effluent analyses are
expressed in mg/1, and as CaCC^ unless
specified.
Test Series
Regenerant, Ib/cu ft
Exhaust, gpm/cu ft
Influent:
Total organic carbon-C
Alkalinity
Magnesium
Calcium
Sodium
Potassium
Ammonia-N
Total non-N cations
Effluent:
Total organic carbon-C
Alkalinity
Magnesium
Calcium
Sodium
Potassium
Ammonia-N
Total non-N cations
Reduction of cations^
A
3.9
6
6.2
357
200
167
229
12
6.4
608
6.9
5.3
17.9
11.7
200.0
12.0
6.4
241.6
60.3
B
3.9
3
17.6
373
189
122
212
16
9.4
539
8.9
7.3
15.9
14.3
188.0
16.0
8.8
234.2
56.5
C
3.4
3
10.9
355
161
154
264
15
7.9
594
7.9
11.0
26.5
27.0
195.0
10.0
7.8
258.5
56.5
D
3.9
3
7.4
367
163
157
245
20
6.6
585
7.9
9.0
28.5
28.5
195.0
9.0
2.6
261.0
55.4
-------
A review of Table 28 shows that there is a great reduction in the
calcium end magnesium content by this process. The total reduction of
influent cations (exclusive of ammonia) ranges from 55.4 - 60.3%. Total
influent cations were about 600 mg/1 (as calcium carbonate), while the
effluent total cations were about 250 mg/1. This very significant
reduction is easily obtained.
Regeneration of this cation exchange resin with sulfuric acid has been
without difficulty. Precipitation of calcium sulfate in the resin bed
has been avoided. Specific analyses for total organic carbon elution
show only analytical and sampling variation throughout the regeneration.
A small excess of regenerant acid, which may require post-treatment
before disposal, is present in the regeneration effluent. Neutralization
of this acid is discussed in a later section of this report. The amount
of neutralization required is minimal due to the high regeneration ef-
ficiency of this resin.
The recommendation is made that weak acid cation exchange resins can be
applied to treat the effluent from secondary sewage treatment plants.
However, weak acid cation exchange resin is only a partial means of
wastewater demineralization. Many applications will require subsequent
treatment or additional demineralization. Some applications may accept
the effluent from this resin without futher demineralization, but
selected processes may be required for removal of phosphate, which is
totally unaffected by this ion exchange process, as well as for the
reduction of ammonia which is only affected slightly by this resin.
76
-------
SECTION 11
WEAK ACID; STRONG ACID CATION EXCHANGE RESIN
PERFORMANCE - HYDROGEN FORMS
This process employs two types of cation exchange resins: strong acid
and weak acid resins. The resins are not mixed, but rather remain in
discrete beds although they may be held in one pressure vessel. During
exhaustion, the water to be treated passes first through the weak acid
resin, then the strong acid resin. The performance of these resins to
separately treat water has been previously discussed in this report.
The advantage to using both forms of cation exchange resin is the
great efficiency obtained during regeneration. Excess acid is required
for the regeneration of strong acid resin. This excess acid, which
would otherwise require neutralization, is of sufficient quality to
regenerate the weak acid exchange resin. As a result, the ion exchange
capacity obtained from the weak acid exchange resin is obtained without
increased operating costs. A second advantage is the reduced require-
ments for alkaline reagents to neutralize waste acid.
Post treatment of the effluent from these two cation resins is required
to produce an acceptable water. Such treatment consists primarily of
weak base anion exchange to adsorb the mineral acid produced. The
weak base resin has previously been discussed in this report. De-
gasification is desirable to eliminate dissolved carbon dioxide from
the product water. Break-point chlorination is desirable to eliminate
residual ammonium nitrogen as well as for disinfection.
We have successfully applied the two types of cation exchange resins
in the hydrogen form, operating in series to treat the effluent from
the activated sludge treatment plant. Accepted techniques for the
ion exchange industry have been applied without difficulty. It is
demonstrated, therefore, that the process can be used in the design of
treatment plants.
The pilot plant studies with these two resins followed two concepts.
First tests were with the resins retained in separate pressure vessels.
The second tests were with both resins retained in one pressure vessel,
although the-two resins were maintained physically separated one from
the other because of their different physical properties.
The tests with resins retained in separate pressure vessels were per-
formed with three beds of resin: a weak acid cation resin bed, and two
beds of strong acid cation resin. The three beds were arranged in two
systems whereby the sewage effluent flowed in parallel to a system
consisting of only a bed of the strong acid exchange resin, and also
77
-------
to a second system with two tanks in series—the first tank containing
the weak acid resin, while the second tank contained the strong acid
resin. The purpose of the parallel study on these two systems of three
resin beds was to determine the merit of using two resins as opposed to
using only the strong acid resin.
Table 29, "Cation Resins Summary," summarizes the performance of these
three beds of cation resin. It. quickly shows that the use of two resins
in series is very desirable rather than using only the strong acid.
Comparing the second and third columns for the two resins in series
with the last column for the one resin only shows that the requirements
to produce 1000 gallons of treated water is less with the two resins.
Less resin (2.19 vs 3.18 cubic feet respectively); less acid (7.7 vs
9.6 pounds respectively); but more rinse water (98 vs 95 gallons re-
spectively) are needed for the systems.
Obviously, Table 29 favors the two resin system. Operating costs can
be further reduced from that suggested in the table. This is because
the tests in support of Table 29 did not use waste acid for regeneration
of the weak acid resin. Such acid reuse is discussed later in this
section.
Tables 30, 31, and 32 are summary sheets for the individual resin beds.
Table 30 is for the carboxylic (weak acid) cation resin. The data shown
is "per cubic foot", although two cubic feet (with a 30" height) were
actually used in the column. This resin treated the sewage effluent
which had been filtered by a dual media filter in the pretreatment system.
As shown, 13 cycles were made in this test series; however, run 6 through
10 were not made on the weak acid resin so that efforts were concentrated
on the strong acid cation resin operating by itself. The table shows
unstable capacities, which are unexplainable. However, if the four
lowest capacities are considered, an average capacity of 18.7 kilograins
per cubic foot were obtained from an acid regeneration dosage equivalent
to 25.8 kilograins. An operational inefficiency is indicated, but this
can be tolerated with regenerant reuse.
The effluent from this weak acid cation resin was then passed through
a 30" bed of strong acid cation resin for additional treatment. The
performance of that strong acid resin is shown in Table 31. The 13 runs
for the series are again numbered on this sheet although several runs
were incompleted. The last two columns in this table show the partial
analysis of a composite drip sample collected from the strong acid
cation resin effluent. This analysis shows the total cations (5th
column) present in the average effluent from this resin. This is
expressed in calcium carbonate equivalents. These cations will be
delivered to service because they are sufficiently low in concentra-
tion to be acceptable.
78
-------
Table 29. CATION RESINS SUMMARY
Requirements to produce 1000 gallons of treated water
Resin Type
Regenerant :
Ibs/cu ft
Utility, %
Capacity:
Kgr/cu ft
gallons
Carboxylic
only
3.9
73
18.7
800
Sulfonic after
Carboxylic
3.0
81
16.0
1062
Sulfonic
only
3.0
56
11.0
314
Needed/1000 gallons:
Resin, cu ft 1.25
Acid, 66° Be 4.9
Rinse water, gal. 70
0.94
2.8
28
3.18
9.6
95
79
-------
Table 30. CARBOXYLIC CATION RESIN (IRC-84) PERFORMANCE
RECAP SHEET - per cubic foot
Regenerated with 3.9 pounds 66 Be sulfuric acid, equivalent to
25.8 kilograins as CaCO_. Exhausted at 3 gpm/cu ft on filtered
sewage effluent.
Run Loading
# mq/1 H/A*
Capacity
gal. Kqr
Endpoint, mg/1
Hardness Alkalinity
1A 344
2A 340
3A 352
4A 328
5A 333
6A Test
7A
8A
9A "
10A
11A 364
12A 364
ISA 356
1.03 1535
1.03 1000
0.97 950
1.09 1650
1.15 1203
series interrupted to
11 n n
11 n n
n n n
1.00 858
0.89 800
0.92 1500
30.9
19.9
19.6
31.6
23.4
concentrate
it
n
11
n
18.3
17.0
31.2
57
61
31
83
83
efforts on
it it
n n
11 n
n tt
20
13
11
32
42
32
38
54
other resin.
n 11
n n
n it
n tt
50
36
40
H/A refers to the ratio of influent hardness to alkalinity.
