Environmental Protection Technology Series
Physical-Chemical Treatment of
Raw Municipal Wastewater
I
55
\
5SSZ
-z.
UJ
CD
/
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Monitoring, Environ-
mental Protection Agency, have been grouped into five series. These
five broad categories were established to facilitate further develop-*
ment and application of environmental technology. Elimination of
traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The five series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the Environmental Protection Technology
Series. This series describes research performed to develop and demon-
strate instrumentation, equipment and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution.
This work provides the new or improved technology required for the
control and treatment of pollution sources to meet environmental
quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the Office of Research and Monitoring,
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
-------
EPA-670/2-73-070
September 1973
PHYSICAL-CHEMICAL TREATMENT
OF
RAW MUNICIPAL WASTEWATER
By
Dolloff F. Bishop
Thomas P. O'Farrell
Alan F. Cassel
Adolph P. Pinto
Contract No. 14-12-818
Project 11010 EYM
Program Element 1B2033
Project Officer
Dolloff F. Bishop
Advanced Waste Treatment Research Laboratory
National Environmental Research Center
Cincinnati, Ohio 45268
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.05
-------
ABSTRACT
Physical-chemical treatment of raw wastewater in a 50,000 to 100fOOO gpd
pilot plant consisted of two-stage lime precipitation with intermediate
recarbonation, filtration, pH control, ion exchange or breakpoint
chlorination for nitrogen removal and carbon adsorption. The complete
system with ion exchange removed 98% of the phosphorus, 95% of the
organics (COD) and 78% of the total nitrogen. With breakpoint chlori-
nation, the complete system removed approximately 98% of the phosphorus,
94% of the organics (COD) and 86% of the total nitrogen.
Lime treatment with approximately 300 mg/1 of CaO increased the waste-
water pH to about 11.5, removed approximately 96% of the phosphorus and
of the BOD, TOC and COD. Recarbonation with 120 mg/1 of CO and
with 5 mg/1 of Fe+++ as a flocculant reduced the pH to 10 and precipi-
tated excess Ca++ as CaCO . The CaCO3 was settled in a second settler.
Dual media filtration (18" of 0.9 mm coal over 6" of 0.45 mm sand)
decreased effluent suspended solids to less than 5 mg/1 and total phos-
phorus to less than 0.15 mg/1 of P. Addition of 10 mg/1 of chlorine to
the influent filter controlled biological growth within the filter and
produced filter runs of greater than 50 hours.
With extensive operator surveillance, the clinoptilolite exchange
mineral reduced the ammonia to less than 1 mg/1 as Ntft-N. Breakpoint
chlorination oxidized the ammonia to N leaving a residual NH -N
concentration of less than 0.4 mg/1. The total nitrogen residual for
the physical-chemical treatment with breakpoint chlorination varied from
2 to 3 mg/1 as N.
The 20 mg/1 of soluble BOD entering the granular carbon columns produced
anaerobic biological growth on the carbon. Carbon loadings as high as
0.3 pounds of dissolved TOC per pound of carbon was obtained during the
study with an effluent TOC of approximately 8 mg/1. Heavy biological
growth developed during the investigation which contributed to high
carbon losses during back wash and heavy H S production. Breakpoint
chlorination ahead of carbon adsorption minimized the biological
activity. The organic loading with breakpoint chlorination was
approximately 0.24 pounds of dissolved TOC per pound of carbon at an
effluent TOC concentration of approximately 10 mg/1.
This report was submitted in partial fulfillment of Project 11010 EYM
and Contract No. 14-12-818 by the Department of Environmental Services,
Government of the District of Columbia under the sponsorship of the
Environmental Protection Agency. Work was completed as of September 1971.
11
-------
CONTENTS
Page
Abstract
List of Figures
List of Tables
Acknowledgments
Sections
I Conclusions j
IT Recommendations 3
III Introduction 4
IV Experimental 7
Pilot Systems 7
Analytical Procedures 24
V Process Operations 26
Clarification and Filtration Operation 26
Ion Exchange Operation 29
Breakpoint Chlorination Operation 28
Carbon Adsorption Operation 32
VI Overall Performance of Physical-Chemical Treatment 40
Organic and Solids Removal 40
Phosphorus Removal 42
Nitrogen Removal 42
VII Adsorption Mechanism 49
VIII Costs 52
IX References 57
X Publications, Presentations and Patents QQ
Hi
-------
FIGURES
No. Page
1 Physical-Chemical Pilot Plant 8
2 Ion Exchange Process 10
3 Breakpoint Chlorination Process 12
4 Daily Removals of COD in Physical-Chemical Treatment 20
5 Daily Removal of Suspended Solids in Lime Clarification 21
6 Daily Removal of Phosphorus in Lime Clarification 22
7 Solids Production and Lime Addition 23
8 Ammonia Removal in Selective Ion Exchange with Four Hour 24
Operator Surveillance
9 Ammonia Removal in Ion Exchange with Continuous Operator 27
Surveillance
10 Daily Ammonia Removal in Breakpoint Chlorination 30
21 Daily Removal of COD in Carbon Adsorption 35
12 Daily Removal of TOC in Carbon Adsorption 36
13 TOC Loading on Activated Carbon 38
IV
-------
TABLES
tfo. Page
1 Pilot Plant Hydraulic Loading 17
2 Operating Variables for Two Stage Lime Clarification 18
3 Selective Ion Exchange and Residual Pollutants 25
4 Breakpoint Chlorination of Lime Clarified and Filtered 29
Raw Wastewater
5 COD Removal in Physical-Chemical Treatment of Raw 33
Wastewater
6 TOC Removal in Physical-Chemical Treatment of Raw 34
Wastewater
7 Organic Loadings on Activated Carbon 39
8 BOD Removal in Physical-Chemical Treatment of Raw 41
Wastewater
9 Suspended Solids Removal in Physical-Chemical 43
Treatment of Raw Wastewater
10 Phosphorus Removal in Physical-Chemical Treatment
of Raw Wastewater
11 Total Nitrogen Removal in Physical-Chemical Treatment 45
of Raw Wastewater
12 Nitrogen Removal with Ion Exchange 46
13 Typical Nitrogen Removal with Breakpoint Chlorination 43
14 Physical-Chemical Treatment Costs 52
15 Breakpoint Chlorination Costs for Lime Clarified 53
Raw Wastewater
16 Design Criteria for Cost Estimate 54
-------
ACKNOWLEDGMENTS
Walter W. Schuk of the Environmental Protection Agency developed the
control system described in the study and is acknowledged with sincere
thanks.
The operation of the pilot system with analytical support was performed
by the EPA-DC Pilot Plant staff under the supervision of the Chief
Operator, George D. Gray, Supervisory Engineering Technician, Robert A.
Hallbrook and Chief Chemist, Howard P. Warner.
The authors wish to acknowledge Battelle-Northwest of Rlchland,
Washington, for the use of their trailer-mounted selective ion
exchange system and to especially thank B.W. Mercer, Senior Research
Engineer, and R.C. Arnett, Research Engineer, of Battelle-Northwest
for the initial operation and the training of the pilot plant staff on
the exchange system.
-------
SECTION I
CONCLUSIONS
1. Physical-chemical treatment consisting of two-stage lime precipi-
tation with intermediate recarbonation, filtration, pH control,
selective ion exchange or breakpoint chlorination and carbon adsorption
removed approximately 92% of the TOC, 96% of the BOD, 95% of the COD,
97% of the suspended solids and 98% of the total phosphorus. With ion
exchange, the system removed 78% of the total nitrogen from the District
of Columbia raw wastewater. With breakpoint chlorination, the system
removed 86% of the total nitrogen. The pollutant residuals for seven-
teen months of treatment averaged approximately 8 mg/1 TOC, 6 mg/1 of
BOD, 16 mg/1 of COD, 4 mg/1 of suspended solids, 0.14 mg/1 of total
phosphorus as P. With ion exchange for nitrogen removal, the total
residual averaged 4.6 mg/1 of total nitrogen. With breakpoint chlorina-
tion for nitrogen removal, the total residual nitrogen averaged 2.8 mg/1
2. In the study, lime precipitation removed approximately 80% of the
TOC, BOD and COD, 88% of the suspended solids, 97% of the total phos-
phorus and approximately 30% of the total nitrogen.
3. Filtration of the lime clarified wastewater increased the overall
removal of pollutants by 1-9%. As examples, the suspended solids
removals increased from 88 to 97%, while the total phosphorus removal
increased from 97 to
4. With minimal operator surveillance, the ion exchange process located
ahead of carbon adsorption, removed 75% of the influent ammonia with an
average residual of 3.0 mg/1 of NH+-N. However, with careful operator
surveillance, the ion exchange process reduced the NH -N residuals to
less than 1 mg/1 (greater than 90% removal) and the total nitrogen
removal to over 85%. The ion exchange process also removed from 15 to
35% of the residual "soluble" organics from the lime clarified and
filtered wastewater and increased the cumulative organic removal to
more than 85%.
5. The breakpoint chlorination of ammonia to nitrogen was carried out
usually in a pH range of 6 to 8. The chlorination process with good
pH control oxidized 10-15 mg/1 of. NH -N to N gas. With good pH
control (steady flow) effluent NH^-N averaged less than 0.4 mg/1. For
the entire study, the effluent NH+-N averaged 0.46 mg/1. NO -N
produced as a by-product was approximately 0.6 mg/1. Essentially no
NCI was detected in the effluent at a reaction pH of approximately 7.0.
The dose weight ratio of Cl:NH -N required for the breakpoint was
approximately 9:1.
6. The experimental system with a 12 minute reactor time exhibited
control problems especially during operation with diurnal flow, and
-------
also with a high influent pH (9-10). Ammonia breakthrough occurred
during periods of poor process control. An in-line mixer with small
reaction detention time is recommended for improved process control
and efficient mixing.
7. NaOH addition neutralized the acid produced during chlorination to
maintain approximately a neutral reaction pH.
8. Chlorination did not appreciably oxidize the soluble organics in
the wastewater.
9. With virgin carbon, adsorption removed between 75 and 80% of the
organics entering the adsorption columns with average residuals of
3.7 mg/1 of BOD and TOC, and 8.2 mg/1 of COD during the first month.
With organic loading and biological growth, residuals after adsorption
increased to 8 mg/1 of BOD and TOC, and 19 mg/1 of COD for the month
before the first replacement of spent carbon. The pattern of the low
organic residuals after an activated carbon replacement gradually
increasing with increasing organic loading on the activated carbon
repeated itself through subsequent carbon replacement cycles.
10. For operation without chlorine addition to carbon column influent,
heavy anaerobic biological growth occurred and H S appeared (2-3 mg/1)
in the effluent. The biological slimes coated the carbon granules,
produced high pressure losses (25 psig across the first column in
24 hours) and caused excessive carbon losses during backwash.
11. The application of breakpoint chlorination ahead of carbon adsorp-
tion minimized the biological activity, eliminated the high pressure
losses across the first column (less than 10 psig in 48 hours) and
produced hydraulic operation in the carbon system comparable to that
obtained during earlier tertiary carbon treatment.
12. The organic loading on the activated carbon without chlorine in
the influent water was 0.3 Ib of dissolved TOC per Ib of carbon at an
effluent TOC of approximately 8 mg/1. The organic loading on the
carbon during breakpoint chlorination (chlorine in the influent to
adsorption) was 0.24 Ib of dissolved TOC per Ib of carbon at effluent
TOC of approximately 10 mg/1.
13. The backwashing sequence with high pH (11.5) water applied to the
adsorption system did not eliminate biological activity in the carbon.
Thus, determination of the effect of the biological activity on the
product quality requires additional work.
14. The solids production for the two-stage lime clarification was
a function of lime dose and wastewater content. Typical sludge
production for a lime dosage of 350 mg/1 was approximately 7.5 Ib of
solids per million gallons of treated water.
-------
SECTION II
RECOMMENDATIONS
1* In future work, the optimum design requirements for lime clarifi-
cation needs to be determined including maximum sedimentation rates and
the effect of reactor solids concentration, pH and temperature on the
rate of precipitation and flocculation of calcium carbonate, calcium
hydroxyl apetite and magnesium hydroxide from the wastewater.
2. In the second stage lime clarification, the effects of flue gas
(8% CO ) on the recarbonation flocculation and second stage sedimen-
tation requires further evaluation for proper system design.
