Ammonia removal from secondary effluent by selective ion exchange with clinoptilolite
by David Irvin McCready
A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE
in Civil Engineering
Montana State University
© Copyright by David Irvin McCready (1974)
Abstract:
The objective of this study was to provide information necessary for the application of .selective ion
exchange to large, scale operation in wastewater treatment. The ammonia-nitrogen removal processes
are described. The advantages and disadvantages of each process are listed. Low temperature (4°C) and
high temperature'(33°) exhaustion may be predicted by Figure 16. The clinoptilolite weight loss after
100 cycles of exhaustion and regeneration ranged from 0.31 to 0.56 percent per cycle for regenerant pH
of 11.0 to 12.5• There was 2.3 percent permanent loss' of exchange capacity after 100 cycles. Organic
fouling temporarily reduced exchange capacity by 11 percent. The least cost of regeneration occurs
with the reuse of calcium oxide regenerant. At pH 11.5 the cost is $0,030/1000 gallons with 0.24 lb
NaCl/galion. STATEMENT OF PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the
requirements for an advanced degree at Montana State University,
I agree that the Library shall make it freely available for
inspection.
I further agree that permission for extensive copying ■
of this thesis for scholarly purposes may be granted by my major
professor, or, in his absence, by the Director of Libraries.
It is
inderstood that any copying or publication on this thesis for
financial gain shall not be allowed without my written permission.
Signature^l^rV/truQ^s.nOKn/i MC
D a t e . 30 . r4-
AMMONIA REMOVAL' FROM SECONDARY EFFLUENT
BY SELECTIVE ION EXCHANGE
WITH CLINOPTILOLITE
by
■ • DAVID IRVIN MCCREADY '
A thesis submitted in partial fulfillment
of the requirements for the degree
of.
MASTER OF SCIENCE
in
Civil Engineering
Approved:
Cha i r m amyExamining Committee
1A
Head/ Major Department
Graduate Dean
■ MONTANA STATE UNIVERSITY
Bozeman, Montana
December, 1974
iii
ACKNOWLE DGHENT
. This investigation was conducted under a training grant from
the U. S . Environmental Protection Agency.administered by Montana
State University, Bozeman, Montana.
Special thanks are due Professor Robert L. Sanks. Professor
Glen L. Martin, Assistant Professor Takashi Asano, Professor
T. T. Williams, Professor A. Hayden Ferguson, and fellow graduate
students are thanked for their aid, criticism, and suggestions.
TABLE OF CONTENTS
Page
VITA ........
. . . . . . . . . . . .
ACKNOWLEDGEMENT
. . . . . . . .
id
............ ■ . .............. .
TABLE OF C O N T E N T S .................
LIST OF TABLES . .
........
. . . ...
. . .
■'iv
. .'.......... ■ . .... . .'. . . . . . . . .
viii
LIST OF F I G U R E S .............• ............... ..
. . . . .
iii
..
. . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONCLUSIONS
x
xii
. . ................. ^
xiii
Chapter
I.
INTRODUCTION . . .............
.
Objectives ...........................
Limitations.
. . . . . . . .
3
. '............ .
. 3
Symbols . . .................................. ..
II.
AMMONIA-NITROGEN REMOVAL M E T H O D S .................
Ammonia Stripping
^ .
4
. . .
5
................. ■ ........ ..
6
Biological Nitrification-Denitrification
Breakpoint Chlorination
Selective Ion Exchange
Land Disposal
.........................
.
. . . . . . . . . . . . . .
..............................
Algae Harvesting
12
. . . . . . . . . . . . . . .
Conventional Ion Exchange
. .: . . . . . . .
Reverse O s m o s i s ..........
I
. . . .............
17
23
23
25
28
32
V
Chapter
Page
Electrodialysis
................... •.................
Electro-oxidation
■Distillation
.111.
........
J . . . . . . . . . . .
Previous Studies . . . .
. . . . . . . . . . .
...................
S t a b i l i t y ................... ..
Ion Exchange Capacity
40
. . . . .
40
. . . . . ;
47
. ; ...............
. . ........
.
. . . . . . . . . .
. ; . . . . . . ........
Regeneration . . . . . . .
.......
EXPERIMENTAL EQUIPMENT AND
METHODS .
.
...............
..........
. . . .
.. . . . . .
. . .
. E q u i p m e n t ...................................... ; . .
Dual Media f i l t e r .................................
. Ion exchange reactors
Feedwater storage
Pumps
. .
Temperature control
Sample collection
. . . .
. . . . .
. . . . . . ’................
.................
. . . . . . . . . . . . . .
. . . . .... . . . .
. . . . . . .. . . .
48
49.
53
..........
Rejuvenation of Fouled Clinoptilolite
Regenerant Renovation
38
38
Structural Properties and Source ........
IV.
.
..........
SELECTIVE ION EXCHANGE LITERATURE
Exhaustion
35
.
55
57
58
61
'6.1.61
61
61
62
. . . . .
62
. . . . .
; 62
Weight measurement.
63
Analysis
63
....................................
vi
Chapter
Page .
Materials
.
Feedwater
... .. .
.. . . . . . . .
. . '. . ..... ....
. .
. . . .. .
. .
63 '
'
63 '
Synthetic wastewater . . . ...' . . ., .... .
63
Clinoptilolite
V..
.....
.
....
LONG TERM S T U D I E S ........ ..
. . ..
.Procedure
Results
. ... .
.
69.
'. .
. . . ■. .
69
;
. 69
................. ................ ..
. 71
Reduction of Ammonia Exchange Capacity During
Cyclic. Operation
. . . . . . .. . .. . . . i
.Procedure
ReSUltS
74
................................ ..
.
75
The EFFECT OF TEMPERATURE.ON EXCHANGE EFFICIENCY
80
. ' .
.
.
.
«
e
.
.
.
- Procedure . . . . . . . . . . . .
.
.
.
. . .
OPTIMIAZTION OF REGENERATION . . . . . .
.
83 ,
86
.. . .
Chemical c o s t s ............ 86
Clinoptilolite Costs
........ . . . . • . '
Regenerant Renovation Costs.
.......... ..
Regenerant Chemical Requirements .
■
..... . ...
R e s u l t s ..................... ................
VII.
.
; .
. . ...
Effect of Regenerant pH on Attrition Rate
VI..
‘ I ,
Bozeman municipal wastewater . . ........
....,.• . . .
88
.
.88
89
vii
Chapter
Page
Regenerant Mixing
andStorage Costs
Costs Without RegenerantReuse
92
. . . .............
95
. . . . . . . . . . . . .
Costs .With Regenerant Reuse, .'........ ..
.'..........
Cost of Regeneration by Computer Program
. . . . . . .
95
^
R E F E R E N C E S ................... L .......... ’..........
APPENDIX A ......................... ............................ 112
APPENDIX B ........................................ .... ..........
APPENDIX C ..................... ................... ............ 119
9
^
viii
. LIST OF TABLES
Table
Page
I « Nitrogen Concentrations in Raw Domestic Wastewater . . .
■2.
Comparison of Nitrogen Removal Processes
I
. . . . . . .
10
3.
Capital and Operating Costs for Breakpoint Chlorination.
11
4.
Capital and Operating Costs for Spray Irrigation
. . .
29
5.
Reverse Osmosis Nitrogen Removal Efficiencies
. . .
33
6 . Previous Investigations of A m m o n i a ..................
7.
. ' .42
Method for Determination of the Ammonia Exchange
Capacity
. . . . . . . . . . . . . . . . . . ... . .
8 . Analytical Methods
............ '................ .
50
64
9.
Average Characteristics of Bozeman Secondary Effluent
.'
66
10.
Typical Characteristics of the Synthetic Wastewater
Produced from Bozeman Tapwater
...................
.
68
11.
12.
13.
14.
Operating Characteristics for the Long Term Attrition
S t u d y ...................... ............. ..
70
Effect of Long Term Cyclic Exhaustion on Ammonia
Exchange C a p a c i t y ......... ...........................
77
Results from McLaren and Farquhar Temperature Effect
S t u d y .......................... ......................
81
Operating Characteristics for the Temperature Effects
Study . ............. .............. .................. '
.82
15.
Cost of Chemicals andClinoptilolite. . . . . . . . .
87
16.
Amount of ChemicalsRequired
91
17.
Amount.of Chemicals Required for Cyclic Calcium Oxide
Regenerant. M a k e u p ........ •.............. .
forRegeneration
.........
93
ix
Table. 18.
Page '
Amount of Chemicals Required for Cyclic Sodium
■ Hydroxide Regenerant Makeup
........... ..
19.
Capital Cost of Regenerant Storage Tank
20.
Capital Cost of Regenerant Mixing Tank
96 .
21. 'Minimum Costs of Clinoptilolite Regeneration . . ,.
22.
94 :
Capital Cost for Brine Disposal by Evaporation -Pond
■ .
.
97
.
100 •
. .
101 .
.
X
LIST OF FIGURES ■
Figure
Page
I.
The Effects of pH and Temperature on the Distribution
of Ammonia and Ammonium Ion in Water . . . . . . . . .
7
'2.
Ammonia S t r i p p i n g ........ .......... ,..................
9
3.
Biological Nitrification and Denitrification. . . . . .
'
1‘
16
I
4.
5.
A Residual Chlorine Curve Showing a Typical
B r e a k p o i n t .................... ......... ............ ..
19
Carbon Adsorption (40 Minutes Detention)
22
6 . Mixed Media or Sand Filtration . . . .
7.
. ... . . ...
...
26
Microstraining ............
8 . Reverse. Osmosis
9.
. ............
31
.........................................
34
E l e c t r o d i a l y s i s ............... ... .......... ..
37
'
10.
Variation of Ammonia Exchange Capacity with Competing
Cation Concentration...................... 52
11.
Relative Effect, of pH onAmmonia Exchange Capacity . . .
54
12.
Exchange Capacity Versus
56
13.
Schematic Illustration for Column Exhaustion with
Bozeman Wastewater . . . . . . ............... ; . . .
14.
15.
16.,
17.
pH
......................
Comparison of Total Attrition Rates with Regenerant
Solutions of pH 7.6, 11.0, 11.5, 12.0 and 12.5 . . .
The Effect of Short Term Cyclic Exhaustion on,
Ammonia Exchange Capacity
...................
. 67
.
. . . .
73
78
Breakthrough Curves for 4, 23, and 33°C Influent
Synthetic W a s t e w a t e r ..................
.
84
Volume of Regenerant Required for 95 Per .Cent
. Ammonia Elution ....................................
.
90
xi
Figure
18.
19.
Page
Cost of Calcium Oxide Regeneration Without
Regenerant R e u s e ................................... •
98
Cost of Sodium Hydroxide Regeneration Without
Regenerant R e u s e ............................... ; '.
99
20.
Cost of Calcium Oxide Regeneration With
Regenerant R e u s e ..................... '.............. 103
21.
Cost of Sodium Hydroxide Regeneration With
Regenerant R e u s e ...............■ ................... . 104
xii
ABSTRACT
' The objective of this study was to' provide information necessary
for the application of .selective ion exchange to large, scale operation
in wastewater treatment. The ammonia-nitrogen removal processes are
described. The advantages and disadvantages of each process are listed.
Low temperature (4°C) and high temperature'(33°) exhaustion may be
predicted by Figure 16. The clinoptilolite weight loss after 100 cycles
of exhaustion and regeneration ranged from 0.51 to .0 ' . 5 6 percent per
cycle for regenerant pH of 11.0 to 12.5- There was 2'.3 percent permanent
loss', of exchange capacity after 100 cycles. Organic fouling temporarily
reduced exchange capacity by 11 percent.. The least cost of regeneration
occurs with the reuse of calcium oxide regenerant. At pH 11.5 the cost
is $0,050/1000 gallons w.ith 0.24 lb NaCl/gallon.
■'
■■ ■
CONCLUSIONS "
Selective ion exchange has several advantages for. the removal of
nitrogen from wastewater including:
ature on removal efficiency,
and (3) favorable economics.
(I) little effect of low temper­
(2) almost complete removal of ammonia,
The objective, of this investigation was
to provide information necessary for using clinoptilolite for ammonia
removal in large scale operations.
The following conclusions were
made regarding long term cyclic operation of clinoptilolite ion
exchange columns:
(I)
The attrition rate for the zone of maximum weight loss
(the bottom of the exchange bed for upflow exhaustion) will range from
0.31 to 0.56 percent per cycle for regenerant pH levels of 11.0 to
12.5.
. (2)
Regeneration at pH 11.5 is the maximum caustic level for
minimum attrition, at lower pH levels regeneration is less efficient
and the attrition rate does not decrease significantly.
(3)
Exposure to 100 cycles of exhaustion with dometic waste-
water and regeneration at pH 11.0 and 12,5 resulted in 2.3 percent’
permanent loss of ammonia exchange capacity.
*
(4)
Organic fouling of clinoptilolite can cause tip to- 11-per- ,
cent temporary loss of exchange capacity in cyclic exhaustion with
filtered secondary effluent unless proper backwashing prior to
regeneration is practiced every cycle.
xiv
The effect of feedwater temperature on clinoptilolite column
exhaustion is summarized as follows:
(5)
Optimum conditions for exhaustion occur at room temper-
o
Exhaustion at temperatures ranging from 4.to. 33.C resulted in
ature.
12.5 to 16 percent change of feedwater volume treated and 0.10 to
0.14 mg/H NH
j
- N change of ammonia leakage concentrations.
'
Low and
high temperature exhaustion performance can be predicted for the
given conditions using Figure 16.
The following conclusions were made regarding the cost of
clinoptilolite regeneration.
(6)
Minimum regeneration cost results with the reuse of
calcium oxide regenerant.
At pH 11.5■the regeneration cost is
$0.030/1000 gallons.
(7)
The cost of sodium chloride, $0.012/lb, is a large pro­
portion of the regeneration costs both with and without regenerant
reuse.
•
CHAPTER I
INTRODUCTION
Historically, wastewater treatment consisted of removal of
settIeabIe solids, biodegradable organics, and bacteria.
But the in­
creasing Intensity of water reuse coupled with increased public re­
action to degradation of aquatic environments has resulted in regula­
tions for the eventual removal of nitrogen, especially ammonia.
Table I lists the range of nitrogen concentrations in raw
domestic wastewater.
TABLE I
NITROGEN CONCENTRATIONS IN RAW DOMESTIC WASTEWATER
(AFTER METCALF AND EDDY, 1972 [35])
Nitrogen type
concentration, mg/£ as N
organic
8-35
ammonia
12-50
nitrite
>0
nitrate
>0
Organic nitrogen, generally in the form of urea, is readily hydro­
lyzed to ammonia during wastewater treatment
[35].
Although there is
some nitrogen removal during conventional biological treatment, it is
not great enough to have a marked effect on the effluent quality.
Future federal effluent standards will limit the ammonia-nitro­
gen concentration in treated wastewater either directly by ammonia con­
centration standards or indirectly by Biochemical Oxygen Demand (BOD)
2
standards.
Rfeasons to limit, thfe wastewater ammonia-nitrogen '
concentration are:
(I.)
Low concentrations of ammonia are toxic to fish and. other .
aquatic life..
Concentrations as low,,as 2.5 mg/f, NH^-N may be toxic
to salmonids [35].
(2)
Ammonia corrodes p.ortland cement concrete■and metals such,
as copper and brass.■
(3)
■
. Ammonia is a plant nutrient.
If it.is the growth-limiting
substance, ammonia addition will accelerate the eutrophication of lakes
and streams.
(4)
Ammonia in natural waters is. easily oxidized to nitrites and
nitrates, thus adding to the oxygen depletion of the-receiving, water.
Oxygen depletion limits the type of aquatic life capable of survival..
Less desirable species of fish are associated with low oxygen levels
[31].
(5)
Nitrate concentrations greater than 60-150 mg/& may cause .
the disease, methemoglobinemia, which can be fatal to humans and live­
stock.
The 1962 United States Public Health Service standards set ■
40 mg/5, as the allowable nitrate concentration for domestic use.
(6)
Ammonia reacts with chlorine during, wastewater disinfection
Increasing the chlorine demand...
Ammonia and oxidized forms of nitrogen in excess restrict
the reuse and disposal .of. wastewater. '
.
.
'
3
Objectives
This study was intended to complement and complete existing
knowledge of ammonia removal.by selective ion exchange using- clinoptilolite.
The general objectives were to establish criteria under
which the most feasible method of.ammonia removal could be chosen and
to define more completely the effects of long term selective ion
exchange operations.
1.
The study had the following specific objectives:
To compare methods of ammonia removal in such terms as cost
and removal efficiency, and explain briefly each process including
inherent advantages and disadvantages.
2.
To compile existing knowledge from previous investigations
of selective ion exchange with clinoptilolite.
3.
To define the effects of long term operation, such as loss
of exchange capacity, attrition rate, organic fouling and temperature
effects more completely, as would be needed for large scale appli­
cation of clinoptilolite to ammonia removal.
4.
To establish a method of easy computation of the minimum
cost of regeneration.
Limltations
(I)
Attrition and organic fouling studies were performed with
Bozeman wastewater.
tapwatcr.•
Temperature effects were studied with fortified
4
(2) . Two exhaustion runs were made in each temperature range.
(3)
The full exchange capacity was restored to the clirioptilolite
every 25 cycles during the attrition studies when the samples were
heated to 600oC to destroy organic coatings.
