TECHNICAL REPORT DATA
(Piecst read Jnttmcitoni on tht revtrtt before completing)
1 REPORT NO 2
EPA/600/D-89/037
3. RECIPIENT'S ACCESSION NO
PB09 *222624
4. title and subtitle
formation AND CONTROL OF non-trihalomethane BYPRODUCTS
6. REPORT DATE
e. PERFORMING ORGANIZATION COOE
7 AUTHORISI
Alan A. Stevens, Leown Moore and Richard J. Miltner
S. PERFORMING ORGANIZATION REPORT NO
RREL-0100
9 performing organization name ano ADDRESS
Drinking Water Research Division
Risk Reduction Engineering Laboratory
U. S. Environmental Protection Agency
Cincinnati, OH 45268
10. PROGRAM ELEMENT NO.
BNC104
11. CONTRACT/GRANT NO
N.A.
12. SPONSORING AGENCY NAME AND AOORESS
Drinking Water Research Division
Risk Reduction Engineering Laboratory
U. S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
Complete
14. SPONSORING AGENCY CODE
EPA/600/14
15 SUPPLEMENTARY NOTES
EPA Project Officer - Alan A. Stevens - FTS 684-7342
Commerical (513)569-7'342
16 ABSTRACT
Hundreds of organic byproducts of chlorination are now known to occur in drink-
ing water along with the trihalomethanes. About twenty of these appear to be found
with sufficient frequency and in sufficient concentration to attract consideration
for regulations. These Include chloral hydrate, chloropicrin, a trichloropropanone,
haloacetonitriles, and haloacetlc acids* Trihalomethane concentrations do not serve
as good predictors of concentrations of these other byproducts because their condi-
tions of formation vary widely. This is especially true when pH is changed.
Treatment strategies for control of these byproducts including the trihalo-
methanes are:
1. Remove the compounds after they are formed
2." Remove precursors
3. Use other disinfectants.
The first option Is not attractive because much of the formation occurs after
the water enters the distribution system. Additionally, several of the compounds
would be difficult to remove by established treatment processes. Current evidence
supports the idea that precursor removal processes effective for trihalomethane
control may be effective for the other byproducts as well.
17 KEY WORDS AND DOCUMENT ANALYSIS
a DESCRIPTORS
t>.IOENT|FIERS/OPEN ENDED TERMS
c. COSati Field/Group
Drinking Water Chlorination
Trihalomethanes
Disinfection Byproducts
18 DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS iThil Report)
UNCLASSIFIED
21 NO Of PACES
2o
20 SECURITY CLASS iTIill pugei
UNCLASSIFIED
22 PRICE
EPA Form 2220-1 (R«t 4-77) PRIVIOU# IDlTION IIOBIOLITt
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse
ment or recommendation for use.
ii
-------
EPA/600/D-89/037
FORMATION AND CONTROL OP NON-TRIHALOMETHANE BY-PRODUCTS
Alan A» Stevens
Leovm A. Moore
and
Richard J. Miltner
Organics Control Branch
Drinking Water Research Qivision
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Trihaloraethanes (THMs) were first regulated in drinking water in 1979.
The regulation of these compounds followed five years of work by this
laboratory and others examining their formation on the bench-scale and
control at the pilot- and full-scale.^- At the same time, examination of
the formation and control of other disinfection by-products*(DBPs)
(mainly resulting from chlorination) as measured by the surrogate
parameter total organic halogen (TOX) was occurring.^TOX data
presented in Fig. 1 suggested the formation of other DBPs whose summed
concentrations likely equalled or exceeded those of the THMs.5 Bench
studies with chlorination of natural water and humic acid-spiked waters
using extraction, capillary column chromatography and mass spectral
analytical procedures detected over 500 DBPs.^ Many of these were
found at ug/L concentrations although most were probably much lower,
and the majority were not identified. A survey of finished waters of
ten U.S. cities confirmed the presence of ug/L concentrations of several
of these.^ The effort of placing rugged analytic procedures on-line
for preservation and routine analyses of these DPBs that are listed in
Table 1 was then taken, followed by studies of the formation, ^ecay and
stability of the non-THM DBPs. With the information gained from these
studies regarding their formation dependency on pH, temperature, bromide
concentration, oxidant concentration, etc., pilot-scale flow through
studies to examine their control began.