3. In the ion exchange process, the brine system which may include
reuse or disposal of the brine solution requires further study. Air
stripping of the ammonia from the brine with subsequent absorption
from the air in phosphoric acid and precipitation as ammonium phosphate
for final disposal should be evaluated.
4. The long term attrition losses of the clinoptilolite exchange
mineral and the long term effects of repeated regeneration on the
exchange capacity of the mineral also needs to be determined.
5. The kinetics of breakpoint chlorination needs further study for
optimum breakpoint reactor design and control.
6. The effect of biological growth in carbon adsorption especially
on carbon losses and H S production needs further study. Alternate
methods other than chlorination for control of the H S production
and slime gorwth should be considered and evaluated. Possible
alternates include up flow carbon columns with oxygen addition or
nitrate addition in pressured downflow columns.
7. Solids handling characteristics such as thickening, dewatering
and lime recovery needs to be studied. Centrifugation for classifica-
tions (separation) of carbonate solids from non-carbonate solids and
the handling of non-carbonate solids in the centrate should be
included in the evaluation.
8. An automated control system for physical-chemical treatment
especially for the breakpoint chlorination process needs to be developed
9. Since the ion exchange and the carbon columns also act as filters,
operation without the dual-media filters should be conducted to deter-
mine whether losses in treatment efficiency occur. The filters should
be relocated to remove suspended solids in the effluent from carbon
adsorption and thus increase treatment efficiency.
-------
SECTION III
INTRODUCTION
In the early work on advanced treatment of secondary effluents/ the
product water from granular carbon treatment contained organic materials
apparently refractory to carbon adsorption (1). Addition of chemical
clarification to the tertiary treatment system (2) for phosphorus
removal eliminated from secondary effluents most of the organics
refractory to adsorption and produced water, after carbon adsorption,
with residual total organic carbon (TOC) concentrations of less than
1 mg/1. Clarification with lime or alum insolubilized the phosphorus
and flocculated the colloidal organic and phosphorus materials that
were too small to efficiently filter, and too large to diffuse into
the pores of the carbon granule. In that early study, primary effluent,
fed to a tertiary treatment system, revealed that clarification and
filtration to remove the colloidal material from the primary effluent
also produced moderately low residual organic concentrations after
carbon adsorption.
Recently, several laboratory and pilot scale studies (3, 4, 5, 6, 7, 8,
and 9) were completed on physical-chemical treatment of raw or primary
wastewater. The physical-chemical treatment in these studies usually
consisted of chemical clarification and dual or multi-media filtration
for solids and phosphorus removal, followed by granular or powered
carbon adsorption to remove the dissolved organics. Physical-chemical
treatment also removed particulate nitrogen by clarification, soluble
organic nitrogen by carbon adsorption, and nitrate, if present, by
biological denitrification on the carbon (9) .
In these studies, the clarification process on either primary effluent
or raw wastewater employed lime precipitation or metallic salts, such
as Fed or alum. In lime precipitation, two treatment options are
available: the two-stage high pH lime process (10), usually above
pH 11.5, with intermediate recarbonation to remove excess calcium ions
from th-3 primary (first) stage, and the single-stage precipitation (8)
at pH as low as 9.5. The high pH lime process was usually applied in
waters of moderate alkalinity (100-200 mg/1 as CaCO ) and produced
efficient solids removal and very low phosphorus residuals of less than
0.2 mg/1 as P. In hard waters with alkalinities above 250 mg/1, the
low pH single-stage lime process produces efficient solids removal and
phosphorus residuals of approximately 0.5 mg/1.
In waters with moderate alkalinities, the low pH lime process requires
supplemental flocculants, such as polymers or Fed , to produce
efficient clarification and phosphorus residuals ox approximately
1 mg/1. Ferric chloride, with or without polymer,- efficiently clarified
raw or primary wastewater (9) and produced phosphorus residuals of
0.6 to 1 mg/1.
-------
After the removal of the colloidal substances by clarification and
filtration, carbon initially physically adsorbed most of the dissolved
organics. In continuous treatment, the build-up of soluble organic
material on the carbon rapidly promoted substantial biological activity
within the adsorption system and degraded the product quality. The
growth also occurred on powdered carbon that was recirculated (11).
The biological activity on the granular carbon particle, as has been
reported (12), increased the carbon loading before regeneration.
However, the metabolic end products and sluffing of excess cell mass
from the carbon degraded the water quality by increasing both the
residual solids and organic concentrations (2). The biological growth
also increased the pressure drop across the carbon column. Expanded
upflow beds of carbon have been employed to minimize pressure losses
and backwash requirements (9), but the microbial end products were
still released into the product water and the slimes coated the carbon
granules.
In domestic raw wastewaters, most of the soluble nitrogen exists as
ammonia or its predecessor urea, and is not removed by clarification
or adsorption. Thus complete physical-chemical treatment requires an
ammonia removal process. Three physical-chemical processes are avail-
able for ammonia removal: air stripping of the water above pH 10.5
(13) , breakpoint chlorination (14) , and selective ion exchange (15) .
In the selection of an ammonia removal process for complete physical-
chemical treatment, air stripping (13) of the ammonia was not effective
in cold weather. The ammonia volatility in an air-water system
decreased with decreasing temperature and the stripping efficiency
decreased sharply for any selected air to liquid loading rate. Calcium
carbonate scaling also gradually decreased the removal efficiency and
increased the maintenance costs. Thus ammonia stripping is not
considered universally applicable because of temperature restrictions.
Breakpoint chlorination (14) and selective ion exchange (15), however,
provided potential all-weather processes for ammonia removal.
The selective ion-exchange process, with a natural zeolite, clinoptilo-
lite, was developed by Battelle-Northwest for ammonia removal (15).
The clinoptilolite exchanged Na+ or Ca ions for NH ions in the waste-
wa ter.
The Battelle ion-exchange process was carried out in columns packed
with the 20 x 50 mesh clinoptilolite. The Clinoptilolite, mined as
a rock from the Baroid Division of National Lead's deposits at Hector,
California was crushed and sieved to a 20 x 50 mesh particle size.
The columns were regenerated by a lime-salt mixture. Concentrated
ammonia was removed from the reused regenerant by air stripping. The
process reduced the ammonia concentrations in wastewater to less than
1 mg/1. Although selective for ammonia, the zeolite also removed
cations present in the water. Studies by Battelle-Northwest (16)
-------
reveal that with an influent Ca++ of 30 mg/1 and NH.-N of 14 mg/1 half
the ion exchange capacity is used by Ca ions.
A recent laboratory study (14) showed that breakpoint chlorination with
proper pH control and mixing provided a physical-chemical method for
removing NH from wastewaters by oxidizing the NH to N gas. The
C1:NH -N dosage weight ratio was in the range 8:1 to 10?1 as compared
with the stoichiometric ratio 7.6:1. The excess Cl for breakpoint,
above that needed for stoichiometric oxidation, varied inversely with
the degree of pre-treatment of the wastewater. The study also showed
that the undesirable side reactions which produces NO was favored by
high reaction pH and that the production of NCI was favored by a
combination of low pH and excess Cl . Hence, to keep the nuisance
residuals to a minimum, the reaction pH was maintained near pH 7.0.
In a later pilot study (17) on secondary effluent, rapid mixing was
necessary to efficiently disperse the Cl in the process water. With
poor mixing, localized high Cl concentrations caused excessive NCI
formation even when NaOH or Ca(OH) was used to maintain the pH at 7-
Thus for successful operation of the breakpoint process with a minimum
of NO3 and NCI, production, the reaction must be well mixed, its pH
2
controlled in the pH range 6-8 and the excess Cl kept to a minimum.
While physical-chemical treatment has been applied to both raw waste-
water and primary effluents, it is most likely to be applied directly
to raw wastewater as the more economical system. Thus, a study of
complete physical-chemical treatment was conducted on domestic raw
wastewater at the EPA-DC Pilot Plant in Washington, D.C. to remove
carbon, phosphorus, and nitrogen. Since very low phosphorus residuals
were needed, the two-stage high pH lime process was employed for clari-
fication. Downflow granular carbon columns were used for removal of
the dissolved organics; selective ion exchange and breakpoint chlori-
nation were separately evaluated for ammonia removal.
-------
SECTION IV
EXPERIMENTAL
Pilot Systems
The initial automated pilot system (Figure 1) consisted of cyclone
degritting, two-stage (high pH) lime precipitation with intermediate
recarbonation, dual-media filtration, pH control, selective ion
exchange, and downflow granular carbon adsorption. The precipitation
process was designed for a nominal capacity of 100,000 gallons per day;
the filtration, ion exchange, and carbon adsorption processes for
50,000 gallons per day. Later breakpoint chlorination was substituted
for the ion-exchange process. When the system employed ion exchange
for ammonia removal, a flow controller impressed a diurnal variation of
approximately 3.2:1 maximum to minimum flow across the clarification
process. Flow splitter boxes and level controllers reduced the flow
from the precipitation process to the hydraulic capacity of the subse-
quent treatment processes while still maintaining the diurnal variation.
Because of process control difficulties, the breakpoint chlorination
process and therefore, the carbon columns were generally operated at a
constant flow of 25 gpm when breakpoint was employed for nitrogen
removal.
In the first-stage of the precipitation process, raw wastewater, lime
slurry and recycled solids were turbine-mixed at 13 rpm (average
reaction time of 40 minutes) in the internal reactor of an upflow
flocculator-clarifier. The lime increased the wastewater pH to about
11.5 and precipitated bicarbonate, phosphate, and magnesium ions from
the water as shown in equations 1-3.
(1)
(2)
Mg(OH) (3)
£
Sludge from the slurry pool at the bottom of the clarifier was recycled
at a rate equal to 10% of the average influent flow. Waste rates of
approximately 1.8% of the influent flow with solids concentrations of
30,000 to 50,000 mg/1 maintained the solids balance of the slurry pool.
The limed water, after sedimentation, flowed through an open channel
to recarbonation. Carbon dioxide below a turbine operating at 171 rpm,
reduced the wastewater pH to approximately 10, and with an average
contact period of 15 minutes, precipitated the excess calcium ions
added in the liming (first) stage according to equation 4.
Ca++ + CO + 20H~ » CaCO + H O (4)
2 -3 &
-------
raw water
recycle
C02
}
c
I
:
IT
IT
1]
TI
IT
IT
TJ
U
LT L^rLr
CARBON ION EXCHANGE
u
V
i
L
C02
« s
i
L
1
T
J L
1
s
FILTERS
FIGURE 1 - Physical-Chemical Treatment Pilot Plant
-------
Five mg/1 of ferric ions, added in the first of the two external 24-
minute flocculation basins, each turbine-mixed at 22 rpm, formed Fe(OH)
according to equation 5 and flocculated the precipitated calcium
carbonate.
Fe + 3HOH ^Fe(OH) + 3H+ (5)
After flocculation, the water flowed to a perpherial feed circular
clarifier. Settled slurry recycled from the bottom of the settler to
recarbonation at a flow rate equal to 10% of the average influent flow
provided nuclei for the calcium carbonate precipitation in the recar-
bonation tank. Wasting at flows approximately equal to 1.9% of the
influent flow maintained the solids balance of the slurry pool in the
settler.
The overflow from the second-stage clarifier, was pumped to a distri-
bution box ahead of the two filters. It then flowed by gravity through
two dual-media filters packed with 18 inches of 0.9 mm coal over
6 inches of 0.45 mm sand. When pressure losses across the filter
exceeded 9 feet of head, a surface wash at 3 gpm/ft and a sequenced
backwash at up to 20 gpm/ft automatically cleaned the filter for ten
minutes and returned it to service.
In operation with selective ion exchange, carbon dioxide added after
filtration, reduced the pH of the filtered water to 7-8 in the pH
control tank. The water then was pumped to the ion exchange followed
by carbon adsorption or to the carbon adsorption processes followed by
ion exchange treatment. The trailer-mounted ion-exchange system was
not fully automated and required continuous operator surveillance for
efficient ammonia removals. Thus, the exchange system was operated
intermittently and not always at peak efficiency.
In the ion-exchange process, the Battelle pilot plant, consisted of
three exchange columns, a stripping tower, regenerant mixing tank, air
blower, and air heater. Figure 2 is a flow diagram showing the ion-
exchange system. The exchange columns were 39 inches in diameter and
eight ft. tall. Each column was packed with approximately 1870 pounds
of zeolite and was operated under a pressure of 35 p.s.i.g.