80
-------
Table 31. SULFONIC CATION RESIN (IRC-120) PERFORMANCE
RECAP SHEET - per cubic foot
Regenerated with 3.0 pounds 66° Be sulfuric acid, equivalent
to 19.8 kilograins as CaCOg. Exhausted at 3 gpm/cu ft on carboxylic
cation resin effluent.
Run
#
IB
2B
3B
4B
5B
Loading Capacity Composite*
mq/1 gal. Kgr Cations Ammonia
Mechanical
Mechanical
312 1000
274 866
284 1066
difficulties
difficulties
18.3 89.6
13.9 118.6
17-7 44.5
2.9
2.4
2.2
Test series interrupted to concentrate efforts on other resin.
gg
10B
11B
123
13B
252
264
256
1007
1066
1120
14.9
16.4
16.8
64.9
79.2
68.1
1.8
2.3
2.7
* Concentration, mg/1 as CaCO-, in composite sample of water treated
by both cation exchange resins - carboxylic and sulfonic.
81
-------
Table 32. SULFONIC CATION RESIN (IRC-120) PERFORMANCE
RECAP SHEET - per cubic foot
Regenerated with 3.0 pounds 66 Be sulfuric acid, equivalent
to 19.8 kilograins as CaCO^. Exhausted at 3 gpm/cu ft on
filtered sewage effluent.
Run
#
1C
2C
3C
4C
5C
6C
7C
8C
9C
IOC
11C
IX
IX
Loading
mg/1
564
584
624
578
584
520
576
596
592
588
600
604
588
Capacity
gal. Kqr
512
430
360
283
323
298
302
312
297
316
310
313
326
16.9
14.7
13.2
9.6
11.1
9.1
10.1
10.8
10.3
10.9
10.9
11.0
11.2
Composite*
Cations Ammonia
102.6
82.9
168.0
38.1
22.4
17.2
18.7
15.4
13.9
16.5
26.4
27.5
31.4
2.5
2.0
3.1
1.6
1.2
0.9
1.0
1.7
1.5
1.5
1.6
1.3
1.8
* Concentration, mg/1 as CaC03, in composite sample of water treated
by sulfonic acid resin only.
82
-------
table 31 also shows the ammonia concentration in the average effluent.
Removal of this ammonia can be accomplished by break-point chlorination.
The average capacity obtained for the last three runs on this resin
was 16.0 kilograins per cubic foot.
Table 32 shows data for the 13 runs which were actually made on the
30" bed of strong acid cation resin treating the filtered sewage
effluent in parallel with the above two resin beds which were operating
in series. This bed of strong acid cation resin was regenerated and
exhausted with the same techniques as were used on the above discussed
bed of strong acid resin, with the exception that the influent waters
were of different composition. The average of the last four results
for the capacity with this resin was 11.0 kilograins per cubic foot.
The last two columns in the table again show the cation content in
the composite effluent. The cation will be delivered to service while
break-point chlorination will remove the ammonia.
Exhaustion and regeneration cycles for these test series were controlled
and calculated with techniques which were previously discussed in this
report.
Further investigation of the applicability of using both weak and
strong acid cation exchange resins was directed toward combining the
resins in one pressure vessel. Such application requires that the
weak acid resin must not become significantly mixed with the strong
acid resin. Possible mixing is easily avoided by selecting a strong
acid cation resin which is denser than the resin normally used. With
this choice of resin, the strong acid resin remains below the weak acid
resin. Therefore, exhaustion must be downflow for the water to contact
the weak acid resin first. Similarly, regeneration with sulfuric acid
must be upflow in order to avoid precipitation of calcium sulfate
within the resin bed. This is accomplished because most of the calcium
is removed by the upper bed of weak acid cation exchange resin. During
regeneration, the eluted calcium is removed from the bed before pre-
cipitation occurs.
The denser strong acid cation exchange resin has performance charac-
teristics which are essentially the same as the resin normally used.
Preliminary studies were made with the two resins held in separate
pressure vessels operated in series for separate evaluation of their
performance prior to combining them in one vessel. In exhaustion, the
water was first treated by the weak acid resin, then by the strong
acid resin. In regeneration, the sulfuric acid was passed upflow
through both resins, first through the strong acid resin then through
the weak acid resin.
In combined bed approach, design considerations much primarily con-
sider the strong acid resin, because of its operating inefficiencies.
A bed heigh of 30" for the strong acid cation resin was used in our
83
-------
studies. The effect of varying amounts of weak acid resin were
considered.
Regeneration was with 2% sulfuric acid upflow through the strong acid
resin, then through the weak acid resin. Precipitation of calcium
sulfate in the regenerant effluent was found to occur about five
minutes (at 80 - 100° p) after leaving the equipment.
Exhaustion was made with the effluent from the dual media filter. The
exhaustion flow rate was based on four gpm per cubic foot of strong acid
cation resin used. Our 30" bed required 2-g- cubic feet of this resin;
therefore, a total flow rate of 10 gpm during exhaustion was used in
these tests. The exhaustion was monitored by means of a recording
conductivity meter, positioned to test the effluent from the strong
acid resin. The exhaustion was considered terminated when the effluent
conductivity abruptly decreased by 500 micromhos from the stable value
throughout the run. This conductivity decrease was caused by a reduction
in the amount of acid produced by cation exchange.
The capacity of the individual resins was determined by methods previously
described, using the volume of water treated and the loading factor. The
loading factor for these studies on the two resins is the net loading
factor during exhaustion, being the difference between the influent and
effluent ions whose concentration was affected by the exchange resin.
Tcbles 33 and 34 contain the results of the exhaustion cycles performed
for these tests. The tests are grouped into several series, according
to the amounts of weak acid resin and regenerant acid used. Stable
performance was easily obtained, so that the test values were averaged
and used to prepare the summaries of Table 36. The effluent quality
during these tests is shown in Table 35. The values are typical of the
results obtained from the analyses of composite samples of the effluent
during exhaustion. As previously discussed, this water will need post
treatment to be of acceptable quality for most uses.
Table 36 summarizes the test series from Tables 33 and 34. Test series
A was run with one cubic foot of weak acid resin (IRC-84) preceding
(in service) 2^- cubic feet of strong acid resin (IRC-122). Utilization
of the acid regenerant was 79%. An inspection of analyses made on
water samples collected during the exhaustion revealed that the weak
acid resin effluent contained an excessive concentration of alkaline
hardness. This was not unexpected because of two factors. First, the
weak acid resin bed height was only 12". Second, the exhaustion flow
rate was too high: 10 gpm per cubic foot of weak acid resin. The test
was made with these two conditions of poor choice to obtain a "base
point" of performance for later comparison.
84
-------
Table 33. Performance of two beds of cation resin in series, (IRC-84
preceding IRC-122). Counter-current flows: downflow ex-
haustion, upflow regeneration. Total cubic foot resin
volume basis. Exhaustion at 4 gpm/cu ft of IRC-122 only.
Run Cap.
No. Gal.