(4)
'
Only 100 cycles of exhaustion and regeneration were performed
in the long term studies.
(5)
All ammonia-nitrogen removal costs are given for a 10 MGD
operation.
(6)
The regeneration costs of clinoptilolite were optimized for
a 10 MGD operation.
(7)
Several assumptions, necessary for optimization of regenerant
cost, are explained in Chapter VII.
Symbols
Symbols and abbreviations are defined in Appendix A.
•
CHAPTER II
AMMONIA-NITROGEN REMOVAL METHODS
Many processes have been developed for the removal of ammonia and
the oxidized forms of nitrogen from wastewater.
Selection of a given
process or combination of selected processes depends on:
(1)
The use to be made of the treated wastwater, which normally
governs the allowable nitrogen content.
(2)
The available means for ultimate disposal of contaminants,
such as nitrogen-rich brines from reverse osmosis.
(3)
The process economics.
(4)
The other waste substances removed.
Some processes remove
only nitrogen, whereas others remove deleterious substances such as
phosphorus and salts.
The five principal ammonia-nitrogen removal methods are ammonia
stripping, biological nitrification-denitrification, breakpoint chlor­
ination, selective ion exchange, and land disposal.
Other methods
which have been investigated for nitrogen removal are reverse osmosis,
conventional ion exchange, electro-oxidation, electrodialysis, algae
harvesting, and distillation.
Some of th.6se methods can be combined,
such as nitrification-denitrification with
breakpoint .chlorination
as a polishing, operation.
All cost estimates for the above processes are based on a 10 MGD
operation with a total ammonia-nitrogen concentration of 20 mg/ft NH^-N.
6
The cost estimates were updated to an Engineering News Record Construc­
tion Cost Index of 2200, estimated to occur in January, 1975 [22].
Ammonia Stripping
The removal of ammonia from wastewater by ammonia stripping in­
volves aeration to remove dissolved ammonia gas.
The solution pH must
be adjusted to greater than 11.0 prior to being pumped to a stripping
tower.
The ammonia solution is allowed to fall downward against an
upflow of air (or at right angles for a side draft system).
Because
ammonia release occurs at the instant water droplets are formed, towers
are designed for continuous droplet formation and coalescence [18].
The circulation of air maximizes the air-water ammonia concentration
differential, thus increasing the removal rate.
The release of ammonia
can occur only when the ammonia is in the dissolved gaseous form.
The
high pH is necessary to shift the equilibrium equation (I) to the right.
NH4+
NH 3
+
H+
(I)
Figure I illustrates the relationship between pH, temperature, and the
form of ammonia.
Most of the ammonia can be maintained in the ammonia
form by maintaining a high organic loading rate on the secondary process
[21].
The ammonia extracted generally escapes into the atmosphere.
wastewater pH must be lowered from approximately pH 10.5 to 8.0 with
acid or other means prior to discharge.
The
7
FIGURE I: THE EFFECTS OF pH AND TEMPERATURE ON THE
DISTRIBUTION OF AMMONIA AND AMMONIUM ION IN WATER
( AFTER CULP AND CULP, 1971 [18]).
/
8
Process advantages.
(I)
This is an efficient process for high
ammonia concentrations (>50 m g /X NH^ - N)- at high pH (>11.0) and high
temperature (>20 C).
Ammonia removal efficiency of 98 per cent may occur
under optimum conditions [20].
(2)
This is the most economical ammonia removal method,. $0.04/1000
gallons., as shown in Figure 2 and Table 2.
It requires relatively
inexpensive equipment, little control, and lime for pH control is par­
tially recoverable [47],
(3)
.
Some phosophorus removal occurs due to the high pH necessary
for the process [20].
Process disadvantages. • (I)
'
The high pH causes carbonate pre­
cipitation which may clog the stripping tower.
In addition.the carbonate
scale may be difficult to remove [18].
(2)
The process removal efficiency drops to 50-60 per cent ■
for average ammonia concentrations (20 mg/Jl NH^ - N) as air temperature
approaches freezing [18].
(3)
Ammonia must be in the dissolved gaseous form to be removed.
(4)
The release of ammonia into the atmosphere may be undesir­
able in urban areas and futhermore the extracted ammonia may redissolve .
in the receiving water, due to its high solubility.
(5)
.
It is difficult to achieve low effluent ammonia conce'ntra- ■
tions, at best 10% of 2-5 mg/& NH^ - N remain in the effluent.
Approx-
9
CAPI TAL
COS T
MIllONS
OF
10.0
DOLLARS
100.0
100.0
10.0
DESI GN
ENR = 2 2 0 0 ,
CAP ACI T Y
ESI .
DATE :
,
MGD
JAN,
1975
FIGURE 2: AMMONIA STRIPPING. CAPITAL COST, OPERATION AND
MAINTENANCE, DEBT SERVICE VS DESIGN CAPACITY (AFTER
SMITH,1968 [47]).
10
TABLE 2
COMPARISON OF NITROGEN REMOVAL PROCESSES
Process
Removal
Cost
References
Remarks
Efficiency, Estimate*
%
$/1000 gal
Conventional
Air Stripping
50-98
0.04-.06
18
Costs for summer
operation
Modified Air
Stripping
(ARRP)
85+
0.06-.12
28
Preliminary cost
estimate
NitrificationDenitrification
50-90
0.12-.13
16,18,36,
43
Conventional
Ion Exchange
80-90
0.60-.70
18,36
Selective Ion
Exchange
85-98
0.12-.18
Breakpoint
Chlorination
100
0.09-.10
18,37,42
Breakpoint
Chlorination
With Carbon
Adsorption
100
0,25-.26
18,37,42
Land Disposal
By Spray
Irrigation
0-91
0.13-.26
21,36,48
Cost dependent on
location
0.04-.06
21,16
Excluding cost of
algae disposal,
cost dependent on
location
Algae Harvesting 40-60
29
Excluding costs of
pretreatment and
brine disposal
Excluding costs of
pretreatment and
brine disposal
May require carbon
adsorption
11
TABLE 2 (CONTINUED)
Process
Removal
Cos t
Efficiency,
Estimate*
X
32-92
Reverse
References
Remarks
1,18,36
Excluding costs of
$ / 1 0 0 0 gal
0.45-.50
p r e t r e a t m e n t and
brine disposal
Osmosis
Electrodialysis
40f
0.27-.30
2,47
Electro­
oxidation
70-95
0.70-.75
37
Distillation
0-100
0.69-1.73
21
E x c l u d i n g cos t of
p r e t r e a t m e n t and
brine disposal
Nitrous acid may
e s c a p e w i t h the
distillate
aCost estimates are based on a 10 MGD operation with an
initial 20 mg/1 NH^-N concentration.Engineering News Record
Construction Index ■ 2200, estimated date January,1975.
12
imately 5-6% of this nitrogen is not in the removable form [18].
A modified ammonia gtripping system (Ammonia Removal and Recovery
Process) has been developed by CH3M Engineers of Redding, California.
This process eliminates the problems of carbonate precipitation, reduc­
ed efficiency at low temperature, and the escape of ammonia into the
atmosphere.
This process utilizes A closed air circulation system, so
no carbonate precipitation can occur after the initial carbon dioxide
in the system precipitates.
This system can maintain high removal
efficiency by exclusion of cold outside air.
The ammonia is concentrat­
ed in an aqueous solution maintained at below pH 6.0 with sulfuric acid.
Although the process cost is 1.5 to 2.0 times the cost of conventional
stripping, costs may be reduced by the sale of ammonia for liquid
fertilizer.
Further process design optimization is planned by use of a
pilot facility operating on lime clarified secondary effluent [28].
Biological Nitrification-Denitrification
This process consists of microbiologically oxidizing all the
ammonia in wastewater to, nitrates (nitrification) and then microbio­
logically reducing the nitrates to nitrogen gas (denitrification) which
is released to the atmosphere.
A three stage biological system is
generally used although the process is adaptable to a two stage system.
The first stage, which removes most of the carbonaceous BOD, can be an
activated sludge or trickling filter process.
The second and third
stages are nitrification and denitrification units., respectively.
13
Nitrification can occur in two steps as represented in
equations (3) and (4).
NH4^ + 1.5 O 2
- Nitrosomonas
„----- 1
NO 2
NO2- + 0.5 O 2
JiltrobacterW
NO3-
+ 2H+ + H 2O
(3)
(4)
Chemosynthetic autotrophic bacteria, such as Nitroscunonas, Nitrosococcus, Nitrosospira, Nitrocystis, and Nitrosogloes, obtain energy for
cell synthesis from the oxidation of ammonia to nitrites.
The nitrites
then serve as an energy source for other chemosynthetic autotrophic
bacteria such as Nitrobacter during nitrite oxidation to nitrate [43].
Nitrification must be complete to assure complete denitrification.
The basic conditions for nitrification are:
dissolved oxygen, (2)
8 hours, and (3)
(I)
at least 0.5 mg/2,
cell detention time in the reactor of at least
a low reactor loading rate (<0.3 lb BOD/lb mixed
liquor suspended solids-day)
decreasing water temperature.
[18].
Nitrification rate decreases with
The reactor volume required at IO0C
would be three times that required at 20°C for a complete-mix system
[35] .
Denitrification is an anaerobic process which occurs when facul­
tative anaerobic bacteria, such as Pseudomonas, Achromobacter. Bacillus,
and Micrococcus use nitrate instead of oxygen as a hydrogen ion acceptor
14
in the election transport system [40].
Denitrification may be
represented by equation (5).
4 NO3' + 4 H+ + 5 CH2O
,pseudomon^
5 C02 + 2N2 + 7 HgO
(5)
During the process, nitrates are reduced to nitrogen gas and carbon
compounds are oxidized.
The basic requirements for denitrification are:
carbon source, (2)
mg/£, and (3)
(I)
an organic
a dissolved oxygen concentration of less than 0.5
a pH of approximately 6.5 [16].
The process proceeds
too slowly to be practical when utilizing the carbon remaining after
secondary treatment, hence the addition of a biodegradable carbon
source is necessary.
Raw sewage has been added as a carbon source,
but this has the limitation of adding unoxidized nitrogen compounds
and additional BOD to the final effluent.
Methanol is the least
expensive carbon source at the present time.
The denitrification rate is also temperature dependent.
Denitrification can occur in three possible designs:
anaerobic
activated sludge, anaerobic filter, and anaerobic lagoon.
An optimum
removal efficiency of 90 per cent is possible with the nitrificationdenitrification system.
Process advantages.
(I)
Ammonia and the oxidized forms of
nitrogen are removed by this process.
15
(2)
The conversion of wastewater nitrogen compounds into
free nitrogen does not pollute the environment.
(3)
The costs and removal efficiency of this process are
competitive with other removal processes, as shown in Figure 3 [36].
(4)
The structures required "are,of simple design.
(5)
This system can be used to either produce a nitrified
effluent or to remove nitrogen compounds.
Thus the process can be
constructed in steps to meet standards of the present and the future
as they are required.
Process disadvantages.
(I)
Nitrification and denitrification-;
rates are reduced significantly by low wastewater temperatures.
(2)
The process will be difficult to control.
Process failure
may occur due to cell washout, short circuiting, inhibiting substances
or failure to maintain anaerobic conditions in the denitrification
unit.
(3)
The process must operate continuously to maintain microbial
populations.
(4)
An exceptionally knowledgeable operator is required to run
the process.
(5)
At optimum removal efficiency, 2-5. nig/Jl nitrogen remains
in the final effluent.
A polishing operation such,as breakpoint
chlorination may be required.
16
100 O
GALLONS
100.0
tre a tm e n t
OF
IOOO
DOLLARS
Ie ta l
CAP I TAL
TREATMENT
COS T
COST
.
MIllONS
CENTS/
6m Wt s e r y le e
1.0
10.0
DESI GN
ENR : 2 2 0 0 ,
CAP ACI T Y
EST.
100.0
,
MGD
DATE = J A N. , 1975
FIGURE 3: BIOLOGICAL NITRIFICATION AND DENITRIFICATION.
CAPITAL COST, OPERATION AND MAINTENANCE, DEBT SERVICE
VS DESIGN CAPACITY (AFTER MONTI AND SILBERMAN,1974 (36]).
17
Breakpoint Chlorination
Breakpoint chlorination consists of using chlorine to oxidize
ammonia to nitrogen gas and to reduce nitrates and nitrites to nitro­
gen gas.
This process has been used extensively for the disinfection
of water and wastewater, however its use to remove nitrogen from
wastewater is a recent innovation.
The addition of chlorine to water forms hypochlorous acid as
shown in equation (6).
Cl 2 + H 2O
HOCl + H+ + Cl"
(6)
The reaction of hypochlorous acid with ammonia to form monochloramine
and dichloramine is shown in equations (7) and (8), respectively.
NH 3 + HOCl
NH2Cl + H 2O
(7)
NH 3 + 2 HOCl
NHCl 2 + H 3O
(8)
The reaction of monochloramine and dichloramine with hypochlorous
acid to produce nitrogen gas is shown in equation (9) [35] .
NH2Cl + NHCl 2 + HOCl
N 3 + H 2O + 2H+ + 4 Cl"
(9)
With mole ratios of chlorine to ammonia up to 1:1, both monochloramine
18
and dichloramine are formed.
function of the pH.
The relative amounts of each are a
Large proportions of dichloramine appear at lower
pH in accordance with the equilibrium equation (10).
2 NH2Cl + H+
^=^5
NH4+ + NHCl 2
monochloramine
(1Q)
dichloramine
Further increases in the mole ratio of chlorine to ammonia result in
formation of dichloramine, as shown in equation (8) ,or the production
of nitrogen gas as shown in equation (9).
These reactions are essen­
tially complete when two moles of chlorine have been added for each
mole of ammonia originally present.
Chloramine residuals usually
reach a maximum when one mole of chlorine has been added for each
mole of ammonia and then decline to a minimum at the chlorine -toammonia ratio of 2:1.
Further additions of chlorine produce free
chlorine residuals as shown in Figure 4.
Chlorination of water to the
extent that all the ammonia is converted to Oxidized free nitrogen or
other gases is referred to as "breakpoint chlorination" because of
the peculiar character of the chlorine residual curve, as shown in
Figure 4.
It is generally assumed that 10 milligrams of chlorine are
required per milligram of ammonia to reach the breakpoint.
Nitrous
oxide and nitrogen trichloride have been identified among the gaseous
products of the breakpoint reaction [42}.
Essentially all the ammonia
can be oxidized to nitrogen gas and the production of other undesir-
19
Mole ratio, CU: NH3-N
05
I
1.5
2
Break point
Combined chlorine
residuals in predominance
Free chlorine
residual
predominant
12
13
Chlorine dosage, mg/I iter
FIGURE 4: A RESIDUAL CHLORINE CURVE SHOWING A TYPICAL
BREAKPOINT. THE AMMONIA NITROGEN CONTENT OF THE WATER
IS 1.0 MG/L (AFTER SAWYER AND MCCARTY, 1967 [42]).
20
able products can be eliminated by pH control [14] .
The hydrochloric
acid formed during chlorination will react with the alkalinity of the
wastewater and the pH drop will be slight [35].
Process advantages; . (I)
Chlorine dosage and pH are the only
control necessary.
(2)
Production of nitrogen gas is not ecologically undesireable.
(3)
This process can achieve almost 100 per cent ammonia-
nitrogen removal.
;
•
Process disadvantages.
(I)
•
•.
The presence of organics, iron (Il),
or manganese (I I ) increases the chlorine demand.
(2)
Carbon adsorption is necessary to remove chloramines" not
destroyed by chlorination,
(3) . This process increases the chlorides in the effluent 200500 mg/&.
This may restrict the reuse or disposal of the final
effluent.
(4)
The cost of breakpoint chlorination ($0,092/1000 gallons),
as shown in Table 3, is comparable to other removal.methods.
However-
the addition of carbon adsorption ($0.16/1000 gallons) as shown in
Figure 5, increases, the cost beyond economic feasibility for removal
of the major portion of ammonia nitrogen concentrations [36]. ' Break­
point chlorination could be used economically to-remove 2-5 mg/fi.ammonia-nitrogen remaining after less efficient but. more economical■
2\
TABLE 3
CAPITAL AND OPERATING COSTS FOR BREAKPOINT CHLORINATION*
(AFTER PATTERSON AND BANKER, 1971 [37])
Item
Value
Total Capital Costs, dollars
Chlorination equipment, chlorine feed
building, chlorine storage building,
contact basin, earthwork, and baffles
Amortized Cost*5 , dollars
Capital Cost , dollars/1000 gal
532,000
49,500
0.014
Operating Costs, dollars/year
Chlorine
Materials
Maintenance and labor
Total Operating Costs, dollars/year
180,000
4,500
100,000
284,500
Operating Cost, dollars/1000 gal
0.078
Total Cost, dollars/1000 gal
0.092
*10 MGD system with 20 mg/1 NH^-N; Engineering News
Record Construction Cost Index of 2200, estimated January,
1975.
^15 year life at 7% interest rate.
22
100.0
Ireolm ent
10.0
CAP I TAL
TREATMENT
COS T ,
COST
,
MIlLONS
OF
DOLLARS
Ie to l
CENTS ' IOOO
GALLONS
100.0
100.0
DESI GN
ENR : 2 2 0 0 ,
CAP ACI T Y
EST.
,
MGD
DATE = J A N . , 1975
FIGURE 5: CARBON ADSORPTION (40 MINUTES DETENTION).