The NH 3 stripping tower was a 43-inch diameter fiberglass tower packed
with seven ft. of 1-inch polypropylene Intalox saddles. The tower
drained into a 200 gal. agitated tank, where lime was added under pH
control. Air for the stripping tower was supplied by the 10-h.p. blower
and was heated by a 60-k.w. heater.
A compact piping cross header, installed underneath the trailer,
contained 36 electrically-operated ball valves, which permitted the
liquid flow to be routed anywhere in the process.
The influent stream, controlled at pH 7.0-7.5, was pumped downflow
through two clinoptilolite beds in series at loading rates from 2.4 to
-------
INFLUENT
CO,
pH
CONTROL
1
COLUMN
1
1
-
-
COLUMN
2
EFFLUENT
NH3
jllitu
HEAT
SALT
LIME
MAKE-UP
TANK
HEAT
STRIPPER
FIGURE 2 - Ion Exchange Process
COLUMN
3
10
-------
2
6.3 gpm/ft . The operation for the three ion-exchange columns was:
1. Two beds were in service downflow while the third was being
regenerated.
2. After the first bed in service approached maximum ammonia
loading (ammonia in effluent was 90% of influent concentra-
tion) , it was taken out of service and the second bed was
placed in the number two position.
3. A column was regenerated by pumping a solution of Na and
T"T
Ca ions (sodium chloride and lime) at a pH of 11.0 upflow
through the zeolite.
The regenerant was recirculated through the bed at a loading rate of
7.2 gpm/ft for three hours to strip the NH ions from the clinoptilolite
as NH OH and to build up the ammonia concentration in the regenerant to
approximately 500 mg/1 NH -N. The regenerant flow was decreased to
3 gpm/ft and then recycled through both the zeolite bed and the stripper
until the NH -N was reduced to 20 mg/1 (average time 12 hours). During
regeneration, the pH was automatically maintained at 11.
A portion of the regenerant solution which remained in the mixing tank
was reused from batch to batch. Following regeneration, the column
was backwashed with ammonia free water to remove the lime deposits.
The air stripper was operated at a liquid rate of 25 gpm with an air
to liquid ratio of approximately 100 ft. of air per gal. of water.
The first five months of operation (82 days) were on lime clarified
and filtered raw wastewater. In July 1970, the ion-exchange unit was
placed after carbon adsorption in the physical-chemical system because
the anaerobic growths on the- carbon were converting organic nitrogen
to ammonia. Thus, more ammonia was available for ion exchange on the
zeolite after carbon adsorption. The air stripper was run with heated
air (80 F) for the first month and with ambient air during the remainder
of its operation. Regenerant solution was not reused during one month
(December) of cold weather operation.
After completing the studies on the ion-exchange system, breakpoint
chlorination was evaluated as an alternate ammonia removal process. The
breakpoint chlorination pilot system in this study (Figure 3) consisted
of a 1200 gallon reactor, 4 feet in diameter, 12 feet tall with appro-
priate pumps, chlorine contactors and automatic controllers. The
reactor vessel, a existing modified pH control tank, was larger than
needed for chlorination. Only the bottom 3 feet (310 gallons) was used
for the chlorination reactor. Mixing was provided by 3 propeller
mixers mounted on a common shaft, driven by a 3 hp motor at 225 rpm.
The pump on the effluent stream recycled 11 gallons back to the reactor.
Chlorine was added to the recirculating water through a Wallace and
Tiernan chlorinator and injector nozzle. The chlorine was added to
process waters with pH control by NaOH addition and with proper mixing
until a point was reached where the total dissolved residual chlorine
21
-------
FEED
NaOH
OR
Ca(OH]2
TO
CARBON
COLUMNS
FIGURE 3 Breakpoint Chlorination Reactor
1 2
-------
reached a minimum and the NH was oxidized chiefly to N gas. The
reaction to nitrogen gas preceded through monchloramine as described
by the following equations.
HOC1 + HC1 (6)
NH Cl + H O + H+ (7)
2NH2C1 + HOC1 » N + 3PIC1 + H O (8)
The overall reaction was:
2NH4 + 3HOC1 »W + 3HC1 + 3H O + 2H+ (9)
Control of pH was needed because the type of chloramine formed and the
ultimate oxidation product depends upon pH. Spectroscopic analysis
(18, 19, 20, 21, 22) indicated that, in the pH range 7-8.5, monochlor-
amine was the chief intermediate product. As the Cl dose exceeded
that for NH Cl formation, the NH Cl was oxidized directly to N gas.
If the pH is allowed to decrease below 7, increasing amounts of
dichloramine were formed (equation 10). In the pH range of 4.5 and
below, trichloramine was the chief product (equation 11). Nitrate was
also produced in small quantities according to equation 12.
NH Cl + HOC1 +NHC1 + H O (10)
£ £ £
NHC1 + HOC1 >NC1 + HO (11)
& *J £
4HOC1 + NH > HNO + 4HC1 + HO (12)
The pH of the water in the reactor was automatically maintained by
controlled pumping of a 1 to 2% NaOH solution into the line ahead of
the chlorine injector. The NaOH feed rate was controlled by an inline
pH probe on the discharged effluent.
A Technicon Autoanalyser was used to continuously monitor the NH -N
concentration in the influent and effluent. The NH -N concentration
in the influent was used to manually set the Cl at a 9:1 C1:N ratio
and the NH -N concentration in the effluent was used to determine
whether breakpoint was achieved. Periodic laboratory measurements of
free and combined Cl were performed on the process effluent to check
pilot plant operation.
In the five months of the study, the feed to the system was lime-
clarified filtered raw wastewater. For the first four months of
operation, the flow rate was a constant 25 gpm. The first month of
operation revealed that breakpoint control was difficult because of
the large reactor detention time of 12 minutes. Adjusting the pH to
7.0 with CO before entering the breakpoint reactor improved process
control.
-------
Hence, after one month of continuous operation with an influent pH of
9 to 10f CO was used in a Kenics in-line static mixer to automatically
preadjust the influent wastewaters pR from 9-10 to 7. The process was
operated continuously for 3 months with CO preadjustment.
Since CO increased base usage, the process was further modified for
the last month of operation. A second chlorinator was installed and
Cl was added in place of CO for preadjusting pH. A diurnal flow
variation with a maximum of 55 gpm and a minimum of 22 gpm was also
impressed across the system.
The flow from either the ion-exchange process or from the pH control
or breakpoint tank was pumped at 50 psig through four downflow carbon
columns each packed with 840 pounds of 8 x 30 mesh granular carbon.
The carbon was supported by a 50 mesh stainless-steel screen on a
gravel bed. The first column was backwashed and surface-washed on a
daily basis with a quantity of water equal to about four percent of
the product water. After four months of operation, the use of a high
pH (11.5) backwash was tested on alternate days with the product water
to minimize biological activity on the carbon. The other columns were
also washed once a week with the high pH water. To stimulate two-stage
carbon treatment, the spent carbon was replaced with virgin carbon in
two columns at a time. In the replacement cycle, the replaced carbon
C2 columns) was located at the end of the carbon system as the second
carbon stage and the partially loaded carbon (2 columns) in the
original second stage was relocated as the first carbon stage. A
flexible piping manifold permitted the counter-current operation and
also allowed the isolation and the backwash of individual carbon
columns without interruption of the process flow. The carbon after
replacement was not regenerated.
Analytical Procedures
In the first four months of the study with ion exchange in the system
all samples for analysis were manually composited over 24 hours. To
reduce the analytical load, samples for Ca , Mg, total phosphorus,
NH , COD, and TOC were then composited over 48 hours for the remainder
of the ion-exchange operation. Samples for COD, TKN, suspended solids,
and NO + NO were always composited over 24 hours. During compositing,
all samples were stored at 3°C to minimize biological activity. In the
operation with breakpoint chlorination, all samples were composited
over 24 hours.
In the laboratory, the total phosphorus were determined by the persul-
fate method (23); BOD by the probe method (24). The Ca and Mg
analyses was measured on a Perkin and Elmer Model 303 Atomic Adsorption
unit (24). Ammonia (24) and nitrate-nitrite (25) were determined on a
Technicon Automatic Analyzer, and the TOC was measured on a Beckman
Carbonaceous Analyzer (26). All other analyses including the Pheno-
phthalein (P) and methyl orange (.'1.0.) alkalinities employed Standard
Methods (27) .
14
-------
The chlorination study required field measurement for NH*-N, which was
analyzed on a programmed Technicon Automatic Analyzer (24). Free and
combined chlorine were measured by the Modified Palin Method (28) with
N, N-diethyl-p-phenylenediamine oxalate as the indicator.
15
-------
SECTION V
PROCESS OPERATIONS
The intake for the pilot plant was located at the head of the main
plant's grit chamber. Solids in the wastewater continually fouled the
intake pumps of the pilot plant. After the first month of operation,
bar screens were placed over the intake. Rags and other refuse clogged
the screens and filtered solids from the water. The pilot plant's
efficient cyclone degritter also removed solids from the influent. Thus,
the feed to the physical-chemical pilot plant contained less solids and
particulate organics than the normal raw plant influent.
The normal operating flows, hydraulic loadings and detention times for
the physical-chemical system are shown in Table 1. While the pilot
plant included only physical-chemical processes, biological activity
occurred within the plant and, as described in this section, altered
the operation of some of the processes.
Clarification and Filtration Operation
Two-stage lime precipitation of the D.C. raw wastewater was a very
stable process as long as the wastewater pH was above 11.3. If the
lime slurry feed to the first stage reactor was interrupted or the
lime concentration reduced, the resumption of the appropriate lime
dose regained the product water quality within a few hours without
biological activity. The average or median monthly operating variables,
chemical requirements, and sludge wasting rates are summarized in
Table 2. The lime CaO dosage and CO dosage for recarbonation during
the entire operation (Table 2) was approximately 300 mg/1 and 120 mg/1,
respectively.
The lime dosage from the slurry tank of a gravimetric lime feeder was
controlled by flow-proportional pH control, flow-proportional alkalinity
control, or flow-proportional conductivity control. The signal con-
trolling the CaO dosage was produced by multiplying a feed-forward
signal proportional to plant flow by a feed back signal from a pH,
alkalinity or conductivity controller. The flow-proportional pH control
provided the best lime feeding system. The same approach was applied
to the CO feed.
The maximum (rain peak) overflow rates in the first stage of clari-
fication was 1450 gal/day/ft ; in the second stage, 1770 gal/day/ft .
The solids wasting usually ranged from 1.7 to 1.9% of the influent
plant flow for the first stage and 1.9-2.1% for the second stage. The
second stage (recarbonation) wasting was pumped to the first stage and
wasted at high pH with the first stage solids. The concentration of
solids in the combined sludge (first stage) varied from 30,000 to
50,000 mg/1. The 10% recycle of sludge from the clarification zones
to the precipitation reactors produced reactor solids varying from
16
-------
TABLE 1
PXLOT PLANT HYTiRAULIC LOADINGS
Lime Precipitation
Stage 1
Dry weather flow-
Rain peak
Primary mixing
Stage 2
Dry weather flow-
Rain peak
Recarbonation
Flocculation
(2 basins)
Filtration
2 Filters
Neutralization
Ion Exchange
Flow
gal./min.
45 min.
70 ave.
105 max.
140
70 ave.
45 min.
70 ave.
105 max.
140
70 ave.
70 ave.
20 min.
35 ave.
52 max.
35 ave.
20 min.
35 ave.
52 max.
Breakpoint Chlorination 25 ave
Carbon Adsorption
20 min.
35 ave.
52 max.
Hydraulic
loading rate
475 gpd/ft*
725 gpd/ft*
1100 gpd/ft*
1450 gpd/ft"
575 gpd/ft*
885 gpd/ft'
1340 gpd/ft*
1770 gpd/ft*
1.7 gpm/ft*
3.0 gpm/ft",
4.5 gpm/ft*
2.4 gpm/ft',
4.2 gpm/ft*,
6.3 gpm/ft*
4.0 gpm/ft~2
7.0 gpm/ft2
10.4 gpm/ft
Detention Time
4.4 hrs.
40 min.
2.8 hrs.
15 min.
24 min. ea.
26.3 min.
30 min.
16 min.
12 min.
27 min.