Series A:
3 1614
4 1547
5 1406
6* 1414
7* 1400
Aver. 1407
Series B:
8 1683
9 1747
10 1626
11* 1570
12* 1648
Aver. 1602
Series C:
13 1713
14 1727
15* 1572
16* 1585
Aver. 1578
Series D:
17 1625
18 1737
19 1666
20 1973
21 1926
22* 1808
23* 1647
24* 1700
25* 1718
26* 1616
Aver. 1698
Series E:
27* 1678
28* 1656
Aver. 1667
IRC-84
Cu ft
84
1
1
1
1
1
1
1-1-
1-1-
iir
i-i-
]i
i-t
i-i-
ii-
i-i-
ii
1s"
IE"
1&
l£
if"
ii
li
ij-
ji
IE
1&
il-
l's"
4
( carboxylic)
Load
mq/1
210
200
210
260
270
265
308
286
270
254
240
247
270
260
278
260
269
296
282
280
258
240
270
270
290
286
280
279
270
294
282
Cap
Kqr
19.8
18.1
17.2
21.6
22.2
21.9
30.3
29.2
25.7
23.3
23.1
23.2
27.0
26.3
27.8
24.1
26.0
28.2
28.6
27.3
29.8
27.0
28.5
26.0
28.8
28.7
26.5
27.7
26.5
28.5
27.5
resin
TH/M**
Infl
0.97
0.91
1.00
0.92
0.89
--
0.87
0.97
1.00
0.97
0.91
—
0.90
0.88
0.88
0.88
--
0.80
0.82
0.79
0.83
0.78
0.76
0.77
0.77
0.84
0.82
—
0.82
0.82
0.82
IRC-122
resin,
Load
mg/1
211
248
256
202
220
211
165
167
178
223
238
230
231
231
206
238
222
190
205
208
171
204
232
238
222
234
261
237
250
262
256
(Sulfonic) |
2 1/2 cu ft \
Cap j
Kqr 1
20.0
22.5
21.0
16.8
18.0
17.4
16.2
17.1
16.9
20.5
22.9
21.7
23.1
23.3
18.9
22.1
20.5
18.1
20.8
20.3
19.7
23.0
24.5
22.9
22.1
23.4
24.7
23.5
24.5
25.4
25.0
System
Capacity
Kqr
39.8
40.6
38.2
38.4
40.2
39.3
46.5
46.8
42.6
43.8
46.0
44.9
50.1
49.6
46.7
46.2
46.4
46.3
49.4
47.6
49.5
50.0
53.0
48.9
50.9
52.1
51.2
51.2
51.0
53.9
52.5
Regeneration
Total Ibs H SO
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
8.3
8.3
8.3
8.3
8.3
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
10.0
10.0
10.0
* Data used for averages
**Ratio of influent total hardness (TH) to methyl orange alkalinity (M),
85
-------
Table 34. Performance of two beds of cation resin in series,
(IRC-84 preceding IRC-122). Counter-cur rent flows:
downflow exhaustion, upflow regeneration. Total
cubic foot resin volume basis. Exhaustion at 4 gpm/
cu ft of IRC-122 only.
Run Cap.
No. Gal.
Series F:
29* 1930
30* 2027
31 1968
Aver. 1980
Series G:
32 2452
33* 2054
34* 2059
Aver. 2056
Series H:
35 2204
36* 2501
37* 2388
38* 2174
Aver. 2354
39 2640
40* 3401
41* 3540
42* 3027
Aver. 3323
IRC-84 (carboxylic)
Cu ft
84
1*
1*
1*
1*
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Load
mq/1
312
296
284
304
316
286
296
291
276
298
302
314
305
346
326
322
320
323
Cap
Kqr
35.3
35.0
32.7
35.2
45.3
34.3
35.6
35.0
27.6
43.6
42.2
39.9
41.9
53.4
64.8
66.7
56.7
62.7
resin
TH/M**
Infl
0.82
0.84
0.78
0.83
0.90
0.76
0.78
0.77
0.81
0.81
0.81
0.81
0.81
0.78
0.84
0.83
0.81
0.83
IRC- 122 (sulfonic)
resin, 2 1/2 cu ft
Load
mq/1
242
246
104
244
112
264
236
250
281
227
230
254
237
216
210
191
218
206
Cap
Kqr
27.4
29.2
12.0
28.3
16.1
31.7
28.4
30.0
37.5
33.2
32.1
32.2
32.5
34.6
41.8
39.6
38.6
40.0
System
Capacity
Kgr
62.7
64.2
44.7
63.5
61.4
66.0
64.0
65.0
65.1
76.8
74.3
72.1
74.4
88.0
106.6
106.3
95.3
102.7
Regeneration
Total Ibs
H?S04
11.25
11.25
11.25
11.25
11.25
11.25
11.25
11.25
12.5
12.5
12.5
12.5
12.5
17.5
17.5
17.5
17.5
17.5
* Data used for averages
**Ratio of influent total hardness (TH) to methyl orange alkalinity (M).
86
-------
Test series B was then run using more ( ITT cubic feet) weak acid resin.
This increased the bed height to 18 inches, while reducing the flow
rate to 6-2/3 gpm per cubic foot of weak acid resin. These changes
resulted in an increased regenerant utilization of 90%. Analyses of
the regenerant effluent revealed that there was no mineral acidity to
be neutralized at the drain. Therefore, the next four test series
(C, D, E, and F) were made with progressively increasing dosages of
acid regenerant. Paradoxically, waste acid was not detected in the
regenerant effluent until very high regenerant dosages were used so
that the utilization during exhaustion was less than 80%. With bette-
acid utilization, waste acid should have appeared in the regenerant
effluent. This anomaly is apparently due to non-productive use of ion
exchange capacity by the water used to backwash and rinse the resin
during regeneration, as well as by the affect of organic ions during
exhaustion and regeneration.
Reviewing the summaries for the test series C through F, it appears
that the acid utilization for run E is inconsistent with the value for
the other series. The likely explanation for this inconsistency is
the fact that series E consisted of only two exhaustion tests, which
at the time appeared to be stabilized. However, such apparently was
not the case. The acid utilization for series C, D, and F was 84-85%,
unaffected by the acid dosages used. Test series G, H, and I were
made with an increased amount (2 cubic feet) of the weak acid resin.
The acid utilization is consistent, and exceeds 85% for these three
series.
Table 36 shows the amount of acid and resin required to produce 1000
gallons of treated water. These values have been used for the further
presentation of cost as shown in Table 37. The following material costs
were used in the calculations for Table 37.
Sulfuric acid, 66° Baume, cj: per pound 1.6
Weak acid resin (IRC-84), $ per cubic foot 50.5
Strong acid resin (IRC-122), $ per cubic foot 23.0
The acid cost is a typical one, which will vary depending on delivery
cost at the plant site. The resin costs are those which have been
established by the manufacturer for industrial users of .these resins.
Table 37 is best understood by inspecting Figures 15 and 16, which
graph the data as a function of the amount of sulfonic acid resin used.
Figure 15 graphs the data for a regenerant dosage of 7^- pounds total of
sulfuric acid, while Figure 16 refers to a regenerant dosage of 11-5-
pounds of sulfuric acid. Figure 15 contains data for the use of "0%
of sulfonic acid resin". This data should not be compared directly
with the other data, because when no sulfonic acid resin is used the
product water quality is entirely different than when sulfonic acid
resin is used.
87
-------
Table 35. Typical Effluent Quality From System
(Analysis in mg/1, as CaC03, of composite sample.)
Test Series A B C D
Acidity
Chloride
Sulfate
Total hardness
Ammonia-N
Sodium
Potassium
TOC-C
130
140
88
1.3
3.7
106
7.3
12
140
150
91
1.1
4.0
100
7.4
9.8
150
160
94
0.6
2.1
96
6.8
11
140
160
89
0.4
3.7
91
5.8
14.0
Test Series E F G H
Acidity
Chloride
Sulfate
Total hardness
Ammonia-N
Sodium
Potassium
TOC-C
160
170
90
0.4
3.2
98.0
6.8
8.8
165
170
91
0.3
3.4
92.0
5.9
11.0
180
170
93
0.3
2.6
91.0
5.4
11.0
175
165
91
0.1
0.9
79.0
4.5
12.0
180
170
93
0.4
1.9
53.0
2.9
10.0
-------
Table 36. CATION RESINS SUMMARY
Requirements to treat filtered treated sewage
Test Series A B C D
Resin, cu ft:
IRC-84 1 il ]£ !i
IRC-122 2& 2£ 2-g- 2£
Capacity:
Gallons
Kilograins
Acid Regenerant:
Lb 66° Be
Kgr equivalent
% utilization
Required for 1000 gal:
Acid, Ib
Rinse, gal.
Resin, cu ft:
IRC-84
IRC -122
Total
1407
39.3
7.5
50.0
79
5.4
54
0.71
1.78
2.49
1602
44.9
7.5
50.0
90
4.7
47
0.94
1.55
2.49
1578
46.4
8.3
55.4
84
5.3
48
0.95
1.59
2.54
1698
51.2
9.0
60.0
85
5.3
44
0.89
1.47
2.36
Test Series E F G H I
Resin, cu ft:
IRC-84 l£ l£ 2 2 2
IRC-122 2£ 2£ 2i 2£ 2%
Capacity:
Gallons 1667 1980 2056 2354 3323
Kilograins 52.5 63.5 65.0 74.4 102.7
Acid Regenerant:
Lb 66° Be 10.0 11.25 11.25 12.5 17.5
Kgr equivalent 66.7 75.0 75.0 83.3 117.0
% utilization 79 85 86 89 88
Required for 1000 gal.
