6000345
INTERIM TREATMENT GUIDE
FOR THE CONTROL OF CHLOROFORM
AND OTHER TRIHALOMETHANES
WATER SUPPLY RESEARCH DIVISION
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
CINCINNATI, OHIO
JUNE 1976
-------
INTERIM TREATMENT GUIDE
FOR THE CONTROL OF CHLOROFORM
AND OTHER TRIHALOMETHANES
by
James M. Symons
Major Contributors
J. Keith Carswell
Robert M. Clark
0. Thomas Love, Jr.
Richard J. Miltner
Alan A. Stevens
Water Supply Research Division
Municipal Environmental Research Laboratory
Office of Research and Development
Cincinnati, Ohio 45268
June 1976
230 ,-
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY 1
INTRODUCTION 5
Background 5
Nomenclature 5
1975 Surveys 7
Surrogate Measurements 8
Position of the Environmental Protection Agency 13
EXTENT OF PROBLEM 15
SUGGESTIONS FOR ALTERNATE TREATMENT 18
Change in Disinfectant 18
Use of Ozone, Chlorine Dioxide, or Chloramine
Instead of Chlorine 18
Performance 18
Cost 18
Recommendations for Specific Treatment Changes 24
Control of Chloroform Potential (Removal of Precursors) 24
Clarification (Coagulation-Sedimentation-Filtration) 25
Granular Activated Carbon 26
Performance 26
Unit Cost 28
Capital Investment 37
Recommendations for Specific Treatment Changes 38
Removal of Chloroform 40
Recommendations for Specific Treatment Changes 42
MONITORING 43
ACKNOWLEDGMENTS 46
REFERENCES 47
APPENDICES 49
-------
INTERIM TREATMENT GUIDE FOR THE CONTROL OF CHLOROFORM AND OTHER TRIHALOMETHANES
IN DRINKING WATER
June 1976
EXECUTIVE SUMMARY
On March 29, 1976 Administrator Train released a public statement that
said in part, "The recent test results and the fact that chloroform is prevalent
in the environment have convinced me that the prudent course of action at this
time is to minimize exposure to this chemical wherever it is feasible to
do so." The statement also said, "EPA will work with cities and States to
evaluate certain modifications to current treatment practices that can reduce
the formation of chloroform during the water treatment process, without
lessening the effectiveness of disinfection. EPA research has shown that
changes in chlorination procedures practiced by some water systems can result
in reductions in the levels of chloroform produced. EPA plans to share these
initial findings on chloroform reduction with the States and some cities
encountering high chloroform levels, in an effort to reduce human exposure as
quickly as possible. This will also allow EPA to gain added information to
support the development of national regulations to limit chloroform levels
in water supplies."
EPA will shortly issue an Advanced Notice of Proposed Rulemaking seeking
public comment on various possible organic regulatory options. If the
Agency choses to issue a new regulation now, it should be finalized this
calendar year to become part of the Revised Interim Primary Drinking Water
Regulations. Between now and effective date of the Revised Regulations,
June 1977, the Agency will be aiding utilities who voluntarily attempt to
reduce the chloroform concentration in their drinking water. The purpose
of this Interim Guide is to provide these utilities with the information they
will need to be able to assess their own particular circumstances in conjunction
-------
- 2 -
with the technical assistance available from EPA. As with all Primary
Drinking Water Regulations, any eventual Regulation related to chloroform will
contain final treatment recommendations.
Chloroform concentrations can be lowered if chlorine is applied to the
water with the lowest possible organic content. Therefore, in locations
where it is feasible, a utility should consider moving the point of application
of chlorine to the point in the treatment process where the water should have
the lowest organic content; after filtration, or after coagulation and
settling, if these unit processes are employed. Improvement in these
clarification processes should also be considered. This should reduce the
chloroform concentration of the finished water somewhat although it will not
be eliminated. Utilities making such a change in disinfection practice
should carefully monitor the microbiological quality of their drinking
water to make sure that it has not deteriorated because of this change in
practice.
Further reduction in chloroform concentration can be obtained if a
disinfectant such as ozone, chlorine dioxide, or chloramine is used instead
of chlorine. These three disinfectants do not produce chloroform, although
they may produce other organic or inorganic by-products that have yet to be
identified, or evaluated for toxicity, or both. Furthermore, chloramine is
a weak disinfectant and should not be used exclusively. Finally, ozone does
not produce a disinfectant residual, thus the addition of chlorine may
also be necessary. If this practice produces a free chlorine residual,
some chloroform will be formed during passage through the distribution
system. Other oxidants such as hydrogen peroxide, potassium permanganate, and
so forth will also be evaluated as alternatives to chlorine.
-------
- 3 -
Water containing very little organic matter can be produced when fresh
granular activated carbon is used as a medium for the adsorption of organic
compounds. This water can then be disinfected with either chlorine, ozone,
or chlorine dioxide and little chloroform or other organic by-products will
be produced because of the small quantity of organic matter available for
reaction with the disinfectant. This treatment technique has the additional
benefit of removing many organic raw water contaminants, thereby providing
consumers with an additional margin of safety. Other adsorbants may also
produce the same effect.
The chief disadvantage of adsorption on granular activated carbon as
a treatment technique is that the adsorption capacity of the material is
limited. For example1, studies with Ohio River water have shown that a
2
30-inch bed of granular activated carbon receiving water at a rate of 2 gpm/ft
is effective for removing the potential for chloroform formation for about 1
month. In other situations where the organic load is higher or lower than
that: in the Ohio River this period of good performance would probably vary
accordingly. In general, however, the use of granular activated carbon for
the control of chloroform precursors means that the frequency of reactivation
will have to be increased over that commonly used when taste and odor control
is the only objective. Other adsorbants will be evaluated for their ability to
control chloroform potential, as well as removing raw water contaminants.
These techniques described above are all preferable to attempting
to remove chloroform once it has been formed as no unit process has yet been
demonstrated to be very effective for chloroform removal. Although all the
information concerning these processes is not known and an extensive research
program is ongoing to refine the information and confirm the results in the
field, enough is known to recommend the use of any of these treatment processes
in certain circumstances at this time.
-------
- 4 -
Costsof these unit processes are difficult to generalize as they vary
widely depending on local circumstances. Table 1 summarizes some example cost
data, however.
TABLE 1
SOME TYPICAL COST DATA
(All Costs in cents per 1000 gal.)
Design Capacity 1 mgd 10 mgd 100 mgd 150 mgd
Average Plant Flow 0.7 mgd 7 mgd 70 mgd 105 mgd
2 mg/£ chlorine, 30 min. contact time 4 1 0.7 0.6
1 mg/£ ozone from air, 20 min, contact time 6 2 0.9 0.8
1 mg/£ ozone from oxygen, 20 min. contact time 8 2 1 0.8
1 mg/£ chlorine dioxide from sodium chlorite,
30 min. contact time 4 211
1 mg/£ chlorine dioxide from sodium chlorate,
30 min. contact time * * * *
Granular Activated Carbon, replacement of sand,
on site reactivation 41 12 6 5
Polymeric Adsorbants (macroreticular resins, etc.) * * * *
Aeration, 30 to 1 air to water ratio
20 min. detention time 22 13 9 9
*Insufficient information available to calculate unit costs at this time.
Discussion of the various treatment techniques in detail including the
estimated cost of treatment, specific suggestions for modification of several
different types of water treatment plants, and recommendations for monitoring
are contained in the Interim Guide that follows.
A supporting document (Appendix) contains a detailed analysis of treatment
at suj
2,3,4
unit process cost , and three papers on the experimentation that supports
the treatment recommendations contained in this Interim Guide.
-------
INTRODUCTION
Background
Reaction of chlorine with certain organic materials to produce chloroform
and related organic by-products has probably been occurring since chlorination
was first practiced as a disinfection procedure for drinking water. The
presence of these compounds in drinking water escaped detection until
recently because, having fairly low boiling points, they were lost during
certain steps in the procedures for performing typical organic analyses of
water by gas chromatography.
Recently, however, both in The Netherlands and in this country ,
investigators developed alternate organic analytic procedures that allowed
the measurement of this type of organic compound. These investigators
used the newly developed analytic procedure to demonstrate that the
concentrations of chloroform and related compounds were generally higher in
finished water than in raw water, indicating that they were being produced
during the chlorination of water. '
Nomenclature
For those readers unfamiliar with organic nomenclature, the following
discussion defines some of the terms used later in the Guide. Although methane
gas is not involved, the reaction of chlorine in water with certain organic
compounds (believed at this time to be primarily humic acids, part of the
group of organic materials associated with decaying vegetation) under certain
conditions produces a group of halogen-substituted single carbon compounds.
These compounds are named as derivatives of methane (CH ) and are listed below.
-------
- 6 -
TABLE I
FORMULAS AND NAMES OF THE TRIHALOMETHANES
1. Cl
I
H-C -Cl
I
Cl
Trichlorome thane
(Chloroform)
2. Br
I
H-C -Cl
I
Cl
CHBrCl
Bromodichloromethane
3. Br
H - C - Cl
i
Br
CHBr Cl
2
Dibromochloromethane
4. Br
I
H-C- Br
i
Br
CHBr,,
Tribromomethane
(Bromoform)
5. I
i
H -C - Cl
i
Cl
CHC12I
Dichloroiodomethane
I
i
H - C - Cl
Br
CHBrClI
Bromochloroiodomethane
7.