Under the 1986 Amendments to the Safe Drinking Water Act (SDWA), allow-
able concentrations of THMs will be reconsidered for regulation in 1991.
Parallel to this, a minimum of twenty-five new contaminants are to be
regulated. These twenty-five likely will Include several DBPs.
Pilot Plant Operation
For the studies reported herein pilot plant operation was utilized to
produce water for bench-top chlorination studies. The pilot plant at
EPA-Cincinnati treats Ohio River water and has been adequately des-
cribed elsewhere.^ Chlorination was not employed on the pilot plant,
because results of continuous flow studies would be extremely difficult
to interpret in the absence of the static bench-top-study information.
In each study, chlorination of raw and undisinfected filtered water
would produce terminal level DBP (analogous to term THM) concentrations,
and consequently, a measure of the effect of pilot-scale treatment on
DBP precursor.^ The pilot plant was operated until steady-state opera-
tion was achieved (typically less than two days) before sampling.
1
-------
Three studies vere conducted vlth the CPA pilot plant treating Ohio
River water. In each, the pU of the clarification process was different.
The three studies were designed to provide a range of pH typical of the
extremes occurring during water treatment*
In the first run (1A), lov pH alum coagulation vas employed. Early THM
studies indicated that optimal removal of THM precursor material occurred
during low pH coagulation and clarification. A previous pilot plant
run predicted this, but did not include analyses of DBPs other than THMs.
Water collected from run 1A was used to determine if other DBPs can
similarly be controlled by low pH coagulation prior to chlorination.
Low pH coagulation was achieved by feeding alum and adding an acid
(HC1) to lower the pH to near 5.7. These were the conditions optimal
for THM precursor control with alum in the previous pilot plant run.
During run 2A, pH was investigated at the other extreme during lime
softening. Water was softened at a pH sufficient to precipitate Mg(0H)2,
near 10.8, so that the high pH effect on precursor control could be
studied.
Ohio River water is not naturally a hard water, but, based on chemical
analyses of the river water for calcium and hardness and alkalinity
characteristics, sufficient chemicals were added to the water upon
arrival at the EPA pilot facility to give the following attempted hard
water quality (all as rag/L CaC03):
total hardness ¦* 240
calcium = 150
magnesium = 90
P alkalinity » 0
-------
Procedures for Chlorlnatlon Experiments
Raw and filtered water samples were collected from the pilot plant runs
1A, 2A, and 3A In 30 to 40 liter quantities*
Aliquots of each sample were buffered to three different selected pH
values (5, 7, and 9.A) by first placing 80 pL> of a buffer solution (a
combination of 0.25 M borate and 0.25 M phosphate) into a 10 L bottle,
then filling to a 4 L mark with either raw water or filtered water*
Either 1.0 N NaOHor 1.0 N H2SO4 was added to the buffered sample, while
stirring and monitoring with a pH meter, until the desired pH was reached.
Each sample was then transferred to a 1 gallon bottle until needed for
further work-
The chlorine demand of each sample, as originally collected, was deter-
mined by a proposed Standard Method^ for the determination of THM
formation potential. The required amounts of chlorine, as determined
above, were then measured into 1 liter bottles, using one bottle per
experimental time period. Three experimental chlorlnatlon time periods
were chosen for most of this work: 4 hours, 2 to 4 days (dictated by
convenience), and 6 to 7 days. This requited a total of 6 bottles per
time period: three bottles containing buffered raw water, chlorinated
at three different pH values, and three more corresponding bottles for
filtered water. Chlorlnatlon reactions were allowed to proceed at 25°C
until the reaction was quenched at the end of the given time periods.