Acid, Ib 6.0 5.7 5.5 5.3 5.3
Rinse, gal. 45 38 46 40 29
Resin, cu ft:
IRC-84 0.90 0.76 0.97 0.85 0.60
IRC-122 1.50 1.26 1.22 1.06 0.75
Total 2.40 2.02 2.19 1.91 1.35
89
-------
Table 37. Costs to produce 1000 gallons of cation
resin treated water. Using 2-g- cu ft of
sulfonic acid resin with varying amounts
of weak acid resin and regenerant.
Regenerant P^SO^, Ibss
per cu ft 122
Total
Resin, cu ft:
IRC-84
Total
% IRC- 122
Needed for 1000 gal:
H2S04, Ib
IRC-84, cu ft
IR-122, cu ft
H2S04, $
IRC-84, $
IRC- 122, $
Resin total, $
Regenerant H2S04, Ibs:
per cu ft 122
Total
Resin, cu ft:
IRC-84
Total
% IRC- 122
Needed for 1000 gal:
H2SO., Ib
IRC-84, cu ft
IR-122, cu ft
H2S04, $
IRC-84, $
IRC- 122, $
Resin total, $
3
7k
0
2.5
100
10.92
--
3.64
17.5
—
83.7
83.7
4-i-
11*
0
2.5
100
16.9
—
3.38
27.0
—
77.7
77.2
3
7k
1
3.5
71.5
5.4
0.71
1.78
8.6
35.9
40.9
76.8
1*
4
62.5
5.7
0.76
1.26
9.1
38.4
29.0
67.4
3
7-g-
1-g-
4
62.5
4.7
0.94
1.52
7.5
47.5
35.0
82.5
4-i-
ll|
2
4"2
55.5
5.5
0.97
1.22
8.8
49.0
28.1
77.1
3
7*
2.2
2.2
0
3.9
1.15
...
6.3
58.1
—
58.1
7
17*
2
4-i-
55.5
5.3
0.60
0.75
8.5
30.2
17.3
47.5
90
-------
Figures 15 and 16 show costs both for the regenerant acid and for the
unamortized capital cost of the resin. The acid cost diminishes continu-
ally to a minimum value as the percent of sulfonic acid resin decreases.
This is because as the amount of weak acid resin increases, greater
utilization of the acid is achieved.
Contrary to decreased cost for regenerant acid with increased amounts
of weak acid resin being used, the unamortized resin cost reaches a
minimum value beyond which it increases as more weak acid resin is used.
This is logical in view of the higher cost for the weak acid resin be-
cause as small amounts of the weak acid resin are added to the system
"free capacity" to treat water is obtained. After sufficient weak acid
resin is added to the system to deplete the excess acid, additional
capacity is not obtained—the weak acid resin addition is merely wasted.
From Figures 15 and 16, it is apparent that the combined cation resin
system should contain the strong acid resin in an amount equal to about
60-80% of the total cation resin. Some slight increase in the un-
amortized resin costs may be justifiable to achieve lower operating
cost for the acid. Unfortunately, the duration of this project has been
insufficient to more clearly define this area in the cost curves. Such
delineation to establish optimum costs would be of interest. The final
choice of the optimum ratio of the two resins can be made only after an
evaluation of plant design, where total operating cost can be evaluated.
Such plant design and evaluation is beyond the scope of this project.
However, the choice was made to use 62-?r% of strong acid resin in the
combined cation resin bed for additional studies under this project.
These studies were made to compare operating characteristics for the two
cation resin system operating in a combined bed with the two resins in
separate beds.
The combined resin bed was made with 1-ijr cubic feet (18" bed height) of
weak acid resin on top of 2g cubic feet (30" bed height) of strong acid
resin in one vessel. Conditions used were the same as those used for
the previous test series B with 1-g- cubic feet of weak acid resin and 2g-
cubic feet of strong acid resin in separate vessels, but regenerated in
series with 7^ pounds total of 66° Baume sulfuric acid.
Table 38, "Performance of Two Cation Resins in One Vessel", summarizes
the test runs for the combined bed of cation resins. The loading shown
in ppm is the difference in the cation content of the composite samples
influent and effluent from the resin bed. The exhaustion was controlled
by conductivity measurement of the effluent, as was previously discussed.
The endpoint was selected when the effluent conductivity decreased 500
microhms from the stable value during service. The average performance
was then used to calculate the requirements of acid and resin to treat
1000 gallons of water by the combined bed. This calculation resulted
in Table 39, "Comparison of Performance", which compares the require-
ments to treat 1000 gallons of filtered sewage effluent with the two
91
-------
100
90
80
Resin
cost, $
70
60
50
a
A
Resin cost
A
A r. i H r n s t.
A
\
20 40 60 80
Amount of sulfonic acid resin used, percent
100
Figure 15. Costs to produce 1000 gallons of water treated by
cation exchange resins. Acid costs are operating
costs; resin costs are unamortized capital costs.
Based on using 2Jjr cu ft of sulfonic acid resin re-
generated with ?£ Ibs of 66° Be H2SQ4.
21
18
.15
Acid
cost,
*
12
92
-------
100
90
30
Resin
cost,
70
60
50
/
/
Resi
/
Q
i cost /* \
\
cr
\ /
^/
/*_
Q
\ 1
I
/
r
Acid cost
\ 1
")
)
\
21
18
15
Acid
cost,
20 40 60 80
Amount of sulfonic acid resin used, percent
100
Figure 16. Costs to produce 1000 gallons of water treated by
cation exchange resins. Acid costs are operating
costs; resin costs are unamortized capital costs.
Based on using 2-g- cu ft of sulfonic acid resin re-
generated with lli Ibs of 66° Be H2S04.
93
-------
Table 38.
Performance of two cation resins in one vessel. Counter-
current flows: downflow exhaustion, upflow regeneration.
1-jjr cu ft weak acid resin on 2§- cu ft strong acid resin.
Regenerated with 7|- Ibs total of 66° Be H2S04 as 2% so-
lution at 2 gpra total flow.
Run Loading Capacity
No^ mq/1 Gal.
1-S 498 1245 36.2
2-S 509 1107 33.0
3-S 495 1146 33.2
4-S 535 1279 40.0
5-S 520 1384 42.2
6-S* 555* 1388* 45.2*
7-S* 572* 1278* 42.8*
8-S Mechanical Difficulty
9-S* 572* 1205* 40.3*
10-3* 576* 1344* 45.2*
Average 569 1304 43.4
*Data used for averages
94
-------
Table 39. COMPARISON OF PERFORMANCE
Requirements to treat filtered treated sewage with separate
beds or layered beds of cation resin.
Test Series, Beds Separate Layered
Resin, cu ft: 1-i- li-
IRC-84 2L 2|-
IRC-122
Capacity:
Gallons 1602 1304
Kilograins 44.9 43.4
Acid Regenerant:
Lb 66° Be 7.5 7.5
Kgr equivalent 50.0 50.0
% utilization 90 87
Required for 1000 gal:
Acid, Ib 4.7 5.8
Rinse, gal. 47 57
Resin, cu ft:
IRC-84 0.94 1.15
IRC-122 1.55 1.91
Total 2.49 3.06
Costs for 1000 gal:
H2S04, $ 7.5 9.3
IRC-84, $ 47.5 58.2
IRC-122, $ 35.0 44.0
Resin total, $ 82.5 102.2
95
-------
cation resins either in a separate vessel or layered into a common
vessel.
Table 39 shows that the operating acid cost as well as capital resin
cost are greater when the two cation resins are contained as layered
beds in one vessel rather than when the resins are maintained in separate
vessels. To treate 1000 gallons of water, 9.3$ of acid are required when
layered beds are used as compared with 7.5
-------
SECTION 12
WASTE REGENERANT DISPOSITION
Ion exchange resins require periodic regeneration when they become
exhausted. Chemicals which are either strongly acidic or strongly
alkaline are required for regeneration of cation resins or anion ex-
change resins respectively. Effective regeneration (to remove the
exchanged ions from the resins) requires the use of regenerant in excess
of the stoichiometric quantity. A large excess is required for the
strong electrolyte resins, while only a small excess is needed for the
weak electrolyte exchange resins.
The design of equipment to control the ion exchange processes should
consider fractionation of the total effluent during regeneration and
rinse. This can greatly reduce the effluent volume to be treated.
Normal regeneration procedures will permit the fractionation into two
portions. One of these contains most of the eluted ions and excess
regenerant in a comparatively high concentration. The second portion
contains the diluted rinse-out of the regenerant. This latter portion
is relatively innocuous, so that disposal should be of minimal concern.