CAPITAL COST, OPERATION AND MAINTENANCE, DEBT SERVICE
VS DESIGN CAPACITY (AFTER MONTI AND SILBERMAN.1974 [36]).
23
nitrogen removal processes.
Conventional Ion Exchange
Ion exchange is a unit process in which ions of a given species
are displaced from an insoluble exchange material by ions of a dif­
ferent species in solution.
This process is commonly used for water
softening and demineralization.
The chemistry of ion exchange may be
represented by equation (11).
NH 4
+ RNA
Exhaustion^
---------Regeneration
RNH 4 + N A +
R represents the exchange material.
(11)
The sodium ion in the exchanger
is displaced by the wastewater ammonium ion.
Regeneration of the ex­
hausted exchange material occurs by passing a concentrated sodium
solution through the exchanger to replace the ammonium ions.
Prefiltration of the wastewater is necessary to prevent fouling
of the exchange material.
The cost of nonselective removal of nitrogen
compounds ($0.65/1000 gallons) provided by conventional ion exchange
resins is rarely justifiable from a pollution control standpoint [36].
The development of ion exchange processes selective for ammonium and
nitrate ions is economically more feasible.
Selective Ion Exchange
Selective ion exchange for ammonia employs a natural zeolite,
24
clinoptilolite, selective for ammonium ions in the presence of common
wastewater ions.
The exhausted clinoptilolite is regenerated with an
alkaline solution of sodium or calcium ions.
The resultant ammonia-
rich regenerant can be ammonia stripped and recycled.
Selective ion
exchange has an optimum ammonia removal efficiency of 98 per cent.
The
cost of this process, $0.12/1000 gallons with regenerant reuse,and up
to $0.18/1000 gallons without regenerant reuse, is comparable to other
methods [29] ..
.
Process advantages. ...(I)
The process untilizes .inexpensive
regenerant solutions of calcium1or ,sodium which may be ammonia'>stripped
and recycled.
(2)
Clinoptilolite, a natural zeolite, occurs in extensive de­
posits and is less expensive than synthetic exchange materials.
The
1971 costs were $4.70/cu ft clinoptilolite and $47.50/cu ft Amberlite
IRC-84.
(3)
Although clinoptilolite has less exchange capacity than
synthetic exchange resins its ammonium ion selectivity is compensating.
Process disadvantages.
(I)
This process cannot remove nonion-
ized ammonia or oxidized forms.of nitrogen from wastewater.
(2)
Prefiltration of wastewater may be necessary to prevent
zeolite fouling and subsequent loss of exchange capacity.
Sand f'iltfa-
25
tion costs about $0.06/1000 gallons as shown in Figure 6 [36].
(3)
posed.
When mineral buildup occurs, the regenerant must be dis­
Disposal of the concentrated brine may present a problem.
(4)
Certain design parameters, such as loss of exchange capac­
ity, were not previously documented.
Land Disposal
Land disposal of wastewater is a potential low cost method of
nitrogen removal.
(I)
The three principle methods of land disposal are:
rapid infiltration, (2)
pverland runoff, and (3)
spray
irrigation.
Rapid infiltration is the high-rate application of wastewater
to a very porous soil.
Overland runoff is the low-rate application of
wastewater to rather impermeable soils with gradual wastewater per­
colation.
Spray irrigation is wastewater application as irrigation
water to promote plant growth with subsequent wastewater infiltration
into the soil..
Spray irrigation is the land disposal system generally used, for
optimum nutrient removal.
Primary effluents have been successfully
applied by spray irrigation.
be saved.
The expense of secondary treatment can '
Wastewater suspended particles are strained out, colloids
and organic matter are absorbed by the soil particles, nutrients are
utilized by vegetation, more complex organic compounds are decomposed
by soil bacteria into simple inorganic compounds, and the ground water
26
DOLLARS
100.0
i o ta I Iro atm o n t
COS T
c a p i t a I coot
CAP I TAL
TREATMENT
COST
,
MI LTONS
CENTS'
OF
IOOO
GALLONS
IOO O
100.0
DESI GN
ENR = 2 2 0 0 ,
CAP ACI T Y
EST.
DATE :
,
MGD
J A N . , 1975
FIGURE 6: MIXED MEDIA OR SAND FILTRATION. CAPITAL COST,
OPERATION AND MAINTENANCE, DEBT SERVICE VS DESIGN CAPACITY
(AFTER MONTI AND SILBERMAN.1974 [36]).
27
is recharged.
ways:
(I)
The majority of the nitrogen removal occurs in three
soil adsorption, (2)
conversion to nitrogen gas.
plant uptake, and (3)
biological
Physical adsorption by soil particles
is the principle method of removal of the ammonium ion [21].
Plant
cover will remove nitrogen compounds as nitrogen is necessary for cell
synthesis.
Biological nitrification-denitrification (similar to the
process previously mentioned) may occur if proper conditions prevail.
If aerobic conditions occur, ammonia may be oxidized to nitrates
by bacteria established in the soil.
The soil must be allowed to rest
periodically after inundation so aerobic conditions occur.
Denitri- .
fication, an anaerobic process,occurs during inundation, when oxygen ■
is depleted by aerobic bacteria.
It also occurs in anaerobic areas
created by the oxygen uptake of plant roots.
convert the nitrates to nitrogen.
Denitrifying bacteria
As in other nitrification-denitri­
fication systems, nitrification must occur before denitrification.
Nitrification is generally complete.
Denitrification may not be
complete due to lack of anaerobic conditions or too high hydraulic
loading rate.
Nitrogen removal varies from 0 to 91. percetit depending on soil
type and depth, design and rate of wastewater application, and vegeta­
tion cover [47].
Cost estimates of land disposal are site specific
because of the inter -relationship of such factors as land availability
and location.
Actual construction costs for land disposal systems are
28
not difficult to estimate,as shownin Table 4,but frequently land
values exceed the construction cost or there exist tradeoffs between
treatment area and pipeline costs [ 3 6 ] . . .
Process advantages.
(I)
Although the Environmental Protection
Agency requires secondary treatment.prior to spray irrigation, it may
not be technically necessary.
(2)
The cost of the process, $0.13/1000 gallons, can be compet­
itive with other nitrogen removal processes, especially if the plant
cover can be sold [21].
Process disadvantages.
(I)
If denitrification or plant uptake
of nitrogen is not complete, the nitrogen will eventually enter the
ground water or surface water.
(2)
Extensive storage' requirements are necesshry when spray
irrigation cannot be practiced during freezing or wet weather.
(3)
This process may not be practical in densely populated
areas because of the high.costs for land or. pipe.systems to low cost
land.
Algae Harvesting
Nitrogen.may be removed from wastewater by algae harvesting. ,In
this process, the soluble nitrogen is assimilated into algae cells,
transformed into algal cell tissue, and subsequently removed from the
29
TABLE 4
CAPITAL AND OPERATING COSTS FOR SPRAY IRRIGATION*
(AFTER POUND AND CRITES,1973[39])
Item
Value
Liquid Loading Rate, in/wk
2.5
Land Used, acres
1030
Land Required, acres*5
1240
Capital Costs0 , $
Earthwork
Pumping station
Transmission
Distribution
100,000
200,000
600,000
1,000,000
Total Capital Cost, $
1,900,000
Amortized Costd , $
176,000
Capital Cost, $/1000 gal
0.048
Operating Costs, $
Labor
Maintenance
Power
Total Operating Costs, $
60,000
90,000
40,000
190,000
Operating Cost, $/1000 gal
0.052
Total Cost, $/1000 gal
0.100
*10 MGD system; Engineering News Record Construction
Cost Index of 2200, estimated January,1975.
^20 percent additional land required for buffer zone.
cCapital costs do not include land costs.
ciIS year life at 7 percent interest rate.
30
wastewater.
The production of algal cells may be represented
by equation [12].
106 CO2 + 81 H 2O + 16 NO3' + HPO4" + 18 H+
+ sunlight — >
C 106 H 181 O45 N 16 P
+ 150 O3
Algae can use either ammonia or nitrates.
(12)
This process requires
shallow aerobic logoons to maximize the penetration of light.
Where
large quantities of nitrate are to be removed, it may be necessary to
supplement the wastewater with a carbon source, such as carbon dioxide,
for increased nitrate removal [21].
per cent nitrogen by weight.
Algal cells are approximately ten
Large amounts of algal cells must be
removed to achieve the desired nitrogen removal.
from the system is not easy or inexpensive.
The removal of algae
Sand filtration at $0.06/
1000 gallons, as s h o w in Figure 7, could increase costs beyond economic
feasibility [36, 47].
Algal cells discharged with the effluent would
exert a high oxygen demand on degradation and release nutrients.
Large
land areas (approximately 15 acres/MGD) are required, however less land
is required than for land disposal.
If long periods of wastewater
storage are required, land requirements will greatly increase.
The nitrogen removal efficiency of this process is very dependent
on the feed characteristics, climatic conditions, lagoon design, and
harvesting method.
The optimum removal efficiency has been estimated
31
100 O
COS T ,
CAP I TAL
TREATMENT
COST
,
M I LLON S
OF
CENTS / IOOO
DOLLARS
GALLONS
100.0
ceplteT celt
I OOO
DESI GN
ENR = 2 2 0 0 ,
CAP ACI T Y
EST.
,
MOD
DATE > J A N . , 1975
FIGURE 7: MICROSTRAININC. CAPITAL COST, OPERATION AND
MAINTENANCE, DEBT SERVICE VS DESIGN CAPACITY (AFTER
SMITH,1968 [47]).
32
as 60 per cent [16].
Higher removal efficiency may be obtained by
constructing multiple ponds for series operation.
Cost estimates for
algae harvesting, as for land disposal, are site specific.
However,,
costs have been estimated as $0.04-.06/1000 gallons [21].
The popul­
arity and economic feasibility of this method will be increased if
the high protein algal cells can be sold as animal feed.
Process advantages.
(I)
The process cost is very low.
Costs
are competitive with those of ammonia stripping.
(2)
The sale of protein-rich algal cells will lower operation
costs.
Process disadvantages.
(I)
The discharge of algal cells would
have a deleterious effect on the receiving water.
(2)
The process cannot be used in cold climates because of '
excessive storage requirements.
(3)
The method is presently unpopular because.of the disposal
problem of the algal cells.
Reverse Osmosis
Reverse osmosis is a nonselective process by which water is
separated from substances in solution by filtration though a semipermeable membrane at high pressure (200-700 psi).
The principal use
of reverse osmosis is demineralization of brackish water.
The amount
of demineralized water produced from the total quantity processed
33
is a function of the concentration of dissolved substances in the
wastewater.
Concentration of these substances occurs as water is
removed during the process.
Precipitates formed during concentration
will coat the membrane surface and effectively reduce the water flux.
The product water quality deteriorates as the recovery ratio increases
above 90 per cent [I].
Recovery ratios of 80-90 per cent appear
feasible in municipal wastewater renovation with proper precondition­
ing of the feed water.
This preconditioning consists of coagulation-
flocculation, and pH control.
The nitrogen removal efficiencies list­
ed in Table 4 are a function of the operating conditions, feedwater
quality, and membrane type.
Intensive study is underway to find
TABLE 5
REVERSE OSMOSIS NITROGEN REMOVAL EFFICIENCIES
(AFTER AEROJET GENERAL CORP, 1969 [I])
Nitrogen type
removal, %
organic
71-92
ammonia
74-87
nitrite
45-90
nitrate
23-92
cheaper, stronger membranes to make the process competitive with other
nitrogen removal methods.
The cost of reverse osmosis is $0.45/1000
gallons as shown in Figure 8 [36].
34
100 O
CAP I TAL
TREATMENT
C OS T
COST
,
M I LI ON S
OF
C E N T S / IOOO
DOLLARS
GALLONS
1 0 0 .0
100.0
DESI GN
ENR = 2 2 0 0 ,
CAP ACI T Y
EST.
,
MOD
DATE i J A N . , 1975
FIGURE 8: REVERSE OSMOSIS. CAPITAL COST, OPERATION AND
MAINTENANCE, AND DEBT SERVICE VS DESIGN CAPACITY (AFTER
MONTI AND SILBERMAN, 1974 [36]).
35
Process advantages.
(I)
The wastewater is demineralized.
Many substances besides nitrogen are removed by this process.
(2)
The process is simple in design and concept.
Process disadvantages.
(I)
Pretreatment of wastewater is •.
necessary as reverse osmosis requires a high quality feed water.
(2)
Membranes are sensitive to pH changes, oxidizing agents,
compaction under pressure, and degradation by bacteria.
The membranes
have a usable life of only a few months and their replacement is ..
:
costly.
(3) '.Nitrogen removal efficiency is variable (23-92 per cent).
(4)
The disposal of the concentrated brine presents a.problem.
Electrodialysis
.
Electrodialysis is a nonselective process by which wastewater
substances are concentrated by the use of semi-permeable membranes.
The principal use of electrodialysis is demineralization of brackish
water.
The application of an electrical potential between two elec.-. ■
trodes causes an electric current to pass through the wastewater
solution.
Cations, such as ammonium ions, migrate toward the. negative
electrode and anions, such as nitrate ions, migrate to the positive
electrode through the permeable membranes.
The water from which these
ions have been collected, flows between the membranes, and is collected '
36
as partially demineralized water.
Recycle of the partially demineral­
ized water through the system is necessary to obtain a high quality
water.
The amount of electrical energy and the membrane surface
area required to demineralize wastewater depends on the amount and
type of substances to be removed.. The process cost is thus dependent
on the amount and type of substance to be removed.
Approximately.
20 per cent of the feedwater is necessary to wash out the concentrated
brine solutions [2].
Electrodialysis has problems similar, to those of
reverse osmosis, such as chemical precipitation with membrane clogging.
Pretreatment, such as chemical precipitation, filtration or carbon
adsorption, will be required for low quality feedwater.
The nitrogen
removal efficiency for electrodialysis is about 40 per cent.
But re­
cycle through the system should result in increased removal efficiency.
The cost of this process is $0.27 - .30/1000 gallons as shown.in Figure
9 [47] .
Process advantages,
(I)
This process demineralizes wastewater.
Many substances besides nitrogen are removed.
(2)
The membranes required have a longer life (5 years) than
those used in reverse osmosis.
Process disadvantages. . (I)
Pretreatment is necessary for low
quality feed water.
(2)
The ultimate disposal of the concentrated brine presents a
37
1000
100.0
CapJteI
DOLLARS
GALLONS
^
CAP I TAL
TREATMENT
COS T ,
COST ,
M I LLON S
OF
CENTS / I OOO
eperailew
malntaitwMa
100.0
DESI GN
ENR = 2 2 0 0 ,
CAP ACI T Y
E ST.
,
MGD
DATE ■ J A N . , 1975
FIGURE 9: ELECTRODIALYSIS. CAPITAL COST, OPERATION AND
MAINTENANCE, AND DEBT SERVICE VS DESIGN CAPACITY (AFTER
SMITH, 1968 [47]).
38
problem.
(3)
Electrodialysis is two times as expensive as nitrification-
denitrification or selective ion exchang.
Electro-oxidation
Electro-oxidation is a process by which wastewater ammonia is
oxidized to nitrogen gas in an electrochemical cell.
The overall
reaction at the anode is a three electron exchange reaction yielding
nitrogen and water as products with almost 100 per cent faradaic
efficiency.
The reaction may be represented by the following equation:
NH 3 + 30H™
---- »
^sN2 + 3H20 + 3e“
(13)
This process appears to have a high ammonia removal efficiency
(approximately 100 per cent).
But in the laboratory only electrodes
containing a significant amount of platinum were capable of converting
ammonia to nitrogen gas.
.75/1000 gallons treated.
The preliminary cost estimate was $0.70 The cost of platinum electrodes is too
high to make this process competitive with other methods
[27].
Distil lat Ion
Distillation
involves driving off water vapor from wastewater by
heating in a still, followed by condensation of the water vapor.
A
variety of distillation processes exist, such as flash, differential,
and vapor compression distillation.
Distillation is unfavorable as a
39
method of nitrogen removal because ammonia would be transferred to the
distillate unless.the wastewater was kept at low pH.
Even under acid
conditions nitrous acid would escape with the distillate
[14]. '
Cost estimates for distillation are $0.69 - 1.73/1000 gallons [21].
Because of high cost
and low removal efficiency, distillation will not
be used for nitrogen removal from wastewater.
CHAPTER III
SELECTIVE ION EXCHANGE LITERATURE
Previous Studies
Ames (1967) was the first to suggest that clinoptilolite be used
to remove ammonia from waste water [9]'.
The clinoptilolite replacement
series of alkali metals and alkaline earth metals was the result of an
early investigation [3].
The order of preference of clinoptilolite'
for various ions decreases in the order, Cs+ >
+
Ba >
+2
Sn
>
4~
NA >
+2
>
CA
4-3
Fe
>
*.»4*3
Al
>
Rb+ >
-, +2
Mg
>
K+ >
Li
+
>.
™ 4+ >
Ames
hypothesized the more closely an ion approached cesium in size, the
more selective clinoptilolite is for that ion.
He also concluded that
the replacement of a given cation is determined by complex relation­
ships between cation size, cation charge, electronic structure, and
sometimes temperature [3].
In a later study, Ames stated the cation
selectivity of clinoptilolite is due to the exchanging cations, ex­
change sites, and structural water [4].