Two procedures were used for stopping the chlorlnatlon reaction at the
various time periods: (1) samples to be analyzed for dlhaloaceto-
nitriles (DHAN), chloropicrin (CP), 1,1,1-trichloropropanone (Ill-TCP),
and trichloroacetonitrile (TCAN) were each poured into a 40 mL glass
vial that already contained about 3 drops of acjaoftlua chloride solution
(5 g NU^Cl/lOO mL); the vial was filled with sasple, shaken,, and then
neutralized to pH 7 by adding either 1.0 N H2SQ4 or 1.0 N NaOH. (2)
For samples to be analyzed for T0X and the other DBPs, the chlorine
residual remaining In the 1L bottle was determined and was then
destroyed by adding a slight excess of sodium sulfite; the pH was then
adjusted to between 5 and 6. Samples for trihalomethanes (THM) and
chloral hydrate (CH) were then taken in separate vials; samples for TOX
and haloacetlc acids (HAA) were also poured into separate 250 mL bottles,
and the pH of the TOX sample was further reduced to 2 by adding nitric
acid. All samples are 6tored at 4 to 6°C until ready for analysis.
Reaultis of Experiments on Effects of pH and Time
For the sake of brevity, only the data from run 1A are presented. The
same trends were observed and general conclusions apply to the other
two runs. These conclusions can be made as follow.
Total Organic Halogen (TOX) (Fig. 2) TOX concentrations were reasonably
Independent of pH in the range of 5 to 9.4; the TOX concentrations
Increased with time, although the 4 hour reaction produced approximately
60% of the TOX that was produced during 7 days reaction. This is
consistent with historical data.^>3|10
Total Trihalomethanes (TTHM) (Fig. 3) The concentration of TTHM
increased with time for each pH value; the TTHM concentrations after
144 hours reaction in raw water, were, for the pH values of 5, 7, and
9,4; 65, 183, and 252 ug/L (as CHCI3), respectively.
3
-------
This trend for trihalomethane formation with pH and time is also well
known* and is presented here for reference* From this trend for THM
formation and that noted above for TOX, one can conclude that the non-
THM portion of TOX decreases with increasing pH of chlorination in
these experiments, also as reported previously (Fig* 1).^»^»10
Trichloroacetic Acid (TCAA) (Fig. A) TCAA concentrations produced at
pH values of 5 and 7 were about equal at any given reaction time (about
50 ug/L at 4 hours and 130 ug/L after 7 days reaction with raw water),
but the TCAA concentrations were always significantly lower at a re-
action pH of 9.4.
This trend is parallel to that mentioned with regard to nonpurgeable
organic halogen and may partially account for it. Note that the abso-
lute concentrations of TCAA are in the range normally expected tor the
TTHMs and, indeed exceed then in some cases.
Dlchloroactlc Acid (DCAA) (Fig.5) DCAA concentrations were essentially
independent of the reaction pH at ail time periods, but an increase
with time was also always observed. Initial thinking is that this
compound is formed by a different mechanism than is TCAA in spite of
the obvious relationship by chemical class. Even if DCAA was an inter-
mediate in the formation of TCAA, the high pH data could only support
this contention if DCAA was stable and unlikely of further reaction at
pH 9.A. Note that concentrations of DCAA, like TCAA, also rivals THM
concentrations in some cases.
Chloral Hydrate (CH) (Fig. 6) The concentration of CH increased with
time for the pH values of 5 and 7, and were in about equal concentra-
tions for both pH values (about 5 ug/L at 4 hours and 25 ug/L after 7
days); CH was, however, initially formed most rapidly at a pH of 9.4
(10 ug/L produced at this pH in the 4 hours period or twice the amount
produced at the lower pH values), but the rate of hydrolysis (decomposi-
tion) of CH at a pH of 9.4 quickly exceeded the rate of formation, and,
consequently, the concentration of CH at the 9.4 pH value decreased
with time (10 ug/L at 4 hours down to less than 2 ug/L after 7 days).
Importantly, CH hydrolyzes (as does Ill-TCP) to form chloroform which
has enhanced formation at high pH (Figs. 1,3).
The competing formation and decay reactions of CH are the most clear
demonstration of the difficulty of predicting the outcome of application
of treatment strategies involving these chemically complex systems.
Even minor pH adjustments in the range of 7 to 9 can dramatically
affect product distribution.