If possible, reuse of the excess waste regenerant would be very desirable
for two reasons. First, reduction of disposal problems would result.
Second, reduction in operating chemical costs would result.
Regenerant acid neutralization.
Acid waste from the regeneration of cation exchange resin is easily
neutralized; however, precipitated sludges and neutral brines must be
disposed of through locally acceptable means.
Waste regenerant acid is present in the greatest concentration and
amount from the regeneration of strong acid cation exchange resin; on
the other hand, a lesser concentration and amount of acid are present
from the regeneration of weak acid cation exchange resin. The best
use of cation exchange resin for wastewater demineralization requires
the use of both strong and weak acid cation exchange resins. As
discussed earlier in this report, the excess acid from the regeneration
of the strong acid resin will effectively regenerate the weak acid cation
exchange resin. As a consequence, the waste regenerant will require
comparatively small additions of alkaline agents for neutralization.
Waste acid neutralization with alkaline agents was studied in the
laboratory. Hydrated lime (calcium hydroxide) and limestone (calcium
carbonate) were used in separate tests to neutralize free mineral
acidity (FMA). Lime was more effective than limestone.
97
-------
Conclusions were reached that lime is efficient, while a larger excess
of limestone is required for neutralization. Hydrated lime dosages can
be adjusted to raise the pH for complete elimination of acid, while
limestone produces a maximum pH of slightly over 6. Hydrated lime can
be used for neutralization of more concentrated acid, while limestone
can only be used on dilute acids with a maximum FMA of about 3000 mg/1
as CaCOo. Economic operation requires controlled addition of hydrated
lime, while limestone additions are "self-adjusting" because of its
lower solubility.
Static beaker tests were separately made for acid neutralization with
hydrated lime and limestone. These were run with a measured amount of
analyzed excess acid being stirred in a laboratory beaker while the
desired dosage of alkaline agent was added. The results can be quickly
compared by inspection of the two graphs, Figures 17 and 18 which show
the results respectively for hydrated lime and limestone. The graphs
show that neutralization with hydrated lime is essentially complete in
10 minutes, while limestone requires about 60 minutes. The graphs show
that 100% stoichiometric dosages of hydrated lime and limestone were
able to raise the pH to about 2.5 and 3.5 respectively. This paradox
is due to experimental error. A stoichiometric addition (quantitatively
equivalent) should produce an equilibrium pH vlue which is higher than
we obtained. The dosage was determined from analysis of the waste acid,
and the purity of the alkaline agents. Experimental error present in
these procedures resulted in a neutralizing dose that was less than
stoichiometric.
Undiluted waste acid was used for the test with hydrated lime. Contrari-
wise, preliminary tests with limestone showed that prior dilution of
waste acid was necessary for neutralization. Chemical analyses of the
neutralization test solutions (before, and after addition of the required
dosage) are shown in Tables 40 and 41 for lime and limestone respectively.
The data supports the above conclusions.
The characteristics of the lime and the limestone used for these tests
are shown in Tables 42 and 43. These tables contain information furnished
by the supplier of the material.
Dynamic neutralization of the waste acid with limestone in a column was
difficult. Calcium sulfate precipitation readily occured. Such precipi-
tation can cause complete solidification of the bed into a non-reactive
mass. Upflow passage of the waste acid through the limestone bed was
mandatory to prevent in-place precipitation of calcium sulfate. The
upflow passage of acid must be of sufficient velocity to expand the
limestone bed about 50%. Such expansion provides two benefits. First,
precipitates are not retained in the bed. Second, limestone particles
are abraded by tumbling action so that reactive surfaces are continu-
ally exposed.
98
-------
10
20 30
Time in Minutes
50
Figure 17.
Static neutralization of waste acid
(FMA 18900 mg/1) with 100, 105 & 110%
stoichiometric lime dosage at 72° F.
99
-------
20
40 60
Time in Minutes
80
100
Figure 18. Static neutralization of diluted waste
acid (FMA 3210 mg/l) with 100, 120 and
150% stoichiometric limestone dosages at
72° F.
100
-------
Table 40
Analyses of Static Lime - Neutralized Acid Regenerant.
Results expressed as grams per liter as calcium carbonate
equivalents, except for pH units and sludge percent.
Stoichiometric Dosages of Lime
pH
Chloride
Free mineral acidity
Sulfate
Magnesium
Calcium
Sodium & Potassium
Total Alkalinity
Sludge, % by:
Weight
Volume
0
1.3
0.06
18.90
35.00
8.00
1.34
6.83
0.13
...
100%'
2.5
0.06
0.70
16.00
7.40
1.00
6.83
2.9
9.0
105%
8.1
0.05
15.50
7.50
1.09
7.00
0.10
6.7
11.8
110%
9.3
0.05
13.50
6.40
0.97
6.60
0.38
7.2
14.2
101
-------
Table 41
Analyses of Static Limestone - Neutralized Acid Regenerant.
Results expressed as grams per liter as calcium carbonate
equivalents, except for pH units and sludge percent.
Stoichiometric Dosages of Limestone
PH
Chloride
Free mineral acidity
Sulfate
Magnesium
Calcium
Sodium & Potassium
Total alkalinity
Sludge, % by;
Weight
Volume
0
1.6
0.01
3.21
6.00
1.28
0.23
1.01
100%
3.9
0.01
0.01
4.00
1.35
1.35
1.12
0.32
2.4
120%
6.0
0.02
3.70
1.34
1.18
1.15
0.01
0.32
2.4
150%
6.6
0.02
3.80
1.38
1.22
1.11
0.02
0.31
1.6
102
-------
Table 42. Characteristics of Lime* Used
Calcium Hydroxide 97.
Insoluble in Hydrochloric Acid
Chloride (Cl) 0.005%
Sulfate (S04) 0.05%
Iron (Fe) 0.03%
Magnesium and Alkalies 1.00
(as sulfate)
Heavy metals (as Pb) 0.003
TOTAL: 98.898
J.I. Baker #1372 (supplier analysis)
103
-------
Table 43. Characteristics of Limestone Used
(For Static Test)
Typical Chemical Analysis
Total Carbonates (Ca, Mg) 95.0%
MgCOg 3.0%
A1203 0.1% to 0.25%
Fe203 0.08 to 0.
Si02 0.3% to 0.<
Mn Trace
Typical Physical Constants
Hardness (Mobs' Scale) 3.0
Specific gravity 2.71
Screen Analysis
Retained on U.S. Screen No. 16 5%
Passing U.S. Screen No. 60 10%
Acid Insolubles, by weight 1.5%
Note: All the above characteristics are issued by Georgia Marble
Company (Supplier).
104
-------
In addition to upflow passage to remove precipitates during limestone
neutralization, such precipitates must be prevented by prior dilution
of the waste acid. Figure 19 presents the results obtained from
dynamic tests with limestone, made with four different dilutions of the
waste acid. The diluted acid solutions were analyzed; the results
appear in Table 44. As shown in Figure 19, the waste acid must be
diluted to avoid neutralization slow-down.
Requirements for neutralization of waste acid are summarized as
follows:
1. Use of both weak-, and strong electrolyte cation exchange
resins will reduce excess acid to a minimum.
2. Regeneration controls should fractionate the effluent to reduce
the volume to be treated.
3. Undiluted acid can be neutralized with hydrated lime. A dose
equal to 105% of the stoichiometric requirement should be
used for neutralization to a pH above 7. This should be by
controlled feed of lime.
4. Dilution of waste acid to a maximum acid concentration of about
3000 mg/1 (as CaCCu) is necessary to neutralization with lime-
stone. This neutralization should direct the acid upflow
through an expanded bed of limestone.
5. Neutralization with lime or limestone yields a neutral
saline solution and sludge, both of which require disposition
by locally approved methods.
Regenerant ammonia recovery.
Waste regenerant ammonia recovery was studied in the laboratory.
Regeneration of weak base anion exchange resins can be accomplished
with 4% solutions of ammonium hydroxide. The effluent during regenera-
tion will then be a solution of excess ammonium hydroxide and its
various salts (chloride, sulfate, etc.)- The concentration of the
effluent solution will approach 4%; great dilution by rinse water
can be avoided by employing fractionating techniques during regenera-
tion. These techniques will simply direct the dilute rinse water
separate from the more concentrated effluent.
The ammonium salts can be converted to ammonium hydroxide by the
addition of caustic materials which will supply the necessary hydroxyl
ions. The resultant ammonium hydroxide will disassociate according to
the following equation to form ammonia gas and water.
