J'/
United States
Environmental Protection
Agency
Water Engineering
Research Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/S2-88/044 Oct. 1988
v°/EPA Project Summary
Alternative Oxidant and
Disinfectant Treatment
Strategies for Controlling
Trihalomethane Formation
Philip C. Singer
To comply with the maximum
contaminant level (MCL) for total
trlhalomethanes (TTHM), many util-
ities have modified their pre-
oxidation and disinfection practices
by switching to alternative oxldants
and disinfectants in place of free
chlorine. To evaluate the impact of
these changes, a research project
was initiated to study several water
treatment plants that had recently
adopted the use of chlorine dioxide,
ozone, potassium permanganate, or
chloramines to partially or fully offset
the use of free chlorine.
The results of the study showed
that total organic halide (TOX)
formation paralleled THM formation
at all eight of the utilities inves-
tigated. The alternative pre-
treatment oxidants and disinfectants
were depleted rapidly as a conse-
quence of the high TOC concen-
trations in the waters examined. This
implies that residual oxidants and
disinfectants will not be carried very
far into the process train, causing
disinfection and oxidation
effectiveness to be reduced. The
case study results presented sug-
gest that many utilities, particularly
those with high TOC concentrations,
will be unable to comply with a
significantly reduced MCL for TTHM's
using only alternative oxidants and
disinfectants without sacrificing fin-
ished water quality. Further research
is recommended before alternative
oxidants and disinfectants can be
promoted for extensive use.
This Project Summary was
developed by EPA's Water Engineering
Research Laboratory, Cincinnati, 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
The final rule establishing an MCL
for total trihalomethanes (TTHM) in
November 1979 was followed in March
1982 by a guidance document for utilities
in which the U.S. Environmental
Protection Agency (U.S. EPA) proposed
treatment technologies that could be
used to control trihalomethane (THM)
levels. Three of the five "generally
available" treatment methods included
the use of chloramines (NH2CI) or
chlorine dioxide (ClOa) as alternative or
supplemental disinfectants or oxidants,
and the substitution of chloramines,
chlorine dioxide, and potassium
permanganate (KMn04.) as pre-oxidants
in place of chlorine. The proposal also
included, as an additional treatment
method for consideration but which was
not identified as "generally available," the
use of ozone (Oa) as an alternative or
supplemental disinfectant or oxidant.
These proposals were subsequently
adopted by the U.S. EPA in February
1983.
Many water treatment plants have
modified their method of disinfection to
include the use of chlorine dioxide,
-------
ozone, or permanganate as pre-
oxidants with free chlorine as the final
disinfectant, and free chlorine as pre-
oxidant and primary disinfectant with
combined chlorine as the final
disinfectant.
When this project began in
September 1983, numerous utilities were
considering a modification in their
oxidant and disinfectant treatment
strategies to comply with the THM
regulation and still provide an
aesthetically acceptable and biologically
safe (pathogen-free) supply of water to
their customers. However, before such
modifications could be encouraged on a
widespread basis, an evaluation was
needed of the extent to which utilities
could comply with the THM regulation by
using alternative oxidants and disin-
fectants, the costs associated with these
changes, and the impact of these
changes on other treatment objectives
and overall finished water quality.
Accordingly, the objectives of this
research project were to determine the
following:
(A) the extent to which water treatment
plants that had recently adopted
the use of chlorine dioxide, ozone,
potassium permanganate, or
chloramines to partially or fully
offset the use of free chlorine had
been able to comply with the THM
regulation;
(B) the impact of these modifications
on other water quality parameters
and other treatment objectives,
such as disinfection, iron and
manganese removal, total organic
carbon (TOC), color and turbidity
removal, total organic halide (TOX)
formation, and control of taste,
odor, and algae; and
(C) the costs of these modifications
and their impact on overall
treatment costs and on the average
consumer's water bill.
In order to address these objectives, this
research project consisted of a series of
case study evaluations of utilities
adopting alternative oxidant and disin-
fectant strategies for controlling THM
formation.