Dichloroacetonitrile (DCAN) (Fig. 7) DCAN was stable only at pH 5, and
its concentration steadily increased with time at that pH. At pH 7,
the DCAN concentration decreased with time after the first sampling
time of four hours (6 ug/L at 4 hours to 2 ug/L after 144 hours) and
hardly formed at all at pH 9.4. At pH 7, a formation-decay competing
reaction phenomenon Is occurring much like CH at pH 9.4. For DCAN at
pH 9.4, this effect is less clear. Nevertheless, DCAN concentrations
are the lowest at high pH, again consistent with historical data.H
Treatment Studies
DBPs may be controlled by changing the oxidant or its application
point, controlling the precursor material that the oxidant reacts with,
4
-------
removing the formed DBPs, or a combination of these.1 This has long
been stated for THM control, although, because of their formation
during the distribution of water, only the first two options are usually
considered. The complexity of the mixture of other DBPs, their differ-
ing physical/chemical characteristics, and differing chemistries of
formation cause this perceived restriction to only two viable treatment
approaches to be reinforced.
Precursor Removal in Pilot Studies
In any discussion of precursors, especially in the case of experiments
reported on herein, where the formation and, in some cases, decay,
depends on several factors, the term precursor concentration has ?n
uncertain meaning. This is the reason that, even for the relatively
simple case of THMs, precursor has been defined as a "formation poten-
tial" that always must be accompanied by a stated specific set of
reaction conditions, especially time, temperature, and pH value. Any
significant differences between these Important reaction conditions for
separate samples make comparisons of absolute "precursor concentrations"
meaningless. Further, one might wonder the value of precursor measure-
ments at all at high pH for CH and DCAN, conditions where decay over-
takes the formation reactions. Nevertheless, these reactions can both
be thought of as taking place both at a slower rate and to a lesser
extent at lowered precursor concentrations (such as resulting from
treatment) just as with a stable DBP's formation. Thusly, the forma-
tion potential concept is used here to define precursor. The results
of formation of DBP from whatever precursor is present in a sample
under the stated reaction conditions are compared in a relative sense
in the following section in a way that leads to some surprising and
encouraging conclusions.
Percent Removals — One Treatment pH, Three Chlorination pH Values At
any specific set of reaction conditions, the relative concentration of
each constituent shown in Figures 2 to 7 was lowered to a similar
extent (Fig. 8). Removal percentages from raw to low pH alum coagu-
lated/filtered water were all within range of approximately 60 to 80%.
If one accepts that the "wobble" of 60-80% removal of precurors is
caused by the variation in formation potential measurements compounded
by the complexities brought about by the differing formation/decay
mechanisms, the ptrcent removal of precursors for any single constitu-
ent can probably serve as a fairly accurate predictor for the percent
removal of the remaining five.
Surprisingly, this generalization seems to hold regardless of the pH of
chlorination even though the absolute values of concentration for each
compound change dramatically with pH.
Pilot runs 2A and 3A resulted in different percent removals of precursor
overall (approximately 50% in run 2A and 40% in run 3A) but exhibited
the same trends as described in tig. 8.
These observations about precursor removal through physical removal
processes are extremely important because this increases the probability
that the vast experience we now have with THM precursor control by
physical removal processes may be transferred to the control of these
other byproducts as well. No such general conclusions, however, can be
drawn about removals of the compounds themselves after they are formed,
nor can we make general statements concerning precursor control by
chemical processes such as ozone oxidation.
5
-------
Alternate Disinfectants
Figure 9 compares the TOX formation by chlorine dioxide and chloramines
with that of chlorine. The THMs are completely controlled with chlorine
dioxide and nearly so with chloramines as applied in this earlier study.
Non-THM organic halogen is also greatly reduced for both disinfectants.
This Information and that from previous Btudles^ would indicate that
the Table 1 chlorination byproducts are probably not a problem when these
other disinfectants are used.