NH40H = NH3 + H20
The ammonia gas is easily volatilized from the solution.
105
-------
the mixing and precipitating zones; flow being then reversed upwardly
to filter through the suspended sludge blanket. The nucleation pro-
vided in the sludge blanket resulted in greater clarity through the
formation of larger particles which did not rise through the upflow
zone of the clarifier. At the designed 15 gallons per minute flow,
the rise rate was 0.57 gallons per square feet per minute (.or 4.5
ft/hour rise) at the overflow weir.
Gaseous carbonation for pH reduction was provided immediately at the
effluent from the clarifier with lime clarification, this pH adjustment
is necessary to prevent precipitation of scale-forming minerals in the
equipment. A retention,tank was provided to collect the carbonated
water. This tank contained a baffle so that precipitated calcium car-
bonate (resulting from interaction of lime and carbon dioxide) would
settle. The discharge from the baffle to the main retention zone was
continually monitored by means of a recording pH meter to insure that
the desired recarbonated pH value was obtained. The retention tank
was a rectangular vessel holding sufficient clarified water for four
hours of ion exchange system testing.
A transfer pump delivered the water from the retention tank through a
totalizing water meter to the filters. This first filter was a dual
media filter, 30" in diameter with a 60" side sheet. This filter con-
tained anthracite with a effective size of 0.8 mm to a depth of 24", on
top of sand with an effective size of 0.5 mm to a depth of 8".
Suitable valving was provided to permit bypass or backwash of this
filter or the following carbon filter if desired. Flow indicators to
verify rates were included in the plumbing.
A granular activated carbon filter was provided for adsorption of or-
ganics from the wastewater. The carbon filter was cylindrical, 54"
in diameter with a 78" side sheet. It contained 80 cu ft of Pittsburgh's
Filtrasorb 400 activated carbon. This was sufficient carbon to provide
a contact time of 40 minutes through the empty column.
The effluent from the pretreatment system was then delivered to the ion
exchange system. The pretreatment system was capable of delivering 15
gallons per minute.
The pilot plant was supplied with potable municipal water for various
needs. An ion exchange water softener was provided so that water free
of hardness was available for preparation of alkaline regenerant solu-
tions.
106
-------
Table 44. Analyses of Acid Waste Before Neutralization
with 36" Limestone Bed
(Results in mg/1 as CaCO , except pH units.)
O
Curve*
PH
Chloride
Free Mineral
Acidity
Sulfate
Magnesium
Calcium
Sodium &
Potassium
A.
1.45
51
4150
4570
135
230
159
B.
1.6
38
3080
6080
1665
1200
249
C.
1.6
51
3080
3550
135
230
159
D.
2.1
51
1280
1725
135
230
159
* See Graph #3
107
-------
Hydrated lime (calcium hydroxide) is an inexpensive, readily available
chemical which will supply the necessary hydroxyl ions for the above
reaction. The requirements for hydrated lime to liberate the ammonia
were studied under varying conditions.
The following four conditions were investigated for vacuum stripping.
1. Lime dosage, equivalent to the ammonia; 90%, 100%, and 110%
of stoichiometric.
2. Temperature of the waste ammonia; 150 and 175° F-
3. Vacuum; 2.5, 10, and 20 inches of mercury vacuum.
4. Flow rate of air sweep through the system: 0.4 and 0.9 cubic
feet per minute.
The waste ammonia was heated to the desired temperature, treated with the
desired lime dosage, subjected to the desired vacuum and air sweep
conditions. The stripped ammonia was recovered in fractions in standard
solutions of acid. The recovery vessel contained methyl orange indicator
for in-place titration of the ammonia. The time required to strip
sufficient ammonia to neutralize the acid in the receiving flask was
determined. This procedures does not permit re-use of the ammonia* The
procedure was used only to study the requirements to strip the ammonia.
The effect of lime dosage on ammonia stripping is shown in Figure 20.
As shown, a stoichiometric dosage of lime was adequate to recover 90%
of the ammonia within 30 minutes. A slight excess of lime (10% maximum)
would provide a desirable excess to insure maximum recovery.
The effect of temperature on ammonia stripping is shown in Figure 21.
Elevated temperatures increased the rate of ammonia recovery. At 175°F,
recovery was obtained in about 30 minutes.
The effect of vacuum on ammonia stripping is shown in Figure 22. At
20 inches of mercury vacuum, 90% ammonia recovery was obtained in 30
minutes. Less vacuum drastically increased the required time.
The effect of air sweep on ammonia stripping is shown in Figure 23.
Increasing the air flow rate increased the rate of ammonia recovery.
The data does not show a great difference in the effect produced by the
two flow rates used. The lower flow rate studied displaced in about
three seconds a volume of air equal to the waste regenerant being
tested. In spite of the high volume ratio for the air sweep, it was
readily obtained by the vacuum source.
These vacuum stripping studies indicate that the best conditions for
ammonia recovery with vacuum will include a lime dosage of 110% (of the
108
-------
Conditions: 20 inches Mercury Vacuum
175° F 0.4 CFM Air Sweep
100
Ammonia
Recovery
(*)
15
30 45
Time in Minutes
75
Figure 20. Effect of Lime Dosage on Ammonia Recovery
109
-------
90
60
Ammonia
Recovery
30
15
30
60 90
Time in Minutes
120
150
Figure 21.
Effect of Temperature on Ammonia Recovery.
Conditions are 20 inches Mercury Vacuum, 0.4
CFM Air Sweep, 100% Stoichiometric Lime Dosage.
110
-------
Conditions: 100% Stoichiometric Lime Dosage
175° F, Air Sweep 0.4 CFM
Ammonia
Recovery
(*)
10 inche 5 Mercury
15
60 90
Time in Minutes
Figure 22. Effect of Vacuum on Ammonia Recovery
111
-------
100
90
Ammonia
Recovery
50
40
Conditions; 175° F 20 inches Mercury Vacuum,
100% Stoichiometric Lime Dosage
10
20 30
Time in Minutes
40
50
Figure 23. Effect of Air Sweep on Ammonia Recovery
112
-------
stoichiometric requirement), a temperature of 175° F, 20 inches of
mercury vacuum, with an air sweep. It is apparent that these
conditions will be more easily obtained through direct distillation
procedures.
Regenerant ammonia recovery studies included distillation. A measured
quantity of waste ammonia was treated with the desired lime dosage,
with steam introduced into the mixture. Ammonia and water were distilled
and surface condensed into a receiving flask which contained a measured
quantity of acid and methyl orange indicator. The time required to
strip sufficient ammonia to neutralize the acid in the receiving flask
was determined.
Two dosages of lime (100% and 120% of the stoichiometric requirement)
were investigated. No significant difference was found in the effect
produced by the two lime dosages. The observed data is shown in
Figure 24. As shown, about 40 minutes were needed to obtain about
90% recovery of the regenerant ammonia with either lime dosage. The
steam requirement to distill this ammonia was approximately 1/3 the
volume of the waste ammonia solution.
It is apparent that waste ammonia is best recovered with distillation
techniques. Such techniques are readily available from suppliers of
distillation equipment. The recommendation of such equipment is
beyond the scope of this project. The requirements for hydrated lime
to liberate the ammonia are a necessary part of this project in order
to predict the operating chemical cost. These costs are outlined in
Section 13.
113
-------
100
80
60
Ammonia
Recovery
40
20
I
120%
Stoichiometric Lime
Stoichiometric Lime
0
10
20 30
Time in Minutes
40
50
Figure 24. Ammonia Recovery by Steam Distillation.
114
-------
SECTION 13
OPERATING MATERIAL REQUIREMENTS FOR ION EXCHANGE PROCESSES
Previous sections have discussed the operating performance of each ion
exchange resin tested. It is the purpose of this section to list the
chemical requirements to produce 1,000 gallons of treated water. Costs
for equipment, labor, building, utilities, land, and other similar
costs are beyond the scope of this project. The costs presented are
predicated on application of the systems to treated sewage effluent
of the composition which has been discussed in this report, as sum-
marized in Table 1.
Product water costs may be varied by establishing quality standards.
Better quality will increase the cost. Poorer quality should reduce
the cost by blending partially treated water with the product.
