vvEPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-81-029 Apr 1981
Project Summary
Removal of Nitrate from
Contaminated Water
Supplies for Public Use
Gerald A. Guter
The general applicability of three
treatment processes for removal of
nitrate from public water supplies are
evaluated: reverse osmosis (RO), ion
exchange, and the combination of RO
followed by ion exchange. The
evaluation consists of using labora-
tory size and field-test equipment to
establish design criteria and operating
experience useful for designing a full-
scale plant of approximately 1 mgd
capacity.
Ion exchange column tests were
conducted with five strong-base anion
exchange resins on nitrate-laden
waters of various anion compositions.
From this work, estimates of product
water quality and the bed volume
capacity for feedwater of any compo-
sition can be made. Also, a working
hypothesis was developed from an
analysis of the data about how the
chemical structure of resins can be
practically altered to obtain nitrate
selectivity.
A 20-inch diameter pilot anion
exchange column containing 4.36 cu.
feet of resin, was designed and
operated for over 1 year. Data from
this column operation are used to
verify estimates of pilot column
performance and to project the cost
for equipment and regenerant for a
well site installation to treat up to 1
mgd. Because of the interim nature of
this report, only preliminary data are
reported on the operation of a 20,000-
gallon per day RO system.
This Project Summary was devel-
oped by EPA's Municipal Environ-
mental Research Laboratory, Cincin-
nati, OH, to announce key findings of
the research project that is fully
documented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
Three treatment processes were
objectively evaluated for removing
nitrate from public water supplies:
reverse osmosis (RO), ion exchange,
and RO followed by ion exchange. Both
laboratory and field equipment were
used to establish design criteria and
operating experience usefulfor
designing full-scale plants of approxi-
mately 3,800 m/day (1-mgd) capacity.
Work emphasizing the ion exchange
process was done during the period July
1978 to April 1980, and work is
continuing on ion exchange and RO.
The remainder of the project will be
discussed in the final report under this
grant. Note that ion exchange experi-
ments are being continued under this
program and could alter conclusions
reported here concerning the cost of the
ion exchange process.
Methods and Materials
All tests were conducted at a well site
(No. 3) owned and operated by the
McFarland Mutual Water Company in
McFarland, California. Nitrate-nitrogen
levels for this water were 16 to 23
-------
mg/L, well above the 10-mg/L maxi-
mum contaminant level. Sulfate levels
were greater than 300 mg/L.
Site improvements were designed for
well No. 3 to accommodate equipment
for conducting both field and laboratory-
scale experiments. Water was supplied
directly to a concrete pad and trailer
from an existing surge tank. All product
and wastewaters were discharged from
the pad directly into the city sewer
system. The 9.1- x 9.8- m (30- x 32-ft)
pad was large enough to accommodate
a field test ion exchange system, a field
test RO system, and a single module RO
system with the necessary tanks for
temporary water storage. A trailer
adjacent to the pad housed a field office
and limited laboratory facilities.
A source of well No. 3 water was
available in the trailer for experimental
tests on various ion exchange resins in
5.1- cm (2-in.) diameter columns.
Synthetic mixtures were prepared and
pumped directly at measured flow rates
through the ion exchange columns. Five
ion exchange resins were selected for
study with the 5.1 - cm (2-in.) laboratory-
size ion exchange columns. The
selection was based on previous work,
which tested 32 commercially available
anion exchange resins for application of
both single-bed and two-bed processes.
Because only the single-bed process
was chosen for this study, tests were
limited to strong-base anion exchange
resins. The previous study and discus-
sions with resine manufacturers indi-
cated that none of the resins would
exhibit exceptional selectivity for nitrate
ion over other major anions
A Culligan HI-FLO 5 Water Softener
Model 1 50* was installed and operated
on the pad at well No. 3 The completely
automatic water softener was converted
to a semiautomatic anion exchanger by
installing an industrial timer and anion
exchange resin (Duolite A-101D).
Conclusions
Engineering Aspects
1. Design parameters have been
developed and tested for a conventional
single-bed ion exchange process with
downflow regeneration to remove
nitrate from well waters. Testing was
conducted using both laboratory
"Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use
columns and a 50.8-cm (20-m.)
diameter pilot column.
2. The study indicates that automatic
ion exchange equipment, which is
commonly used by the water softening
industry, can be adapted for nitrate
removal. The equipment can be
installed at a well site for direct
treatment of well water a nd operated on
demand without storage.
3. The selected resin was effective
for nitrate removal at loading rates
above 48.9 m/h (20 gpm/ft2) of bed
area (1.38 bed volumes/min). This rate
was the upper limit of the test equip-
ment used. Such high flow rates bring
costs for equipment and resin quantities
to low practical levels
4 Capital equipment costs for an ion
exchange system to treat half of a 3,800
mVday (1-mgd) production well are
estimated to be less than $90,000
installed. This estimate is based on
moderate nitrate-nitrogen levels of less
than 14 mg/L in well water, sulfate
levels of less than 200 mg/L, and
blending of treated and raw water to
produce a product containing less than
10 mg/L nitrate-nitrogen.
The corresponding equipment cost
estimate for a system to treat all water
from a 1-mgd production well is less
than $150,000. This figure is based on
high nitrate levels in raw water (about
23 mg/L as nitrate-nitrogen), high
sulfate levels (about 300 mg/L, and ion
exchange treatment to reduce nitrate-
nitrogen to less than 10 mg/L without
blending.
5. A significant operating cost for the
process is the cost of sodium chloride
used as a resin regenerant. A method is
presented to estimate the sodium
chloride requirements for regenerating
the resin used in nitrate removal from
waters of various compositions.