H- C - I
i
Cl
CHC1I2
Chlorodiiodomethane
H - C - Br
Br
CHBr I
Dibromoiodome thane
H -C - Br
CHBrI
Bromodiiod ome thane
10. I
I
H-C -I
I
I
CHI
Triiodomethane
(lodoform)
-------
- 7 -
Under typical circumstances the trihalomethanes produced in drinking water
are dominated by compounds 1 and 2 above, but compounds 3 and 4 have been
frequently found and 5 has been detected. The bromine- and iodine-containing
trihalomethanes have been shown to be formed by chlorine oxidizing any bromide
and iodide in water to bromine and iodine, these halogens then reacting with
organic matter to form the corresponding trihalomethanes. Fluorine-containing
trihalomethanes are neither formed during fluoridation nor can chlorine
oxidize fluoride to fluorine to produce them.
1975 Surveys
The increased interest in the organic content of drinking water generated
by studies of the New Orleans, Louisiana finished water, plus the information
in the literature cited ' prompted the Administrator of the U.S. Environmental
Protection Agency to announce a National Organic Reconnaissance Survey (NORS)
on November 8, 1974. The purpose of this Survey was, in part, to determine
on a nationwide basis, the conditions under which trihalomethanes were formed
during water treatment. To accomplish this objective the raw and finished
water in 80 water utilities across the nation was sampled and the concentration
of compounds 1 through 4 in Table I determined on each sample. Note, in an
effort to somewhat simulate passage of water through a distribution system,
these finished water samples were not dechlorinated at the time of collection,
but the trihalomethane formation reaction was allowed to proceed during sample
shipping, although the samples were iced.
This Survey, carried out during February to April 1975 confirmed that
all of the chlorinated drinking waters investigated contained some chloroform,
ranging from less than 0.2 ug/& (ppb) to 311 yg/£. One utility surveyed
ozonated as the only treatment and had <0.1 yg/Jl chloroform in its drinking
Q
water . A companion Survey (carried out during the same time period by
-------
EPA'g Region V Laboratory) of 83 utilities in the upper mid-West yielded
very similar data ranging from a chloroform concentration of <1 yg/A to
Q
366 yg/£. Combining these two surveys, Figure it shows the median chloroform
concentration to be 20 yg/£, with 10 percent of the drinking waters containing
more than 105 yg/£ of chloroform for this season. Most of the finished
waters also contained some of the other three trihalomethanes measured.
Because 80 percent of the utilities surveyed were surface sources while only
20 percent of the Nation's community supplies are surface sources and many
utilities using ground water as a source do not practice chlorination, these
data should not be taken as national statistics. A similar survey, the
National Organics Monitoring Survey (NOMS) is being made during 1976 in
113 locations with samples collected in the Spring, Summer and Fall to
determine the seasonal variation in trihalomethane concentrateons.
Prior to these Surveys some concern was expressed that other chlorinated
compounds detectable by this analytic technique were formed during chlorination.
This possibility was examined for three other compounds, 1,2-dichloroethane,
carbon tetrachloride,and methylene chloride. Results showed that these
compounds were not formed during chlorination.
Surrogate Measurements
Because the analysis for chloroform is a gas chromatographic procedure
requiring skilled operators and about one hour to complete, a simple rapid
surrogate measurement that would predict chloroform concentrations seemed
desirable. Chlorine reacts with some organic precursors to form chloroform
and related products, therefore a test that would measure the precursor
concentrations in the raw water would be useful for anticipating finished water
chloroform concentrations. No direct test for trihalomethane precursors
exists, so a test for general organic content was considered as an
-------
- 9 -
400
300
200
100
O)
50
«J
Z
O
u
-s.
ee.
O
u.
O
ot
O
25
10
COMBINED NORS AND REGION 2 SURVEY
152 UTILITIES
122 SURFACE, 3"J GROUND
165 SAMPLES, FEB.-APRIL, 1975
5 10 20 40 60 80 90 95
PERCENT OF SAMPLES EQUAL TO OR LESS THAN GIVEN CONCENTRATION
FIGURE 1. FREQUENCY DISTRIBUTION OF CHLOROFORM
-------
- 10 -
alternative. The difficulty with using a general organics tests as a
measure of chloroform precursors is that the precursor concentrations
are not a constant percentage of the general organic content.
Nevertheless, in the report of the National Organics Reconnaissance
o
Survey three general organic tests were proposed as surrogate parameters,
non-purgeable total organic carbon* (NPTOC), ultra-violet absorbance, and
emission fluorescence scan. Because turbidity interfered with the two
optical measurements, ultra-violet absorbance and fluorescence, they were
o
performed on the finished, rather than the raw water. The report suggested
that raw water NPTOC concentrations were related to total trihalomethane
concentrations in the finished water.
To review this suggestion, these data were statistically analyzed by
three different methods; Spearman Rank Correlation; Linear Regression
Analysis, and Log-Log Regression Analysis. The high level of confidence of
the Spearman Rank Correlation Coefficients in Table II shows that the level
of the surrogate measurements does rise and fall as the chloroform concentration
rises and falls. Table III, however, shows that although the linear
regression correlation coefficient for NPTOC and ultra-violet absorbance with
chlorofrom concentration is fairly high, the percent of the chloroform
concentration variation explained by the variation in NPTOC concentration
or ultra-violet absorbance is fairly low, indicating that other factors are
important in determining the chloroform concentration in a given water.
Finally, the scatter of the data is shown by the magnitude of the 95 percent:
confidence limits around the mean, again indicating that these surrogate
measurements are poor predictors of chloroform concentrations.
*That portion of the total organic carbon concentration that remains in a
sample after the carbon dioxide has been purged out under acid conditions.
-------
- 11 -
TABLE II
SPEARMAN RANK CORRELATION OF SURROGATE MEASUREMENTS
Chloroform
Surrogate
Number
of Observations
Correlation
Coefficient
Level of
Confidence
Raw Water
NPTOC
Finished Water
Ultraviolet
Absorbance
Finished Water
Fluorescence
82
81
82
0.57
0.48
0.42
>99%
>99%
>99%
TABLE III
LINEAR REGRESSION OF SURROGATE MEASUREMENTS
% of CHC10
95% Confidence
Variation Explained Limits Around
Chloroform
Surrogate
Raw Water
NPTOC
Number
of Observations
82
Correlation
Coefficient
0.74
by Surrogate
Variation
54.5
the Arithmetic
Mean (43.7 yg/1
- 79.8 yg/£
Finished Water
Ultraviolet
Absorbance
Finished Water
Fluorescence
81
82
0.54
0.13
29.2
l.i
- 99.9 pg/£
- 117.4 yg/£
-------
- 12 -
In an effort to improve the usefulness of these surrogate measurements
as chloroform concentration predictors, they were reanalyzed after making a
log transform. The data in Table IV shows that making a log transfer does,
in general, improve the usefulness of these surrogate measurements, but not
sufficiently to be considered satisfactory. Therefore, the previous
Q
suggestion of the relation of raw water NPTOC concentrations and finished
water total trihalomethane concentrations was overstated, although the data
from the 1976 Survey will be tested for this possibility. The possible
correlation of these general organic parameters and the organics - carbon
adsorbable test yielding the carbon chloroform extract concentration will
be attempted with the 20 other specific organic chemicals being measured
in NOMS.
TABLE IV
LOG-LOG REGRESSION OF SURROGATE MEASUREMENTS
Chloroform
Surrogate
Number of
Observations
Correlation
Coefficient
Percent of CHC13 95% Confidence
Variation Explained Limits Around
by Surrogate the Geometric
Variation Mean (16.2 \ig/i)
Raw Water
NPTOC
Finished Water
Ultraviolet
82
0.67
45.5
234 to 1.7
Absorbance
Finished Water
Fluorescence
81
82
0.51
0.41
26.0
16.8
372 to 0.7
480 to 0.5
One other possibility would be to make a multi-variant analysis to take
into account some of the other factors thought to influence chloroform
production such as the presence of free or combined chlorine residual, raw
-------
- 13 -
water chlorination and so forth. Even if this would be successful, the
resultant equation might be too'complicated to be useful. The Division of
Public Health at the University of Massachusetts is attempting to make a
multi-variant analysis realting common water treatment measurements such as
turbidity, color, pH and so forth with chloroform concentrations as a
method of estimating historical chloroform concentrations from historical
water treatment plant data. If successful, these data will be correlated with
cancer incidence.
Although further research on the three surrogate analytic procedures
might improve them, at this time, the best method of determining the chloroform
concentration is to obtain the necessary equipment and technical staff to
perform the analysis directly. This is particularly true because special
analytic equipment and skilled operators are required to make the surrogate
measurements. These analyses do have a place in water treatment, however.
If an organic removal unit process is being used by a utilitys these general
organic content measurements are good process control determinations.