He suggested the structural
water of clinoptilolite is not as firmly bonded to the aluminosilicate
framework as for other less selective zeolites.
This results in a
sieving effect on cations entering the clinoptilolite framework.
Ames characterized the exchange kinetics of clinoptilolite by.
consideration of particle and film diffusion rates [5].
The exchange
capacities of various samples of clinoptilolite were determined and ..
reasons for their variability were suggested ,[8]. -Syrithhtic ■ ■
41
clinoptilolite was produced by subjecting- silica and aluminum compounds
to a hydrothermal process at high temperature-(300OC) and pressure
(15000 psi)
[7].
No cost estimates are available, but synthetic
clinoptilolite is likely to cost more than natural clinoptilolite.
The results of Ames' early studies did much to characterize clinoptil­
olite and enable its later application to wastewater ammonia removal.
Mercer, ejt aJL were the first to conduct small column and pilot
plant studies using clinoptilolite for ammonia removal [15, 33, 34].
The potential low cost and the ammonium ion selectivity of clinoptil­
olite appeared to make this process favorable for use in ammonia re­
moval from domestic wastewater.
shown in Table 6, indicated:
(I)
Results of the small column studies,
Considerably less regenerant volume
was required for sodium regeneration than for calcium regeneration.
(2)
A 20 percent greater volume of feedwater could be treated to
breakthrough with sodium regenerated clinoptilolite.and (3) Regenerant
,solutions used and ammonia stripped for three cycles were as effective
as fresh regenerant [15] .
There was a 25 per cent loss of exchange
capacity due to organic fouling of the clinoptilolite when unfiltered
secondary effluent was used [34].
There was also a high leakage and
shallow breakthrough due to channeling In the exchanger bed.
Results of tests performed with the mobile pilot plant indicate:
(I)
The ammonia leakage prior to breakthrough could be maintained at
0.7 mg/£ NHg -N.
(2)
The operationof two columns in series resulted
TABLE 6
PREVIOUS INVESTIGATIONS OF AMMONIA REMOVAL WITH CLINOPTILOLITE
Reference
Type of
Influent
Mercer
• t #1
Average
Influent
NH3-N,mg/1
Exchange
Reactor
Depth, in
Exhaustion Throughput to
Rate ,
Breakthrough, BV
BV/hr
(Breakthrough
NH3-N, mg/1)
Tertiary
Tertiary
Clarified
Rta Sewage
15.0
15.0
12.0
72 (single)
56(aeries)
56(series)
6.5-9.7
(downflow)
150 (1.0)
250 (I.0-2.0)
232 (1.4)
94
97
93
Battelle
MAaorial
Inatitute
[151
Coagulated
Secondary
Untreated
Secondary
11.0
24(series)
220(1.1)
98
15.0
24(single)
20
(downflow)
16.6
(upflow)
Mercer
et al
[331
Coagulated
Secondary
Untreated
Secondary
10.4
24(series)
16.6
(upflow)
200-360
(C/C.-0.1)
Koon and
Kaufman
[29]
Fortified
Tapwater
Coagulated
Secondary
20.0
72(series)
180(1.7)
17.6
36(series)
16-19
(downflow)
12-15
(downflow)
[341
Average N H - N
Removal, Z3
140(1.5)
100-200
(0.39-.94)
97
91-97
TABLE 6 (CONTINUED)
Average
Influent
NH3-N,nig/1
Exchange
Reactor
Depth, in
Exhaustion
Bate,
BV/hr
Fortified
Tapwater
2.0-3.O
24(aingle)
20
(downflow)
Fortified
Tapwater
14.0
6.6-26.7
(downflow)
17 (single) 6.6-26.7
(downflow)
Reference
Type of
Influent
Braico
117 ]
McLaren
and
Farquhar
[32]
70.0
17 (single)
Throughput to
Breakthrough1BV
(Breakthrough
NH 3-N 1 mg/1)
300(0.50)
130-260
(2.0)
66-106
(2.0)
Average N H - N
Removal, ZJ
90
Hot
Available
44
in a 60 percent increase in the utilized column exchange capacity.
and (3)
A batch regeneration technique might be-utilized in which 2
to 4 BV (Bed Volume) of regenerant were recycled through the column
until saturated with ammonia,■ thei ammonia stripped immediately and
reused [15].
Mercer and his co-workers concluded efficient ammonia
removal combined with regenerant reuse increased .the potential for
selective ion exchange.
Koon and Kaufman presented a most comprehensive report■on ■
selective ion exchange characteristics, exhaustion, regeneration,
theory, process performance, design.criteria, and cost estimates [29].
Clinoptilolite was investigated with, the objective of optimizing its
application to ammonia removal from wastewaters.
The study included
multiple cycle pilot plant operations at three municipal wastewater
treatment plants as well as small column studies.
Particular attention
was given to cation interference with exhaustion performance and with
minimum cost of regeneration,.
An average of 95.7 per cent ammonia re­
moval was obtained with three municipal wastes each of which contained
about 20 mg/& NH^ - N.
Ammonia leakage of less than 0.5 mg/2 NH^ - N
is feasible but only with shorter runs and greater regeneration re- ■
quirements.
The 1971 cost of ammonia removal was estimated as $0,082/
1000 gallons with regenerant reuse.
Braico evaluated the potential use of clinoptilolite for ammonia
removal from fish hatchery water [17] .
As ammonia is toxic, to fish in
45
concentrations as low as 2.5 .mg/5- NH^ - N, an effective means of
removal is necessary for reuse of water at hatcheries.
this small column study were:
(I)
The results of
The column effluent ammonia con­
centration could be limited to 0 . 5 'mg/i NH^ - N prior to breakthrough.
(2)
The ammonia capacity of clinoptilolite is not linearly dependent
on the influent competing ion concentration.
(3)• Room temperature
(23°C) investigations may be used to predict results at the temperature
range of salmonid propagation (12.5°G).
(4)
The performance of
clinoptilolite in removing ammonia was superior to that of the syn­
thetic exchangers, Amberlite 200 and Amberlite IRC - 84.
and (5)
The 1972 cost of .ammonia removal for a 5 MGD hatchery with influent
ammonia concentration of 2.1 mg/S, NH^ - N w a s estimated to be $0.031/.
1000 gallons with regenerant reuse.
Braico concluded selective ion
exchange was superior.to other methods of ammonia removal from hatchery
wastewater because of its lower cost, higher removal, greater control­
lability, and.compactness.
■McLaren and Farquhar further investigated the use of .clinoptil­
olite as a means of wastewater ammonia removal [32].
They performed
small column studies to determine the effects of temperature, flow
rate, and ammonium ion concentration on the ammonia capacity and
breakthrough, characteristics.
They made these conclusions:
(I)
The
influent ammonium' ion concentration has the greatest effect on the
ammonia capacity of clinoptilolite.
(2)
High flow rates significant-
46
Iy decrease the ammonia capacity.
the ammonia capacity.
(4)
(3)
Low temperatures do not effect
The volume of..solution treated to break­
through is inversely proportional to the influent ammonium ion con­
centration, and (5)
Batch equilibrium data at 23°C provides a good
estimate of the ammonia capacity obtained in low temperature column
operations over a wide range of ammonium ion concentrations.
McLaren
and Farquhar concluded the use of clinoptilolite should be considered
when essentially complete ammonia removal at low temperature is re­
quired for wastewaters containing ammonium ions in concentrations up
to and possibly beyond 70 mg./5, NH^ -. N.
• ■ Sims and Little developed a technique utilizing clinoptilolite
for improving the efficiency of the nitrification process during '
wastewater treatment [46] .
Clinoptilolite, added to the activated'
sludge unit, selectively adsorbs ammonium ions from the wastewater
and provides a surface for the attachment of nitrifying bacteria.
These bacteria oxidize ammonium ions to nitrate ions.
The removal of
the adsorbed, ammonium ions from the zeolite by the nitrifying bacteria
regenerates the zeolite and enables repeated zeolite use. . The results
of this study indicated:
(I)
The nitrification efficiency of the
activated sludge unit was increased.
(2)
The activated sludge unit
was better buffered against shock loads of raw sewage, and (3)
effluent with a lower BOD was produced.
An
.'The addition of clinoptil­
olite to the activated sludge process did not require modification, in
47
design or use of the unit.
It is therefore possible to adapt the
activated sludge process to selective conversion of the ammonium ion
to. nitrate.
Structural Properties and Source
Clinoptilolite is mineralogically classified as a member of the
silicate group.
It is defined as a hydrated aluminosilicate of
univalent base which can be reversibly dehydrated to varying degrees
without undergoing a change in crystal structure and which is capable
of undergoing cation exchange [10].
Clinoptilolite is structurally
quite similar to the zeolite heulandite.
The largest known deposit of clinoptilolite in the United States
is near Hector, California.
John Day Formation in Oregon.
There is another large deposit in the
'Clinoptilolite. is a common material
found in bentonite clay deposits throughout the United States and the
world [7].
Clinoptilolite is generally found with 5 to 15 per cent
Impurities■such as quartz,'feldspar^ hnd clay [7].
Stoichiometrically clinoptilolite is derived from the formula
(SiC^)U with periodic substitution of aluminum atoms for silica atoms
with sufficient alkali metal and alkaline earth cations to maintain
electroneutrality.
SiO^
A1™
The zeolite framework consists of tetrahedral
5“
and AlO^.
combined, into crystal structures [10].
The alkali
metal and alkaline earth cations are exchanged during the ion exchange
48
process.
, I
The extent to which the cations within the zeolite may be ex­
changed depends on the nature of the clinoptilolite .crystal, such as .
the size of the pore openings and the degree to which channels within
the crystal are interconnected [43].
Because the rigid, three-
dimensional crystal lattice contains definite sized' openings into the
ion cages, clinoptilolite exhibits ion sieve properties to a much
greater extent than conventional ion exchange resins.
In more dense
zeolites some ions are completely excluded -from the channels.
However
in the "open" zeolites, all alkali metal and alkaline earth cations
have access to passageways, although partial sieving action is ob­
served due to the stripping of hydrated water from the ion as it enters
the zeolite opening [44].
The preference of a zeolite for an ion is
a function of the energy with which the water of hydration is bound
to both the cation and the zeolite, the size of the ion, and its
valence [25].
'•
,Stability.
The instability of zeolites toward acids and alkalis
has been long known because zeolites have been used extensively in
water soften Lng.
zeolites [3].
Clinoptilolite is. more acid resistant than other
Treatment of clinoptilolite with acid in 1.0 N and high­
er concentrations resulted in progressive displacement of aluminum
from the zeolite fromewofk leaving only hydroxylated silica [12].
49
Samples of Clinoptilolite exposed for two days to a 20% sodium hydrox­
ide solution had a 70% weight loss [13].
Weight loss during this
period was assumed due to attack by the alkali.
Other attrition
studies have been made to determine the stability of clinoptilolite
in the presence of caustic solutions [29] .
In small column tests.
.
designed to simulate 100 exhaustion and. regeneration cycles, attrition
rates of 0.25, 0.35, and 0.55% per cycle were measured for exposure to
pH 11.5, 12.0, and 12.5 solutions, respectively.
The rates of mechan­
ical attrition (flowrate of 17 BV/hr, upflow) were 0.15, 0.25, and
0.48 per cent per cycle for exposure to pH levels of 11.5, 12.0, and
12.5, respectively.
Although the attrition rate decreased after ex­
posure to 100 simulated cycles, Koon and Kaufman concluded in actual
application the required replacement rate may not decrease after 100
cycles of exhaustion and regeneration [29].
Ion exchange capacity.
The total exchange capacity of clinop­
tilolite ranges from 1.6 to 2.0 meq/g as measured by various invest­
igators [5, 8, 9, 13, 26].
The value of the exchange capacity depends
on the method used for determination, as well as the ions involved.
Koon and Kaufman's method of exchange capacity determination is de­
scribed in Table 7 [29].
50
TABLE 7
METHOD FOR DETERMINATION OF AMMONIA EXCHANGE CAPACITY
(AFTER KOON AND KAUFMAN,1971 [29])
^ - .. .
1.
'.
....... .
=
'
Exhaust 2.0 gram samples of oven dried clinoptilolite
with 0.50 M NH^Cl solution at 50 ml/hr for 48 hours.
2.
Rinse samples with distilled water at 50 ml/hr for
24 hours.
3.
Elute ammonia from clinoptilolite with a 0.5 M NaCl
solution at pH 7.0 recycled at 50 ml/hr for 72 hours.
4.
Determine the ammonia concentration of the elutant
by direct nesslerization [41].
51
Barrier found that an exchange capacity value of 1.83 meq/g
corresponds to 98 per cent of the possible exchange capacity [13].
Thu s
' practically all the exchange sites in clinoptilolite are
accessible to alkali-metal ions.
•
The acid wash employed by Arnes to
-
-
increase the exchange capacity by dissolving acid soluble impurities
within the crystal structure does not affect the exchange capacity
[29].
The total exchange capacity is identical for sodium and calcium
based clinoptilolite, but the breakthrough exchange.capacity is more
■
than twice as great for sodium base clinoptilolite.
Structural factors
which affect, both selectivity and diffusion of ions through the zeolite
appear to restrict the mobility of the larger calcium' ion more than
the sodium ion.
This results in superior exchange•kinetics for sodium
relative to calcium regardless of whether these ions are entering the
zeolite or are being displaced from it [29].
The ammonia exchange capacity is of more interest than the total
exchange capacity.
The ammonia exchange capacity can be estimated
from the cationic strength of the influent wastewater as shown in'
Figure 10.
Koon and Kaufman observed the ammonia exchange capacity
to decrease sharply with increasing competing cation concentrations
to a cationic strength of about 0.01 moles/liter [29]!.
Increases in
cationic strength above this value continued to decrease the exchangecapacity but to a much'lessor degree.■ .
-
52
■Total Anmonla Exchange
Capacity
o l i d Phase Aamonia Conce
16.4-19.0 mg/t
^-E ffe c tiv e Anmonia Exchange Capacity
(to I mg/t NH3-N in e ff lu e n t)
0.005
0.01
0.015
Cationic Strength, 1/2% ]m ,:!, moles/t
FIGURE 10: VARIATION OF AMMONIA EXCHANGE CAPACITY WITH COMPETING
CATION CONCENTRATION (AFTER KOON AND KAUFMAN, 1971 [29]).
53
Exhaustion.
The pH of wastewaters is of primary importance in
removing ammonia by ion exchange.
ranges from 5 - 1 0 .
at high pH.
The pH of domestic wastewater
Ammonia is predominantly in a nonionized form
Hydrogen ions compete with ammonium ions for. exchange
sites at low pH.
The optimum conditions for ammonia exchange.exist
between pH 4.0 and 8.0 as shown in Figure'11.[29].
There is little
variation in ammonia exchange between these values, but ammonia
exchange decreases rapidly outside this range.
Ames investigated the effect of particle size on cesium ex- .
change in clinoptiltilite [3].
The breakthrough volume to C/Co = 0.05
for 18 x 60 mesh particles (U.S. Standard Sieve) was 70 percent of
that for 60 x 100 mesh particles.
10 x 18 mesh particles.
Leakage occurred immediately using
As a balance between exchange kinetics and
headless, 20 x 50 mesh particles have been used in most studies.
The optimum flow rate for exhaustion, is the greatest flow rate
which does . not result in a sacrifice of exchange kinetics.
In studies
to determine the effect of flow rate on exchange kinetics of 20 x 50
mesh, particles, the breakthrough curve became unfavorable, when the"
flow, rate was increased bom 20 to 30 BV/M [15].. Hence 20 BV/hr is
considered to be the maximum flow rate for clinoptilolite exhaustion.
Ion exchange columns are exhausted upflow to minimize bed
clogging with unfiltered wastewater.
If columns are exhausted down-
at Indicated pH
54
o
o
Z
a
■M
IO
Z
I
----- Equation 18
o normalized to pH 6.0
PO
Z
I
<r
3=
I
Z
cr
Z
cr
o 2-hr equilibration
a Column to saturation
pH
FIGURE 11: RELATIVE EFFECT OF pH ON AMMONIA EXCHANGE
CAPACITY (AFTER KOON AND KAUFMAN, 1971 [29]).
55
flow to minimize mechanical attrition the feed must be of high quality.
Exhaustion is continued until the ammonia leakage (the ammonia
concentration in the column effluent) becomes■significant.
The volume
of throughput depends on the exchanger bed depth, column arrangement,
initial influent ammonia concentration, competing cation concentration,
pH, allowable breakthrough, flow rate, and level of regeneration.
The exchange capacity of a single ion exchange column will be
more completely utilized by increasing the exchanger bed depth or by
operating two column's in series.
In the two column series operation
the second column may be completely exhausted by placing it before a
freshly regenerated column.
Regeneration.
,
The ineffectiveness of ammonia adsorption at high
pH corroborates.observations that regeneration of ammonia based.clinoptilolite is best accomplished at a high pH [29].
The ammonia exchange
capacities at various pH levels are shown in Figure 12.
The regenerant
pH is the controlling factor in determining the amount of regenerant
required to remove ammonia from clinoptilolite.
It is hypothesized
that nonionized ammonia formed at a high pH is more able to diffuse
through the zeolite pores than'the ammonium ion [29].
This coupled
with- the maintenance of a maximum differential between solid arid solu­
tion phase ammonium ion concentrations at high pH results in an in­
creased rate of ammonia elution.