Phenol Study - List Modification
The current list (Table 1) of the target byproducts for regulation con-
tains brominated and mixed bromine/chlorine species of trlhalomethanes
and haloacetonitriles. These are known to form in bromide containing
waters when they are chlorinated. Logically, the analogous mixed halo-
and bromoacetlc acids might also form. As a test of this idea, phenol,
which giveu TCAA in high yield as percentage of TOX,^® was chlorinated
in the presence of bromide ion under typical formation potential re-
action conditions. Given the qualifications that several standards
were (and still are) not available and that some reference mass spectra
were not available, interpretation of the data indicated bromodichloro-
acetic acid (BDCAA), dibromochloroacetic acid (DBCAA), bromochloroacetic
acid (BCAA), and tribromoacetic acid (TBAA) were all formed.^ The
same array of products has since been seen when humic acids were chlori-
nated under similar conditions. The data Indicate that the bromo- and
mixed haloacetic acids probably should merit regulatory consideration
to remain consistent with THM and DHAN precedents. Further work might
also result in finding analogous products for Ill-TCP, CP and CH.
Summary of Implications for Treatment Strategies
1. The most Important chemical variable to consider in chlorination by-
product formation is pH. Yields of nearly all halogenated organics can
usually be either maximized or minimized by controlling the pH at which
the various reactions occur, although dichloroacetic acid (DCAA) seems
to be an exception to this rule. Table 5 summarizes this formation in a
qualitative way.
The most obvious implication for water treatment is the direct trade-off
between THM control (low pH) and control of most of the other byproducts
(high pH). While DCAN, Ill-TCP and CH are not likely to be problems at
pH above 8-9 and TCAA above pH 10, under these conditions THM would be
maximized. In general, the reverse is true at the lower end of the
normal pH range of drinking water treatment (pH 5).
2. Precursor control may prove to be similar for all of these chlori-
nation byproducts through physical removal mechanisms. At least the
current data support this for conventional treatment. Experiments with
adsorpclve and membrane process are underway or anticipated. No such
conclusion about oxidative removal or modification processes can be
drawn from these data.
3. Analytical methods for DBP and preservation or dechlorination
agents for samples were not addressed in any detail. The analyses for
several of the DBP, especially the haloaclds, are not easily put on
line in the laboratory. We experienced considerable difficulty with
the current procedures, delaying this work. Now that procedures have
6
-------
been worked out, things should be simpler, but we do expect that compli-
ance monitoring might be somewhat of a problem.
The sample dechlorination procedures used for the DBP were different,
depending on the compounds to be measured* Ammonium ion was used for
CP and the DHANs because commonly used sulfite destroys these compounds.
This raises the possibility that SO2 application (a common water treat-
ment process) may be used la eome way to control CP and DHANs.
Likewise, the fact that HH4CI-stops the formation of these substances
supports the contention that these compounds will not be formed by
chloramlnation.
4. Temperature and chlorine dose were not investigated for their
effects on formation of DBP* This must still be done. As the THMs,
the conclusion that higher temperatures will lead to higher concentra-
tions of DBP at a faster rate is logical. Alternatively, however,
hydrolysis rates for CH, DHANs, and possibly CP and the haloacids are
also likely to Increase, possibly having the opposite effect on the
presence of these species.
Chlorine dose may also be important. Although chlorine dose appears to
have little impact on THM formation, it does affect TOX significantly
(Fig. 1). This may be reflected in increased concentrations oE non-THM
DBP with increased chlorine dose.
5. Concern concurrent with an oxidant's potential to form or not to
form DBPs is its potential to control microbiologic contaminants during
water treatment. Strategies for the control of DBPs must also ensure
control of regj]ated microbiologic entitles as well as the compounds
under regulaiJry consideration. Hence, any change in oxidant type,
location, l> dose should examine both DBP and microbiologic levels.
7
-------
REFERENCES
1. Symons, James M., Stevens, Alan A., Clark, RoberfcH., Geldreich,
Edwin E., Love, 0. Thomas, Jr., and DeMarco, Jack, "Treatment
Techniques for Controlling Trlhalomethanes la Drinking Water",
originally published by U.S» EPA, Municipal Environmental Research
Laboratory, Drinking Hater Research Division, and is now available
from the American Water Works Association (1982)*
2. Fleischacker, S. J., Randtke, S. J*, "Formation of Organic Chlorine
in Public Water Supplies- J. AWWA (3), 132 (1983).