Of particular interest in connection
with Objective B was the impact of these
changes on TOX formation. TOX is a
collective parameter representing the
concentration of all organic halides that
has found increased use as a surrogate
parameter for other halogenated disin-
fection by-products which might prove
harmful to man.
Procedure
When this project began in
September 1983, large water utilities
supplying more than 75,000 consumers
had been required to be in compliance
with the 0.10 mg/L MCL for TTHM by
November 1981.'The utilities of inter-
mediate size serving between 10,000
and 75,000 consumers had been
required to initiate monitoring of their
finished water in the distribution system
in November 1982 and to be in
compliance with the 0.10 mg/L MCL by
November 1983. Because the start of
this research project coincided with the
THM compliance schedule for these
utilities of intermediate size, and because
it was believed that these smaller utilities
might experience more difficulty in
successfully implementing these alter-
native oxidant and disinfectant programs,
this project focused on utilities serving
between 10,000 and 75,000 consumers.
At the start of the project, letters
were sent to directors of water supply
programs in each of the 50 states as well
as to each of the U.S. EPA regional
offices informing them of the nature of
this research project and requesting their
recommendations of water utilities within
their jurisdiction that might be case study
candidates for this project. All of the EPA
regional offices and 36 states responded
with various recommendations. Tele-
phone calls to many of these utilities and
follow-up correspondence describing
the nature of the investigation were
initiated. Utilities with THM problems that
were contemplating the use of alternative
oxidant and disinfectant strategies or that
had only recently changed their oxidant/
disinfection scheme were identified.
Preliminary field visits were made to
several of these candidate utilities to
review the nature, quality, and variability
of the source water, the type of treatment
provided and the record of chemical
usage, and the quality and variability of
the finished water. Based upon these
preliminary visits, an analysis of the
available data, and, in some cases,
preliminary chemical analysis of the THM
concentration and THM formation
potential of the water, selection of the
utilities for detailed case study evaluation
was made.
In selecting utilities for evaluation, an
attempt was made to choose at least two
utilities using or proposing to use each of
the four alternatives (KMnO4, CI02, 03,
and NHaCI). Consideration of chlor-
amines was not emphasized at the
expense of the other options even
though it was, by far, the principal
alternative disinfection strategy bei
contemplated by the utilitii
recommended. Consideration was a)
given to utilities from differe
geographical regions and to utiliti
drawing water from different types
sources (i.e., rivers, lakes, groundwate
Utilities in the midst of implementing
alternative oxidant/disinfectant modific
tion were considered ideal candidates 1
this study since they could provide tl
most suitable data base for making
meaningful "before and after" evaluatii
of the impact of the change. Finall
travel and scheduling logistics played
major role in the final selection, and tl
majority of the utilities selected for tl
detailed case study evaluation were fro
the southeastern United States.
The eight utilities examined in deti
in this research project, along wi
selected characteristics of the
operations, are shown in Table 1. Tf
eight utilities are distributed among fix
states (Florida, Indiana, Virginia, Nor
Carolina, and South Carolina), and all b
Chesapeake served between 10,000 ar
75,000 consumers. One utility use
groundwater as a source of water suppl
the others used river or im-pounde
water. Two utilities attempted to solv
their THM problems usin
chloramination, three changed to chlorir
dioxide as a pre-oxidant, two use
permanganate as a pre-oxidant, an
one used ozone for pre-treatment.
Samples were collected by th
research team from the various treatmei
plants from December 1983 throug
December 1985 on approximately
quarterly basis in order to includ
operations over all four seasons of th
year. TOC and UV samples were take
for raw, settled, and filtered water i
order to measure organic carbon remov<
through the conventional treatment trair
Terminal TTHM and TOX samples wer
collected primarily for raw and settle
water in order to measure the extent c
precursor removal by coagulation an
settling. Instantaneous TTHM and TO!
measurements were made on raw
settled, filtered, finished (tap), and rep
resentative distribution system sample
in order to monitor the progression c
trihalomethane and overall organic halidi
formation.