1 Chlorination and subsequent adsorption usually operated at a
constant rate of 25 gpm during system operation with breakpoint
Chlorination.
27
-------
TABLE 2
CD
Month Alkalinity
mg/lCaCO
Mar. 70
Apr. 70
May
June
July
Aug.
Sept
Oct.
Nov.
Dec.
Feb.
Mar.
Apr.
May
June
July
Aug.
70
70
70
70
. 70
70
70
70
71
71
71
71
71
71
71
126
131
129
147
153
161
142
147
135
127
134
127
129
140
OPERATING VARIABLES
Lime Dose First
Stage
mg/lCaO pH
11.8
11.7
289
360
320
280
370
300
240
259
251
266
310
306
375
338
11.7
11.7
11.4
11.4
11.5
11.4
11.3
11.4
11.4
11.3
11.4
11.3
11.3
11.4
FOR TWO-STAGE LIME CLARIFICATION
Recarbonation Second Neutralization First
with CO Stage with CO Stage
Waste
mg/ICO pH pH % of
Inf.
10.4 7.1
10.7 7.2
236.0
192.0
121.0
74.3
107.0
95.6
63.0
72.5
126
93
173
138
124
110
104
10
10
10
10
10
10
10
9
10
10
10
10
10
9
.1
.1
.1
.1
.1
.2
.1
.7
.0
.0
.0
.0
.1
.8
7
7
7
7
7
7
7
7
-
-
-
7
7
7
-
.0
.3
.3
.5
.5
.7
.3
.4
.0
.0
.0
2.8
1.8
1.7
1.7
1.8
1.9
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
Second
Stage
Waste
% of
Inf.
2.1
2.0
2.0
2.0
2.0
2.0
1.9
1.9
1.9
1.9
1.9
1.9
1.8
1.8
1.8
-------
3000 to 5000 iag/1 for operation with the above sludge wasting rates.
The 5 mg/1 of Fe+++ produced effective flocculation in the second
stage with suspended solids carryover in the second stage overflow of
about 20 mg/1.
As the water temperature increased in the spring and summer, microbial
growths occurred throughout the second stage clarifier at pH 10 but not
in the first stage at pH 11.5. The growths, however, did not reduce
clarification removal efficiencies.
The daily performance of the clarification system (Figure 4, 5, 6)
indicated the stability of the two stage lime clarification process.
The process typically removed 80% of the organics, 96% of the phos-
phorus and 88% of the suspended solids. The wide variations in the
influent concentrations of phosphorus, organics (COD) and suspended
solids did not significantly alter process performance or reliability.
The sludge production (Figure 7) in the process varied as a function
of wastewater alkalinity, influent phosphorus and magnesium concen-
trations, influent suspended solids and lime dose. For a wasting rate
of 1.8% from the first stage and with the maximum solids concentration
in the waste slurry, the sludge production was approximately 7.5 pounds
of total solids per thousand gallons of treated water (900 mg/1).
With, efficient lime clarification and nominal filter loadings of
3.0 gpm/ft , filter runs of more than 50 hours occurred in the early
spring. With increasing water temperatures, the growth of biological
slimes at pH 10 within the filters and in the effluent from lime
clarification gradually decreased the length of the filter runs to
less than 12 hours. Ten mg/1 of chlorine, added to the influent of
the filters to control the slimes restored the filter runs to about
50 hours.
Ion-Exchange Operation
The ion-exchange system exhibited mechanical problems and required
continual surveillance for optimum operation. Because of mechanical
problems, the unit was operated for 120 days during the ten month test
period. The average run lasted 24 hours before switching columns.
With four hours of operator surveillance, ammonia removals (daily
composites) for the intermittent operation (Figure 8) averaged 75%
with residuals of 3.0 mg/1 NH -N. Small amounts of COD and phosphorus
were removed from the wastewater by the zeolite bed (Table 3). The
suspended solids, however, increased and an increased turbidity appeared
in the effluent. The turbidity was produced by lime deposits from poor
regenerations.
The large fluctuations in the daily effluent NH concentrations revealed
the effects of minimum manpower for the ion-exchange operation.
Electrical valve malfunctions were frequently responsible for mixing
29
-------
K5
O
456
342
228
Q
O
INFLUENT
114
0
CLARIFIED
MAY
JUNE
JULY
FIGURE 4 - Daily Removals of COD in Physical-Chemical Treatment
-------
(D
O
cn
Q
UJ
Q
~ZL
LU
CL
230
184-
138
92
46
0
MAY
JUNE
JULY
FIGURE 5 - Daily Removal of Suspended Solids in Lime Clarification
-------
Q_
CO
Z)
or
o
i
Q_
CO
O
I
Q_
II
10
9
8
7
6
5
4
3
2
I
0
INFLUENT
MAY
JUNE
JULY
FIGURE 6 - Daily Removal of Phosphorus in Lime Clarification
-------
g
x
10-
RAW WASTEWATER
o
^
c/>
OQ
8
z
g
u
D
0
o
cr
a
O
t/1
200
250 300 350
LIME DOSE. MG/L C&O
FIGURE 7 - Solids Production and Lime Addition
400
-------
18
16
14
12
10
8
6
2
0
I I
-ION EXCHANGE
J I
0 10 20 30 40 50 60 70 80 90 100 110 120
DAYS
FIGURE 8 - Ammonia Removal in Selective Ion Exchange with Four Hour
Operator Surveillance
-------
TABLE 3
SELECTIVE ION EXCHANGED AND RESIDUAL POLLUTANTS
March
April
May
June
July
2
August
2
September
2
December
IN
53
54
44
42
30
16
13
20
COD
mg./l.
OUT
42
41
38
38
26
19
23
17
TOTAL
mg./l
IN
.45
.27
.18
.18
.13
.10
.11
.09
P
OUT
51
16
16
10
08
09
11
08
SUSPENDED
mg./l
IN
3.8
12
5.5
3.4
5.0
6.8
3.0
6.7
SOL.
OUT
7.1
6.4
21.0
5.1
9.0
7.0
3.6
24.4
1 Ion exchange was operated intermittently
2 Ion exchange placed after carbon adsorption.
25
-------
of the regenerant and service cycle streams, a major contributor to the
poor removal.
Ammonia removals from the D.C. lime clarified raw wastewater, however,
(2-hour grab samples) averaged 95% during periods of good operation
with 24-hour surveillance (Figure 9). The results on the seventh day
in Figure 9 were excluded from the calculation of the average ammonia
removal because of incomplete regeneration of the replacement column.
In the period of efficient operation, the average run lasted 30 hours
before regenerating the lead column.
The most important variable for controlling the process was the pH during
regeneration. With a low pH (less than 10.5), not all of the ammonia
was removed from the bed and made available for air stripping. A high
regenerant pH (greater than 12.0) left deposits of lime on the zeolite,
which later during a service cycle, raised the pH of the feed and
converted NH to unionized NH OH. The pH monitoring of lime solutions
required frequent checks of the pH electrodes to prevent their scaling.
Thus the majority of the daily operational problems were caused by a
malfunctioning pH meter and lime scaling difficulties.
During the first five months of operation, the water contained approxi-
mately 18.1 mg/1 of BOD and 45 mg/1 of COD, with potential for biological
growth on the zeolite. The pressure drop across the bed increased on
every run and the flow had to be reversed for one minute at least twice
a run. The high differential pressure, however, was not cumulative
from run to run as the high pH regeneration minimized growth. Approxi-
mately 0.16 pounds of NaOH and 0.5 pounds of CaO were employed in the
regenerant brine for each thousand gallons of water treated.
When ion exchange was placed after carbon adsorption, influent BOD
averaged only 5.8 mg/1; the COD, 16 mg/1. Microscopic examination of
the regenerant solution, however, showed evidence of biological growth
in the columns. The growth did not increase the pressure drop through
the bed. The biological growths, apparently leaking from the adsorption
system, fouled the exchange minerals and increased the 12 hours regen-
eration time by about 20%. During the early part of regeneration
cycles, larger quantities of lime (approximately a 10% increase) were
also unexpectedly needed to maintain the regenerant solution at pH 11.0.
The ammonia removal capabilities of the zeolite were not affected.
After 90 days of operation, the stripping tower packing was plugged with
calcium carbonate scale and lime. The lime scale was soft and easily
removed. The calcium carbonate scale was hard and had to be removed by
agitating the packing in water. The lime found in the stripping tower
was caused by the erratic operation of the pH controller. The pH
control problem was also responsible for the large quantities of back-
wash water (3-6% of total service flow) required after each regeneration.
26
-------
20
* 16
CO
CD
12
0
1 2 3
5
DAYS
7 8 9 10
FIGURE 9 - Ammonia Removal in Ion Exchange with Continuous
Operator Surveillance
-------
Breakpoint Chlorination Operation
Breakpoint chlord.na.tion removed chiefly NH -N with the organic nitrogen
and dissolved organics (TOC) generally unaffected by chlorination
(Table 4). In the first month of operation with an influent pH of
9 to 10, the process control was difficult. The pH periodically cycled
usually between 6-8 but occasionally the cycle exceeded the 6-8t pH
range. Moderately high ammonia breakthrough (Table 4) averaging
approximately 0.9 mg/1 occurred during this period. The oxidation of
ammonia in the first month of operation also produced slightly more
nitrate with an average value of 0.8 mg/1 than subsequent operation.
In earlier work (14) high nitrate production occurred at high reaction
pH.
The breakpoint with the influent wastewater pH at 9-10, however, reduced
the base (NaOH) used to neutralize the acid produced by chlorination
from the stoichiometric (equation 9) requirement of 1.5 pounds of NaOH
per pound of chlorine to 0.9 pounds of NaOH per pound of chlorine. The
alkalinity in the wastewater at pH 9-10 neutralized some of the acid
from the breakpoint.
In the next three months with CO- preadjustment of the pH to 7, the
controller consistently maintained the pH at 7.0. With effective pH
reaction control, the daily NH ~N variation in the effluent (Figure 10)
revealed consistently good NH -N removal. The average influent and
effluent concentrations for this period were 11 mg/1 and 0.4 mg/1 of
NH -N. The NO in the effluent averaged 0.6 mg/1 as N.
The NH -N removals represented a significant improvement over the first
month's operation. However, the NaOH usage increased to 1.8 pounds per
pound of chlorine (Table 4). The increased base requirement_occurred
because the CO converted the OH alkalinity at pH 10 to HCO alkalinity
which was unavailable for neutralizing acid at pH 7. Since the stoichio-
metric base requirement is 1.5 pounds of NaOH per pound of Cl , the
excess CO also appeared to exert an additional base demand.
During the last stage of operation with chlorine to preadjust the pH
and with the system on a diurnal flow, control of the breakpoint process
was very difficult. The daily average NH -N variation in the effluent
(Figure 10) however, did not reveal the NH -N breakthrough since sampling
was discontinued with loss of breakpoint. The monthly average NH -N
concentrations in the influent and effluent were 9.7 mg/1 and 0.5 mg/1.
While the use of Cl in pH preadjustment probably reduced base require-
ments, the base usage during this stage of operation could not
accuratedly be determined because of frequent loss of breakpoint. Thus
an improved control system was needed to operate the split chlorine
treatment approach and to handle the diurnal cycle.
28
-------
f. TABLE 4
Breakpoint Chlorination
of
Lime Clarified and Filtered Raw Wastewater
Month
Flow, gpm
Influent
pti
P. alk, mg/1
M.O. , alk. , mg/1
NH -N, mg/1
TKN-N , mg/1
NO -N, mg/1
TOC, mg/1
pH Adjustment
pH after Adj .
Reaction Conditions
pH
C1:N, Ib/lb.
NaOH:Cl, Ib/lb.
Effluent
M.O. , alk. , mg/1
NH+4-N, mg/1
TKN,- mg/1
NO -N, mg/1
TOC, mg/1
April
25
9.9
70
140
12.6
14.2
0.16
22.6
None
9.9
7.0
9:1
0.9
52
0.9
4.8
0.8
23.1
May
25
9.3
45
84
11.6
14.1
0.02
19.9
co2
7.0
7.1
9:1
1.8
100
0.4
2.5
0.7
20.1
June
25
9.3
54
96
10.8
12.2
0.0
16.7
co2
7.0
7.2
9:1
1.8
124
0.4
2.0
0.6
18.8
July
25
9.2
31
82
10.4
11.9
0.0
17.0
co2
7.0
7.3
9:1
1.8
118
0.4
1.8
0.5
18.8
August
22.5-55.5
9.0
30
80
9.7
10. 7
0.0
18.3
C12
7.0
1
9:1
1
60
0.5
2.11
0.5
16.9
pH cycled with poor process control; breakpoint was frequently
lost. Data represents periods of successful breakpoint; base
requirement and pH not accurately determined.