Position of the Environmental Protection Agency
The release of the National Cancer Institute chloroform carcinogenicity
report caused the Administrator of EPA to make a public statement on
March 29, 1976 suggesting that water utilities voluntarily take what steps
they could to reduce the chloroform concentration in their particular
drinking water. To aid these utilities in this effort, the Agency has
prepared the Interim Treatment Guide that follows and will offer volunteering
utilities technical assistance. In addition, EPA will issue an Advance Notice
of Proposed Rulemaking to solicit public comment and information regarding
alternative regulatory strategies for organics in drinking water.
-------
- 14 -
The purpose of the Interim Guide is to provide utilities with the
information they will need to be able to assess their own particular
circumstances and contains treatment suggestions for water utilities desiring
to reduce the concentration of chloroform in their drinking water, summarizing
data on cost and effectiveness. A companion document to the Guide (Appendix)
details the unit process cost information and the research data to support
2 3
the treatment suggestions ' . Any final Regulation of chloroform will,
of course, contain treatment recomendations.
The Administrator's statement dealt with the problem of chloroform in
drinking water because at the present time chloroform is the only trihalomethane
that has been tested for carcinogenicity, and other physiological effects. The
other trihalomethanes measured in the three Surveys may, however, eventually
also be classed as health hazards. For example, retrospective epidemiological
studies are currently underway in an effort to assess the impact of the
concentrations of chloroform and the other trihalomethanes on cancer rates in
exposed populations. Therefore, the treatment research, although emphasizing
chloroform, investigated treatment techniques for removing four of the
2 3
trihalomethanes. ' Thus a utility desiring to remove bromine-containing
compounds, will have the benefit of the available research information on that
topic. For example, fourteen utilities in the NORS and Region V Survey
had concentrations of bromine-containing trihalomethanes in their finished
water that exceeded the chloroform concentration.
-------
- 15 -
EXTENT OF PROBLEM
Although 90 percent of the water utilities sampled in the NORS and
Region V Surveys had wintertime concentrations of chloroform in their
drinking waters less than 105 yg/£, a utility manager in a location not
yet sampled might want to know the concentration of chloroform and other
trihalomethanes in that particular drinking water, particularly in warmer
weather. The analytic procedure has been published and can be performed
12
onsite if qualified staff analists and proper equipment are available.
Although an initial investment of from $7,000 to $8,000 is necessary,
purchase of such equipment will allow a water utility to monitor this
important parameter frequently, thereby providing their consumers with an
additional assurance of safety.
If this is not possible, many State, EPA Regional, or qualified private
laboratories can perform this analysis. Using these laboratories does
involve shipping of samples, and some delay in receiving results, but is
a satisfactory method of operation. Note: Because the trihalomethane
format-ion reaction will continue -in the sample bottle -If chlorine is present,
sodium thiosulfate should be added to dechlorinate the sample upon collection
if the concentration of trihalomethane at the time of collection is
desired. If the potential for additional trihalomethane formation, such as
might occur during distribution, is to be investigated, the sample should
be stored without the dechlorinating agent for a time and at a temperature
and pH similar to that occurring in the distribution system and then
4
dechlorinated. Reference 4 is available as an Appendix to the Interim Guide.
-------
- 16 -
Another method for estimating which utilites might produce water with
high chloroform concentrations would be a comparison with the nine utilities
having the highest chloroform concentrations in the National Organics
Reconnaissance and Region V Survey, approximately the upper 10 percent.
These data, see Tables V and VI indicate that high chloroform concentrations
result when surface or shallow ground water with a high NPTOC concentration
and a high chlorine demand is dosed with enough chlorine to produce a
high free chlorine residual, particularly if the water is somewhat basic.
Water utilities with similar characteristics would be expected to have
finished water with relatively high chloroform concentrations. Controlled
experiments have confirmed that these factors tend to enhance the reaction
2 13
of chlorine with precursor to produce trihalomethanes '
-------
- 17 -
TABLE V
ANALYSIS OF NINE UTILITIES HAVING HIGH CONCENTRATIONS OF CHLOROFORM IN THE
NATIONAL ORGANICS RECONNAISSANCE SURVEY
Range Average
Chlofororm Concentration 103 yg/£ - 311 yg/£ 177 yg/£
Total Chlorine Dose 4.3 mg/£ - 18 mg/£ 9.0 mg/£
Combined Residual 0-1.7 mg/£ 0.5 mg/£
Free Residual 0-2.7 mg/£ 1.2 mg/£
Raw Water NPTOC* 4.5 mg/£ - 19.2 mg/£ 8.4 mg/£
Finished Water NPTOC 2.3 mg/£ - 12.2 mg/£ 4.7 mg/£
Chlorine Demand (Total Dose -
Total Residual) 2.8 mg/£ - 15.7 mg/£ 7.3 mg/£
Finished Water pH 7.3-9.5 (one unknown)
Number of Utilities with the following characteristics:
River Source - 5 Old Granular Activated Carbon - 2
Lake or Reservoir Source - 3 Raw Water Chlorination - 6
Shallow Ground Water Source - 1 Settled Water Chlorination - 3
Filtration - 9 Post-Chlorination - 6
Precipitative softening - 4
*Non-purgeable total organic carbon
TABLE VI
ANALYSIS OF NINE UTILITIES HAVING HIGH CONCENTRATIONS OF CHLOROFORM IN THE
REGION V SURVEY
Chloroform Concentration 127 yg/£ - 366 yg/£ 203 yg/£
Total Chlorine Dose 4.5 mg/£ - 13 mg/£ 7.4 mg/£
(2 unknown)
Number of Utilities with the following characteristics:
River Source - 8
Lake Source - 1
Raw Water Chlorination - 3
Post-Chlorination - 7 (2 unknown)
-------
- 18 -
SUGGESTIONS FOR ALTERNATE TREATMENT
In general terms, the reaction to produce chloroform is as follows:
Chlorine + Precursors ——> Chloroform + Other Trihalomethanes
This implies that three approaches to chloroform concentration control
are available:
1. Change disinfectant.
2. Treat to reduce the precursor concentration prior to chlorination
(control chloroform potential).
3. Treat to reduce the chloroform concentration after formation.
In these studies the use of ozone, chlorine dioxide and chlorine
and ammonia were evaluated as'techniques for changing disinfectants. For
control of chloroform potential (precursor concentration) coagulation-
sedimentation, adsorption on powdered- and granular activated carbon,
ozonation, and the use of chlorine dioxide were investigated. Adsorption
on powdered- and granular activated carbon, ozonation, aeration and
the use of chlorine dioxide were studied as methods for chloroform removal.
Change in Disinfectant
Use of Ozone, Chlorine Dioxide, or Chloramine Instead of Chlorine
Performance
When used as a disinfectant neither ozone, chlorine dioxide, nor chloramine
produced measurable quantities of trihalomethane. Although this appears
favorable, the use of ozone does not produce a disinfectant residual to be
carried throughout the distribution system. Further, the health hazard,
if any, of the by-products of the reaction of ozone with organic matter
occurring in water is not known. The same situation, in general, exists
with chlorine. Except for trihalotnethanes, chloramines and chlorophenols,
little is known about the by-products formed during chlorination.
-------
- 19 -
Finally, if chlorine is used to provide a disinfectant residual following
ozonation, trihalomethanes will then be formed and little improvement is
gained when compared to chlorination. Combined chlorine has been suggested
as a possibility, but its ability to control general aftergrowth in the
distribution system is questionable, although it will control total coliform
organisms, under some circumstances.
Chlorine dioxide, on the other hand, does produce a residual, which is
an advantage. A disadvantage is the possible toxicity of the organic
by-products resulting from the reaction of chlorine dioxide with organic
matter in water, again, similar to possible undiscovered problems with
chlorine. Furthermore, a few citations in the literature have indicated
concern over toxicity of chlorite, a possible product of the reactions of
14 15
chlorine dioxide when added to natural water. '
Another problem with chlorine dioxide is its generation. The reaction
of sodium chlorite (NaCIO ) and sulfuric or hydrochloric acid will produce
chlorine dioxide without chlorine being present, but this reaction is
inefficient. Therefore, chlorine dioxide is usually generated by reacting
sodium chlorite with chlorine. Because this reaction proceeds better
at low pH, excess chlorine is usually added to reduce the pH. This produces
a chlorine dioxide and chlorine mixture. The quantity of excess chlorine
in chlorine dioxide can be reduced by adjusting the pH with acid and
carefully controlling the ratio of chlorine to sodium chlorite. Although
data indicate that the resultant chloroform concentration will be lower
when chlorine and chlorine dioxide are used together when compared to the
use of chlorine alone, trihalomethanes will not be absent if excess
chlorine contacts the water.