The strength (molarity) of the re­
generant caustic is limited by the attrition rate of the clinoptilolite
Ammonia Exchange Capacity,
56
10.0
11.0
12.0
13.0
pH
FIGURE 12: AMMONIA EXCHANGE CAPACITY VERSUS pH
(AFTER KOON AND KAUFMAN,1971 [29])
57
in caustic solutions.
An increase in the regenerant NaCl concentration beyond a certain
value at a particular pH had no effect on the volume of regenerant
required [29].
For regeneration at pH 12.0 and 12.5, no benefit was
realized by using a NaCl concentration greater than 0.10 Ib/gal.
In- "
creasing the NaCl concentration beyond 0.17 lb/gal at pH 11.5 produced
no improvement in regeneration performance.
The flow rate had no effect on regeneration efficiency over the
range of 4 to 20 BV/hr [29]..
Flow rates of 25 BV/hr produced minor im­
pairment of regenerant performance.
Regeneration at 30 BV/hr resulted
in unacceptable performance as only 63 per cent of the ammonia in the
zeolite was eluted with the same volume of regenerant.
A freshly regenerated ion exchange column should be rinsed to
remove NaCl and a caustic before use.
The volume of the product rinse
water required varies from 5 to 15 BV depending on the buffering
capacity of the water.
Clinoptilolite beds which have been exhausted.
.
with unfiltered secondary effluent will require backwashing before
regeneration as well as rinsing.
Rejuvenation of Fouled Clinoptilolite
The fouling of clinoptilolite.can be minimized by adequate pretreatment , such as chemical coagulation and filtration.
Ion exchange
materials are too expensive to be used as wastewater filters.
The
58
accumulation of colloid's on the surface of exchangers can be minimized
by good backwash techniques.
It may be possible to rejuvenate fouled
exchangers by an acid wash followed by a caustic, NaCl wash.
extensive organic fouling, a backwash with 0.25
chlorite solution may be helpful [19].
After ..
0.5 per cent hypo­
The accumulation of iron can '
be minimized by the addition of a small quantity of a reducing agent
to the regenerant solution.
Colloidal iron accumulated on the ex­
changer may' be removed by an acid wash.
removed by a caustic wash:
Colloidal silica may be.
It is necessary, to limit the strength of
the caustic or acid washes employed because of their previously
mentioned effects to the clinoptilolite structure.
Experience has
shown that fouling problems are best handled by good preventive
measures, i.e., water pretreatment, adequate backwashing, and early
rejuvenation [30].
Regenerant Renovation
Renovation of the regenerant solution is more feasible than
disposal of the regenerant solution after its initial use because of
the difficulty of disposal of the brine,.
In most instances it is less
expensive to reuse regenerant solutions than to discard after initial
use.
Koon and Kaufman estimated the 1971 total cost of ammonia re­
moval to be $0.134/1000 gallons where regenerant is used only once and
and $0,082 with regenerant reuse [29].
59
After regeneration of the exhausted clinoptilolite, the regen­
erant is a concentrated ammonia solution at a high pH.
These condi­
tions are ideal for ammonia removal from the regenerant by ammonia
stripping.
The stripped regenerant solution is as efficient as fresh
regenerant in restoring exchange capacity of clinoptilolite.
Ammonia
concentrations in the regenerant up to 14 m g /'Jl NH„ - N have no dele-
3
terious effects on regeneration [15].
sidered, regeneration costs are
When regenerant reuse is con­
..sensitive to the NaCl concentration
of the regenerant and are significantly influenced by ammonia.strip­
ping costs and by pH which causes zeolite attrition.
The buildup, of mineral concentration in the regenerant will
require eventual disposal of the regenerant solution.
Cations other
than ammonia are sorbed by clinoptilolite during exhaustion with wastewater.
These ions are eluted from the clinoptilolite during, regen­
eration and remain in the regenerant solution;
the caustic, it is usually 90 per cent pure.
If lime is used as
The 10 per cent impur­
ities it contains are magnesium and other alkali salts which further
increase the mineral
concentration.
The anion concentration also
increases as chemicals are added during regenerant makeup.
Disposal
of the regenerant will bd necessary when the clinoptilolite is no ■
longer completely regenerated, i.e., shorter exhaustion runs or higher
ammonia leakage are noticed;
.
Regeneration may be accomplished with two successive regenerant
,
60
solutions to reduce the regenerant volume necessary [34].
Re­
circulation of the first regenerant solution through the exchange •
column should continue until the solution ammonia concentration is
maximized.' The ammonia concentration ought to be in excess of .600
mg/£
- N.
This solution is then stripped to reduce the ammonia
concentration to. less than 10 mg/ 1 NHg - N. The second freshly
stripped solution used to complete regeneration ought to contain no
more than 100 to 150 mg/SL NHg - N after its initial use to allow.
recycle in the following exhausted column without, ammonia stripping.
Regenerant renovation by electrolytic destruction of the ,
ammonia has been investigated [34].
This process is the same as the
electro-oxidation process mentioned previously. ■
. Regenerant renovation
by this process is less dependent on temperature than ammonia strip­
ing but it costs more.
CHAPTER IV
EXPERIMENTAL EQUIPMENT AND METHODS
The experimental phases of this study'involved exhaustion runs
using synthetic wastewater and Bozeman municipal wastewater.
The
study utilizing the synthetic wastewater was performed in the Envir­
onmental Engineering Laboratory, Cobleigh Hall, Montana State Uni­
versity.
The studies utilizing municipal wastewater were performed
in the Montana State University Mobile Environmental Engineering
Laboratory located at the Bozeman Wastewater Treatment, Facilities.
Equipment
Dual media filter.
A 3 in ID Corning pyrex glass column with an
18 in. bottom layer of white sand (Permutit Company, No. Q665) and an
18 in. top layer of anthracite (Permutit Company, No. 2 size) served
as a dual media filter.
Ion exchange reactors.
Corning pyrex glass columns served as
exchange reactors in all studies.
either upflow or downflow.
The columns could be operated
The temperature effects study utilized a
15 in.x I.in clinopti.Iolite bed, but all other studies used a 1.4 in.
x ] in.clinoptilolite bed.
The exchange columns were designed to
prevent loss of clinoptilolite during exhaustion or regeneration.
feedwater storage.
Calibrated 30 gallon polyethylene
62
containers served as reservoirs for the column influent.
Pumps.
a
Jabscoi "Water Mule" pumped the secondary effluent to a
reservoir in the Mobile Laboratory.
A Sigmamotor peristaltic pump,
using H in. x7/16 in.Tygon tubing, pumped the effluent through the
dual media filter into another reservoir.
with a Minarik W53 controller, using
A 16- channel tubing pump
.in.x 3/16 in.Silastic tubing,
pumped the filtered effluent through the exchange columns' for ex­
haustion and the regenerant solutions for regeneration.
The temperature effects study utilized a 4- channel Sage pump
using V in.x 3/16 in,Silastic tubing.
The Sage and 16- channel pumps
were calibrated to deliver the desired flow rates during exhaustion
and regeneration.
The Silastic tubing could be stretched to decrease
its diameter to regulate' flow so that all columns received about the
same flow rate.
Temperature control.
An environmental cabinet capable of main­
taining uniform temperature stored the equipment and feedwater during
(ho
;md 330C exhaustion studies.
A Fisher Low Temperature
Incubator stored the equipment and feedwater during the 40C exhaustion
studies.
Temperature was measured with a VWR Scientific mercury ther­
mometer accurate to + I0C.
Sample collection.
Exchange column effluent and wastewater
samples were collected manually during constant surveillance in 300 m&
63
glass stoppered bottles.
Weight measurement.
The clinoptilolite and liquids were measured
to 0.05% and 0.3% accuracy, respectively;
Analysis.' The chemical and physical analyses, as shown-in-Table
8, were performed in accordance with Standard Methods for the Exam­
ination of Water and Wastewater and the Environmental Engineering
Laboratory Manual [11, 41].
The Environmental Engineering Laboratory
Manual methods are' similar to those in Standard Methods.
Materials
Feedwater.
Bozeman municipal wastewater.
Secondary effluent
with characteristics typical of those in Table 9,was filtered, as
shown in Figure 13, prior to use.
Synthetic wastewater.
water, simulated wastewater.
Ammonium, chloride.added to Bozeman tap*Typical characteristics of the synthetic
wastewater are shown in Table 10.
ClliioptLlolite.
Previously crushed and sieved 20 x 50 mesh
(I). S . Standard Sieve) clinoptilolite from Hector, California'deposits
was used in all studies.
64
TABLE 8
ANALYTICAL METHODS
Analysis
Alkalinity
Ammonia
Description
a
EELM- Potentiometric titration to pH 4.3
with 0.020 N H-SO.
2 4
EElM- Direct nesslerization. Measured
colorimetrically with Bausch and Lomb
Spectronic 20. Wastewater samples required
distillation prior to nesslerization.
Calcium
EELM- Titrimetric with EDTA. Hydroxy
napthol blue indicator.
Chloride
EELM- Titrimetric with Hg(NO3)2 .
Diphenylcarbazene indicator.
Magnesium
EELM- Total hardness value less calcium
value.
Nitrate
EELM- Phenoldisulfonic method. Measured
colorimetrically with Bausch and Lomb
Spectronic 20.
Nitrite
SM- Diazotization method. Measured
colorimetrically with Bausch and Lomb
Spectronic 20.
Organic Nitrogen
EELM- Kjeldahl digestion, distillation,
then direct nesslerization. Measured
colorimetrically as ammonia with Bausch
and Lomb Spectronic 20.
pH
EELM- Potentiometrically. Measured with
Corning pH meter.
Potassium
EELM- Flame spectrophotometry. Measured
with Hitachi Perkin-Elmer Spectrophoto­
meter .
Sodium
EELM- Flame spectrophotometry. Measured
with Hitachi Perkin-Elmer Spectrophoto­
meter.
65
TABLE 8 (CONTINUED)
Analysis
Total Hardness
Description 3
EELM- Titrimetric with EDTA. Erichrome
black i-methyl red indicator.
aSM refers to Standard Methods [11]. EELM refers to
Environmental Engineering Laboratory Manual [Al].
66
TABLE 9
AVERAGE COMPOSITION OF UNFILTERED BOZEMAN SECONDARY EFFLUENT *
Constituent
Value
Total Solids, mg/1
Total Dissolved
Total Suspended
830
770
60
Settleable Solids, ml/1
b
BOD (5 day, 20°C), mg/1
62
Total Nitrogen, as N mg/1
Organic
Ammonia
Nitrites
Nitrates
20
b
20
b
b
Alkalinity, as CaCO^ mg/1
240
pH, units
7.3
Temperature, eC
12
Sodium, mg/1
60
Magnesium, mg/1
16
Calcium, mg/1
50
Potassium, mg/1
10
Total Cation, meq/1
4.41
Total Anion, meq/1
4.50
Composition determined from two samples August 2,1974.
Value is approximately 0.
Primary
Clarifier
Activated
Sludge
Secondary
Clarifier
Chlorination
O'
'-j
Polyethylene
Reservoir
Dual Media
Filter
Polyethylene
Reservoir
Ion Exchange
Columns
0 —
1 —
2 —
3 —
PUMPS
Jabsco Water Mule
Sigmamotor Peristaltic
16 Channel Tubing
FIGURE 13: SCHEMATIC ILLUSTRATION FOR COLUMl EXHAUSTION WITH BOZEMAN WASTEWATER
68
TABLE 10
TYPICAL CHARACTERISTICS OF THE SYNTHETIC WASTEWATER
PRODUCED FROM BOZEMAN TAPWATER a
Constituent
Concentration, mg/1
Original
Ammonia as Nitrogen
0.4
Fortified
30. - 33.0
Sodium
10.0
10.0
Calcium
20.0
20.0
Magnesium
11.0
11.0
Potassium
1.0
1.0
Total Cation, meq/1
1.44
3.78
Total Anion, meq/1
1.53
3.87
a
Characteristics were derived
f r o m two
sets of
data.
CHAPTER V -
LONG. TERM STUDIES
O
The practical use of clinoptilolite for ammonia removal depends •
on its performance after many cycles of exhaustion and.regeneration.
Previous investigations by Koon and Kaufman [29], Mercer et^ al [15,
33, 34] and Braico [17] indicate the following potential, problems of
long term operation:
(1)
Weight loss of clinop.tilolite due to mechanical and
chemical attrition.
(2) ■ Gradual.reduction in ammonia exchange capacity..
(3)
Organic fouling of the clinoptilolite.
Effect of Regenerant pH on Attrition Rate
Procedure.
Five ion exchange columns were exhausted with
filtered Bozeman secondary effluent under the conditions listed in.
Table 11.
The column regenerated at pH.7.6, the natural pH of the
topwater, served as a control.
•minimize bed clogging.
for 100 cycles.
The. columns were exhausted upflow to
The columns were exhausted arid regenerated
After 25.,. 50, 75, and 100 cycles, the columns were
dismantled and the clinoptilolite removed,. The ammonia exchange
capacity was.determined for the.columps regenerated at pH 11.0. and
12.5 but.nrit for the columns regenerated at pH 7.6, 11.5, and 12.0.
Heating the clinoptilolite. to 600°C for one hour removed organics
70
TABLE 11
OPERATING CHARACTERISTICS FOR THE LONG TERM ATTRITION STUDY
Item
Columns, number
Value
5
Cllnoptllollte weight, g
20.00
Depth of zeolite bed, cm
3.5
Total cycles, number
100
Exhaustion rate, BV/hr
200(upflow)
Regeneration rate, BV/hr
Regenerant pH, units
Regenerant volume, BV
Regeneration level
NaCl, lb/ cu ft zeolite
NaOH, lb/ cu ft zeolite
pH 7.6
pH 11.0
pH 11.5
pH 12.0
pH 12.5
Influent
Cycle time, hr
Exhaustion
Regeneration
2 0 (downflow)
7.6,11.0,11.5,12.0,12.5
20
32.0
0
0.17
0.32
0.66
1.63
Bozeman wastewater,
as shown in Table 9
5.0
1.0
71
so an accurate weight determination could be made.
The columns were exhausted at 200 BV/hr upflow to simulate the
effects to the bottom zone of the exchanger bed which would likely
develop the maximum amount of attrition.
large scale application a.re 36 in deep.
Ion exchange columns for
A flow rate of 15 BV/hr
results in 5.6 gpm/sq ft through the zeolite bed.
So 200 BV/hr
(1.5 gpm/sq ft), the maximum flow rate attainable by this system,
was chosen to simulate the bottom layer.
A five hour exhaustion
time was chosen because it resulted in the most convenient cycle
time.
Exchange columns are exhausted about 6 to 8 hours during
single column operation and 8 to 11 hours during series column
operation in practical application.
.
The simulated and the actual
mechanical attrition rates would be proportional to the ratio of the
exhaustion cycle times.
The mechanical attrition was minimized during regeneration by
reduced particle movement (downflow operation), and lower flow rate
(20 BV/hr).
The chemical attrition due to exposure to caustic
solutions causes weight loss during regeneration, thus the regener­
ation cycle time is of more importance.
In practice regeneration
cycle time varies from I to 2 hours, depending on flow rate.
.Regeneration cycle time was one hour for this study.
Results.
The results of the attrition tests are shown in.
72
Figure 14.
The mechanical attrition due to the friction between the
clinoptilolite particles was the weight loss in the control column.
The chemical attrition due to the caustic regenerant was the weight
loss difference between each exchange column and the control.
These
results are similar to those of previous investigations, i.e., high
regenerant pH resultsin high attrition rate.
The small difference
in the attrition rates at pH 11.0 arid 11.5.could be due to ,experi•m ental error.
The clinoptilolite particles could have weaker zeolite surface
bonds which are readily broken.
The same type of structural bonds,
farther from the particle surface, are more protected and not so
readily broken.
This would result in high initial weight loss with
•
a, decrease in the weight loss rate during cyclic exposure, as were the
results of this study.
Koon and Kaufman concluded that harder, more
caustic resistant impurities, such as quartz and feldspar, remaining
in the clinoptilolite would, also reduce the attrition rate [29].
These impurities would remain in the exchanger bed while the
clinoptilolite is being broken into smaller pieces'and washed away,
hence the results would be similar.
No tests were made in this study
to determine the presence of impurities or their weight loss.
■ ;
There does not appear to be any.leveling off of the attrition
rate at 100 cycles.
The attrition rate would gradually reduce if the
curves in Figure 14 were extended, until 0.15% per cycle is reached.
73
Weight Lose,
%/cycle
0.90
Number of Cycles
FIG U RE 1 4 :
COMPARISON OF TOTAL A T T R IT IO N RATES FOR
REGENERANT pH LEVELS OF 7 . 6 ,
1 2 .5
1 1 .0 ,
1 1 .5 ,
1 2 .0 ,
AND
OVER 1 0 0 CYCLES OF EXHAUSTION AND REGENERATION
74
This appears .to be the minimum attrition rate possible.
These results indicate that regeneration at a pH of 11.5 is the
maximum caustic level for the minimum attrition rate, at lower pH the
attrition rate does not decrease significantly.
The total attrition
rate for the high weight loss zone will be 0.32, 0.31,.0.40, and 0.56
per cent per cycle for pH levels of 11.0, 11.5, 12.0, and 12.5,re­
spectively.