29
-------
CO
o
14
13
12
II
10
O»
E 9
ro
X
o
DAILY VARIATIONS
INFLUENT
EFFLUENT
10 20
APRIL
30
10 20 30
MAY
10
20 30
JUNE
10 20 30 10
JULY AUG
DAYS
FIGURE 10 - Daily Ammonia Removal in Breakpoint Chlorination
-------
Carbon Adsorption Operation
The carbon adsorption process included two different periods of
operation, one without chlorine in the influent wastewater (selective
ion exchange or no nitrogen removal) and one with chlorine in the
influent wastewater (breakpoint chlorination). In the period without
chlorine, the soluble BOD (20 mg/1) in the influent wastewater caused
heavy anaerobic biological activity on the carbon columns. This
activity produced appreciable R2S production (2-3 mg/1 of H.2^) i-n the
product water and objectionable odors, relatively high pressure losses
(25 psig in a 24 hour period) across the lead column and high carbon
losses during backwash into the carbon decant tank. Since carbon
recycle from the decant tank to the backwashed carbon columns was not
employed because of the heavy slimes, carbon losses exceeded 25%
between the carbon replacement cycles.
Normally the lead carbon column was backwashed once a day by the
following procedure:
2
1. 10 minutes of backwash at 2 gpm/ft .
2. 10 minutes of backwash at approximately a 17%
bed expansion (10-11 gpm/ft2) with a simultaneous
surface wash at 3 gpm/ft2.
3. 10 minutes of backwash at 29% bed expansion
(13-16 gpm/ft2 as a function of water temperature)-
Each of the remaining three columns were backwashed by the above
procedure once a week. Beginning in July 1970, a modification in the
backwash procedure was evaluated for controlling the biological activity
on the carbon columns. In July, after the replacement of the first
half of the carbon (2 columns) with virgin carbon, the backwash water
during the period (10 minutes) with surface wash contained on alternate
days 300 mg/1 of NaOH in an attempt to control the biological activity
in the lead column. The remaining columns were backwashed once a week
with the sodium hydroxide added during the surface wash period.
The backwash with sodium hydroxide was continued until the end of
October 1970 and produced a pH of 9 in the backwash effluent. The
use of the high pH backwash water, however, did not reduce the pressure
drop (25 psig per day) across the lead column, and did not control the
H2S production (2.4 mg/1 of H£ in a September 1970 test). The plant
operation exhibited high backwash carbon losses of about 13% between
carbon replacement cycles.
In November 1970 after a second carbon replacement cycle, the NaOH in
the backwash water was increased to 600 mg/1 in the surface wash period
and the time of backwash surface wash increased from 10 minutes to
30 minutes. The pH in the effluent washwater increased from 9 with the
300 mg/1 dose to 11.3 with the 600 mg/1 dose. Unfortunately, the
wastewater temperature began to decrease significantly and the evaluation
31
-------
of the very high backwash sequence fox controlling biological activity
was complicated by the decreasing temperature. While the amount of H-2S
and the plant's odor problems decreased, the pressure drop in the lead
carbon column remained high. In late December 1970, a support plate in
the lead carbon column ruptured and accurate carbon losses could not be
evaluated. In addition, with the replacement of carbon in November,
the residual organics in the carbon effluent did not exhibit as large
a decrease in concentration as expected. After only fourteen days of
operation with the new carbon charge, the TOO and COD unexpectedly
rapidly increased. While the increase occurred with an accompanying
increase in influent COD, the residual was considerably higher than
that obtained for similar carbon loading in previous operation. The
low removal continued through December 1970. The decrease in organic
removal appeared to have been caused by a combination of an increase
in influent organics, a decrease in water temperature, and a fouling
(or scaling by CaCOj) of the carbon with the high pH backwash water.
While recognizing that the backwash system on the carbon columns was not
ideal and could be improved by redesign, the results of the biological
activity (H2S, slime, etc.) indicated a problem within physical-
chemical treatment of raw wastewater that must be considered in the
system design. The carbon columns may have to be converted to upflow
expanded beds to eliminate or reduce the need for backwashing or employ
a suitable technique such as breakpoint chlorination ahead of adsorption
to control the biological activity.
In the second period of operation, breakpoint chlorination ahead of
the adsorption system supplied combined and free chlorine (1-10 mg/1)
in the influent to the carbon column. The presence of the chlorine
minimized biological activity and controlled but did not completely
eliminate H-S odors. Sufficient anaerobic activity persisted in the
carbon columns to denitrify the small (0.6 mg/1) amount of nitrate
produced by the breakpoint process. Essentially complete dechlorination
required approximately three quarters of the detention time of the
carbon columns.
The pressure drop in the carbon columns with the chlorinated influent
did not exceed 10 psig in 48 hours and backwashing of the lead column
was performed once every 48 hours. The previous high carbon losses
and the heavy biological growth did not occur and the adsorption column
operation behaved similarly to that of earlier carbon studies on
tertiary physical-chemical treatment.
Without chlorine in the influent water, the organic residuals (TOC and
COD) in the carbon column effluent revealed a repeating pattern in
which low residuals (Tables 5 and 6), (approximately 8 mg/1 of COD and
3 mg/1 of TOC in the first month of the first replacement cycle)
occurred during initial operation with virgin carbon and then gradually
increased (22 mg/1 of COD and 8 mg/1 of TOC in the last month of the
first replacement cycle) before the replacement of one half of the
carbon (Figures 11 and 12). A similar pattern occurred in the second
32
-------
TABLE 5
COD REMOVALS IN PHYSICAL CHEMICAL TREATMENT
OF
RAW WASTEWATER
Month
March 70
April 70
May 70
June 70
July 70
Aug. 70
Sept . 70
Oct. 70
Nov. 70
Dec. 70
Feb. 71
March 71
April 71
May 71
June 71
July 71
Aug. 71
Raw
mg/1
347
311
299
302
282
289
261
351
323
308
282
297
300
284
262
263
250
Ion exchange
2
Ion exchange
Clarified
mg/1 %Rem
66.
56.
52.
48.
40.
43.
59.
58.
57.
63.
60.
63.
64.
55.
55.
44.
51.
2
8
5
5
4
6
1
1
6
9
4
1
7
2
2
2
5
81
80
83
84
86
85
77
84
82
79
78
79
78
81
79
83
79
or breakpoint
relocated
Filtered
mg/1 %Rem
53.9
51.8
45.2
44.9
36.1
45.1
48.6
55.0
50.2
55.6
49.3
57.1
54.9
50.5
37.8
38.6
46.1
84
82
85
85
87
84
81
84
85
82
82
81
82
82
86
85
82
chlorination
after carbon
Nitrogen Rem.
mg/1 %Rem
41.
39.
38.
48.
25.
22.
23.
16.
65.
66.
55.
48.
51.
3
2
4
2
6
92
0
-
-
82
-
-
I3
I3
I3
23
88
85
87
88
91
92
91
95
78
77
79
82
79
Adsorbed
mg/1 %Rem
8.2
12.7
16.4
18. 7
7.8
15.4
16.5
22.2
17.1
20.3
24.4
17.1
13.6
15.2
18.8
20.5
98
96
95
94
97
94
94
94
95
93
91
94
96
95
93
92
or none
adsorption
n /-»
j_
_!_.
_ J_
Breakpoint chlorination employed in P.C. treatment
Note: Ion exchange process operated intermittently
% Removals are accumulative.
33
-------
TABLE 6
TOO REMOVAL IN PHYSICAL-CHEMICAL TREATMENT
OF
RAW WASTEWATER
Month
Mar-
Apr -
May
June
July
Aug .
Sept
Oct.
Nov.
Dec .
Feb.
Mar.
Apr .
May
June
July
Aug.
70
70
70
70
70
70
. 70
70
70
70
71
71
71
71
71
71
71
Raw
mg/1
118
102
114
85
78
96
118
119
122
112
96
106
94
92
95
86
72
Clarified
mg/1
25.5
22.8
18.8
18.1
17.6
17.5
23.1
24.3
24.0
26.6
21.6
24.8
24.5
22.5
20.7
19.6
20.4
% Rem
78
77
84
79
78
82
80
80
80
76
78
77
74
76
78
77
72
Filtered
. mg/1
20.1
19.9
16.8
18.5
17 .3
18.4
22.3
22.1
25.7
25.2
21.6
22.8
22.6
19.9
16.7
17.0
18.3
% Rem.
83
81
85
78
78
81
81
81
79
78
78
79
76
78
82
80
75
Nitrogen Rem
mg/1
14
14
13
14
11
7 .
7.
9.5
23.
20.
18.
18.
16.
.9
.8
.5
.5
.8
62
72
2
I3
I3
83
83
93
% Rem.
87
85
88
83
85
92
93
92
75
78
80
78
77
Adsorbed
mg/1
3.7
4.9
8.1
8.3
5.2
6.1
8.3
7.5
10.1
10.1
12.1
9.1
6.2
7 .4
8.9
10.1
% Rem
97
95
93
91
93
94
93
94
92
91
87
91
93
92
91
88
1 Ion exchange or breakpoint chlorination or none
2 Ion exchange relocated after carbon adsorption
3 Breakpoint chlorination employed in P.O. treatment
Note: Ion exchange process operated intermittently
% Removals are accumulative
34
-------
RAW WASTEWATER
o>
E
Q
O
O
50-
40-
30
20
0
0
0
! i
I , I
2 3 40123
THROUGH-PUT, MG
FIGURE 11 - Daily Removal of COU in Carbon Adsorption
4 5
-------
RAW WASTEWATER
O>
E
o
o
h-
0
0
3 40
THROUGH-PUT,
MG
FIGURE 12 - Daily Removal of TOG in Carbon Adsorption
-------
carbon replacement cycle.
During the use of the very high pH backwash procedure (600 mg/1 NaOH)
in the third replacement cycle, however, the adsorption process did not
exhibit the normal removal pattern. The residual COD and TOC .in the
first month of operation (November 1970) after the second carbon re-
placement were unusually high and averaged 17 and 10 mg/1, respectively.
With the discontinuance of the high pH backwash in December 1970, the
average residual COD and TOC of 17 and 9 in the last month (March 1970)
before the third carbon replacement were the same as in November and
thus indicated that the suspected fouling (CaCO scale) did not remain
on the carbon.
The presence of chlorine in the influent to the adsorption process did
not alter the basic pattern of increasing organic residuals in the
effluent with increasing carbon loading. The average COD and TOC
residuals (Tables 5 and 6) increased, respectively, from 13.6 mg/1 and
6.2 mg/1 in the first month (April) with chlorination and after the
third carbon replacement to 20.5 mg/1 and 10.1 mg/1 in the last month
(July 1971) of the carbon study.
The adsorption process with a normal average column hydraulic loading
of 7 gpm/ft2 exhibited organic loadings as described in Figure 13 and
Table 7. In the operation without chlorine in the influent, the
organic loadings increased with increasing biological activity on the
carbon columns from 0.13 Ib of TOC/lb of carbon (0.4 Ib of COD/lb of
carbon) at the first replacement to 0.30 Ib TOC/lb of carbon (0.74 and
0.70 Ib of COD/lb of carbon) for the next two replacements. With
chlorine in the influent, the carbon loading was 0.25 Ib of TOC/lb of
carbon at the end of the study. Further work is needed to determine
whether the reduced biological activity with breakpoint chlorination
will produce a substantial decrease in the activated carbon loading.