-------
- 20 -
The Pulp and Paper Industry, the largest single consumer of chlorine
dioxide, generates relatively high volumes of chlorine dioxide via processes
using lower cost sodium chlorate (NaCIO ). Two of these processes involve
a reaction mixture containing equal amounts of sodium chlorate and sodium
chloride (NaCl) and an excess of sulfuric acid to yield chlorine dioxide
and chlorine in a molar ratio of 2 to 1. The major difference between the
two systems is engineering design. This difference results in one
system having greater yields and improved waste stream handling. The Pulp
and Paper Industry has also used two other systems. One of these systems
is based upon reacting sodium chlorate with sulfur dioxide (SO ) and sulfuric
acid to yield chlorine dioxide that is contaminated with chlorine and sulfur
dioxide. The other system makes use of a reaction mixture containing sodium
chlorate, methanol, and sulfuric acid to yield chlorine dioxide. The systems
making use of sodium chlorate, sodium chloride, and sulfuric acid produce
the highest yields and are the most cost effective for the Pulp and Paper
Industry. It should be pointed out that all four of the major processes
that have found use in the Pulp and Paper Industry to produce chlorine
dioxide operate on a scale larger than would probably be necessary for
drinking water disinfection. The economics of scaling down these processes
are unknown at this time.
Combined chlorine is not as reactive as free chlorine for the
formation of chloroform. Therefore, if a utility should add ammonia
in conjunction with chlorine addition or shortly thereafter such that
no free chlorine residual ever existed for very long, chloroform formation
should be low. The mere presence, however, of a combined chlorine residual
in the finished water does not assure that free chlorine was not present
sometime earlier during the treatment of the water. For example, many
-------
- 21 -
water utilites in the National Organics Reconnaissance Survey that had
a finished water with only combined residual did have substantial quantities
of chloroform in their water. In these cases, free chlorine residual must
have been reacting with chloroform precursor at some time during the
treatment of the water, In spite of this, combined chlorination as a
primary disinfectant was not discussed in detail in this sub-section
because not enough is known about its disinfecting power to judge its
value at this time. Combined chlorine may, however, have a potential
as a secondary disinfectant to provide a residual in the distributuion
system following ozonation.
Cost
In an effort to compare the cost of alternate disinfectants, calculations
were made assuming no disinfection facilities at a water treatment plant.
Therefore, costs of disinfection equipment, plus the cost of a disinfectant
contact chamber is included for chlorine, ozone and chlorine dioxide.
The summary of these data, Table VII, shows that the cost of all three
disinfectants are similar. Of course, in an existing plant, a change to
ozone might mean the abandoning of some of the existing chlorination facilities,
so on that basis the costs in Table VII might not be applicable. On the
other hand, a water treatment plant now using chlorine could add chlorine
dioxide capability at a small incremental cost.
-------
- 22 -
TABLE VII
ESTIMATED COST OF DISINFECTION*
All Costs in Cents/1000 gallons
Design Capacity
Average Daily Flow
Chlorine 15«f/lb
0., generated by air
0 generated by oxygen
Chlorine 15«
-------
- 23 -
In this cost analysis, a lower ozone and chlorine dioxide dose was
compared to the chlorination dose on the basis that both of these disinfectants
are more effective than chlorine and therefore less disinfectant would be
required. During the experiments described in Reference 3, for example, 0.5 mg/£
of either ozone or chlorine dioxide was sufficient to adequately disinfect
the effluent from the dual-media filter in the pilot plant. In a later
experiment 0.3 mg/£ of C10~ reduced the standard plate count population to
less than 1 per 1 ml. A dose of 1.3 mg/£ was required for chlorine to accomplish
the same disinfection, although the pilot plant was not performing too well
at this time, so this might not be the minimum chlorine dose. Finally, using
a lower disinfectant dose should cause the formation of less non-trihalomethane
organic by-products, although this has not yet been demonstrated experimentally.
To determine •' the cost of disinfection to a typical household, a family
of four was assumed to use 200 gallons of water per day. If this rate of
consumption was steady throughout a calendar year, the annual usage would be
73,000 gallons. Therefore, multiplying any disinfection cost in cents per
1000 gallons time 73 would produce an estimate of the annual cost to a
typical household for a given treatment process. Using the costs in Table VII
chlorination contributes from about $0.40 to $2 per year to the water bill of
an average household, depending on treatment plant size. On the same basis,
changing to ozone would make these figures about $0.50 to $4 per year,
while the use of chlorine dioxide would cost from about $0.70 to $2 per
year. Note: Because of the influence of local conditions these costs should
be considered approximate.
-------
- 24 -
Recommendations for Specific Treatment Changes
Although research on these and other alternate disinfectants is still
in progress, enough is known about their effectiveness
that any utility currently chlorinating that desires to lower their
trihalomethane concentration should consider these alternate disinfectants.
Care should be taken, however, during the transition period to insure the
microbiological quality of the finished and distributed water. Utilities
currently using chlorine dioxide for taste and odor control should consider
using it both for taste and odor control and disinfection. They should
attempt to generate the chlorine dioxide with as little chlorine as possible
in it for the maximum reduction in trihalomethane concentration.
Utilities currently adding ammonia to create chloramines should review
their practice and reduce the elapsed time between the addition of chlorine
and ammonia to the minimum compatible with good disinfection. Shortening
the contact time of free chlorine residual should reduce the formation of
trihalomethanes and if mixing is relatively good, disinfection should
, 16
be adequate.
Control of Chloroform Potential (Removal of Precursors)
Control of chloroform potential was attempted by five treatment
techniques prior to chlorination — adsorption on powdered- and granular
activated carbon, the use of chlorine dioxide, ozonation, and coagulation-
sedimentation. Unrealistically high doses of powdered activated carbon only
resulted in partial control of chloroform potential, the use of chlorine
dioxide was only moderately successful, and very high doses of ozone were
required to produce measurable results. Clarification and adsorption on
granular activated carbon, on the other hand, were successful.
-------
- 25 -
Clarification (Coagulation-Sedimentation-Filtration)
Data developed during this study show that less chloroform is formed
if chlorine is added to water with the lowest possible organic content
(highest quality). For example, filtration of Ohio River water through
filter paper reduced the chloroform potential 26 percent while in the pilot
plant chlorinating settled water rather than adding chlorine to the rapid
mix reduced the chloroform potential 15 percent. Two field studies have
demonstrated similar favorable results. A limited attempt was made to enhance
coagulation and sedimentation by pre-ozonating raw water prior to coagulant
addition, but this was not too successful. Therefore the quickest and the
least expensive method of maintaining low chloroform concentrations in
finished water, would be for a utility to chlorinate the highest quality water
possible.
If water is filtered, the highest quality water is filter effluent. In
many water treatment plants, however, chlorine cannot be added at this point
because insufficient contact time is present to permit adequate disinfection
before use, unless additional contact tanks were constructed. Therefore,
chlorination of settled water just prior to filtration may be the best
alternative. This also has the advantage of having a disinfectant pass
through the filters, thereby keeping them cleaner.
Chlorination of coagulated and settled water has disadvantages, however.
One is that the concentration of chloroform and other trihalomethanes will
be reduced, but not eliminated in the finished water and will continue to
be formed during distribution. Secondly, other organic by-products produced
during chlorination may or may not be reduced in concentration, but will
not be eliminated. Finally, the absence of a disinfectant at the beginning
of treatment may cause problems because of the growth of algae, slimes
-------
- 26 -
and higher forms in the early part of water treatment plants. Periodic shock
chlorination or an alternate disinfectant could possibly control these
problems, but at those times some chloroform formation might occur.
Granular Activated Carbon
Performance
When fresh, granular activated carbon will adsorb most trihalomethane
precursors so that following granular activated carbon treatment chlorination
can be practiced without forming much trihalomethane;
will reduce the possibility of producing hitherto unknown organic
by-products during disinfection because little organic matter will be
present with which any disinfectant can react, and,
will, beyond removing chloroform and trihalomethane precursors,
produce water with a low overall concentration of organic matter, thereby
increasing the likelihood of the removal of raw water organic contaminants
that may be of health concern now or in the future.
In addition, fresh granular activated carbon will adsorb trihalomethanes
that have been formed by chlorination practiced prior to granular activated
carbon treatment (see next Sub-section) .
In spite of the advantages listed above, this treatment technique is
not without its disadvantages. The performance noted above can be achieved
when granular activated carbon is fresh, but this effectiveness does not
last for a long time. For example, in a plant treating water similar in
character to the Ohio River (the source used in these experiments), removing
the filter media, replacing it with 2-3 feet of granular activated carbon,
and operating the filters at conventional approach velocities of 2-3 gpm/sq ft,
would provide the excellent performance described above from the granular
activated carbon for about one month. This time period will vary if water
-------
- 27 -
with an organic content much higher or much lower than that in the settled
water used in these studies, about an NPTOC of 1 mg/1 in the winter, 1.1 mg/Jl
in the spring, 1.4 mg/£ in the summer and 1.3 mg/£ in the fall, is applied
to granular activated carbon. Within these limits, however, with respect to
performance, granular activated carbon adsorption is the best technique
for chloroform potential control yet investigated.
As noted above, although not directly related to the chloroform problem
in drinking water, granular activated carbon has the ability to adsorb many
other organics. Because adsorption is not complete, however, some uncertainty
exists relating to the exact organic content of the effluent from fresh
granular activated carbon beds. At this time, measurement of the total
organic carbon content of fresh granular activated carbon bed effluent is not
possible. Although in these studies non-purgeable total organic carbon
concentrations in fresh granular activated carbon bed effluents are relatively
low, commonly less than the 0.1 mg/Ji detection limit of the analytic methods,
the concentration of the "purgeable" total organic carbon fraction in fresh
granular activated carbon bed effluents is not known, although it is not
expected to be high.