The rate of mechanical attrition was relatively stable
at 0.15 per cent per cycle.
These results indicate zeolite replace­
ment costs are significant.
This bottom zone will be completely
worn away in 180 to 320 cycles. . The attrition rates for the rest of
the exchanger bed are probably less than those predicted by Koon
and Kaufman [29].
Reduction of Ammonia Exchange Capacity during Cyclic Operation
Ion ,exchange materials'can'lose their effectiveness due to the
accumulation of materials on their surfaces and in their pores.
This
acctnnuIutlon of material or fouling, is the result of adsorption of
coJ. Ioicls or h igh molecular weight species, which are not completely
removed during the normal back wash and regeneration steps.
The sub­
stances that could foul clinoptilolite are iron, clay, manganese,
grease, and other organic materials.
The accumulation of materials
reduces exchanger effectiveness by reducing the fate of ion exchange
and the blocking off of exchange sites [30]. 'Exchanger fouling be­
75
comes apparent with less column throughput, I. e., loss of exchange
capacity.
Procedure.
tfhe exchange capacity was determined during the
attrition studies, after 25 cycles for all samples. At 50, 75, and
IOO cycles, the exchange capacity was determined only for the samples .
regenerated at pH 11.0 and 12.5.
As. it was necessary to heat the
o
'
■ ■
samples to 600 C to destroy organics so that the weight could be
accurately determined, the full' exchange capacity was subsequently ■
returned to the samples every. 25 cycles.
These tests were performed
to determine the permanent loss of exchange capacity during long term
use.
Ariother series of exhaustion and regeneration cycles were per­
formed to determine the temporary short term loss of exchange capacity.
One exchange column, containing 20.0 g clinoptilolite, was exhausted
and regenerated under the conditions listed in Table 11.
erant was a 0.25M
weighing
NaCl solution at pH 11.5.
The regen­
Samples of clinoptilolite
1.5 gram (wet) were removed from the column after 5, 10,
15, 20, 25, and 30 cycles.
The ammonia exchange capacity of the wet
samples was determined as shown in Table 7.
The samples were then
o
heated to 600 C for one hour for accurate determination of their
weight.
Results. The results of the long term cyclic exhaustion on ■
.
76
exchange capacity are shown in Table 12,
The results of short, term
cyclic exhaustion and regeneration are shown in Figure 15.
After 100
cycles, there was a permanent loss of exchange capacity of 0.04 meq/g
or 2.3%.
The regenerant pH did not affect the loss of exchange '
capacity as the loss' was similar the pH 11.0 and 12.5, as shown in
Table 12.
The clinoptilolite had 0.20 meq/g or 11 per cent temporary
loss of exchange capacity after 20 cycles, as shown in Figure 15.
The clinoptilolite must have been fouled by organics as heating
restored the exchange capicity.
Heating samples at 100°C for one
hour resulted in partial restoration, while heating samples at 600°C
resulted almost complete restoration.
The heating to 600°C caused a
5% weight loss to the clinoptilolite and changed its color from light
green to light brown.
The inability to restore the complete ammonia
exchange capacity could be due to substances, such as iron and.
manganese, irreversibly blocking pore channels.
Impurities in the
clinoptilolite contributing to the exchanger mass but not the exchange
capacity would also cause the exchange capacity to appear reduced
after 100 cycles by ■ occupying.an increased fraction of. the zeolite
mass.
Heating clinoptilolite to restore exchange capacity would not
be feasible for large.scale application.
Organic fouling.causes significant temporary reduction of
ammonia exchange capacity to clinoptilolite. ' Clinoptilolite did not
have significant permanent loss of ammonia exchange capacity at. 100
77
TABLE 12
EFFECT OF LONG TERM CYCLIC EXHAUSTION ON
AMMONIA EXCHANGE CAPACITY 8
PH
11.0
12.5
Number of
Cycles
Exchange Capacity,
meq/g
0
1.80
25
1.78
50
1.77
75
1.79
100
1.76
0
1.80
25
1.79
50
1.77
75
1.77
100
1.76
The exchange capacity was not determined for columns
regenerated at pH 7.6 , 11.5 , 12.0 .
Ammonia Exchange Capacity, milliequivalents/grai
78
Number of Cycles
FIGURE 15: THE EFFECT OF SHORT TERM CYCLIC EXHAUSTION
ON AMMONIA EXCHANGE CAPACITY
79
cycles of exhaustion and regeneration.
CHAPTER VI
THE EFFECT OF TEMPERATURE ON EXCHANGE EFFICIENCY
Two previous investigations sought to determine if column
operation at room temperature (23°C) could be used to predict oper­
ation at low temperature.
Braico concluded low temperature (12.5°C)
exhaustion and regeneration was nearly the same as that at room
temperature [17].
Slightly more feedwater was treated at low temper­
ature than at room temperature.
due to feedwater variance.
conclusion, namely:
He concluded this might have been
McLaren and Farquhar drew a similar
low temperature exhaustion could be approximated .
by room temperature exhaustion data [32].
They drew their con­
clusion partially from the data given in Table 13.
There was no
explanation for the results: 270 BV feedwater to breakthrough at
12°C, but 140 BV feedwater to breakthrough at 2°C.
As a result of these conflicting data, it was deemed appropriate
to investigate the temperature effects on ammonia exchange efficiency.
Procedure
An exchange column containing 200.0 grams of clinoptilolite. .
was exhausted with the synthetic wastewater listed in Table 10 under
the conditions listed in Table 14.
made at each temperature.
Only two exhaustion runs were
The equipment and feedwater was stored in
an environmental control cabinet to maintain a constant temperature
81
TABLE 13
RESULTS FROM MCLAREN AND FARQURAR TEMPERATURE EFFECT STUDY
(AFTER MCLAREN AND FARQURAR,1974 [32])
Value
Characteristic
Feedwater Temperature, 0C
2
12
Ammonia Concentration, mg/1 NH^-N
14.0
14.0
Exhaustion Rate, BV/hr
13.33
13.33
Volume Treated to Breakthrough, BV
(2.0 mg/1 NH 3-N)
140
270
82
TABLE 14
OPERATING CHARACTERISTICS FOR THE TEMPERATURE EFFECTS STUDY
Item
Value
Cllnoptllolite weight, g
200.0
Depth of zeolite bed, cm
42.0
Exhaustion rate, BV/hr
20(downf low)
Regeneration rate, bv/hr
15(upflow)
Regenerant pH, units
12.5
Regenerant volume, BV
30
Rinse volume, BV
15 (distilled)
Regeneration level
NaCl, Ib/cu ft zeolite
32.0
NaOH, Ib/cu ft zeolite
1.63
Influent
synthetic wastewater
as shown in Table 10
Operating temperatures,°C
4,23,33
83
during exhaustion.' The effluent ammonia concentration was measured
every 50 BV (2.5 hours) until 0.5 mg/& NH^ - N breakthrough.
Care
was taken to regenerate the column to the same condition, for every .
exhaustion run.
After regeneration the column was rinsed with
distilled water.
Results
The results of the exhaustion tests are shown in Figure 16.
Equation 14 shows the thermodynamic relationship between temper­
ature and equilibrium constant in which K = the equilibrium constant,
R = the universal gas constant, T = temperature, and AF = Gibbs Free
Energy.
.
■ AF = -RT In K
(14)
For a.given Gibbs Free Energy, which must be negative for the reaction
to proceed, the equilibrium constant will increase for a decrease in
temperature.
Based on information from McLaren and Farquhar, a
temperature drop will increase the ammonia capacity of clinoptilolite
when selectivcty coefficients are large [32].
A decrease in ex­
change capacity would be expected.for a temperature increase.
The overall effects of temperature bn ion exchange are complex
and not yet systematically studied. • In general, diffusion coefficients
in ion exchangers' increase with increasing temperatures? ion-pair
Effluent Ammonia Concentration, mg/1 NH
84
0.70
0.50
.
Symbol
0.00
Throughput, Bed Volumes
FIGURE 16: BREAKTHROUGH CURVES FOR 4, 23, AND 33 C
INFLUENT SYNTHETIC WASTEWATER
formation, and solvation are usually ,discouraged by -an ■.increase in '
temperature, as shown in Figure.I' [25];
Hence the selectivity would-
be 'expected to decrease .with increasing .temperature.'.
'
■'
■■■■■■■
-
-•
'
.
"
■
■■
■■
.
.
Ames reported that the effect of a temperature increase on the '
equilibrium constants was practically neglibible'between. 25°C and 70°C
and exchange isotherm's at the two temperatures •were nearly identical
[8],
But Figure'16'shows optimum clinoptildlite exhaustion at 2'3°C.
The observed phenomenon may be explained by an increasing fractidn of !
the ammonia in the N
H
form with decreasing solution temperature and
the ion exchange columns, may. not be in true equilibrium during
exhaustion.
There was a 16 percent decrease in volume treated to break­
through at 33°C as compared to exhaustion at .23 C.
There was a '12.5
percent decrease in volume treated to. breakthrough, at 4.C as compared.
O
to. exhaustion at 23 C.
There is only about a 4 percent increase in-
the percentage of ammonia in the N H . - form.
* O'
' ''
There were also higher
o
initial leakages, of ammonia- for the .4 C .and 33 C '.exhaustions .
The 12.5 to 16 percent decrease in volume treated for.lower
and higher,temperature exhaustion could be due partially to experi­
mental error.,
By using. Figure 16, low and high'temperature exhaustion
CHAPTER VII
OPTIMIZATION OF REGENERATION COST
The total process cost of ammonia removal with clinoptilblite •
depends on many factors, one of which is the cost of regeneration of
the exhausted zeolite.
The cost of regeneration depends on such
factors as chemical cost, zeolite replacement rate, and regenerant
recycle or disposal after initial use.
Regeneration costs in this
study are for a 10 MGD plant.
Chemical costs
Regeneration cost was previously optimized using sodium chloride
with either lime (calcium oxide) or sodium hydroxide as the base
[29].
Regeneration is much more efficient with sodium ions than
calcium ions.
sodium ions.
Sodium chloride is the
least expensive source of
There is little change in performance between the use
of sodium hydroxide or lime as the b a s e
pH [29].. Lime- is much less expensive.
to increase the regenerant
Although the use of ..lime
increases the calcium concentration in the regenerant, this has no
effect on regeneration efficiency [29] .. The chemical prices listed
in Table 15 were obtained from Van Water and Rogers, I n c . chemical
distributers in Billings, Montana [24].
Transportation charges were.
for shipping the chemicals: in bulk 140 miles.to Bozeman, Montana.
a7
TABLE 15
COST OF CHEMICALS AND CLINOPTILOLITE
Materials
Cost
Sodium Chloride
Base price, $/ton
140 miles transportation, $/ton
Total cost, $/ton
13.00
11.00
24.00
Lime (Calcium Oxide)
Base price, $/ton
140 miles transportation, $/ton
Total cost - 93% pure, $/ton
Total cost per unit of
CaO ,
$/ton
35.00
21.00
56.00
60.25
Sodium Hydroxide (50% liquid)
Base price, $/ton
140 miles transportation, $/ton
Total cost, $/ton
Total cost per unit of
NaOH,
$/ton
125.00
46.00
171.00
342.00
Clinoptilolite a
Base price, $/cu ft
1200 miles transportation, $/cu ft
Crushing and storage
, $/Cu ft
Total cost per unit of
product, $/cu ft
3.50
0.81
0.87
5.18
aBased on 50% recovery of clinoptilolite as 20 x 50
mesh material.
bUpdated to January,1975 from estimates by Koon and
Kaufman [29].
Clinoptilollte Costs
Clinoptilolite is marketed by the Baroid . Division of the National
Lead Company, Houston, Texas.
It is only available in the -4 mesh
size and the -28 mesh size [38].
As 20 x 50 mesh particles have been
determined as having the most desirable qualities for ion exchange,
crushing the -4 mesh size is necessary.
The -4 mesh size costs
$75/ton ($1 .75/cu ft) f.o.b. Newberry, California for minimum ship­
ments of 35 tons.
The cost of crushing and storing 20 x 50 mesh from
-4 mesh particles, updated from a previous study, was estimated as
$0.87/ou ft[29].
Approximately 50% of the clinoptilolite would be re­
covered as 20 x 50 mesh particles [29].
It was assumed the clinop­
tilolite would be shipped to Bozeman, then crushed.
The total cost
per cubic foot of clinoptilolite is shown in Table 15.
It was assumed the bottom 4 inch clinoptilolite layer or 6 per
cent of a 12 ft x 15 ft x 6 ft exchange reactor would have the high
attrition rate predicted in the previous chapter.
The rest of the
zeolite bed (94 per cent) would have an attrition rate approximately
that predicted by Koon and Kaufman, as listed in Chapter III [29].
Regenerant Renovation Costs
A 10 MGD treatment plant would require ammonia stripping facili­
ties capable of treating approximately 100,000 gallons of regenerant
every six hours.
The total treatment costs of ammonia stripping are
99
estimated as $0.08/1000 gallons from Figure 2;.
:
Regenerant chemical requirements
■ Koon and Kaufman have previously optimized regenerant volume,
N a C l •concentration, and pH; as shown in Figure 17 [29].
The amount
of chemicals for initial regenerant preparation, shown in Table 16,
is derived from Figure 17.
The ammonia exchange capacity of the
clinoptilolite is 0.44 meq/g (9.4 eq NH.'- N/cu ft), as ,predicted
from Figure 10 for an influent wastewater cationic strength of 0.006
moles/liter.
T h e .exchange capacity of the clinoptilolite would be
completely exhausted, during series column operation.
Two schemes were considered for regeneration:,: (I) ' The regen­
erant solution is used once and disposed, and (.2) The,regenerant is.■
recycled after ammonia stripping. When the regenerant solution is
used once and disposed, for solution preparation one simply adds, the
amount of chemicals from Table 16 to achieve the desired pH and NaCl
concentration.
One proceeds similarly for the initial regenerant
preparation when the regenerant is to. be recycled.
Regenerant makeup ■
after ammonia stripping is,more complex.
When the regenerant is reused the pH will drop both as the
ammonia in the clinoptilolite is absorbed by the regenerant and as the ■
regenerant is ammonia stripped.
From,data of ammonia stripping
operations evaluated by Culp and Culp [18] and laboratory experiment- -
Ammonia
95%
Volume Regenerant Required for
Elution, gal/eq NH3-N Removed
pH 1 1 . 5
VO
O
pH 1 2 . 5
0.0
0.10
0.20
0.30
0.40
0.50
0.60
0.70
Salt Concentration, lb NaCl/gal
FIGURE 17: VOLUME OF REGENERANT REQUIRED FOR 95 PERCENT ELUTION (AFTER KOON
AND KAUFMAN, 1971 [29])«
91
TABLE 16
AMOUNT OF CHEMICALS REQUIRED FOR REGENERATION
Regenerant
Volume of Regenerant
Regenerant Required for
Composition, for 95% NH3-N Elution,
pH
lb NaCl
gal
11.5
12.0
12.5
gal
eq NH3-N removed
95% NH3-N Elution,
lb NaCl3
CU ft
lb Ca(0H)ab lb NaOHac
CU ft
CU ft
0.049
42.0
19.4
0.55
0.67
0.10
28.0
26.2
0.37
0.45
0.17
23.0
36.6
0.30
0.36
0.24
17.0
38.4
0.23
0.31
0.73
17.0
116.0
0.23
0.31
0.00
32.0
0.0
0.049
25.0
11.5
0.68
0.82
0.10
21.0
19.7
0.57
0.69
0.17
16.0
25.5
0.44
0.53
0.24
15.0
33.8
0.41
0.50
0.00
17.5
0.0
e
1.43
0.049
16.0
7.4
1.10
1.34
0.10
11.0
10.4
0.76
0.92
0.17
11.0
17.6
0.76
0.92
0.24
10.0
22.6
0.69
0.84
0.73
10.0
68.6
0.69
0.84
d
a9.4 equivalents NH^-N/cu ft Clinoptilolite
^0.0014, 0.0029, and 0.0073 lb C a (OH)^ / gal regenerant
required for pH 11.5, 12.0, and 12.5, respectively.
cO.00169, 0.0035, and 0.0087 lb NaOH/ gal regenerant
required for pH 11.5, 12.0, and 12.5, respectively.
^'^Regeneration with NaOH only.
1.03
92
ation, it was estimated 32, 50, and 57 percent of the original base
=
requirements are necessary after ammonia stripping to increase the
regenerant pH to 11.5, 12.0 and 12.5 respectively.
Sodium chloride requirements for calcium oxide regenerant reuse .
were estimated as the stoichiometric amount of ammonia eluted from the
zeolite during regeneration (9.4 eq NH 3 - N/cu ft).
Sodium chloride
requirements for sodium hydroxide regenerant reuse were estimated as
t h e 'stoichiometric amount of ammonia eluted from the zeolite less the
equivalent amount of sodium added as sodium hydroxide to restore the
original solution pH.
Koon and Kaufman predicted a loss of regenerant volume due to
incomplete regenerant removal from the exchanger bed after regeneration
and evaporation during ammonia stripping [29].
Regenerant chemicals
and volume would be lost when the regenerant remaining in the exchange
column is washed away during column rinsing.
would be lost during evaporation.
Only regenerant volume
This loss would require about 2
percent of the original regenerant chemical requirements for chemical
makeup and 5 percent makeup of regenerant volume.
The amount of '
chemicals required for cyclic regenerant makeup are shown in Tables
I7 and 18.