Ion exchange resins must be backwashed with water before each regenera-
tion for two reasons. First, to loosen and hydraulically classify the
resin to maintain low pressure losses in the system. Second, to remove
insoluble material which has been removed by filtration on the resin
bed. The water used for this backwash should be clear: it can be
the normal influent water to the ion exchange system. After use then,
this water and its turbidity can be returned upstream in the process
for clarification and reuse. The backwash water, then, is not a
significant cost factor. Costs presented in this section are based
on the following chemical and resin prices.
Chemical or Resin $/Ton
-------
Table 45. Chemical requirements for strong acid cation and
weak base anion exchange system. To produce
1000 gallons of treated water.
Cation exchange resin (IRC-120) component:
Regeneration level, Ib/cu ft 135
Resin, cu ft 5.56 3.13 2.38
Acid, 66° Be sulfuric:
lb 5.6 9.4 11.9
-------
Data presented in earlier sections about each resin has been
abstracted into this table. Acid neutralization for the cation
resin regenerant was based on the amount of unused acid and the
optimum excess dosage of neutralizing agent. The total chemical
costs of regenerant acid and lime neutralization ranges from
9.48 to 23.8 cents per 1,000 gallons of treated water.
Similarily shown for weak base resins are the cost for ammonia
(without reclaim) and the cost of lime to release the ammonia
for recovery based on a lime dosage of 105% of the stoichiometric
equivalent. Thus, it is shown in the table, for IRA-93 resin,
the cost to produce 1,000 gallons of treated water ranges from
4.3 to 5.7 cents for ammonia without its recovery; or, 1.6 to
1.4 cents espectively for lime to recover the ammonia.
Weak acid cation exchange resin system. (One bed system.)
This system uses only one resin for demineralization to accomplish
a partial reduction of ionic contamination approximately equal to
the alkalinity of the influent sewage. The filtered sewage present
at Elgin, Illinois and total ionized inorganic salts at a concentration
of about 600 mg/1. The weak acid cation exchange resin separately was
able to reduce this to about 250 mg/1. Post treatment is required,
primarily for reduction of carbon dioxide, ammonia, and phosphate
levels.
Table 46 lists the chemical requirements to produce 1000 gallons of
treated water from only'this one resin. The total chemical cost
(acid plus lime for neutralization) is 6.7
-------
Table 46. Chemical requirements for weak acid cation
exchange system. To produce 1000 gallons
of treated water.
Cation exchange resin (IRC-84) component:
Regeneration level, Ib/cu ft 3.9
Resin, cu ft 1.00
Acid, 66° Be sulfurics
Ib 3.9
$ (at $32/ton) 6.2
% utilization 84
Water, gallons fors
Regenerant solution 67
Rinse-out 54
Waste acid (fractionated), gal. 80
Acid neutralization with:
lime hydrated, 105% stoichiometric:
Ib 0.50
$ (at $18/ton) 0.45
limestone, 120% stoichiometric:
Ib 0.77
$ (at $6/ton) 0.23
Total cost, acid and lime, <(; 6.7
Weak acid - strong acid cation resin and weak base anion resins system.
(Three bed system).
This system uses three ion exchange resins for demineralization. The
system is capable of greatly reducing the concentration of all ionic
contamination; however, economical operation requires continued operation
until the concentration of an inorganic constituent reaches an undesirable
level in the product water.
Estimated costs for this system are subject to greater variation than
are the previously estimated costs. This is because the costs for the
present system depend gravely upon the ratio of the amounts of the two
cation resins which are used. This is because as the amount of weak
acid resin increases, operating costs are decreased. Contrariwise, as
the amount of weak acid resin increases, the unamortized resin cost
reaches a minimum value beyond which it increases as more weak acid
resin is used. It may be economically desirable to use more weak acid
resin beyond this "minimum resin cost" point to achieve maximum reduc-
tion in operating costs. Selection of the optimum ratio of cation
resins is beyond the scope of this project and should be clarified during
detailed pilot plant studies on this system.
118
-------
Table 47 shows the cost to produce 1000 gallons of demineralized water
with this three bed system. For this estimate, we have used data
collected for cation resin used in the ratio of 1-g- cubic feet of
weak acid resin to 2% cubic feet of strong acid resin. A regeneration
dosage of three pounds of acid per cubic foot of strong acid resin used.
Unamortized resin costs were based on the resin costs established by the
manufacturer for industrial users of these resins. Table 47 shows that
the operating costs vary from 11.8
-------
Table 47. Costs to produce 1000 gallons of
demineralized water with three
resins.
Weak acid cation resin:
Resin, cu ft 0.94
Regenerant: reuse of that
from strong acid resin
Water, gallons for:
Regenerant dilution 81.0
Strong acid cation resin:
Resin, cu ft 1.52
Acid, 66° Be sulfuric:
Ibs 4.7
$ 7.5
Water, gallons for:
Regenerant solution 28
Regenerant rinse 47.0
Weak baseanion resin:
Regeneration recovery Without With
Resin, cu ft 0.75 0.66
Regenerant Ammonia, 100%, Ibs 1.29 1.72
Regenerant Ammonia, 100%, $ 4.3 5.7
Lime, Ibs 0 1.6
Lime, $ 0 1.5
Water, gallons for:
Regenerant solution 8 11
Regenerant rinse 65 75
Operating costs, t/lOOO gal:
Without ammonia recovery 11.8
With ammonia recovery 9.0
Unamortized resin costs:
Without ammonia recovery:
Cation resin, weak, $ (@ $50.5) 47.5
Cation resin, strong, $ (@ $23.0) 35.0
Anion resin, weak, $ (@ $70.5) 53._0
Total, $ 135."5
With ammonia recovery:
Cation resin, weak, $ (@ $50.5) 47.5
Cation resin, strong, $ ( $23.0) 35.0
Anion resin, weak, $ (@ $70.5) 46._5
Total, $ 129.0
120
-------
Table 48. Comparison of systems to produce 1000 gallons of water
by ion exchange.
System
2 Resin A
2 Resin B
1 Resin
3 Resin
Resin
Type
Strong Cat.
(IRC- 120)
Weak An.
(IRA-93)
Strong Cat.
(IRC- 120)
Weak An.
(IRA- 68)
Weak Cat.
(IRC-84)
Weak Cat.
(IRC-84)
Strong Cat.
(IRC- 120)
Weak An.
(IRA-93)
Resin
cu ft
3.13
0.75
With
3.13
0.66
With
1.0
0.94
1.52
0.75
With
Regen't
Ib/cu ft
3.0
1.72
Rinse
gal.
110
65
Total
ammonia recovery
3.0
1.72
110
34
Total
ammonia recovery
3.9
Re-used
3.00
1.72
ammonia re<
54
—
47
65
Total
;overy
Regen' t
cost, $
0.166*
0.043
0.209
0.184
0.166*
0.037
0.203
0.187
0.064*
Nil
0.075
0.043
0.118
0.093
Unamort' d
Resin, $
71.990
52.875
124.865
124.865
71.990
56.430
128.420
128.420
50.500
47.500
35.000
53.000
135.500
129.000
*Includes cost of limestone to neutralize excess acid.
121
-------
SECTION 14
ACKNOWLEDGEMENTS
The authors, Ed Kreusch and Ken Schmidt, gratefully acknowledge the
varied assistance received from many sources in the completion of
this project. Financial support was received from the Water Quality
Office, Environmental Protection Agency; previously the. Federal
Water Quality Administration, Department of the Interior at the
start of the project. The guidance of the Project Officer, Mr.
Richard Dobbs and his superior Mr. Jesse Cohen of the Robert A. Taft
Water Research Center in Cincinnati, Ohio was a firm foundation for
the project's inception.
Mr. Giles McVey, representing the Elgin Sanitary District, was
congenial in the support of this project by freely permitting the
erection and operation of the pilot plant on their property.
The services of Messrs. Cliff Skoning, Dean Schwark, and Ralph Large
in the construction of the pilot plant, equipment assembly, and
actual operation of the pilot plant were tantamount to successful
completion of the project.
The services of the various departments of the parent company,
Culligan International Company, are gratefully recognized. Their
assistance, which was beyond their responsibilities in support of
the commercial organization, embraced many areas which were beyond
the fields of specialization of the authors.
Of particular importance is the investigation of waste regenerant
disposition by Mr. Farouk Husseini. Mr. Doug Rossburg contributed
necessary support by expeditious analysis of samples.