Nevertheless, granular activated carbon has the ability to adsorb many
specific organics of current concern even when partially exhausted for the
removal of general organic compounds as measured by the NPTOC test. For
example, taste and odor causing compounds are removed for many months following
the breakthrough of NPTOC. Further, several years ago, partially exhausted
granular activated carbon was shown to remove dieldrin, lindane, 2,4,5-T,
DDT, and parathion. Finally, granular activated carbon removed 30 pg/£ of
naphthalene spiked into Cincinnati, Ohio tap water for eight months even
though other organics were penetrating the bed long before that time. This
-------
- 28 -
is not true for chloroform, however, which is much more water soluble than
naphthalene. Chloroform at a detection limit of 0.1 yg/£ penetrated a
granular activated carbon bed treating Cincinnati, Ohio tap water about ten
days to two weeks before NPTOC penetrated at a detection limit of 0.1 mg/£.
Therefore, although some uncertainties exist relating to the complete organic
content of the effluent from fresh granular activated carbon beds, past
evidence indicates that non-polar synthetic organics with low solubility in
water are well adsorbed by granular activated carbon. Also, as noted
above, granular activated carbon can be very effective for removal of
precursors of chloroform and other trihalomethanes thereby reducing the
potential for their formation upon chlorination.
Unit Cost
Attempts have been made to estimate the additional cost of water
treatment using granular activated carbon. The details of the assumptions used
in the computer program developed from the data in the "Technology Transfer
Process Design Manual for Granular Activated Carbon Systems" are presented
in Table VIII. Once calculated, these costs were then analysed to determine the
influence of the following eleven variables on the estimated cost: reactivation
frequency, percent granular activated carbon attrition loss per reactivation,
granular activated carbon cost, fuel cost, wage rate, electric power cost,
interest rate, design life, load factor, construction cost index,and wage
price index. An analysis was also made of the cost of off-site reactivation
for small plants and the cost of using granular activated carbon after
filtration rather than as a combination filtration-adsorption medium.
The influence of the eleven variables on cost on one size plant are
presented in Table IX as an example. Note: Studies have not yet been
-------
- 29 -
completed to determine the loss in adsorption capacity, if any, of granular1
activated carbon for chloroform potential removal during reactivation. If
the loss in capacity were high, this would increase the final cost.
TABLE VIII
FACTORS USED IN GRANULAR ACTIVATED CARBON COST CALCULATIONS
Hydraulic Loading Rate = 2 gal/min/sq ft
Contact time =4.5 min (Apparent)
Activated Carbon attrition loss per reactivation cycle = 10%
No loss in adsorptive capacity during reactivation
Volume of granular activated carbon per 1 mgd filter >= 865 cu ft
Reactivation frequency = once per month at 100% of design capacity,
once per 1.4 months at 70% of design capacity (see Reference 1 for discussion)
Activated carbon used as replacement for granular filter media
On-site reactivation furnace, multiple hearth
Plant production 70% of design capacity (A typical yearly average)
-------
- 30 -
TABLE IX
FACTORS FOR ADJUSTING UNIT COSTS FOR GRANULAR ACTIVATED CARBON TREATMENT
FOR A 100 mgd PLANT
Part A
ADDITIVE FACTORS
Effect on Operational Costs
Item
Granular Activated Carbon Loss
per Reactivation Cycle, %
Granular Activated Carbon Cost,
cents/pound
Fuel cost, $/Million BTU
Power Cost, cents/kw-hr
Direct Hourly Wage Rate, $/hr
Wage Price Index
Effect on Capital Cost
Item
Granular Activated Carbon Cost,
cents/pound
Construction Cost Index
Value
15
10
5
54
38
19
1.89
1.26
0.63
1.5
1.0
0.5
7.785
5.190
2.595
2.672
1.781
0.891
54
38
19
3.851
2.567
1.248
Factor (see text for explanation)
+0.295
0
-0.147
+0.169
0
-0.220
+0.040
0
-0.038
+0.006
0
-0.006
+0.159
0
-0.159
+0.006
0
-0.005
+0.169
0
-0.220
+0.495
0
-0.510
-------
- 31 -
PART S
MULTIPLICATIVE FACTORS
Effect on Operational Cost Value Factor (see text for explanation)
Item
Reactivation Frequency, weeks between 3 1.459
6 1.000
9 0.721
Hydraulic Load Factor, % 50 1.052
70 1.000
100 0.924
Effect on Capital Costs
Item
Amortization Period, yrs. 10 1.509
20 1.000
30 0.854
Interest Rate, % 10.5 1.273
7 1.000
3.5 0.730
Reactivation Frequency, weeks between 3 1.215
6 1.000
9 0.915
Hydraulic Load Factor, % 50 1.289
70 1.000
100 0.764
In Table IX three sets of values are given for each of the eleven
factors that influence the final unit cost. The middle value is the one
used to calculate the unit costs presented in Table X. For each variable the
lower value is about 50 percent of the middle value and the upper value is
about 150 percent of the middle values. The factors in Table IXA when
multipled times the appropriate costs in Table X yields the change in final
unit cost caused by the change in that factor. An example follows-
-------
- 32 -
For instance, changing the hourly wage rate from $5.190/hour to $7.785/hour
would increase the operating unit cost for a 100 mgd plant by [4.5^/1000 gallons
(from Table X) x 0.159]= 0.72^/1000 gallons, while reducing the construction
cost index from 2.567 to 1.248 would reduce the capital unit cost by [1.5^/1000
gallons (from Table X) x -0.51] = -0.77«f/1000 gallons. To individualize
unit costs, the changes for each factor in the "A" section of Table IX
from the middle value must be added and subtracted from the appropriate
cost in Table X, yielding the final new unit cost.
TABLE X
ESTIMATED UNIT COST OF GRANULAR ACTIVATED CARBON TREATMENT
(See Tables VIII and IX for Assumptions)
Design capacity, mgd 1 10 100 150
Capital unit cost 19.5 3.5 1.5 1.1
Operating unit cost 21.5 8.2 4.5 4.0
Total unit cost 41.0 11.7 6.0 5.1
The factors in Table IXB are used slightly differently. Here, rather
than add or subtract costs from the costs in Table X, the factors are
multiplied in sequence to yield an overall multiplying factor that is
applied to the resultant unit cost after the additive factors from Table IXA
have been applied.
Because of the number of variables involved,each with its own impact
on total cost, calculation of a universally applicable unit cost is impossible.
To aid utilities calculate their own unit cost, Reference 1 contains eleven
monographs each of which shows the impact on total cost for the variable over
a wide range for four different plant sizes, Table IX being a brief example.
Several examples are also presented in Refrence 1. This should allow a utility
to estimate its own total unit cost using local conditions for each variable.
-------
- 33 -
Using the middle values in Table IX, the unit costs were calculated that
appear in Table X. For a 100 mgd plant, using all the lower factors in
Table IX, the total unit cost drops from 6«(/1000 gallons to less than 2(^/1000
gallons. On the otherhand, using all the high factors in Table IX, the total
unit cost rises to over 17 ^/lOOO gallons. Although either of these extremes
is unlikely, this shows the sensitivity of these unit costs to local conditions,
The cost of granular activated carbon treatment with on-site reactivation
for a 1 mgd plant is very high, 41^/1003 gallons, see Table X. These
costs could be lowered to more reasonable levels if the exhausted granular
activated carbon was transported to a central reactivation facility. For
example, the cost of granular activated carbon treatment could be reduced to
about 16(^/1000 gallons by transporting the granular activated carbon as far
as 100 miles to a reactivation facility of the size equivalent to that required
at a 10 mgd water treatment plant. This makes the cost for a 1 mgd facility
approach that of a 10 mgd plant. If granular activated carbon treatment
is desired for small water treatment plants, regional reactivation facilities
will be essential. Although need and not size will be the final determining
factor as to which utilities might use granular activated carbon treatment, to
give some prospective as to national impact, the Inventory of Public Water
Supplies lists about 420 utilities using surface water as a source with a
design capacity of 10 mgd or greater, serving about 77 million people and about
265 utilities with ground water sources in the same size range, serving
about 34 million people.
-------
- 34 -
Another factor that must be considered in the overall cost of granular
activated carbon treatment is the major capital expense of the on-site
furnace. Currently four types of furnaces are available, the multiple-
hearth furnace, the fluidized bed furnace, the infra-red furnace and the
rotary kiln furnace. At this time experience in this country is with the
multiple-hearth furnace. Table XI summarizes the estimated cost of these
types of furnaces.
TABLE XI
ESTIMATED CAPITAL COST OF GRANULAR ACTIVATED CARBON REACTIVATION FURNACES
Furnace Type Capacity Estimated Total Cost
Multiple-Hearth 5,000 Ibs/hr $4.2 million
Infra-red* 5,000 Ibs/hr $0.8 million
Rotary Kiln 5,000 Ibs/hr See note
Fluidized Bed* 5,000 Ibs/hr $1.2million
*Because furnaces of this size have not yet been manufactured these
estimates are very preliminary.
Note: Insufficient information is available to estimate a cost for this
type of furnace.