3. Stevens, Alan A., Symons, James M., "Alternative Disinfection
Processes", from Organic Carcinogens in Drinking Water, Edited by
Neil M. Ram, Edward Calabrese, and Russell F. Christian, John Wiley
& Sons, Inc., (1986).
4. Stevens, Alan A., Moore, Leown A., Slocum, Clois J., Smith,
Bradford L., "Chlorinated Humlc Acid Mixtures Establish Criteria
for Detection of Disinfection Byproducts in Drinking Water", In:
Suffet, 1. H., MacCarthy, P. Eds, Aquatic Humlc Substances,
Advances in Chemistry Series 219, American Chemical Society,
Washington, DC, 1989.
5. Stevens, Alan A., Moore, Leown A., Slocum, Clois J., Smith,
Bradford L., Seeger, Dennis R., Ireland, John C., "By-Products of
Chlorlnatlon at Ten Operating Utilities'*, presented at the Sixth
Conference on Water Chlorlnatlon: Environmental Impact and Health
Effects, Oak Ridge, TH, May 3, 1987; to appear in Proceedings.
6. Stevens, Alan A., Mlltner, Richard J., Moore Leown A., Slocum,
Clois J., Nash, H. D. Reasoner, D. J., Bernan, D., "Detection and
Control of Chlorlnatlon Byproducts in Drinking Water", Proceedings
of the Conference on Current Research in Drinking Water", EPA/600/
9-88/004, 242-256 (1988).
7. Stevens, Alan A., Symons, James M., "Measurement of Trihalomet'nane
and Precursor Concentration Changes", JAMWA 69, 546 (1977).
8. Sawyer, C. N., McCarty, P. L. "Chemistry for Sanitary Engineers",
2nd Ed., McGraw-Hill, New York, NY (1967).
9. To be published in the 17th Edition of Standard Methods for the
Examination of Water and Wastewater-
10. Reckhow, David A., Singer, Philip C., "Mechanisms of Organic
Halide Formation During Fulvic Acid Chlorlnatlon and Implications
with Respect to Preozonation" in Water Chlorlnatlon: Environmental
Impact and Health Effects, Vol. 5, R. L. Jolley, W. A. Brungs, J.
A. Cotruvo, R. B., Cummins, J. S. Matt ice, and V. A. Jacobs, Eds.
Lewis Publishing 1229-1257 Chelsea, MI, 1985.
11. Trehy, M. L., Bieber, T. S., "Dihaloacetonitriles in Chlorinated
Natural Waters", in Water Chlorlnatlon: Environmental Impact and
Health Effects, Vol. 4, R. L. Jolley, et al. Eds (Ann Arbor, MI;
Ann Arbor Science Publishers, Inc.), Ann Arbor, MI, 1983.
12. Ireland, J. C., Moore, L. A., Pourmoghaddas, H. Stevens, A. A.
Biomedical and Environmental Mass Spectrometry, 17, 483-486 (1988).