37
-------
CO
0.5
o
oo
cr
oo
0.4
0.3
o 0.2
00
-1 0.1
0
0
NO CHLORINE
EFF. TOO ( )
CHLORINE
46 8 10 12 14 16
FLOW, MILLION GALLONS
18 20
FIGURE 13 TOC Loading on Activated Ca-bon
-------
TABLE 1
ORGANIC LOADINGS ON ACTIVATED CARBON
First Stage
Second Stage
Date
3-10-70
6-28-70
10-26-70
3-16-71
7-30-71
Operation
Startup
Replacement
Replacement
Replacement
End
Flow
MG
4
9
14
18
0
.2
.2
.6
.8
Ib
Ib
0
0
0
COD
Carbon
-
.41
.74
. 70
3
Ib TOC
Ib Carbon
-
0.13
0.30
0.30
0.25
Ib
Ib
COD
Carbon
-
0.15
0
0
.16
.34
3
Ib TOC
Ib Carb<
-
0.066
0.063
0.17
0.12
First stage included first 2 carbon columns
Second stage included last 2 carbon columns
The flow is the accumulated flow in millions of gallons
Residual chlorine in the breakpoint process prevented
accurate measurement of COD
-------
SECTION VI
OVERALL PERFORMANCE OF PHYSICAL-CHEMICAL TREATMENT
Organics and Solids Removal
Lime clarification removed approximately 80% of the organics (TOC, BOD,
and COD) from the wastewater (Tables 5, 6, & 8). Filtration increased these
removals only slightly (1-4%) producing water containing 15-20% of the
original organics. Ion exchange then removed from 15 to 35% of these
residual ("soluble") organics, increasing the accumulated organic
removals to more than 85%. Breakpoint chlorination, however, when
substituted for the ion exchange did not increase the accumulated
organics removal.
With virgin carbon, the adsorption system in the first month removed
75% of the TOCf 78% of the BOD, and 80% of the COD entering the columns
with average residuals of 3.7 mg/1 of BOD and TOC and 8.2 mg/1 of COD.
Adsorption increased the overall organic removals to 97-98%.
During typical operation with high organic loading and normal biological
growth on the activated carbon in the fourth month, the carbon removed
55% of the influent (soluble) TOC, 22% of the BOD , and 56% of the COD
with average residuals of approximately 8 mg/1 for BOD and TOC and
19 mg/1 for COD. The overall organic removals through the plant
decreased to between 91 and 94%. The spent carbon (the first half of
the carbon) was then replaced at the end of the fourth month (June 1970).
The replacement of the spent carbon with virgin carbon reduced the
organic residuals for the fifth month (July 1970) to approximately the
same level as those experienced during the first month of operation
(March 1970). As the organic loading on the carbon increased, the
pattern of decreasing overall organics removal efficiencies was again
noted. The relocation of the ion-exchange process in August to a
position following .the carbon adsorption process did not significantly
increase the removal of organics from the adsorption column effluent
nor alter the overall removal efficiencies.
Substitution of breakpoint chlorination for ion exchange also did not
alter the basic overall organic removal pattern although the final
effluent BOD from adsorption with prior chlorination for nitrogen
removal was slightly lower than earlier operation and averaged about
5 mg/1. For the entire operation, the physical-chemical plant removed
approximately 95% of the COD, 92% of the TOC and 96% of the BOD with
average residuals of 16 mg/1 of COD, 8 mg/1 of TOC, and 6 mg/1 of BOD.
Laboratory ozonation (29) of the effluent from the carbon columns with
50 mg/1 of O reduced the COD to 4 mg/1, the TOC to 4 mg/1 and the BOD
to zero.
As with organic removal, lime clarification produced consistently good
solids removal. During the study, lime clarification produced an average
40
-------
TABLE 8
BOD REMOVAL IN PHYSICAL^CHEMICAL TREATMENT
Of
RAW WASTEWATER
Month
Mar.
Apr.
May
June
July
Aug.
Sept
Oct.
Nov .
Dec .
Feb.
Mar.
Apr.
May
June
July
Aug.
70
70
70
70
70
70
. 70
70
70
70
71
71
71
71
71
71
71
Raw Clarified
mg/1 mg/1 % Rem.
142
126
158
111
99
98
115
131
141
166
134
152
165.3
152.4
129
131
123
31.4
28.3
26.1
18.1
13.0
16.2
19.8
28.5
29.7
31.9
32.3
28.6
36.4
28.1
19.3
20.7
20.1
78
78
83
84
86
83
83
78
80
81
76
81
78
82
85
84
83
Filtered
mg/1 % Rem.
23
24
19
15
11
13
20
21
23
29
22
-
21
-
-
-
-
.7
.3
.4
.1
.8
.8
.6
.6
.9
.3
.7
-
.2
-
-
-
-
83
81
88
86
88
86
82
84
83
82
84
87
Nitrogen Rem
mg/1 % Rem.
16.7
18.6
12.6
9.6
7.8
4.32
4.0
'4.4
87
85
92
91
92
95
97
98
--
Adsorbed
mg/1 % Rem
3.7
6.4
6.5
7.5
3.0
4.7
6.3
7.4
8,3
8.1
8.0
5.8
4.6
4.6
4.7
7.6
98
95
96
93
97
96
95
94
94
93
94
96
97
97
97
94
__
1 Ion exchange or breakpoint chlorination or none
2 Ion exchange relocated after carbon adsorption
3 Breakpoint chlorination employed in P. C. treatment
Note: Ion exchange process operated intermittently
% Removals are accumulative
42
-------
suspended solids removal of 88% as shown in Table 9. Filtration
increased the removal to 97%. Further decreases in suspended solids
did not occur across the remaining plant processes. The overall
removals continued to average approximately 97% with about 4 mg/1 of
residual suspended solids.
Phosphorus Removal
Lime clarification consistently removed 97% of the total phosphorus
from the raw wastewater. In March of 1970, lime clarification reduced
the total phosphorus (Table 10) to 0.45 mg/1 as P. Treatment through
adsorption further reduced the total phosphorus to 0.21 mg/1
(98% removal). As the water temperature increased in the summer, the
total phosphorus residuals gradually decreased to 0.17 mg/1 after
clarification, and to 0.08 mg/1 after complete treatment. With the
onset of colder weather, the phosphorus removal again decreased slightly
with residuals of about 0.2 mg/1. For the complete study, physical-
chemical treatment removed 98% of the total phosphorus with an average
phosphorus residual of 0.14 mg/1.
Nitrogen Removal
The raw wastewater in the District of Columbia contained chiefly organic
nitrogen and ammonia with usually less than about 0.07 mg/1 of nitrate
nitrogen. Lime precipitation removed about 30% of the total nitrogen
(TKN and nitrate) from the water (Table 11) . The nitrogen removed in
clarification was approximately 75% organic (particulate) and 25%
(8% of the total original nitrogen) ammonia (Table 12). The small but
consistent ammonia removal across the clarification process probably
occurred by stripping from the water at high pH into the air.
Filtration increased the total nitrogen removal to about 35%. Ion
exchange (clinoptilolite) essentially removed only ammonia from the
wastewater. While with careful operator surveillance, the exchange
process (Figure 9) reduced the NH -N to less than 1 mg/1 with removals
greater than 90%, when unattended the ammonia residuals in the ion
exchange effluent increased. For the first five months with the
exchange process ahead of adsorption, the clinoptilolite removed approxi-
mately 75% of the ammonia entering the process with nitrogen residuals
averaging 2.6 mg/1 of NH -N and 2.3 mg/1 of other nitrogen (4.9 mg/1
of total N, Table 12). Little additional removal of nitrogen occurred
in the adsorption process.
In the first five months, the ammonia concentrations in the water from
the ion-exchange processes consistently increased after carbon adsorp-
tion. The soluble organic nitrogen entering the carbon columns
(2.3 mg/1) apparently was partially converted by the biological activity
on the carbon to ammonia nitrogen. Thus, for maximum nitrogen removal
potential, the exchange process was relocated after carbon adsorption
for August, September and December 1970. The exchange process only
42
-------
TABLE 9
SUSPENDED SOLIDS REMOVAL IN PHYSICAL-CHEMICAL TREATMENT
OF
RAW WASTEWATER
Month
March 70
April 70
May 70
June 70
July 70
Aug. 70
Sept. 70
Oct. 70
Nov. 70
Dec. 70
Jan. 71
Fei> . 71
March 71
April 71
May 71
June 71
July 71
Aug. 71
Ion
Ion
3 ..._
Raw Clarified
mg/1
180
160
164
159
144
139
136
168
173
163.4
147
156
164
159
162
146
162
exchange
exchange
7 j_ _7
mg/1 %Rem
19.3
22.7
12
6.3
8.9
17.5
23.5
25.5
25.0
24.4
9
14.9
13.2
22.5
34.3
21.7
32
89
86
93
96
94
87
83
85
85
85
94
90
92
96
79
85
80
or breakpoint
relocated
,, 1 * j_' _
Filtered
mg/1
4.6
10.3
5.1
4.6
4.3
4.2
4.2
4.7
6.9
9.2
2
4.9
4.5
4.3
4.3
5.9
4.9
%Rem
98
94
97
97
97
97
97
97
96
94
99
97
97
97
97
96
97
chlorination
after carbon
~ 7 , ~ J -.' ~
Nitrogen Rem~
mg/1
7.1
6.4
21
5.1
9
7.02
3.62
2
4.03
7.53
2.53
3.73
3.33
or none
%Rem
96
96
87
97
94
95
97
98
95
98
98
98
Adsorbed
mg/1
2.7
4.6
6
4.3
4
5.9
4.9
3.3
4.4
5.9
2.0
2.8
3.0
3.8
3.0
3.6
"~
%Rei
99
97
97
97
97
96
96
98
98
96
99
98
98
98
98
98
"
adsorption
~r> n
Note: Ion exchange process operated intermittently
% Removals are accumulative
43
-------
TABLE 10
PHOSPHORUS REMOVAL IN PHYSICAL-CHEMICAL TREATMENT
OF
RAW WASTEWATER
Month
March
April
70
70
May 70
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Feb.
March
April
70
70
70
70
70
70
70
71
71
71
May 71
June
July
Aug.
71
71
71
Raw
mg/1
8.7
8.5
8.9
8.5
8.1
8.0
8.2
8.1
8.4
8.3
7.5
7.6
8.1
7.3
7.0
6.9
6.7
Clarified
mg/1 %Rem
0.45
0.30
0.25
0.16
0.17
0.18
0.25
0.25
0.30
0.37
0.29
0.32
0.36
0.30
0.46
0.27
0.30
95
97
97
98
98
98
97
97
96
96
96
95
96
96
93
96
95
Filtered
mg/1 %Rem
0.31
0.18
0.1'.
0.14
0.14
0.11
0.16
0.18
0.22
0.22
0.26
0.27
0.39
0.23
0.19
0.18
0.23
96
98
98
98
98
99
98
98
97
97
97
96
96
97
97
97
97
Nitrogen Rem^
mg/1 %Rem
-
0
0
0
0
0
0
-
-
0
-
-
0
0
0
0
0
.16
.16
.10
.08
.092
.11
2
.08
.373
.323
.IB3
.173
.163
98
98
99
99
99
99
99
,
96
96
97
97
98
Adsorbed
mg/1 %Rem
0.21
0.16
0.12
0.14
0.07
0.08
0.09
0.11
0.11
0.12
0.23
0.13
0.21
0.23
0.14
0.14
_____
98
98
99
98
99
99
99
09
99
99
97
98
97
97
98
98
__
Ion exchange or breakpoint or none
2
Ion exchange relocated after carbon adsorption
Breakpoint chlorination employed in P.C. treatment
Note: Ion exchange process operated intermittently
% Removals are accumulative.
44
-------
TABLE 11
TOTAL NITROGEN REMOVAL IN PHYSICAL^CHEMICAL TREATMENT
OF
RAW WASTEWATER
Month
Mar.
Apr .
May
June
July
Aug.
Sept
Oct.
Nov.
Dec.
Feb.
Mar.
Apr.
May
June
July
Aug.
70
70
70
70
70
70
. 70
70
70
70
71
71
71
71
71
71
71
Raw
mg/1
22.9
21.0
21.7
21.2
19.8
20.4
21.9
23.0
22.8
23.1
22.4
23.6
23.0
21.4
20.3
18.8
17 .9
Clarified
mg/1 % Rem.
15.5
15.6
15.1
14.0
12.9
13.5
15.5
16.1
16.0
17 .3
14.9
16.8
15.9
16.4
14.0
13.9
13.6
32
26
31
34
35
34
24
30
30
25
33
29
31
23
31
33
24
Fil tared
mg/1 % Rem.
16.4
15.8
13.5
12.3
12.9
14. 8
15.5
15.9
16.9
14.7
15.2
14.2
14.1
12.2
11.6
10.7
28
24
36
38
37
32
33
30
27
34
36
38
34
40
38
40
Nitrogen Rem
mg/1 % Rem.