Another analysis has been to determine the cost of constructing and
operating granular activated carbon contactors as a unit process following
filtration. The assumptions used in this calculation are given in Table XII.
-------
- 35 -
TABLE XII
ASSUMPTIONS FOR POST-ADSORPTION CONTACTORS
Item
Number of contactors
2
Hydraulic loading (gal/min/ft )
Diameter contactors (ft)
Depth of contactors (ft)
3
Volume of granular activated carbon (ft )
Apparent Contact time (min.)
Reactivation frequency (months) at 70% of
capacity
Activated carbon attrition loss (% per
reactivation)
Financial Assiinintinn - See Middle Values Table
10 mgd
8
5.4
12
20
18,096
9
2.9
5
IX
100 mgd
28
5.5
20
20
175,930
9
2.9
5
Although more capital intensive, calculations indicate that using
granular activated carbon in a post-filtration mode is only slightly more
expensive than replacing the granular media in the existing filter boxes with
granular activated carbon, see Table XIII. The reason the overall costs are
so close is that post-filtration contactors can be constructed of any size
and shape and could be made deeper. The longer contact time would permit longer
periods between reactivations, thus reducing the operating cost. This
reduction in cost, plus the reduction in operating cost that would accrue
from lowering the percentage of activated carbon loss per reactivation cycle
because the handling of the granular activated carbon from these contactors
would be facilitated, overcomes the increased cost of constructing new facilities.
-------
- 36 -
COMPARISON OF CAPITAL AND OPERATING COSTS FOR EQUIVALENT ADSORPTION/FILTRATION
AND POST-FILTRATION ADSORPTION SYSTEMS FOR 10 AND 100 MGD AT 70% LOAD FACTOR
TABLE XIII
All Costs in Cents/1000 Gallons
Operating and
Capital Costs Maintenance Cost Total Cost
10 100 10 100 10 100
mgd mgd mgd mgd mgd mgd
Adsorption/Filtration 3.5 1.5 8.2 4.5 11.7 6.0
Post-Filtration
Adsorption 8.2 4.3 5.4 2.4 13.5 6.7
*These costs will vary at different locations so should be considered approximate,
All of these unit costs should be compared to typical water treatment
costs. EPA has nearly completed a study that shows the average cost (not
price which is reflected in the consumer's water bill) of drinking water in
eleven major utilities to be about 43 cents per 1000 gallons. Twelve percent of
these costs are for treatment, with the balance for acquisition of water,
pumping, salaries of employees, administration, amortization of distribution
systems, and other nontreatment costs. Using the previously discussed value
of 73,000 gallons of water usage per year for a household of four at 43 cents
per 1000 gallons the annual cost for water would be about $31, about $4 of which
would be for treatment. Because of other charges, many annual water bills might
be higher than this value.
The use of granular activated carbon adsorption as an additional treatment
process for controlling organic contaminants would add from $4/year to $9/year
to the water bill for a household of four for treatment plants in the 10 to 150
mgd range, using the data in Table X. Of course, in locations where the
organic content of the water is greatly different from that used in the studies
upon which these costs are based or the factors are greatly different from the
-------
- 37 -
middle values in Table IX, the cost for treatment using granular activated
carbon will be different.
Capital Investment
As can be seen from Table XIII, a separate contactor system is much
more capital intensive than replacing activated carbon in the filter shell.
Exploring these investment costs in more detail is worthwhile. To this
point in the discussion all costs have been presented as unit costs, which
presents one view of costing. In building these systems, however, utilities
must raise a considerable amount of initial capital. Table XIV summarizes
for 1, 10 and 100 mgd plants the principal required and the total payback
cost for both the option of replacing sand with activated carbon in the
filter shell and using separate activated carbon contactors.
TABLE XIV
INVESTMENT COSTS FOR GRANULAR ACTIVATED CARBON TREATMENT AS A REPLACEMENT FOR SAND
AND IN A SEPARATE CONTACTOR SYSTEM (All Costs in Thousands of Dollars)*
Design Capacity 1 mgd 1Q mgd IQQ mgd
Filter Filter Filter
Item Shell Contactor Shell Contactor Shell Contactor
Principal 610 857 994 2,328 4,247 12,207
Total Cost
(Principal & Interest) 1,100 1,543 1,788 4,190 7,665 21,973
*Based on 7% interest and 20yr. amortization period.
As can be seen from Table XIV a separate contactor system requires a much
greater capital investment when compared to replacing sand in the filter shell.
Estimates of capital investments for larger treatment plants will show economics
of scale that are greater for the systems replacing the sand in an existing
filter shell than for post-filtration adsorbers.
-------
- 38 -
Recommendations for Specific Treatment Changes
The "potential for the formation of trihalomethanes can be reduced somewhat
by coagulation and sedimentation, adding chlorine dioxide, large doses of
powdered activated carbon or ozone and can be eliminated by the use of
adsorption on fresh granular activated carbon. Water utilities employing
unit processes for turbidity removal, color removal, or both, or practicing
precipitative softening should attempt to improve these unit processes by
altering coagulants, dosages, pH, use of polyelectrolytes and so forth to
maximize the removal of chloroform potential, as measured by the procedure
described in Reference 4 in the Appendix, and then chlorinate after these
unit processes. With precipitative softening plants, chlorination should be
after recarbonation, if practiced, so that chlorine is not added to water
2
with a high pH. Care must be taken to insure microbiological safety of
the distributed water if alternations in the point of application of chlorine
is practiced. Also any alteration of pH of the finished water may cause
corrosion problems, so this potential should be evaluated prior to any
change.
Little information is available on what type of organic removal would
be expected in a plant using ion exchange resin or natural zeolite for
softening. In the National Organics Reconnaissance Survey only two utilities
employing zeolite ion exchange for softening were included. Both had low
concentrations of chloroform in the finished water. In one case, this
was not unexpected because the raw water NPTOC concentration was low,
but at the other utility, this was not the case, and the low chloroform
concentration in the finished water was somewhat surprising. Further
research is necessary to evaluate synthetic ion exchange resins and zeolites.
-------
-39 -
Data show that powdered activated carbon is not particularly
efficient, either for removing chloroform or chloroform precursors.
Nevertheless, a water treatment plant that has the capability of feeding
powdered activated carbon should consider increasing the dose and possibly
adding it at several points through the treatment plant to increase its
effectiveness. This may not be the complete solution to a given problem,
but if the dose could be raised to a very high level, powdered activated
carbon should help to control chloroform concentrations. This may, however,
create a sludge disposal problem. Prior to using this technique on a full-
scale, jar tests should be run, with the chloroform potential determined
according to the technique described in Reference 4 for various doses of
powdered activated carbon.
Some utilities currently have equipment for generating chlorine
dioxide so that it can be used as an odor control process. In these
circumstances, chlorine dioxide reacts with phenol and prevents the
formation of disagreeable chlorophenolic tastes and odors upon chlorination.
Utilities with such equipment should consider using chlorine dioxide
for both odor control and disinfection, if they desire to reduce the
chloroform concentration in their finished water. Even if the chlorine
dioxide is generated such that it is contaminated with chlorine, research
data show that the chlorine dioxide will have an effect on the chloroform
precursors and therefore lower quantities of chloroform should result.
Utilities comtemplating this practice should make some tests on their
water to determine the chlorine dioxide demand of the water to produce
an adequate residual for good microbiological kill. During these tests,
the quantities of chlorite formed should be measured to assure that it is
not present in quantities near 1 mg/£.
-------
-40-
Although potassium permanganate has yet to be evaluated for chloroform
precursor removal, water utilities employing this chemical for taste and
odor control should run some jar tests and evaluate the effect of various dosages
of potassium permanganate on the chloroform formation potential as
determined by the technique described in Reference 4. This should allow
these utilities to evaluate whether or not potassium permanganate will be
effective for the control of chloroform concentration.
Some utilities currently have granular activated carbon adsorbers
in the treatment plant for the purpose of controlling organics that previously
had been creating consumer taste and odor complaints. Analysis of several
of these water utilities in the National Organics Reconnaissance Survey and
Region V Survey indicated that these adsorbers were not controlling the
chloroform concentrations in the finished waters except for a short
initial period of operation. To improve performance, these utilities
should reactivate their granular activated carbon more frequently to obtain
its maximum benefit. Making this change in operation should allow these
utilities to produce the water nearly free of trihalomethane and low in
general organic content.
Removal of Chloroform
Four techniques were studied to remove chloroform from water after
it has formed - the use of adsorption on powdered- and granular activated
carbon, aeration and ozonation. Unrealistically high doses of powdered
activated carbon and ozone were requied to effect substantial, although not
complete, removal of chloroform and other trihalomethanes. Preliminary
cost estimates indicate that both of these processes would be prohibitively
expensive for the removals obtained.
-------
-41 -
Both aeration and adsorption of granular activated carbon were
effective for chlorofrom removal under certain circumstances. Granular activated
carbon treatment is, however, effective beyond just removing chloroform.
Because this was discussed above it will not be discussed
again. To determine the approximate cost of a diffused-air aeration system
for the removal of about 90 percent of the chloroform formed during water
treatment, an aeration basin constructed following final disinfection with
a 20-minute detention time and an air-to-water ratio of 30 to 1 was assumed.