8
-------
Table 1
DPBs for Analyals In Pilot Plane Studies
chloroform
bromodlchloroaethane
dibromochloronethane
bromoform
dlchloroacetonltrlle
bromochloroacetonltelle
dlbromoacetonitrlle
trichloroacetonitjrlle
chloroacetlc acid
dlchloroacetic acid
trichloroacetic acid
chloral hydrate
chloroplcrln
l,lAl-trlchloropropanone
TABLE 2
RUN 1A PILOT PLANT OPERATION: LOW pH ALUM COAGULATION
Hater Quality3
clear
Parameter raw** flocculated settled filtered'5 well
temperature, °C 11.1-26.2 17.2 17.1
pH, units 7.58 5.3-5.8 5.79 5.93 5.89
turbidity, ntu 35-55 2.77 0.13 0.18
alum dese, rag/L 40.5
HC1 dose, rag/L 16.3
arange as tuean value
bvrater sampled for organic analyses
9
-------
TABLE 3
RUN 2A PILOT PLANT OPERATION: HIGH pH COAGULATION (LIKE SOFTENING)
Water Quality*
Ohio settled pH £11-
Rlver^ rawW® flocculated softened adjusted tered^
temperature, °C
27.4
26.3
25.5
25.9
pH, units
8.2
7.96 10.6-10.8
10.72
8.6
8.45
turbidity, ntu
27
7-27
0.76
0.12
hardness0 103
235
67
64
calciumc
75
144
37
34
nagnesiu3ic
28
91
30
20
total alkalinityc
48
98
66
30
coagulant dose, og/L
33.7
lime dose, mg/L
186
soda ash dose, mg/L
117
NaOH dose, ng/L
66.2
HC1 dose, mg/L
24.2
arange or mean value
^vater sampled for organic analyses
Cmg/L as CaCO^
^before spiking
efolloving spiking
TABLE 4
RUN 3A PILOT PLANT OPERA'"ION: CONVENTIONAL pH ALUM COAGULATION
Water Quality3
raw'1
flocculated
settled
filteredD
temperature, °C
28.1
28.0
26.7
26.9
pH, units
7.5
6.9
7.05
6.98
turbidity, ntu
5.4-11
0.75
0.09
alun dose, mg/L
25.5
arange or mean value
^water sampled for organic analyses
10
-------
TABLE 5
SUMMARY OP DBP FORMATION CONDITIONS
^9cmm
By-Product
pH of Chlorlnation
pH 5
pH 7
pH 9.4
TTHM
lover formation
higher formation
TCAA
similar formation
lover formation
DCAA
similar formation - perhaps slightly higher at pH 7
MCAA
concentrations below 5 ug/L, trends not discernible
DBAA
concentrations below 1 ug/L, trends not discernible
CH
similar formation
forms within 4
hours; decays
over time to less
than 5 ug/L
CP
concentrations below 1 ug/L trends not discernible
DCAN
higher formation
forms within 4
hours; then
decays over
concentrations
belov 2 ug/L
trends not dis-
cernible
time to less
than S ug/L
BCAN
concentrations belov 2 ug/L trends not discernible
DBAN
concentrations below 0.5 ug/L trends not discernible
TCAN
not detected
Ill-TCP
higher formation
concentrations
below 2 ug/L
trends not
discernible
not detected
11
-------
600-
500
400
300
200
100
0
100
200
8.1 mg Clj/L
20 mg Clj/L
pH:
oi «g
a u
-------
FIGURE 2. THE VARIATION OF TOX WITH pH AND TIME
4 48 144
Reaction Time (Hours)
FIGURE 3. THE VARIATION OF TTHM WITH pH AND TIME
pH - 5 pH - 7 pH - 9A
kNWWWI C
144
Pesclion Time (Hours)
13
-------
FIGURE A. THE VARIATION OF TCAA WJTH pH AND TIME
pH-5 pH-7 pH-9.4 ¦¦ Kau
[
Reaction Time (Hours)
FIGURE 5. THE VARIATION OF DCAA WITH pH AND TIME
Reaction Time (Hours)
14
-------
FIGURE 6. THE VARIATION OF CHLORAL HYDRATE WITH pH AND TIME
Reaction Time (Hours)
FIGURE 7. THE VARIATION OF DCAN WITH pH AND TIME
Reaction Time (Hours)
IS
-------
FIGURE 8. PERCENT FORMATION POTENTIAL REMOVAL FOR COMPOUNDS
AFTER 48 HOURS CHLORINATION (RAW --> FILTERED)
TOX 100'
TTHM
TCAA
DCAA
CH
DCAN
Y//////A
pH of Chlorinaiion
-------
600
500
400
300
200
100
0
100
200
Chlorine
20 mg/L
Reaction Time
Scale (hrs)-
r 0
2
LL
i- 72
144
I
1
L
Chloramlnes
22.9 mg/L
n
Chlorine Dioxide
20.7 mg/L
pH:
A jsl —
¦si a In
in —
io in in
Ut yl —
t* tk U)
e 9. A comparison of Che formation of NPOX and THJls
j) at 20 C (68*f) in distilled water solutions of 5 rag
acid/L dosed with various disinfectants.
17
------- |