Because anion exchange resins are
quite selective for sulfate ion, the
presence of sulfate m raw water
decreases the efficiency of the resin in
absorbing nitrate. In this study,
however, sulfate was easily removed
from the spent resin by the sodium
chloride regenerant m nearly stoichio-
metric proportions, whereas excess
regenerant is required for nitrate
removal. Nonetheless, the overall effect
of sulfate is to increase the salt required
to remove nitrate per unit quantity of
water treated. This study also confirmed
that large quantities of regenerant (320
kg/m3 or 20 Ib/ft3 of resin) are required
to remove most of the nitrate from the
spent resin. Not all nitrate need be>
removed, however, to reduce nitrate-'
nitrogen levels in treated water to less
than 10 nng/L.
For the McFarland wells, the salt
costs for lowering nitrate-nitrogen
levels to 7 to 10 mg/L ranges from an
estimated 1.90/1,000 gal of blended
water (or S6.10/acre-ft)forwellNo. 2 to
100/1,000 gal of treated water (or
$32.50/acre-ft) for well No. 3. Water
from the latter well represents a
particularly difficult water to treat as
nitrate-nitrogen concentrations are
near 23 mg/L, and sulfate levels are
above 300 mg/L. Nitrate-nitrogen
concentrations in well No. 2 are near 14
mg/L, and sulfate levels are near 200
mg/L. Salt requirements for waters of
other compositions are given in the
report text.
6. To achieve efficient nitrate
removal, good brine and influent flow
distribution are essential and may
require modifications of commercially
available softening equipment. A
method of declassification (thorough
mixing) of the resin after downflow
regeneration should also be incorporated
in the regeneration cycle.
7. During the regeneration cycle,
wastewater is produced that is rich in
sodium sulfate, chloride, and nitrate.
Continuous operation of well No. 2
would produce more than 45.4 m3
(12,000 gal) of wastewater/day. Con-
tinuous operation of well No. 3 would
produce an average of 146.4 m3 (38,686
gal) of wastewater/day.
Theoretical Aspects
1. Previous studies have shown that
the higher resin selectivity for sulfate
over nitrate gives rise to some chroma-
tographic sulfate enrichment in the
upper portion of spent columns and
nitrate enrichment in the lower portion.
In studies using McFarland well No. 3
water, such chromatographic separa-
tion was observed only for microporous
Type I resins. Microporous Type II and
macroporous resins appeared to have
nearly equal nitrate and sulfate selec-
tivities. The difference in behavior can
be attributed to the greater porosity of
microporous Type I resins and/or the
lesser steric requirements of the Type I
resins. This effect has practical signifi-
cance because it indicates that nitrate
selectivity might be increased over
sulfate selectivity by decreasing
porosity and modifying the structure
about the quaternary ammonium ion. To
-------
be of practical use, resin selectivity for
nitrate must be made to exceed that for
sulfate to reverse the order of the
enriched portions of the spent resin
column
2. Although ion exchange can be
used with minimal salt requirementsfor
some waters, the use of sodium chloride
as a regenerant is an obvious disadvan-
tage for extensively employing the
conventional single-bed ion exchange
process because of waste disposal
requirements Spent brine can be
separated into sulfate-rich and nitrate-
rich fractions for recovery purposes, but
little effort has been spent on developing
recovery processes or on using alterna-
tive regenerants that could make
recovery or reuse more practical.
Recommendations
1. Pilot column studies should be
extended and conducted on waters of
various compositions to verify the
method of estimating engineering
design parameters and process require-
ments These studies should include
Type II resins because of their greater
capacity and potential for upflow
regeneration.
2. Efforts should be made to reduce
regenerant requirements to the lowest
level practicable Such reduction can be
achieved by recycling portions of brine
and brine rinse as well as backwash
waters. Upflow regeneration should
also be studied as a method to achieve a
low nitrate leakage into the column
effluent. This method would make all
treated water blendable with raw water
and could reduce salt requirements per
unit of water produced. The use of RO in
conjunction with ion exchange is
another approach to reducing regener-
ant requirements RO can reduce the
total dissolved solids (TDS) load on the
resin and may in some cases provide a
brine useful for resin regeneration.
3. A demonstration plant of 1,900-to
3,800-m3/day (0.5- to 1-mgd) capacity
should be installed and operated to
obtain actual operational experience
regarding reliability, health, safety, and
costs.
Although the ion exchange process
has been used for many years for
industrial applications and for removing
hardness from domestic supplies, no
significant operating experience has
been obtained on a full-scale domestic
system for nitrate removal Use of the
process for this purpose cannot be
considered as a standard engineering
application until the process has been
successfully demonstrated on a full
scale.
4. Efforts to synthesize nitrate-
selective resins should be continued to
make the process more attractive to
sulfate-laden waters. In such studies,
close attention must be paid to the
regeneration requirements of the new
resins.
This interim report is the seventh of a
series of quarterly progress reports in
fulfillment of a contract under Environ-
mental Protection Agency Grant R-
805900-01 to the McFarland Mutual
Water Company of McFarland, California.
The reader should be aware that ion
exchange experiments are being
continued under this program and could
alter conclusions concerning cost of the
ion exchange process reported herein.
Gerald A. Guter is with Boyle Engineering Corporation, Bakersfield, CA 93302.
Richard Lauch is the EPA Project Officer (see below).
The complete report, entitled "Removal of Nitrate from Contaminated Water
Supplies for Public Use," (Order No. PB 163 206; Cost: $11.00. subject to
change! will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at: •
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
> US GOVERNMENT PRINTING OFFICE 1961-757-012/7074
-------
. Postage and
United States Center for Environmental Research Fees Paid
Environmental Protection Information Environmental
Agency Cincinnati OH 45268 Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED Third-Class
Bulk Rate
LUtl W TILLO
V t;PA
S DKAKbORN ST
CHICAGO TL 60604
------- |