In addition, the monthly operating
reports for the utilities were analyzed fc
the periods immediately preceding ani
following the modifications to assess thi
impact of the alternative oxidanl
disinfectant treatment program. The cost;
of implementing the modifications wen
calculated from chemical dosages an<
-------
Table 1
Utilities Selected for Detailed Evaluation
Utility
Service Average Flow (in
Population million gal/day) Source of Water
Alternative Oxidant/Disinfectant Strategy
Chester Metropolitan Water 18,000 3.0
District; Chester, SC
Bloomington Water Dept; 53,000 12.4
Bloomington, IN
Lancaster County Water and 28,000 1.8
Sewer Authority;
Lancaster, SC
Monroe Water Dept.; 15,000 7.0
Monroe, NC
Palm Beach County Water 60,000 6.6
Utilities Dept.;
West Palm Beach, FL
Wilmington Public Works 55,000 8.0
Dept; Wilmington, NC
City of Belle Glade; 20,000 4.5
Belle Glade, FL
Chesapeake Dept. of Public 77,000 8.4
Utilities; Chesapeake, VA
River Moved point of C\2 addition to post-sedimentation; CIO?
applied to raw water.
Lake Added ammonia after sedimentation to convert free chlorine
to combined chlorine residual.
Impoundment Moved point of CI2 addition to post-filtration;
applied to raw water.
, KMnO4
Lake
Wells
River
Lake
River
Moved point of C\% addition to post-sedimentation; relied on
KMnO4 application to raw water.
Split part of C>2 addition between raw water and post-
sedimentation; added ammonia post-sedimentation to
produce combined chlorine residual.
Moved point of CI2 addition to post-sedimentation; added
KMnQ4 to raw water.
Moved point of C/2 addition to post-filtration; installed two-
stage ozonation for raw water and ore-filtration application.
Coniunctive use of C\z ar>d C'Oj pre- and post-filtration;
installed air-stripping towers prior to distribution.
operating costs provided by utility
personnel.
The full report describes the
treatment facilities for these eight utilities,
presents the historical record of THM
compliance monitoring, and presents and
discusses the results of the research
team's field sampling visits. The impact
of the alternative oxidant/disinfectant
treatment modifications on finished water
quality, treatment plant operations and
performance, and cost are evaluated. In
addition, the complete data set for
samples collected and analyzed by the
research team as part of this research
project is presented, and correlations
between trihalomethane and total organic
halide concentrations are explored.
Results and Discussion
Of the eight utilities examined, two
successfully reduced the extent of THM
formation to unequivocably demonstrate
compliance with the MCL for total
trihalomethanes. The other six reduced
THM formation significantly but either
were unable to clearly demonstrate that
they consistently met the requirements
of the THM regulation as a result of the
modifications, or encountered other
difficulties in treatment plant operations
or in producing an acceptable finished
water. The two successful utilities were
Bloomington, IN, and Palm Beach
County, FL, both of which adopted
chloramination to halt THM formation.
While Bloomington experienced no
adverse impacts as a result of the
modification, Palm Beach County did
encounter some deterioration in finished
water quality, notably an increase in
color of the finished water causing the
utility to periodically exceed the MCL of
15 color units, which is a primary
standard in the State of Florida. The cost
of implementing the modifications at both
utilities was negligible; in fact, Palm
Beach County experienced a decrease in
chemical costs.
Of the other six utilities, Chester, SC,
and Wilmington, NC, both appeared to
comply with the TTHM standard as a
result of the modifications, but only
barely so. Both utilities moved the point
of chlorine addition to post-sedimen-
tation; Chester applied chlorine dioxide
to the raw water at the flash mix basin
while Wilmington applied potassium
permanganate at the raw water pump
station, 26 miles from the treatment
plant. Neither utility experienced any
serious adverse impacts on finished
water quality or on treatment plant
operations as a result of the
modifications in oxidation/disinfection
practice. At Chester, the turbidity of the
finished water deteriorated somewhat
after the utility switched to chlorine
dioxide. The cost of changing from pre-
chlorination to chlorine dioxide pre-
treatment had a negligible impact at
Chester; the chemical costs increased by
$0.035/1,000 gal, which amounted to
1.27% of the total water cost or an
increase of $1.76 per residence per
year. At Wilmington, the cost of
permanganate pre-treatment increased
the monthly-average chemical costs by
only $0.011/1,000 gal.