4
6
4
4
5
4
4
5
5
3
2
2
2
.5
.0
.3
.4
.3
.72
.4
.o2
.63
.23
.63
.33
.63
80
72
78
79
73
77
80
78
75
85
87
88
85
Adsorbed
mg/1 % Rem
3.1
4.2
9.7
8.1
9.2
11.9
13.7
14.8
13.5
15.2
13.4
14.3
4.6
2.6
2.5
2.6
87
80
41.7
37
36
41
34
40
39
80
88
88
86
1 Ion exchange or breakpoint chorination or none
2 Ion exchange relocated after carbon adsorption
3 Breakpoint Chlorination employed in P.C. treatment
Note: Ion exchange process operated intermittently
% Removals are accumulative
45
-------
tfcv
TABLE 12
NITROGEN REMOVAL WITH ION EXCHANGE
Month Influent Clarification Filtration Ion Exchange Adsorption % Removals
1970 Total N NH -N Total N NH -N Total N NH -N Total N NH -N Total N NH -N Total N NH -
mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1
March
April
May
June
July
Aug .
Sept.
Dec.
23.4
20.7
21.0
20.5
20.6
19.7
21.9
23.1
13.7
11.3
12.8
12.7
12.1
12.3
13.3
18.5
17 .1
15.9
14.8
13.4
13.0
13.1
15.5
17.3
12.8
10.9
12.1
11.0
10.4
11.4
13.0
15.1
16.1
15.1
13.9
13.1
12.2
11.6
14.8
16.9
13.0
10.9
11.7
10.6
10.8
10.9
12.0
13.9
4.1
5.8
4.9
4.5
5.3
4.1
4.4
5.0
1.7
2.7
2.9
2.6
3.3
2
3.3
2
3.1
2
4.3
3.2
4.1
5.2
5.1
4.8
11.2
13.7
15.2
3.0
3.8
4.0
3.6
3.6
11.2
12.7
14.1
86
80
75
75
77
79
80
78
78
66
69
72
70
73
77
77
1 Data represents only periods of operation with the Ion Exchange system on stream
2 Ion exchange located after adsorption
-------
removed about 75% of the influent ammonia with an average residual of
3.6 mg/1 of NH -N, but the amount of other (organic) nitrogen in the
exchange effluent (product water) decreased from an average of 2.3 mg/1
to 0.9 mg/1i Thus with careful operator surveillance, the ion-exchange
process could have reduced the total nitrogen to approximately 2 mg/1
and produced a total nitrogen removal of about 90%. As operated,
however, the total nitrogen removal by physical-chemical treatment
with the ion exchange averaged 78%.
Total nitrogen removal (Table 11) by physical-chemical treatment without
an ammonia removal process (October 1970, November 1970, February 1971,
and March 1971) averaged about 39%. The total nitrogen removal after
filtration during the same period was about 33%. Thus carbon adsorption
increased the accumulated total nitrogen removal by only 6%.
The use of breakpoint chlorination for ammonia removal (Tables 11 and 13)
produced an overall total nitrogen removal of about 86% during the
study. The average residual total nitrogen of about 2.8 mg/1 as N
contained approximately 1 mg/1 of ammonia nitrogen. Thus a second
application of chlorine after carbon adsorption could be employed to
reduce the total nitrogen to about 2 mg/1.
47
-------
TABLE 13
TYPICAL NITROGEN REMOVAL WITH BREAKPOINT CHLORINATION
TKN-N
mg/1
NH+-N
4
mg/1
NO -N
mg/1
NO -N
mg/1
Total
N
mg/1
Ace.
% Rem.
of N
Raw
Clarified
Fil tared
Breakpoint
Carbon
20.3
14.0
12.2
2.0
2.5
12.9
12.3
10.8
0.4
1.0
0
0
0
0.6
0
0 20.3
0 14.0 31
0 12.2 40
0 2.6 87
0 2.5 88
1 June 1970
48
-------
SECTION VII
ADSORPTION MECHANISM
The results from this work and from earlier work (2) suggest a. simple
two phase mechanism for the carbon treatment process. Initially the
carbon physically adsorbs most of the soluble organics in the wastewater
including the highly-soluble poorly-adscrbable but biodegradeable
organics in the clarified raw wastewater. The adsorbed organics on the
carbon surface are readily available for biological degradation which
begins quickly on the carbon columns.
In continuous operation, the developing "fixed film" biological activity
removes by biosorption portions of the highly-soluble poorly-adsorbable
organics and reduces their breakthrough in the carbon column effluent.
The biological growths, however,- coat the carbon surfaces and block the
pore structure of the carbon granules thus reducing sites for physical
adsorption. The heaviest growth occurs at the beginning of the adsorp-
tion column and gradually spreads toward the end of the column. Without
oxygen added to the wastewater, the biological activity is anaerobic and
produces H S from the sulfates in the water. The carbon near the end of
the column continues to physically adsorb organics from the water but the
biological activity releases biological end products and cellular
material into the water.
As shown in the earlier work (2), the collodial biological cellular
material appears as turbidity in the wastewater and is not readily ,
adsorbed on the downstream carbon granules. The collodial cell material
is too srna.ll to be filtered by the carbon granules and too big to diffuse
into the pores of the carbon. The collodial material and gradually
increasing amount of soluble organics appear in the column effluent.
As shown in the current study, the total concentration of organics (BOD,
TOC and COD) increases in the effluent with organic loading. Since the
biological activity occurs indefinitely, the effluent quality without
carbon replacement should eventually deteriorate until the physical
adsorptive capacity of the carbon is exhausted or is physically isolated
from the aqueous phase through blockage of the pore structure by biolog-
ical slimes. At this point, the effluent quality should stabilize at
a level related to the efficiency of the "fixed film" biological activity.
With the biological activity, the carbon loading without replacement can
increase indefinitely but this loading really represents biological
removal rather than physical-adsorption of organics. Thus the apparent
loading as shown in this work, greatly exceeds the typical physical
adsorption loading developed from batch carbon isotherm tests (2).
Even with the increased loadings, the desirability of biological activity
in the adsorption system has not been determined. While the carbon may
accelerate the development of biological activity, especially in the
-------
early operation by providing an adsorbed source of biodegradable organics,
the fully developed film May function satisfactorily without the adsorbed
organics on the carbon. Indeed if the pores of the carbon becomes clogged
with cellular debris, much, of the active surface and the adsorbed organics
would become unavailable for rapid interaction with the biological film.
Study over an extended time of parallel systems, one with a fixed film
biological column followed by carbon adsoprtion and the other a carbon
column with the biological activity supported on the carbon granules,
is needed to evalute any synergistic effects between the physical-
adsorption process and the biological process.
If "fixed film" biological activity can be continuously supported on
carbon granules, it is likely also to develop on less expensive support
media of similar size. Thus a more economical approach may be to gradu-
ally develop a fixed film biological reactor on an alternate media and
reserve the carbon treatment for chiefly physical-adsorption of the
materials leaving the ''fixed film" biological column. As the final
process in the treatment system, a filtration step with chemical floc-
culation should be added to remove the collodial cell debris passing
through the carbon column and thus improve the final product quality.
These physical-chemical treatment systems,- without a process such as
chlorination to control biological activity becomes a combination of
physical-chemical and biological processes and presents several unknowns
in terms of systems reliability. The mechanical handling of large scale
"fixed film" columns with heavy biological growth on materials the size
of carbon granules with either upflow or downflow operation and the
control of the H S production without chlorination have not been demon-
strated.
In preliminary studies in the EPA-DC pilot plant, the use of oxygen to
prevent the H S production in the downflow carbon columns produced very
heavy growths and caused high pressure losses and excessive backwashing
requirements. While more efficient backwashing/surface wash systems
may be developed to minimize growth accumulation and H S production,
-these also require large scale demonstrations.
50
-------
SECTION VIII
COSTS
The total estimated water costs (Table 14) for physical-chemical treat-
ment with solids disposal in a 300 MOD plant is 32.6$ per thousand
gallons (30). The costs include recalcination of the calcium carbonate
from the second sedimentation stage for lime recovery (approximately
50% recovery).
The cost of approximately 23$ per thousand gallons for carbon, phosphorus
and particulate nitrogen removal were based on processes with known
chemical technology and with developed large scale equipment. In contrast,
the selective ion exchange costs of approximately lOf per thousand gallons
required an estimation of the bulk production costs and the life of the
clinoptilolite. The ion-exchange process also included that of an untried
acid adsorption step to remove the ammonia from the stripping air in the
regenerant step. Finally, ion exchange technology has not been employed
at the scale of large wastewater treatment plants. Incineration data
was not available and estimates of performance were employed.
The estimated cost of breakpoint chlorination (Table 15) for nitrogen
removal is 7.32$ per thousand gallons. The estimate reveals that break-
point chlorination is a high operating and low capital cost process. The
chlorination cost is very sensitive to NH concentration increasing very
markedly with increasing NH concentration. The design criteria used in
developing the cost estimates are provided in Table 16.
51
-------
TABLE 14
PHYSICAL-CHEMICAL TREATMENT
COSTS
(Cents per Thousand Gallons)
Influent Pumping Lime Filtration Solids Carbon Ion
S Grit Removal Treatment Disposal Adsorption Exchange TOTALS
i-n
FUEL
ELECTRICITY
CHEMICALS
SUPPLIES
O & M LABOR
2
CAPITAL CHARGES
-
0
-
0
0
0
.1
.1
.4
.9
0.
0.
2.
0.
0.
4.
5
4
2
2
6
1
0.
0.
a.
i.
-
3
1
5
0
1.2
.1
.1
.1
O.9
1.5
0.
1.
0.
4.
-
3
-
9
6
8
0.5
2.6
1.4
0.6
4.6
1.7
1.4
5.2
3.8
3.6
16.9
Totals 1.5
8.0
1.9
7.6
9.7
Plant size 300 MGD based on June 1970 costs
Annual capital cost computed at annual rate of 8% including
interest and amortization.
-------
TABLE 25
BREAKPOINT CflLQRINAT,ION COSTS
FOR LIME CLARIFIED RAW WASTEWATER
Cents/1000 gals
Electricity (Mechanical Mixing) 50
**
Chemicals
Chlorine (135 mg/1) 4.22
Lime (1 Ib./lb Cl^ 1.55
Supplies .05
O & M Labor .60
***
Capital charges .60
Total 7.32
* Plant size 300 million gal/day, 15 mg/1 NH -N in,
cents/thous. gallons
** Chlorine costs $75/ton; Lime costs $24/£on
*** Annual capital cost computed at annual rate of 8% including
interest and amortization.
C1:N weight ratio 9:1
-------
TABLE 16
DESIGN CRITERIA FOR COST ESTIMATE
GRIT CHAMBERS
Detention Time (Avg), min.
LIME TREATMENT
Rapid Mix Detention Time (Peak), sec.
Lime Dose, mg/1
Lime Recovery
Return Solids as percent of flow
Flocculation Detention Time (Peak), min.
First Stage Sedimentation Overflow
Rate (Peak), gpd/ft
Second Stage Sedimentation Overflow
Rate (Peak), gpd/ft
Sedimentation Detention Time (Avg), hr.
Recarbonation or Stabilization
Detention Time (Peak), min.
Ferric Chloride Dose, mg Fe /I
Lime & Recarbonation Thickeners
Feed Rate (Peak), ft day/ton
Centrifuge Feed Rate for 36" Dia.
Bowl and 8% Feed Solids, dry ton/day
Calciner Feed Rate,
ION EXCHANGE
wet ton
furnace-day
14.7
30
350
All of 2nd
stage CaCO.