Regenerant mixing and storage costs
A regenerant storage tank would be required for regenerant reuse.
93
TABLE 17
amount
of
chemicals
required
Sodium Chloride
Calcium Hydroxide3
lb NaCl
cu ft zeolite
lb C a (OH)^
cu ft zeolite
0.049
2.17
0.203
0.10
2.52
0.137
0.17
3.04
0.111
0.24
3.12
0.075
0.73
7.04
0.075
0.049
1.78
0.374
0.10
2.19
0.313
0.17
2.48
0.242
0.24
2.90
0.225
0.049
1.58
0.682
0.10
1.73
0.471
0.17
2.09
0.471
0.24
2.34
0.428
0.73
4.64
0.428
Regenerant Composition
d
H
11.5
12.0
12.5
FOR CYCLIC CALCIUM OXIDE REGENERANT MAKEUP
lb NaCl
gal
a Calcium oxide would be added for regenerative
purposes; 1.00 lb calcium oxide produces 1.32 lb calcium
hydroxide in aqueous solution.
94
TABLE 18
AMOUNT OF CHEMICALS REQUIRED FOR CYCLIC
SODIUM HYDROXIDE REGENERANT MAKEUP
Regenerant Composition
Sodium Chloride
Sodium Hydroxide
lb NaCl
cu ft zeolite
lb NaOH
cu ft zeolite
pH
lb NaCl
gal
11.5
0.049
2.06
0.248
0.10
2.45
0.167
0.17
2.99
0.133
0.24
3.07
0.115
0.73
7.00
0.115
0.00
0.00
0.567
0.049
1.54
0.451
0.10
2.00
0.380
0.17
2.32
0.291
0.24
2.78
0.275
0.00
0.00
0.887
0.049
1.23
0.831
0.10
1.49
0.570
0.17
1.85
0.570
0.24
2.12
0.520
0.73
4.42
0.520
12.0
12.5
95
This tank could be used for regenerant mixing as well.
The tank cost
is estimated as $0.00045/1000 gal wastewater as shown in Table 19.
Only a small regenerant mixing tank would be required if the regen­
erant is not reused.
This cost is estimated as $0.00042/1000 gal
wastewater, as shown in Table 20.
Costs without regenerant reuse
The costs of calcium oxide and sodium hydroxide regeneration
without reuse are shown in Figures 18 and 19.
A large proportion of
the regeneration cost is due to the sodium chloride necessary for
regenerant preparation as shown in the sample calculations in Appendix
B.
The cost of sodium chloride increases proportionately for each
cost curve in Figures 18 and 1 9 as the regenerant sodium chloride
concentration increases while other costs change only slightly.
accounts for the curves being approximately parallel.
This
The minimum
cost of regeneration without reuse, listed in Table 21, is with calcium
oxide at pH 12.5 with 0.049 lb NaCl/gal for $0,080/1000 gallons.
. The cost of regenerant brine disposal must also be considered
when the regenerant is not reused.
The cost of brine disposal by
evaporation pond is estimated as $0,019/1000 gallons as shown in
Table 22.
Costs with regenerant reuse
The costs.of calcium oxide and sodium hydroxide regeneration
96
TABLE 19
CAPITAL COST OF REGENERANT STORAGE TANK
(AFTER PAGE,1963 [37])
Characteristic
Value
Tank Capacity3 , gal
250,000
Total Capital Cost*1, $
Amortized Costc , $/year
Capital Cost, $/1000 gal wastewater
33,000
3,060
0.00045
Required for 25 gal regenerant/equivalent
NH^-N removed.
k Purchase and installation costs for a cypress
tank; Engineering News Record Construction Cost Index
of 2200, estimated January,1975.
c 15 year life at 7% interest.
97
TABLE 20
CAPITAL COST OF REGENERANT MIXING TANK
(AFTER PAGE,1963 (37])
Characteristic
Value
Tank Capacity *, gal
2,000.
Total Capital Coatb , $
31,000
Amortized Cost^ $/year
2,840
Capital Cost, $/1000 gal wastewater
0.00042
a Mixing tank required for operation without
regenerant reuse.
b Purchase and installation cost for stainless
steel tank; Engineering News Record Construction
Cost Index of 2200, estimated January, 1975.
c
15 year life at 7% interest.
98
0.90
Regeneration Cost, $/1000 gal
0.70
0.30
0.00
Regenerant Salt Concentration, lb NaCl/gal
FIGURE 18: COST OF CALCIUM OXIDE(LIME) REGENERATION
WITHOUT REGENERANT REUSE
Regeneration Cost, $/1000 gal
99
0.20
0.0
0.1
Regenerant Salt Concentration, lb NaCl/gal
FIGURE 19: COST OF SODIUM HYDROXIDE REGENERATION
WITHOUT REGENERANT REUSE
100
TABLE 21
MINIMUM COSTS OF CLINOPTILOLITE REGENERATION
Characteristic
Calcium Oxide without Reuse
Cost, $/1000 gal
0.082
pH 12.5
0.049 lb NaCl/gal
Calcium Oxide with Reuse
0.030
pH 11.5
0.24 lb NaCl/gal
Sodium Hydroxide without Reuse
0.112
pH 12.0
0.0 lb NaCl/gal
Sodium Hydroxide with Reuse
pH 11.5
0.24 lb NaCl/gal
0.038
101
TABLE 22
CAPITAL COST FOR BRINE DISPOSAL BY EVAPORATION POND
(AFTER PATTERSON AND BANKER,1971,[37])
Characteristic
Value
Brine Disposal Ratef MGD
2.4
Land Required, acres
300
Capital Costs, $
Land^
300,000
Earthwork and Construction
400,000
Pumping and Pipe System
Total Capital Cost, $
Amortized Costf $/year
Capital Costf $/1000 gal wastewater
50,000
750,000
69,000
0.014
3Required for regenerant volume of 32.0 gal/eq NH--N
removed, 10 MGD wastewater treatment plant.
^Land costs estimated as $1000/acre.
C15 year life at 7% interest.
^Engineering News Record Construction Cost Index of
2200, estimated January,1975.
102
with reuse are shown in Figures 20 and 21.
A. large proportion of the :
regeneration cost is due to the calcium oxide and sodium hydroxide
costs, as well as the sodium chloride costs.
As the regenerant pH
increases the amount of base necessary for regenerant makeup increases.
The cost curves in Figure 21 are spread more than those in Figure 20
because sodium hydroxide is six times as expensive as calcium .oxide.
The minimum cost of regeneration with reuse is $ 0 , 0 3 0 / 1 0 0 0 gallons
with calcium oxide at pH 1 1 . 5 using 0 . 2 4 lb NaCl/gal.
These costs are the cost of regeneration per cycle.
The cost of
the initial regenerant solution preparation is negligible if the.
regenerant is reused many times.
The calcium and magnesium ions in
the regenerant will precipitate as carbonates during ammonia stripping.
The chloride concentration of the regenerant may increase to the point
that it limits the solubility of the sodium chloride.
But this would,
likely.occur only after several hundred cycles of reuse.
The cost of
regenerant brine disposal is estimated as $0.00015/1000 gallons, when
the regenerant is recycled 300 times.
Cost of regeneration by computer program
The computer program in Appendix C was developed to estimate the
cost of regenerating clinoptilolite.
in the comments.
All program variables are defined
The optimum values of the regenerant pH, salt
concentration, and volume (program variables-PH, A, B), determined by
Regeneration Cost, $/1000 gal
103
pH 11.5
Regenerant Salt Concentration, lb NaCl/gal
FIGURE 20 : COST OF CALCIUM OXIDE(LIME) REGENERATION
WITH REGENERANT REUSE
104
Regeneration Cost, $/1000 gal
0.10
Regenerant Salt Concentration, lb NaCl/gal
FIGURE 21: COST OF SODIUM HYDROXIDE REGENERATION
WITH REGENERANT REUSE
105
Koon and Kaufman, are listed in Table 16.
It would be necessary to
develop new Tables 15 and 16 as chemical costs (G, Al, A J , AQ) and
tapwater chemical composition (C, AP) would likely change.
It is also
necessary to know the ammonia concentration (F) of the wastewater.
If
the ion exchange reactor dimensions change it will be necessary to reestimate the fractions of the zeolite beds (AN, AO) with the two
attrition rates (D, E ) .
All other program variables
AM, AR, AS, AT, AU, A V , AW) remain constant.
regenerant conditions and
costs.
(D, E , H, A K , AL,
The program lists the
REFERENCES
1.
Aerojet General Corporation.
Reverse Osmosis Renovation of
Municipal Wastes. Dept. No. ORD 170 (Washington, 0. C.:
Federal Water Quality Control Administration, 1969).
2.
American Water Works Association. Water Quality and Treatment.
3rd E d ., (New York: McGraw-Hill Book Company, 1971) ;
3.
Ames, L. L. "The cation sieve properties of clinoptilolite."
Ammerican Mineralogist 4 5 :689-700 (1960).
4.
Ames, L. L. "Cation sieve properties of the open zeolites
chabazite, mordenite, erionite, and clinoptilolite."
American Mineralogist 4 6 :1120-1131 (1961).
5.
Ames, L. L. "Effect of base cation.on the cesium kinetics of
clinoptilolite." American Mineralogist 47:1310-1316 (1962).
6.
Ames, L. L.
"Mass action relationship's of some zeolites in
the region of high competing cation concentration". American
Mineralogist 4 8 :868-882 (1963) .
7.
Ames, L. L. "Synthesis of a clinoptilolitelike zeolite."
American Mineralogist 4 8 :1374-1380 (1963) .
8.
Ames, L. L. "Some zeolite equilibria with alkoli metal cations."
American Minerologist 4 9 :127-145 (1964).
9.
Ames, L. L. "B eolitic removal of ammonium ions from agri­
cultural and.other wastewaters." 13th.Pacific Northwest
Industrial Waste Conference Proceedings. (Pullman, Washington:
Washington State Univeristy, April, 1967)
10.
Omphett, C . B .
1964) .
Inorganic Ion Exchangers, (New York: Elsevier,
11.
APHA,' AWWA, WPCF. Standard Methods for the Examination of Water
and Wastewater, 13th E d ..{New York: American Public Health
Association, 1971) .
12.
Barrer, R. M., and M. B . Makki, "Molecular sieve sorbents from
clinoptilolite". Canadian Journal Chemistry 42:1481-1487
(1964).'
----- :
—
:---------------------
107
13.
Barrier, R. M., R. Papadopoulas, and L.V.C. Rees, "Exchange of ,
sodium in clinoptilolite by organic cations." Journal Inorg.
Nucl.Chemistry 29:2047-2063 (1967).
14.
Barth, Edwin F . "Perspectives on wastewater treatment processes'physical-chemical and biological.'/ Journal Water Pollution
Controll Federation 43flO ;2189-2194 (Nov. 1971).
15.
Battelle Memorial Institute. Ammonia Removal from Agricultural
Runoff and Secondary Effluents by Selected Ion Exchange.
Report no. TWRC-S (Cincinnati, Ohio: Federal Water Pollution
Control Administration, March, 1969) .
16.
Bauman, R. E. "Nitrogen Control in the Midwest." Presented
at EPA Design Seminar, (Kansas City, Missouri 1971).
17.
Braico, Robert D . . Ammonia Removal from Recycled Fish Hatchery
.Water. Masters Thesis (Bozeman, Montana: Montana State
University, 1972) .
I
18.
Culp, Russell L. and Gordon L. Culp. Advanced Wastewater
Treatment, (New York: Van Nostrahd Reinhold Company, 1971).
19.
Dorfner, Konrad, Ion Exchangers, Properties and Application,
(Ann Arbor, Michigan: Ann Arbor Science Publishers Inc.,
1972) .
20-. . Eckenfelder, W. W." Jr. Water Quality Engineering for Pro­
fessional Engineers, (New York: -Barnes and Noble, 1970).
21.
Eliassen, Rolf and George Tchobanoglous, "Removal of nitrogen
and phosphorus from wastewater." Environmental Science and
Technology 3(6) .•536-541 (June 1969) .
22.
Engineering News Record 193(15):28 (October 3, 1974).
23.
Evans, Sheldon, "Nitrate removal by ion exchange." Journal
Water Pollution Control Federation 45(4):632-636 (April
1973) .
.
24.
Glueckert, Ben, Technical Coordinator, VanWaters and Rogers,
Inc., Billings, Montana^ Private communication (October 28,
1974) .
25.
Helfferich, Friedrich, Ion Exchange, (New York: McGraw-Hill,;
1962) .
I
108
26.
Howery, Darryl G., and Henry C. Thomas, "Ion exchange on the
mineral clinoptilolite." Journal Phys. Chemistry 69(2) :531537, 1965.
27.
Ionics, Incorporated.. The Electro-oxidation of Ammonia in
Sewage to Nitrogen. Eept. No.I70IOEED (Washington, D. C.:
Environmental Protection Agency, 1970).
28.
Kepple, Larry G. "Ammonia removal and recovery becomes
feasible." Water and. Sewage Works 121(4):42-43 (April 1974).
29.
Koon, John and Warren Kaufman. Optimization of Ammonia Removal 1
by Ion Exchange using Clinoptilolite. SERL REPORT No. 71-5
(Berkeley, California: University of California, 1971).
30.
Kunin, Robert, Elements of Ion Exchange.
Publishing Corporation, 1960)..
31.
McGauhey, P. H. Engineering Management of Water Quality.
(New york: McGraw-Hill Book Company, 1968) .
(New York: Reinhold
32.. McLaren, James and Grahame Farquhar. "Factors affecting ammonia
removal by clinoptilolite." Jour. Env. Eng. D i v ■, Proc.
Amer. Soc. Civil Engr.. EE 4(99):429-445 (August .1973).
33.
Mercer, B . W., L. L. Ames, C . J . Touhill, W. J. Van Slyke, and
R. B . Dean. "Ammonia removal from secondary effluents by
selective ion exchange." J . Water Pollution Control Fed.
i2:R95-R107 (February 1970).
34.
Mercer, B . W., R. C. Arnett and R. B. Dean, "Optimization of
Column Performance for Ammonia Removal from Wastewater Using
Selective Ion Exchange." Wastewater Reclamation and Reuse
Workshop Proceedings. (Berkeley, California: University of
California, June 25-27, 1970).
35.
Metcalf and Eddy, Inc. Wastewater Engineering.
McGraw-Hill Book Company, 1972).
36.
Monti, Randolph P ...and Peter T . Silbcrman, "Wastewater system
alternatives: What are They...and What Cost?" Water and
Wastes Engineering ll(3):32-40 (March 1974).
(New York:
109
37.
Page, John S. Estimator's Manual of Equipment and Installation
Costs, (Houston: Gulf Publising Company, 1963).
38'.
Patterson, W. L. and R. F. Banker* Estimating Costs and Man- ■
power Requirements for Conventional Wastewater Treatment
Facilities^ ■ (Washington, D. C.: Environmental Protection
Agency, 1971).
3
9
Pettier, Robert. Transportation Department Coordinator,
Baroid Division, National Lead Company, Houston, Texas.
Private communication (October 25,. 1974) .
40.
Pound, C. E. and R. W. Crites. Wastewater Treatment and Reuse
by Land Application Volume I - Summary, Rept. No.660/273-0069. (Washington, D. C .: Environmental Protection
Agency, 1973).
41.
Reeves, Thomas G.
"Nitrogen removal: A literature review."
Journal Water Pollution Control Federation 44(10): 1895-1908
(October 1972).
42.
Banks, Robert L. Environmental Engineering Laboratory Manual,’
2nd E d ., (Bozeman, Montana: Montana State University, 1974).
43.
Sawyer, C. N . and P. L. McCarty.
Chemistry for Sanitary
Engineers, 2nd E d ., (New York: McGraw-Hill Book Company, 1967)
44. 1 Sawyer, Clair N. "Nitrification and Denitrification Facili­
ties." Presented at EPA Design Seminar, (Kansas City,
Missouri: 1971)
45.
Sherry, Howard S. The Ion Exchange Properties of Zeolites,
(New York: Marcel Dekker, 1969).
46.
Shindala, Adnan.' Evaluation of Current Techniques for Nutrient
Removal from Wastewater, (State College, Mississippi, 1971).
47.
Sims, Ronald and Linda Little.
"Enhanced nitrification by
addition of clinoptilolite to tertiary activated sludge
units." Environmental Letters 4(1):27-34 (1973).
'
48.
:
:
—
Smith, Robert.
"Cost of conventional and advanced treatment
of wastewater." Journal Water Pollution Control Federation
40(9):1546-1574 (September 1968).
HO
49.
Welch, Eugene B. and Demetrios E. Spyridakis. "Treatment
Processes and Environmental Impacts of Liquid Waste Disposal
■ on Soil." Fourth Environmental Engineers' ConferenceProceedings (Bozeman, Montana: Montana State University,
February 15-16, 1973).