Monroe, NC, relied on potassium
permanganate for pre-treatment and
moved the point of chlorine addition to
post-sedimentation, but did not achieve
compliance with the MCL as a result of
these modifications. Belle Glade, FL,
implemented two-stage ozonation in
place of pre-chlorination and reduced
THM formation significantly from
concentrations approaching 1,000 pg/L to
concentrations below 200 ng/L, but the
utility was still not in compliance with the
MCL for total trihalomethanes. The color
of the finished water improved and TOC
removal increased by about 5%, but
periodic growths of algae were observed
in the recarbonation basins after making
the pre-treatment modifications. A
noteworthy observation is that, as a result
of the switch from pre-chlorination to
pre-ozonation, the distribution among
THM species shifted. Before the change,
chloroform constituted an average of
87% of the TTHM's, while after the
change, chloroform constituted an
average of only 40% of the TTHM's. The
remaining 60% were distributed among
the various brominated THM species.
From a cost standpoint, it was difficult to
discern any differences in operating
costs after Belle Glade converted to
two-stage ozonation. The principal
increase in cost appeared to be the
capital costs of the installation, which
-------
amounted to an amortized annual cost of
$0.03/1,000 gal.
The last two utilities, Lancaster, SC,
and Chesapeake, VA, experienced mixed
results after adopting an alternative
oxidant/disinfectant program involving
chlorine dioxide. Both lowered THM
production significantly and, at times,
achieved running annual averages of less
than 100 u,g/L for total trihalomethanes,
but neither utility operated consistently
for a long enough time to judge the
effectiveness of the modified program.
Both raw waters had excessive oxidant
demands, and both utilities used up to 6
mg/L of chlorine dioxide for pre-
treatment. This resulted in high levels of
chlorite in the finished water, ap-
proaching 3 mg/L at times for both
utilities. Additionally, Lancaster County
was plagued with manganese problems
in their finished water after modifying
their treatment program, and Ches-
apeake had difficulty carrying a residual
disinfectant in their distribution system
without resorting to high levels of post-
chlorination, resulting in excessive THM
formation. Chesapeake installed air-
stripping towers prior to the high-
service pumps feeding the distribution
system but, while these seemed to expel
effectively the volatile THM's, they did
not help the utility achieve compliance
with the MCL because of continued
formation of THM's in the distribution
system. Additionally, the air-stripping
towers had no effect on the concentration
of the non-volatile halogenated
disinfection by-products that comprised
about 70% of the TOX concentration.
Results from this research project
demonstrate that TOX formation closely
parallels THM formation for all of the
utilities investigated. The instantaneous
TTHM concentration in the finished
water, including the distribution system,
was strongly correlated with the
instantaneous TOX concentration in the
finished water. For 166 pieces of data,
the correlation coefficient was 0.885. The
TOX/TTHM ratio in the treated water was
about 3.4:1 for surface waters treated by
conventional coagulation, settling, and
filtration at near-neutral pH values. On a
chlorine-equivalent basis, the THM's
comprised approximately 26% of the
total organic halide concentration. The
concentration of non-volatile organic
halides, such as di- and tri-chloro-
acetic acid, was collectively approxi-
mately three times the concentration of
TTHM on a chlorine-equivalent basis.
In waters subjected to precipitative
softening at alkaline pH values, the
TTHM's comprised about 39% of the
total organic halide concentration in the
finished water, on a chlorine-equivalent
basis. A reduction in the extent of THM
formation as a consequence of modifying
the oxidant/disinfectant program resulted
in essentially a parallel reduction in the
extent of TOX formation.
Conclusions and Implications
Alternative pre-treatment oxidants
and disinfectants were depleted relatively
rapidly, particularly in waters having TOC
concentrations greater than 5 mg/L. The
implications of this rapid rate of depletion
are:
- that it will be difficult to carry
residual oxidants through the pre-
treatment process train;
- that disinfection effectiveness will be
reduced as a result of the decrease
in "Cxt" for disinfection, i.e.,
concentration of disinfectant (C)
times contact time (t). This rapid rate
of depletion will also impact the
effectiveness of the oxidant for
oxidizing taste and odor compounds,
organic color, and iron and
manganese.