10
10
1500
2000
2.5
10
4
20
96
600
Number of Stages in Service
Direction of Flow
2
down
54
-------
TABLE 16
(CONTINUED)
ION EXCHANGE-Continued
Type of flow gravity
2
Hydraulic Loading (Peak), gpm/ft 10
Pressure Drop Across Bed
1st Stage, Ib/in 15
2nd Stage, Ib/in 10
Empty Bed Contact Time per Stage, min. 3.37
Time Each Stage in Service, hr. 24
Time in Regeneration Cycle, hr. 24
Clinoptclolite Capacity, gmN/gal 10
Clinoptololite Bulk Density, Dry, Ib/ft 50
Number of Regeneration Stages 2
Time per Regeneration Stage, hr. 4
Hydraulic Loading for Regeneration, gpm/ft 6
Type of Flow for Regeneration up
Wash Volume after Regeneration, bed volumes 12
2
Hydraulic Loading for Washing, gpm/ft 6
Air-to-Liguid Ratio for Ammonia
Stripping, ft /gal 200
Hydraulic Loading for Ammonia
Stripper, gpm/ft 2
Ratio of Ammonia Absorption Spray
( (NH ) SO with excess H SO ) to gas, gal/ft 1/480
4 <& 4 ^4
FILTERS
2
Hydraulic Loading (Avg) , gpm/ft 3
Head Loss, ft. 9
55
-------
TABLE 16
(CONTINUED)
FILTERS-Continned
2
Backwash Rate, gprn/ft 20
Backwash Time, min/day 10
2
Surface Wash Rate, gpm/ft 3
ACTIVATED CARBON
Number of Stages 2
Direction of Flow down
2
Hydraulic Loading (Avg) , gpm/ft. 3
Empty Bed Contact Time per Stage, min. 20
2
Pressure Drop per Stage, Ib/in 10
Carbon Loading, °' , 0.15
gm Carbon
Carbon Bulk Density, Ib/ft 27.5
Carbon Loss per Regeneration, % 5
2
Backwash Rate, gpm/ft 15
2
Surface Wash Rate, gpm/ft 3
Backwash Time, min. 10
56
-------
SECTION IX
REFERENCES
1. Joyce, R.S., Allen, J.B., and Sukenik, V.A. , "Treatment of
Municipal Wastewater by Packed Activated Carbon Beds." Jour.
Water Poll. Control Fed., 38, 813 (1966).
2. Bishop, D.F., et al., "Studies on Activated Carbon Treatment."
Jour, Water Poll. Control Fed., 39, 188 (1967).
3. Bishop, D.F., "Advanced Waste Treatment Research at the FWPQA-DC
Pilot Plant." Presented at the FWPCA Technical Workshop,
Fredericksburg, Va., May 13, 1969.
4. Eager, D.G. , and Reilly, D.B., "Clarification-Adsorption in the
Treatment of Municipal and Industrial Wastewaters." Jour. Wate.r
Poll. Control Fed., 42, 794 (1970).
5. Molof, A.H., and Zuckerman, M.M., "High Quality Reuse Water from
a Newly Developed Chemical-Physical Treatment Process." Presented
at the 5th International Water Pollution Research Conference,
San Francisco (July 1970).
6. Stamberg, J.B., Bishop, D.F., Warner, H.P., and Griggs, S.H.,
"Lime Precipitation in Municipal Wastewater", Chem, Engr. Progr ,
Symposium Series 107, 67, 310 (1970).
7. Stander, G.J., and Van Vuuren, L.R.J., "The Reclamation of Potable
Water from Wastewater." Jour. Water Poll. Control Fed., 41, 355
(1969).
8. Villers, R.V. . Berg, E.L., Brunner, C.A., and Masse, A.N.,
"Treatment of Primary Effluent by Lime Clarification and Granular
Carbon." Presented at the 47th Annual Meeting of ACS, Toronto
(May 1970).
9. Weber, W.J., Hopkins, C.B., and Bloom, Jr., R., "Physicochemical
Treatment of Wastewater." Jou^p. Water Poll. Control Fed., 42, 83
(1970) .
10. O'Farrell, T.P., Bishop, D.F., and Bennett, S.M., "Advanced Waste
Treatment at Washington, D.C." Chem. Engr. Prog., Symposium
Series 97, 65_, 251 (1969).
11. Burns, D.E., "Physical-Chemical Treatment of Municipal Wastewater."
Thirteenth Progress Report, FWQA, Contract No. 14-12-585
(August 1970) .
57
-------
12. English, J.N. et al., "Removals of Organics from Wastewater by
Activated Carbon." Presented at the 67th National Meeting of
the AIChE, Atlanta (February 1970).
13. O'Farrell, T.P., Frauson, P.P., Cassel, A.F., and Bishop, D.F. ,
"Nitrogen Removal by Ammonia Stripping", Jour. Water Poll. Control
Fed. 44_, 1527 (1972).
14. Pressley, T.A., Bishop, D.F., Roan, S.G., "Ammonia-Nitrogen
Removal by Breakpoint Chlorination", Environmental Science and
Technology, 6_, 622 (1972).
15. Mercer, B.W., et al., "Ammonia Removal from Secondary Effluent by
Selective Ion Exchange." Jour. Water Poll. Control Fed., 42, R 95
(1970) .
16, Mercer, B.W., private communication.
17. Cassel, A.F., Bishop, D.F., Pressley, T.A., "Physical-Chemical
Nitrogen Removal from Municipal Wastewater", AIChE, Symposium
Series 124, 68, 56 (1972).
18. Chaplin, R.M., "The influence of pH upon the Formation and
Decomposition of the Chloroderivatives of Ammonia", JACS, 53,
912 (1931).
19. Carbett, R.E., Metcalf, W.S., and Soper, F.G. , "Studies of
N-halogeno-compounds, Part IV", J. Chem. Soc., London, 1927
(1953) .
20. Czeh, F.W., et al., "Determination of Mono, Di and Trichloramine
by Ultraviolet Adsorption Spectroscopy", Anal, Chem. 33, 705
(1961).
21. Moore, E.W., Water and Sewage Works, "Fundamentals of Chlorination
of Sewage and Wastes", 98, 3 (1951).
22. Palin, A.T., "Study of the Chloroderivatives of Ammonia and
Related Compounds with Special Reference to their Formation
on the Chlorination of Natural and Polluted Waters", JAWWA,
44, 8 (1952).
23. Gales, M., Julian E., and Kroner, R., "Method for Quantitative
Determination of Total Phosphate in Water." Jour, of Am. Water
Wks. Assoc., 58, 1363 (1966).
24. "FWPCA Methods for Chemical Analysis of Water and Wastes."
U.S. Dept. of the Interior, Fed. Water Poll. Control Adm.,
Cincinnati (November 1969).
-------
25. Kamphake, L., Hannah, S.r and Cohen, J., "Automatic Analysis
for Nitrate by Hydrazone Reducation." Water Res., I_, 205
(1967) . ~"
26. Schaeffer, R.B., et al., "Application of a Carbon Analyzer in
Waste Treatment." Jour. Water Poll. Control fed., 37, 1545
(1965) .
27. "Standard Methods for the Examination of Water and Wastewater."
12th ed., American Public Health Association, New York (1965).
28. Water and Chlorine Residuals #1, study #34, Analytical Health
Service (1969).
29. Roan, S.G., Bishop, D.F., Pressley, T.A., "Ozonation of
Clarified Wastewater", Presented at the 68th National Meeting
of the AIChE, Houston, Texas, March, 1971.
30, Bishop, D.F., O'Farrell, T.P.,Stamberg, J.B., Porter, J.W..,
"Advanced Waste Treatment Systems at the FWQA-DC Pilot Plant",
AIChE Symposium Series 124, 68, ll (1972).
59
-------
SECTION X
PUBLICATIONS, PRESENTATIONS AND PATENTS
Publications
Bishop, D.F., O'Farrell, T.P., Staniberg, J.B., "Physical-Chemical
Treatment of Municipal Wastewater", Jour. Water Poll. Control Fed.
44_, 361 (1972).
Bishop, D.F., O'Farrell, T.P., Stamberg, J.Bi, Porter, J.W.
"Advanced Waste Treatment Systems at the FWQA-DC Pilot Plant", AIChE
Symposium Series 124, 68, 11 (1972).
Cassel, A.F., Bishop, D.F., Pressley, T.A., "Physical-Chemical Nitrogen
Removal from Municipal Wastewater", AIChE, Symposium Series 124, 68,
56 (1972).
O'Farrell, T.P., Bishop, D.F., "Lime precipitation in Raw, Primary
and Secondary Wastewater", AIChE, Symposium Series, 124, 68, 43 (1972).
Presentations
O'Farrell, T.P., Stamberg, J.B., Bishop, D.F., "Carbon Adsorption
of Lime Clarified Raw, Primary and Secondary Wastewaters", Presented
at the 68th National Meeting of the AIChE> Houston, Texas, March, 1971.
Pinto, A. and Bishop, D.F., "Breakpoint Chlorihation of Lime Clarified
and Filtered Raw Wastewater", Presented at the 73rd National Meeting
of AIChE, Minneapolis, August, 1972 ±
Patent Applications
Bishop, D.F., Pressley, T.A., and Cassel, A.F. , "Wastewater
Purification by Breakpoint Chlorination and Carbon Adsorption", Patent
Pending, Serial No. 178310 (Sept. 7, 1971).
Pressley, T.A., and Bishop, D.F., "Nitrogen Removal from Wastewaters
by Breakpoint Chlorination", Patent Pending, Serial No. 175902
(Aug. 30, 1970) .
Schuk, W.W., Pressley, T.A., and Bishop, D.F:, "An Automatic Control
System for the Safe and Economical Removal of NH by Breakpoint
Chloriaation", Patent Pending, Serial No. 251777 (May 9, 1971).
60
OU.S. GOVERNMENT PRINTING OFFICE:1974 546-315/226
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
/. Report No,
3. Accession No.
w
4. Title
PHYSICAL-CHEMICAL TREATMENT OF RAW MUNICIPAL WASTEWATER
7.
B1shop> QOH Off p., O1 Parrel 1, Thomas P.,
Cassel . Alan F. , and Pinto, Adolph P.
Organization
EPA-DC Pilot Plant
5000 Overlook Avenue S.W.
Washington, D.C. 20032
ENVIRONMENTAL PROTECTION AGENCY
inf Bait
g. F'::rioTtning QrtfKaisttfoti
10. Project No.
11010 EYM
11. Contract/Grant No.
14-12-818
Type ^i Repot i and
Period Covered
15, Supplementary Notes
Environmental Protection Agency Report Number
September 1973. '
EPA-670/2-73-070,
is. Abstract Physical-chemical treatment of raw wastewater consisted of two-stage lime
clarification with intermediate recarbonation, filtration, pH control, ion exchange
or breakpoint chlorination, and carbon adsorption. Lime treatment with approximately
300 mg/1 of CaO increased the wastewater pH to 11.5 and removed 96% of the phosphorus
and 80% of the organics. In the second stage, recarbonation with 120 mg/1 of C02 and
mineral addition of 5 mg/1 of Fe+++ reduced the pH to 10.0 and precipitated excess Ca"1"*"
as CaC03. Dual media filtration decreased effluent suspended solids and total phosphorus
to less than 5 mg/1 and 0.15 mg/1 as P, respectively. Addition of 10 mg/1 chlorine to
the filter influent controlled biological growth within the filter and produced filter
runs of greater than 50 hours. With extensive operator surveillance, the clinoptilolite
exchange media reduced the NH3 to less than 1 mg/1 as NHj-N. Breakpoint chlorination
oxidized the NH3 to N2» leaving a residual NHo-N concentration of less than 0.4 mg/1.
The 20 mg/1 of soluble BOD entering the granular carbon columns produced anaerobic
biological growth on the carbon, which contributed to heavy H£S production and high
carbon losses during backwash. Breakpoint chlorination ahead of carbon adsorption
minimized biological activity.
The complete physical-chemical system, with ion exchange, removed 98% of
the phosphorus, 95% of the organics (COD) and 78% of the total nitrogen. With
breakpoint chlorination, the complete system removed 98% of the phosphorus, 94% of the
oraamcs (COD) and 86% of the total nitrogen.
17a. Descriptors
Wastewater Treatment
Anaerobic Conditions
*Filtration
*Adsorption
,. *I on,.Ex change
17b. Identifiers
*Physical-Chemical Treatment
Lime Clarification
Carbon Adsorption
Breakpoint Chlorination
Recarbonation
Biochemical Oxygen Demand
Calcium Carbonate
Chemical Oxygen Demand
Colloids
Ammonia
Organic Loading
*Flocculation
Phosphorus
Nitrogen
Lime
Sedimentation
17c. COWRR Field & Group
05D
18. Avail ability
(Report)
(Pag?)
St,
. of
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 2O24O
Abstractor Kent S. KJsenbauer
I institution ENVIRONMENTAL PROTECTION AGENCY
WRSIC 1O2 (REV. JUNE 1971)
------- |