123
-------
SECTION 15
DEFINITIONS
ALKALINITY
ANION
BACKWASH
BED DEPTH
BED EXPANSION
BED VOLUME
BREAKTHROUGH
Capacity to neutralize acids. In
water, most alkalinity is due to the
water's content of bicarbonates,
carbonates, or hydroxide. The
alkalinity is normally expressed in
terms of calcium carbonate equivalents.
An ionic particle which is negatively
charged.
Reverse (normally upwards) flow through
a bed of mineral or ion exchange resin
to remove insoluble particulates and to
loosen the bed.
The height of mineral, or ion exchange
resin in a column.
The amount of expansion given to a bed of
mineral or ion exchange resin, by upflow
passage of water. It is usually expressed
as a percent of the unexpanded bed.
The amount of mineral, or ion exchange
resin, in a column.
Refers to the concentration of a partic-
ular ion, or other substance in the
effluent from a treatment system.
Breakthrough occurs when the effluent
concentration rapidly increases. Normally,
when the breakthrough concentration
reaches about 10% of the influent concen-
tration, exhaustion has occurred.
125
-------
CALCIUM CARBONATE
EQUIVALENT
CAPACITY
CATION
COCURRENT
COMPOSITE SAMPLE
CONDUCTIVITY
COUNTERCURRENT
DEMORALIZATION
DOWNFLOW
An expression for the concentration of
constituents on a common basis for ease
of calculation. Conversion of the
quantity expressed "as calcium carbonate"
to "as another form" requires multipli-
cation by the ratio of the chemical
equivalent weight of the desired form to
that of calcium carbonate. For example,
80 mg/1 of magnesium as calcium carbonate
becomes 44.4 mg/1 (80 x 12.2/20) as
magnesium.
The quantitative ability of a treatment
component or system to perform. With ion
exchange systems, this quantity is
expressed as kilograins per cubic foot.
An ionic particle which is positively
charged.
Operation of a column of ion exchange
resin or other mineral, with the service
cycle and the regeneration cycle per-
formed in the same direction, both either
upflow or downflow.
A sample collected to be representative
of a water flow which continues for an
extended period of time.
Ability of water to conduct electricity;
it is the reciprocal of resistivity.
Conductivity is measured in reciprocal
ohms per centimeter. Water with a low
concentration of ionic solids will have
very low conductivity.
Operation of a column of ion exchange
resin or other mineral, with the service
cycle and the regeneration cycle per-
formed in opposite directions.
Reduction of the ionic content of water.
Direction of flow of solutions through
ion exchange, or mineral bed columns
during operation; in at the top and out
at the bottom of the column.
126
-------
DRIP SAMPLE
EFFLUENT
ELUATE
FLUENT
ELUTION
ENDPOINT
EXHAUSTION CYCLE
FMA
FREE MINERAL ACIDITY
9P9
GRAINS PER GALLON
GRAIN
A composite sample collected by slow
continuous sampling of a flowing stream.
The solution which emerges from a
component or system.
Effluent during regeneration of an
ion exchange resin. (See "Elution").
Influent regeneration solution to an
ion exchange resin. (See "Elution").
The removal of an adsorbed ion or ions
from an ion exchange resin during regen-
eration by using solutions containing
relatively high concentrations of other
ions. This latter solution is called the
eluant. During elution, the eluant
removes the adsorbed ions from the ion
exchange resin; the effluent solution
which contains the eluted ions is then
called the eluate.
The achievement of exhaustion. With ion
exchange resins, the endpoint of the
service cycle is at 10% breakthrough.
The function of a process component in
the service cycle. The regenerated form
of a weak base resin without adsorbed
acids.
Strong acids, which in water are formed
principly by chloride or sulfate ions
when the water has been treated by a
cation exchange resin in the hydrogen form.
A unit of concentration (weight per
volume) that is used in the ion exchange
industry. (See "GRAIN".) One gpg is
numerically equal to 17.1 mg/1.
A unit of weight, being numerically equal
to l/7000th of a pound. (See "GRAINS PER
GALLON".)
127
-------
gpm
gpm/cu ft
gpm/sq ft
HARDNESS
ION EXCHANGE RESIN
kgr
KILOGRAINS
kgr/cu ft
Gallons per minute.
Gallons per minute per cubic foot of ion
exchange resin or other mineral in a
column.
Gallons per minute per square foot of
cross-sectional area.
The sum of the calcium and magnesium ions,
although other polyvalent cations are
included at times. Hardness is normally
expressed in terms of calcium carbonate
equivalents.
An insoluble material which can remove
ions by replacing them with an equiva-
lent amount of a similarly charged ion.
A unit of weight (l,000 grains) equal to
1/7th of a pound.
Kilograins (expressed as calcium
carbonate) per cubic foot of ion exchange
resin.
LEAKAGE
LIME
MICROMHOS
mg/1
MILLIGRAMS PER LITER
NEUTRALIZATION
The amount of unadsorbed ion present in
the effluent of a treatment component.
Lime refers to compounds of calcium.
Hydrated lime is calcium hydroxide. Lime
which is not hydrated is referred to as
quick lime, which is calcium oxide.
Unit of measurement of electrical con-
ductivity.
A unit of concentration referring to the
milligrams weight of a solute per liter
of solution. The term is approximately
equal to the older "part per million"
term.
Mutual reaction of acids and alkalies
until the concentrations of hydrogen and
hydroxyl ions in solution are at the
desired value which is usually approxi-
mately equal.
128
-------
ppm
PARTS PER MILLION
REGENERANT
REGENERATION
RINSE
SALT SPLITTING
SERVICE CYCLE
SLUDGE
SLUDGE BLANKET
SOFTENING
UPFLOW
WEAK ACID RESIN
WEAK BASE RESIN
A unit of concentration, which in the
water treatment industry equals one part of
solute in one million parts by weight of
solvent. It is approximately equal to the
more precise term mg/1.
A solution of relatively high ionic con-
centration used to restore an ion exchange
resin to its desired ionic form.
Restoration of an ion exchange resin to
its desired ionic form.
The removal of excess regenerant from an
ion exchange resin.
The conversion of neutral salts to their
corresponding acids or bases.
The use of a process component to perform
its desired function.
Settled precipitates of large amount.
A layer of sludge which is suspended by
upflow passage of water.
Removal of the hardness (calcium and
magnesium ions) from water.
Direction of flow of water upwardly through
a component.
A cation exchange resin which cannot split
neutral salts.
An anion exchange resin which cannot split
neutral salts, but will merely absorb free
mineral acidity.
129
-------
SELECTED WATER i. Report No.
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
2. 3. Accession No.
w
4. Title , „ . „ .
5. Report Date
WASTEWATER DEMINERALIZATION BY ION EXCHANGE, 6'
. *. Performing Organization
7. Author(s) Report No.
Kreusch, E., and Schmidt, K.
10. Project No.
17040EEE
9. Organization
„,,._. 11. Contract/Grant No.
Culligan International Company
Northbrook, Illinois 60062
13. Type of Report and
Period Covered
12. Sponsoring Organization
15. Supplementary Notes
Final report to Water Quality Office of Environmental Protection Agency,
December, 1971.
16. Abstract
Pilot plant studies conducted on secondary treated (activated sludge process)
sewage have demonstrated the feasibility of wastewater demineralization by
ion exchange.
Lime treatment to reduce phosphate is unnecessary. Filtration through
dual media filters and activated carbon filters is desirable.
Partial demineralization of sewage containing a significant portion of
alkaline salts is simple with one bed of weak acid (carboxylic) cation
exchange resin. Complete demineralization requires at least two resins:
strong acid (sulfonic) cation exchange resin and weak base anion exchange
resin. The use of the weak acid cation exchange resin as a third resin
will reduce operating costs and waste regenerant acid.
This report was submitted in fulfillment of Project 14-12-599, sponsored
by the Water Quality Office.
17a. Descriptors
*Ion exchange, ^Tertiary Treatment, ^Demineralization, Acid neutralization,
Activated Sludge, Activated carbon
17b. Identifiers
Desal process, Ammonia Recovery, Elgin Illinois sewage treatment,
Lime Clarification
17c.COWRR Field & Group 03A, 05D, 05G
18. Availability 19. Security Class.
(Report)
20. Security Class.
(Page)
21. No. of Send To:
Pages
,,,
&**•
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
WASHINGTON. D. C. 20240
Abstractor Ed Kreusch \institution Culligan International Company
VRSIC 102 (REV. JUNE 197l) *U.S. GOVERNMENT PRINTING OFFICE: 1972-484-483/93 1-3
------- |