Accordingly, based upon the results of
the case studies, it can be concluded
that many utilities will not be able to
comply with a significantly reduced MCL
for TTHM's using only alternative
oxidants/disinfectants and conventional
treatment without sacrificing overall
finished water quality.
Recommendations
Before endorsing the widespread
application of alternative oxidants and
disinfectants for controlling trihalo-
methane formation in drinking water,
additional work needs to be done to
better characterize and understand the
behavior of these chemicals. The
limitations of their use as well as their
beneficial properties need to be
determined, particularly for ozone and
chlorine dioxide, which, until recently,
have not been extensively used for
drinking water treatment in the United
States. As more utilities adopt the use of
ozone and chlorine dioxide, detailed
evaluations such as those reported in this
investigation should be conducted and
published so that others can learn from
the successes and failures and, thereby,
minimize additional failures in the future.
Of specific interest regarding
chlorine dioxide and ozone are the
kinetics of their reaction with impurities in
water, especially humic material, which
comprises most of the total organic
carbon content of most natural waters.
Such reactions are responsible for the
rapid rate of depletion of these stror
oxidants, which in turn limits the
disinfecting potential. These reactior
can produce a variety of disinfection b;
products that, for the most part, have ni
been identified nor have any associate
adverse health effects been determinei
Disinfection kinetics and by-produi
identification are areas of critical intere
and intense research in the water supp
field at this time.
With many utilities beginning 1
employ a variety of different oxidants an
disinfectants during the course c
treatment, e.g., permanganate treatmei
of raw water, chlorine dioxide treatmei
of settled water, and free or combine
chlorine treatment of filtered water,
critical need exists for accurate an
precise measurement of the individu;
oxidant and disinfectant residuals. Wit
current analytical technology, particularl
for routine water treatment laborator
use, it is difficult to distinguish among th
various residual species.
Specific research questions genei
ated by this research project are:
- What happens to the residual chlorit
when water pre-treated wit
chlorine dioxide is post-treated wit
chlorine? Is chlorate the principj
product of this reaction, or i
additional chlorine dioxide produce'
to act as an oxidant and disinfectant
a second time?
- What is the mechanism responsibl
for the shift in speciation towan
brominated THM species when pre
chlorination is replaced by pre
ozonation? How is the extent of thi
redistribution influenced by thi
TOC/Br formation in waters will
appreciable bromide concentrations'
- What is the most effective means c
ensuring the oxidation and retentioi
of manganese in waters containini
high concentrations of TOC whei
pre-chlorination is replaced b;
alternative oxidants and disinfectant:
for THM control? What factors an
responsible for the retention an<
release of manganese from filte
beds?
- How do water treatment plan
operators establish an optimal pre
oxidant dosage to provide for th<
effective control of iron am
manganese, taste and odor, ant
color, as well as for disinfection, am
how do they control this dos<
operationally? In the case of pre
chlorination, free chlorine residual;
were used to establish th(
appropriate pre-chlorine dose, li
the case of the rapidly depletet
-------
pre-oxidants chlorine dioxide,
ozone, and permanganate, it is
difficult to provide a sufficient pre-
treatment dose to carry a residual
oxidant very far into the treatment
train.
It is recommended that these specific
questions be addressed and answered
before the widespread use of alternative
oxidants and disinfectants is promoted.
The full report was submitted in
fulfillment of Cooperative Agreement CR
811108 by the University of North
Carolina under the sponsorship of the
U.S. Environmental Protection Agency.
-------
Philip C. Singer is with the University of North Carolina, Chapel Hill, NC27514.
Benjamin W. Lykins, Jr., is the EPA Project Officer (see below).
The complete report, entitled "Alternative Oxidant and Disinfectant Treatment
Strategies for Controlling Trihalomethane Formation," (Order No. PB 88-
238 9281 AS; Cost: $32.95, 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:
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
EPA/600/S2-88/044
0000329 PS
OTECTION AGENCY
------- |