I
appendix
112
APPENDIX A
GLOSSARY
Symbol
Definition
BOD
biochemical oxygen demand
BV
bed volume
0C
degrees, Centigrade
ENRCCl
Engineering News Record
Construction Cost Index
g
gram
gpm/sq ft
. gallons per minute per square
foot
hr
. hour
ID ’
inside diameter
in
inch
i
liter
lb
pound
M
molarity, moles per liter of
solute
meq
milliequivolent
MGD
million gallons per day
mg
milligram
a IzI
m^ is tjgg cation concentration of
the i
species, 2, is the va­
lence of the particular cation
milliliter
113
GLOSSARY (CONTINUED)
Symbol
Definition
m&
mill! .liter
N
normality, equivalents per liter
of solute
psi
pounds per square inch
QNH4 - N
concentration of the ionic
ammonia in the solid phase
(the exchanger)
114
APPENDIX B
I. Cost of calcium oxide regeneration with reuse
pH 12.0
0.17 lb NaCl/gal
16.0 gal regenerant/eq NH3-N removed
0.44 lb Ca(OH)^Zcu ft zeolite
1.00 lb CaO ■ 1.32 lb Ca(OH)g
; costs were estimated
using Ca(OH)2
NaCl Costs
IA
5.43 eq NH3-N x 0.129 lb NaCl
1000 gal
IB
2%
x
0.17 lb NaCl
gal
x
$0,012
lb NaCl
x
eq
x
$0,012
lb NaCl
16.0 gal
eq NH3-N
x
$0.0083
1000 gal
5.4:3 eq N H - N
1000 gal
$0.00354
1000 gal
Ca(OH)2 costs
2A
2B
50%
2%
x
0.44 Ib-Sa-(OH)„
x
cu ft zeolite
x
$0.0228
lb Ca(OH)2
x
0.44 lb Ca(OH)„
x
cu ft zeolite
x
$0.0228
lb Ca(OH)2
cu ft
9.4 eq
x
5.43 eq NHL-N
1000 gal
$0.00288
1000 gal
cu ft
9.4 eq
x
5.43 eq N H - N
1000 gal
$0.000115
1000 gal
Ammonia stripping costs
3.
$0.08
1000 gal
x
5.43 eg N H - N
1000 gal
x
16.0 gal
- $0.00694
eq NH3-N 1000 gal
115
Cllnoptllollte replacement costs
4A 6%
x
0.0040
x
5.43 eg NML-N
1000 gal
x
cu ft x
9.4 eg
$5.18
cu ft zeolite
=
$0.00072
1000 gal
4B 94%
x
0.0035 x
5.43 eg N H - N
1000 gal
x
cu ft x
9.4 eg
$5.18
cu ft
-
$0.00987
1000 gal
Storage tank costs
-
$0.00045
1000 gal
Total Cost
-
$0.0323
1000 gal
II. Cost of calcium oxide regeneration without reuse
pH 12.0
0.17 lb NaCl/gal
16.0
gal regenerant/eg NH3-N removed
0.44 lb Ca(OH)3Zcu ft zeolite
NaCl cost
I. 0.17 lb NaCl
gal
x
16.0 gal
eg NH3-N
$ 0.012
5.43 eg NH--N
1000 gal
-
$0.17710
lb NaCl
2. 0.44 lb Ca(QH )3
x
cu ft zeolite
$0.0228
lb Ca (OH)0
1000 gal
C U
ft
9.4 eg
5.43 eg N H -N
1000 gal
-
$0.00570
1000 gal
116
Cllnoptllollte replacement costs
3A 6%
x
0.0040
x
5.43 eg N H - N
1000 gal
x
cu ft x $5.18
9.4 eq
cu ft
-
$0.00072
1000 gal
3B 94%
x
0.0035 x
5.43 eq N H - N
1000 gal
x
cu ft x $5.18
9.4 eq
1000 gal
-
$0.0098
1000 gal
Mixing tank coat
4'
Total Cost
-
$0.00042
1000 gal
-
$0.1934
1000 gal
III. Cost of sodium hydroxide regeneration with reuse
pH 11.5
0.24 lb NaCl/gal
17.0
gal regenerant/eq NH3-N removed
0.31 lb NaOH/cu ft zeolite
NaCl costs
IA 2%
x
0.24 lb NaCl
gal
x $0.012
lb NaCl
IB ( 5.43 eq N H - N
1000 gal
cu ft x
9.4 eq
x
$0.012
lb NaCl
x
17.0 gal
x
eq NH3-N
5.43 eg N H - N
1000 gal
$0.00532
1000 gal
- t 0.31 lb NaOH x 5.43 eq N H - N x
cu ft
1000 gal
0 2 % + 2%) x 11.3 eg____
]) x 0.129 lb NaCl
lb NaOH
eq
$0.00735
1000 gal
117
NaOH costs
2A 2%
x
0.31 lb NaOH
C U
x
ft
5.43 eq N H - N
1000 gal
X
C U
$0,171
lb NaOH
2B 32% x
ft
X
9.4 eq
$0.00063
1000 gal
0.31 lb NaOH
C U
ft
x 5.43 eq NH--N
1000 gal
$0,171
lb NaOH
X
C U
ft
X
9.4 eq
-
$0.01008
1000 gal
Ammonia stripping cost
3 $0.08
1000 gal
x
5.43 eq N H - N
1000 gal
x
17.0 gal
eq NH
3"
N
$0.00738
1000 gal
Clinoptilollte replacement costs
4A 6% x 0.0031 x 5.43 eg N H - N x cu ft x $5.18 - $0.000558
1000 gal
9.4 eq
cu ft
1000 gal
4B 94% x 0.0025 x 5.43 eq NHn-N x cu ft x $5.18 - $0.00705
1000 gal
9.4 eq
cu ft
1000 gal
Storage tank cost
5. *
Total Cost
- $0.00045
1000 gal
- $0.0388
1000 gal
IV. Cost
of sodium hydroxide regeneration without reuse
pH 11.5
0.24 lb NaCl/gal
17.0
gal regenerant/eq NH3-N removed
0.31 lb NaOH/cu ft zeolite
118
NaCl cost
I. 0.24 lb NaCl
gal
x 17.0 gal
eq NH3-N
x
5.43 eq N H - N
1000 gal
$ 0.012
$0.26580
lb NaCl
1000 gal
NaOH cost
2. 0.31 lb NaOH
C U
x
ft
cu ft x
9.4 eq
5.43 eq NH--N
1000 gal
$0.171
lb NaOH
x
$0.0127
1000 gal
Cllnoptllollte replacement costs
3A
6%
x
0.0031
x
5.43 eq N H - N
1000 gal
x
3B
94% x
X
$0.000558
ft
1000 gal
0.0035 x
5.43 eq NH--N
1000 gal
$5.18
C U
ft
9.4 eq
$5.18
C U
C U
x
cu ft
9.4 eq
x
$0.00987
ft
1000 gal
Mixing tank cost
$0.00987
1000 gal
Total Cost
$0.3044
1000 gal
x
119
OOO OOOO o o o o o o o o o o o o o o o o o o o o o o o o o o
APPENDIX C
DlVlC MCCREADY CE590 NOVEMBER 1 , 1 9 7 4
OPTIMIZATION of CLINOPTILOLITE REGENERATION COST “
■ PCLNDS OF SODIUM CHLORIDE PER GALLON OF REGENERANT
■ GALLONS OF REGENERANT PER EQUIVALENT OF AMMONIA REMOVED
■ POUNDS OF CALCIUM HYDROXIDE REQUIRED PER CUBIC FOOT OF CL INOPTI LOLI TE
" ATTRITION RATE FOR HIGH WEIGHT LOSS ZONE OF ThE EXCHANGER
• 0 0 3 1 , . 0 0 4 0 , AND . 0 0 5 6 FOR PM 1 1 , 8 , 1 2 - 0 , AND 1 2 . 5 , RESPECTIVELY
E " ATTRITION RATE FOR MAJORITY OF THE EXCHANGER
• 0 0 2 5 , . 0 0 3 5 , An D . 0 0 8 5 FOR PH 1 1 . 6 , 1 2 . 0 , AND 1 2 . 6 ,RESPECTIVELY
F - EQUIVALENTS OF AMMONIA PER 1000 GALLONS OF WASTEWATER,20 MO/L NH3-N
O ■ CCST OF SODIUM CHLORIDE,DOLL ARS PER POUND
H • RECIPROCAL OF THE AMMONIA EXCHANGE CAPACITY OF CLTNOPTILOLITE,,
CUBIC FEET PER 9 . 4 EQUIVALENTS
Al • COST OF CALCIUM HYDROXIDE, DOLLARS PER POUND
AJ - COST OF CL I NOPT I LOL I TE, DOLLARS PER CUBIC FOOT
AK ■ COST OF AMMONIA STRIPPING,DOLLARS per 1000 gallons
AL ■ CONVERSION FACTOR ,
POUNDS PER EQUIVALENT CF SODIUM CHLORIDE
AM . ESTIMATE OF THE REGENERANT VOLUME LOST DURING REGENERATION
AN ■ ESTIMATE OF THE FRACTION OF THE ZEOLITE BEO WITH THE HIGH
ATTRITION RATE
AO ■ ESTIMATE OF THE FRACTION OF THE ZEOLITE BCD WITH THE ATTRITION RATE
PREDICTED BY
KOON ANQ KAUFMAN
AP • POUNDS OF SODIUM HYDROXIDE REQUIRED PER CUBIC FOOT OF CL I NOPTI LOLI TE
AQ ■ COST OF SODIUM HYDROXIDE, DOLLARS PER POUND
AR - FRACTION OF THE ORIGINAL AMOUNT OF BASE REQUIRED FOR
REGENERANT MAKEUP at PH 11*5
AS • FRACTION OF THE ORIGINAL AMOUNT OF BASE REQUIRED FOR
REGENERANT MAKEUP AT PH 12«0
AT ■ FRACTION OF THE ORIGINAL AMOUNT OF BAlE REQUIRED FOR
REGENERANT MAKEUP AT PH 1 2 - 5
AU • EQUIVALENTS PER POUND SODIUM HYDROXIDE
AV - CAPITAL COST OF REGENERANT STORAGE TANK, DOLLARS PER 1000 GAL
AU ■ CAPITAL COST OF REGENERANT MIXING TANK, COLLARS PER 1000 QAL
DIMENSION
PHllSI20),
PHl 2 0 12 0 1 ,
PH12BI20)
F • 5.430
G ■ 0* 0 1 2
H • 0.1063
AT - 0 . 0 2 2 8
AJ ■ 5 . 1 8
AK ■ 0 . 0 8
AL ■ 0 . 1 2 9
AM ■ 0 . 0 2
AN ■ 0*06
AC - 0 . 9 4
AQ " 0 . 1 7 1 0
"
A
B
C
D
120
AS
AT
ALi
Av
AU
C
C
C
• CiSC
■ C•57
■ 11 • 3
« C i COOAS
■ C i OOOAZ
CCST CF' SODIUM HVDROxIOF REOCKER AT 10N--REU8E
NRI TE 11 CS* 606 I
DC 10 I • 1<5
REAOI ICS, 900 I AaBi APi DaE
PHl ISI ! ) • A, H»AM, F* 0+ AP*AM,F,AQ*k+ ANvO#FvH»AJ* AO»E*FAH»AJ+ IF-I
I AP»F*Ha I Am+ AR IAAU I I *AL*0+ AX*F*B/1000A ARa APaF a Haa O ♦ AV
WRITEIIO816OOI PHl l SI I I a A
10 CONTINUE
CO PO I • I aS
RFAD(ICSiSOO) AaB aAPaDaE
PHl PCf I I- Aa Raamaf » 0 a aP a AMaP a AQa Ha ANaOaF aHaa J a aOa Ca F aHa AJA ( F - i
IAP aF a Ha I AMAARlAAUI I a ALaOa AKAFaB/IOOOA a S a APaF aHaa O ♦ AV
NR ITE I 1 0 8 a 6C1 I PHlPC I I I a A
PO CONTINUE
DO SC I ■ I a 6
RFAOHCSiSOCI Aa Ba AP a D1E
PHlPS I I I - AAHAAMAFAQ+ aP aaMaF aa Oa H* ANAOa F aHaa J+ aOaEaF a Ha AJ* | F . (
I APAFa Ha I AMAAR I AAU I I AALaO+ AKAFaB/ICCOA ATa APaF a Haa O ♦ AV
NRITEIICB aSOPI
PHlPSI I I aA
30 CONTINUE
CCST OF SODIUM HVDROXIOF REGENERATION""— NO REUSE
CO AC N " I A5
RFAOIICS aSOOI Ai Bi APi Di E
PHl l SI I I ■ AaBa FAG a APa Ha FAAQ A ANADa F a Ha AJ A AOaE*F a Ha AJ AAW
WRITE I 1 0 8 , 6 0 3 I PHl l SI I I iA
40 CONTINUE
DC SC I - I aS
REACI1C5 a9CCI AARiAPiDAr
PHlPC I I I ■ AaHaF ag aaP a Ha F aa O a ANADaF a Ha AJ a aOa Ea F a Ha Aj aaw
WRITE I 1 0 8 , 6C4 I PHlPO I I I AA
SO CONTINUE
CO 60
I - 1,6
READ I 1 0 5 , 9 0 0 I Ai Bi APi Di E
PHlPS I I I ■ A a B a F a G a A P a h a F a a O a ANAD a F a H a AJ a AO a F a F a H a AJ AAW
WRITE 1 1 0 8 , 6 0 5 I PHlPB I I I , A
6C CONTINUE
COST OF CALCIUM HYDROXIDE REGENERATION-— REUSE
WRI TE 11 0 8 , 6 0 7 I
CO 70 I " 1 , 5
R EAOH0 5 , SCOI Ai B, C O , E
121
c
1AK*F*B/1COO*C*H*F*M¥AR * av
WRI TE( I OSj 6C0 ) P H l l B ( I ) ,A
70 CONTINUE
DC 80
I - 1,4
RE* 0 ( 1 0 5 , 9 0 0 I A , B , C , 0 , E
PHl SC( I I » AM*A*B*F,o + C»H*F,AI, AM+ AN*0¥F*H*AJAA0*E»F¥H¥AJ¥F*AL¥QA
1AK¥F¥B/ 1000¥C¥H¥F#AI*AS ♦ AV
WRITE( 1 0 8 , 6 0 1 I PHl SO( I I , A
80 CONTINUE
DO 90
I ■ 1,5
RFAD 11 0 3 , 9 0 0 ) A , B , C D , E
PH1S5I I ) - AM¥A¥B¥F,o +C¥H¥F»A!,AM+ AN¥ D, F¥H¥AJ+AOa E¥F¥H¥AJ+F¥ AL¥Q +
I AK¥F¥B/ 1000¥C¥H¥F, A1, AT ♦ AV
WRl TEI 108, 60S) P Hl S S ( I ) ,A
SC CONTINUE
COST OF CALCIUM HYDROXIDE AFGENERATI CN--NO REUSE
CO 100 I - 1 , 5
READ ( 1 0 5 , 9 0 0 I A, B, C D , E
PHl l BI I I - AaBaF¥ Q ♦ Ca HvF a a I ♦ anada F vhv AJ ♦ A0¥E*F*Ha AJ +AW
WRITE I 1 0 8 , 6 0 3 I PHl l SI I I , A
ICC CONTINUE
DO HO
I - 1,4
READ 11 0 5 , 9 0 0 I A , Bi Ci D, E
PHlSO I I I - A , B a F a q ♦ C a H v F a a I + AN a D a F v h a AJ A AC a E a F a H a AJ ♦ AW
WRITE 1 1 0 8 , 6 0 4 I PHl SC( I I ,A
HO CONTINUE
CO ISO
I - 1,5
RF AC ( 1 0 5 , 9 0 C I A , B , C , 0 , E
PHlSS I I I - A a B a F a Q + C a H a F A A I + AN a D a F v h a AJ ♦ AO a F a F a H a AJ ♦ AW
WRITE I 1 0 8 , 6 0 5 I PHlSBI I I ,A
ISO CONTINUE
600 FORMAT I 5 x » ' REUSE REGENERATION COST AT PH 1 1 . 5 • * / 1 0 0 0 GAL-1, F10.
1 6 , BX, 'POUNDS NACl/GAI REGENERANT-' , F I O. 6 I
601 FORMAT I 5x« 'REUSE REGENERATION COST AT pH IS-O • « / 1 0 0 0 G A L - ' , F10.
1 6 , 5X, 1POUNDS NACL/GAL REGENERANT"', F10. 6 I
tCS FORMAT I bx, 'REUSE REGENERATION COST AT pH I S - S - */1000 GaL- ',Flo.
1 6 , SX, 'POUNDS NACL/GAI REGENERANT - ' , F10 . 6 I
6C3 FORMAT I 5 x , ' NO REUSE REGENERATION COST PH I l - S */1000 QAL- ', FlO •6
I i 5x,' PCUn DS NACL/GAL REGENERANT - ' , F l O . * )
604 FORMAT! 5 x , ' NO REUSE REGENERATION COST RH 1 2 - 0 «/1000 GAL-', Fl0.6
I, 5x.'POUNDS NACL/GAL REGENERANT - ' , F 1 0 . 6 I
6C5 FORMAT I b x « ' NO REUSE REGENERATION COST ph 1 2 - 5 */1000 QAL- ', F10.6
I , 5X, 'POUNDS NACL/GAL REGENERANT - ' , F 1 0 . 6 I
606 FORMAT I IX, ' COST OF SODIUM HYDROXIDE REGENERATION-' I
607 FORMAT I IX, ' COST OF CALCIUM HYDROXIDE REGENERATION-' I
STOP
EMO
3 1762 iuu I*+' ^
N378
M138
cop .2
McCready, David I
Ammonia removal from
secondary effluent by
selective ion exchange ...