Table XIV shows the cost of this form of treatment is reasonable, from about
$6.50/yr.to about $16/year additional cost for water for a household of four,
assuming a household of four uses 73,000 gallons of water annually.
TABLE XIV
ESTIMATED COST OF AERATION*, 20-MINUTE DETENTION TIME, 30 TO 1 AIR-TO-WATER RATIO
Design Capacity, mgd 1 10 100 150
Average Daily Flow, mgd 0.7 7 70 105
Total Unit Cost, «
-------
-42 -
the chloroform concentration had increased to slightly over 120 yg/£.
Further, aeration might raise the dissolved oxygen level in the water,
thereby aggrevating corrosion problems in the distribution system. Therefore,
a treatment technique for the control of precursor concentrations or
changing to an alternate disinfectant is better than merely removing
chloroform from finished water by aeration.
Recommendations for Specific Treatment Changes
Aeration is frequently practiced for the removal of such reduced
materials as ferrous iron, manganeous manganese, and hydrogen sulfide.
This aeration step is frequently before chlorination in the treatment
process. Therefore, although trihalomethanes can be removed from water
during aeration, this would not be the case in the situation described
above. If treatment plants practicing aeration for taste and odor
control have the aeration process located following chlorination, some
chloroform will be lost to the atmosphere. Data have shown that higher than
usual air to water ratios will be necessary to achieve good removal of chloroform,
therefore the utility that has the capability of varying the intensity of
aeration should maximize it to help improve removal of chloroform.
-------
- 43 -
MONITORING
A utility attempting to control the concentration of chloroform and
other trihalomethanes should consider a monitoring program along with any
changes in treatment. This monitoring program should be developed so
that concentrations of chloroform and other trihalomethanes can be
determined, thereby evaluating the effectiveness of the changes in treatment
that were made to control chloroform. In all cases the chloroform potential
of the water prior to any treatment should be determined by the technique
described in Reference 4 in the Appendix.
The analytic technique for chloroform, involving purging the chloroform
from a sample with an inert gas prior to introduction into a gas chromatograph,
11 12
with a halogen specific detector has been described in the literature. '
Although this determination is not difficult, qualified technicians
are required to produce precise and accurate results. Any gas chromatographic
procedure requires some skill and knowledge to perform properly and this is
no exception. If such qualified technicians are not available on a
utility's staff, samples may have to be shipped to another laboratory where
they can be analyzed. This is less desirable than having the analytic
capability in the utility, but is an acceptable solution to the problem of
monitoring.
If a utility is making these analyses itself, samples should be collected
at various stages of the water treatment process to determine where and what
quantity of chloroform is being formed. Further, because the concentration
of trihalomethanes may change from the treatment plant to the point of use,
samples should be collected at various places in the distribution system to
determine the increase, if any, in chloroform concentration as the water
moves from the treatment plant to the consumer. Samples should be
-------
dechlorinated with sodium thiosulfate to avoid changes prior to anaiysis.
Studies have demonstrated that precursor concentrations change
during different seasons in some raw waters, thereby changing the ultimate
chloroform concentration. Therefore, tap water samples should be taken
frequently enough to allow a utility to be aware of changes in raw water
such that they know the eventual chloroform concentration reaching
the consumer at any time.
The initial cost of the equipment to make chloroform analyses is in the
range of $7,000 - $8,000. Once this initial purchase is made, the analysis
cost will mostly be the cost of the analyst's time. A typical analysis for
chloroform requires from 45 minutes to 1 hour with about 6 samples being
able to be processed in a typical working day, considering the time for
standardization of the gas chromatographic detector response.
A water utility practicing organic removal may also wish to monitor
the effectiveness of those unit processes. This is usually done by measuring
the total organic carbon concentration before and after such a unit process.
At the present time, equipment is available that will measure the portion of
TOG remaining in a sample that has been purged of carbon dioxide under
acidic conditions. The detection limit of this equipment is approximately
0.1 mg/Jl, with a precision of about - 0.1 mg/£. The cost of such equipment
is about $10,000 and also requires the same type of qualified technician
needed to make a chloroform analysis for reliable operation.
-------
-45 -
Another analytic determination currently under development would measure
total organic halogens (TOH). If the development of this test is successful,
it could be used advantageously to determine the concentration of all of the
halogen-containing organics (non-natural) in a single analysis. A variation
of this procedure that measures most of the total organic chlorine (TOC1) is
used in Europe to monitor the performance of granular activated carbon beds.
Samples are collected from within the bed at a point about six inches above
the bottom. When the TOC1 concentration begins to rise in these samples,
breakthrough is near and the granular activated carbon is reactivated.
A similar approach could be taken using NPTOC measurements.
-------
-46 -
ACKNOWLEDGMENTS
The authors wishes to thank those who reviewed this Guide for all of
the helpful suggestions. These reviewers were: J.K. Carswell, R.M. Clark,
J. DeMarco, O.T. Love, Jr., A.A. Stevens, Gordon G. Robeck, J. Cotruvo,
G. Goad and J. Hoffbuhr. The author also expresses his appreciation to
Ms. Maura M. Lilly who typed the five drafts and the final version so
promptly.
-------
-47 -
REFERENCES
Machisko, J.
1. Clark, R.M., Guttman, D.,/and Crawford, J., "Cost Calculations of Water
Treatment Unit Processes," Water Supply Research Division, Municipal
Environmental Research Laboratory, U.S. Environmental Protection Agency,
Cincinnati, Ohio (March 1976), Appendix 1.
2. Stevens, A.A., Slocum, C.J., Seeger, D.R. and Robeck, G.G., "Chlorination
of Organics in Drinking Water," Proceedings of Conference on the
Environmental Impact of Water Chlorination, Oak Ridge, Tennessee,
October 22-24, 1975, and submitted to the Journal of the American Water
Works Association for publication, Appendix 2.
3. Love, O.T., Jr., Carswell, J.K., Stevens, A.A., Miltner, R.J. and Symons, J.M.,
"Treatment for the Prevention or Removal of Chlorinated Organics in
Drinking Water," to be submitted to the Journal of the American Water
Works Association, Appendix 3.
4. Stevens, A.A., "Determination of Chloroform Formation Potential in Water,"
To be submitted to the Journal of the American Water Works Association,
Appendix 4.
5. Rook, J.J., "Production of Potable Water from a Highly Polluted River,"
Water Treatment and Examination, 21, Part 3, 259-274 (1972).
6. Bellar, T.A., Lichtenberg, J.J. and Kroner, R.C., "The Occurrence of
Organohalides in Chlorinated Drinking Water," Jour. AWWA, 66:12:703,
(December 1974).
7. Rook, J.J., "Formation of Haloforms During Chlorination of Natural
Water," Water Treatment Exam., _2!3:2:234 (1974).
8. Symons, J.M., Bellar, T.A., Carswell, J.K., DeMarco, J., Kropp, K.L.,
Robeck, G.G., Seeger, D.R., Slocum, C.J., Smith, B.L., and Stevens, A.A.,
National Organics Reconnaissance Survey for Halogenated Organics,
American Water Works Association, 67, 634-647 (Nov. 1975).
-------
- 48 -
9. Region V Joint Federal/State Survey of Organics and Inorganics in
Selected Drinking Water Supplies, U.S. Environmental Protection
Agency, Chicago, Illinois, 60604, June 1975, Draft.
10. Report on the Carcinogenesis Bioassay of Chloroform, Carcinogen
Bioassay and Program Resources Branch, Carcinogenesis Program, Divisior
of Cancer Cause and Prevention, National Cancer Institute, Bethesda,
Maryland.
11. Bellar, T.A. and Lichtenberg, J.J., "The Determination of Volatile
Organic Compounds at the yg/£ Level in Water by Gas Chromatography,"
Journal of the American Water Works Association, 66, 739 (Dec. 1974).
12. Stevens, A.A. and Symons, J.M., "Analytical Considerations for
Halogenated Organic Removal Studies," Proceedings American Water
Works Association Water Quality Technology Conference, pp. XXVI-1 -
XXVI-6 (December 1974), American Water Works Association, Denver,
Colorado (1975).
13. Rook, J.J., "Haloforms in Drinking Water, Journal of the American
Water Works Association, 68, 168-172 (March 1976).
14. Samdal, J.E., "Water Treatment and Examination in Norway, Water
Treatment and Examination, 21, 309-314 (1972).
15. "Clinical Toxicology of Commercial Products." Gleason, Geosselin, Hodge
and Smith, 3rd Edition (1969).
16. Kruse, C.W., Oliveri, V.P. and Kawata, K., "The Enhancement of Viral
Inactivation by Halogens," Water and Sewage Works, 118, 187-193 (June 1971)
17. Process Design Manual for Granular Activated Carbon Adsorption,
Technology Transfer, U.S. Environmental Protection Agency, October 1971.
-------
APPENDIX 1
Clark, R.M., Guttman, D., Machisko, J., and Crawford, J., "Cost Calculations
of Water Treatment Unit Processes," Water Supply Research Division, Municipal
Environmental Research Laboratory, U.S. Environmental Protection Agency,
Cincinnati, Ohio (March 1976).
-------
THE COST OF REMOVING CHLOROFORM AND OTHER TRIHALOMETHANES
FROM DRINKING WATER SUPPLIES
by
Robert M. Clark
Daniel L. Guttman
John L. Crawford
John A. Machisko
Water Supply Research Division
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem .solu-
tion and it involves defining the problem, measuring its impact, and searching
for improved technology and systems for the prevention, treatment, and manage-
ment of wastewater and solid and hazardous waste pollutant discharges from
municipal and community sources, for the preservation and treatment of public
drinking water supplies, and to minimize the adverse economic, social, health,
and aesthetic effects of pollution. This publication is one of the products
of that research; a most vital communications link between the researcher and
the user community.
Trihalomethanes in general, and chloroform - a known carcinogen - in
particular, are found in drinking water as a direct consequence of the practice
of chlorination, a long established public health practice for the disinfection
of drinking water. EPA would like to minimize the drinking water consumer's
exposure to trihalomethanes at reasonable cost. This report is devoted to
the presentation of the results from a research study which examines the
costs of the various treatment technologies suited to the removal and control
of trihalomethanes in drinking water.
Francis T. Mayo
Director
Municipal Environmental Research Laboratory
iii
-------
PREFACE
The Safe Drinking Water Act of 1974 has intensified the public awareness
and interest in the quality of drinking water as delivered to the consumer's
tap. The Act establishes a set of enforceable, health-related regulations
and a set of non-enforceable esthetics-related guidelines for drinking water.
For each health-related standard the Act establishes an associated Maximum
Contaminant Level (MCL) that must not be exceeded. The Act also contains
provisions for the EPA Administrator to take various courses of action when
a contaminant is detected for which no MCLs have been established. Trihalo-
methanes in general, and chloroform, recently determined to be a carcinogen
in drinking water, in particular, are examples of such contaminants.
Trihalomethanes are found in drinking water as a direct consequence of
the practice of chlorination, a long established public health practice for
the disinfection of drinking water supplies. Recent research has demon-
strated that the concentration of chloroform and related compounds is
generally higher in finished than in raw water, leading to the conclusion
that they are being produced during the chlorination process.
Acting on these findings, Russell Train, Administrator of the U. S.
Environmental Protection Agency, directed that EPA work with cities and
states to evaluate certain modifications to current treatment practices that
can reduce the formation of chloroform during the water treatment process,
without lessening the effectiveness of disinfection. Part of this effort
has been the preparation of a document, entitled "Interim Treatment Guide
for the Control of Chloroform and Other Trihalomethanes in Drinking Water."
The "Guide" has been prepared in an attempt to present EPA's knowledge con-
cerning the removal and control of chloroform and other trihalomethanes in
drinking water. It covers such items as: changing the point of chlorine
application to reduce chloroform concentrations; the use of alternative
disinfectants, such as ozone or chlorine dioxide; and the use of granular
activated carbon as a medium for the adsorption of organic compounds.
Appendix I of the "Guide" presents cost information with respect to the
use of granular activated carbon, ozonation, aeration, and chlorine dioxide
for trihalomethane removal. This report was originally Appendix I of the
"Interim Treatment Guide for the Control of Chloroform and Other Trihalo-
methanes," and provides an in-depth examination of the costs related to the
above-mentioned techniques.
iv
-------
ABSTRACT
This research effort was conducted to provide an in-depth examination
of the costs associated with the use of activated carbon, ozonation, aeration,
and chlorine dioxide for removal of trihalomethanes. It is intended as
support information for the "Interim Treatment Guide for the Control of Chloro-
form and Other Trihalomethanes."
The costs presented in this report are intended for the development of
planning estimates only and not for the preparation of bid documents or
detailed cost estimates. Exact capital and operating costs are highly vari-
able from location to location within the United States, even for plants of
the same size and design. These costs are presented in such a way as to
enable the planner to make adjustments to the reported costs when local
information is available. Standardized levels for a selected set of design
parameters are assumed and sensitivity analysis is performed for the majority
of the parameters.
Because chlorine is associated with the formation of trihalomethanes,
several technological alternatives which may be used in lieu of or in
combination with chlorination are examined. Costs are presented for the
chlorination process itself. Costs are also calculated for ozonation,
chlorine dioxide, aeration, and granular activated carbon. An in-depth
analysis of the costs associated with granular activated carbon systems is
presented. This analysis includes costs both with and without separate
contactor systems and an examination of the possible cost savings associated
with regional regeneration.
-------
CONTENTS
FOREWORD ill
PREFACE iv
ABSTRACT v
FIGURES vii
TABLES xi
METRIC CONVERSION TABLE xiii
ACKNOWLEDGMENTS xiv
INTRODUCTION 1
COST DETERMINATION 2
Basis of Cost Estimates 3
COST OF CHLORINATION 4
COST OF CHLORINE DIOXIDE 8
COST OF OZONATION 19
Cost of Ozone from Air 19
Cost of Ozone from Oxygen 26
COST OF AERATION 26
COST OF GRANULAR ACTIVATED CARBON 26
The Cost of GAG as Filter Media Replacement 48
Additive Modifications 69
Multiplicative Modifications 74
Regional Reactivation 74
Separate Contactor System 80
Capital Investment 90
Labor Costs for GAG Systems 90
SUMMARY AND CONCLUSIONS 92
REFERENCES 95
vx
-------
FIGURES
Number Page
1 Total Unit Cost for Chlorination Versus Plant Size 6
2 Capital and 0 & M Costs for Chlorination Versus Plant Size . 7
3 0 & M Cost for Chlorination Systems Versus Cost of Chlorine 9
4 0 & M Cost for Chlorination Systems Versus Direct Hourly
Wage Rate 10
5 0 & M Cost for Chlorination Systems Versus Wholesale Price
Index 11
6 Amortized Capital Cost for Chlorination Systems Versus
Chlorine Contact Time 12
7 Amortized Capital Cost for Chlorination Systems Versus
Interest Rate 13
8 Amortized Capital Cost for Chlorination Systems Versus
Construction Cost Index 14
9 Amortized Capital Cost for Chlorination Systems Versus
Amortization Period 15
10 0 & M Cost for Chlorination Systems Versus Chlorine Dosage. . 16
11 Amortized Capital Cost for Chlorination Systems Versus
Chlorine Dosage 17
12 Total Unit Cost for Chlorine Dioxide Versus Plant Size ... 21
13 0 & M Cost for Chlorine Dioxide System Versus Plant Size . . 22
14 Total Unit Cost for Ozonation (Air) Versus Plant Size .... 24
15 Amortized Capital and 0 & M Costs for Ozonation (Air)
Versus Plant Size 25
VII
-------
FIGURES (Cont.)
Figure Page
16 0 & M Cost for Ozonation (Air) Versus Direct Hourly
Wage Rate 27
17 0 & M Cost for Ozonation (Air) Versus Wholesale Price
Index 28
18 0 & M Cost for Ozonation (Air) Versus Electric Power Cost . . 29
19 Amortized Capital Cost for Ozonation (Air) Versus Ozone
Contact Time 30
20 Amortized Capital Cost for Ozonation (Air) Versus Construc-
tion Cost Index 31
21 Amortized Capital Cost for Ozonation (Air) Versus Interest
Rate 32
22 Amortized Capital Cost for Ozonation (Air) Versus
Amortization Period 33
23 0 & M Cost for Ozonation (Air) Versus Ozone Dose 34
24 Amortized Capital Cost for Ozonation (Air) Versus Ozone
Dose 35
25 Total Unit Cost of Ozonation (Oxygen) Versus Plant Size . . 37
26 Amortized Capital and 0 & M Costs for Ozonation (Oxygen)
Versus Plant Size 38
27 0 & M Cost for Ozonation (Oxygen) Versus Liquid Oxygen Cost . 39
28 Amortized Capital Cost for Ozonation (Oxygen) Versus
Liquid Oxygen Cost 40
29 Construction Cost for an Aeration Basin Versus
Volume of Basin 43
30 Annual 0 & M Costs for Air Supply Versus Standard Cubic Feet
per Minute Throughput 44
31 Construction Cost for Air Supply Versus Standard
Cubic Feet per Minute Throughput 45
Vlll
-------
FIGURES (Cont.)
Figure Page
32 Total Unit Cost Versus Plant Size 50
33 Amortized Capital and 0 & M Costs Versus Plant Size 51
34 Total Unit Cost for a 100 mgd Plant Versus Time Between
Reactivations in Months 52
35 Total Unit Cost for a 100 mgd Plant Versus the Product
of Time Between Reactivations in Months and Capacity Factor 53
36 0 & M Costs Versus Direct Hourly Wage Rate 55
37 0 & M Cost Versus Carbon Loss per Reactivation Cycle 56
38 0 & M Cost Versus Fuel Cost 57
39 0 & M Cost Versus Wholesale Price Index 58
40 0 & M Cost Versus Electrical Power Cost 59
41 Amortized Capital Cost Versus Construction Cost Index .... 60
42 Amortized Capital Cost Versus Amortization Interest Rate . . 61
43 Amortized Capital Cost Versus Amortization Period 62
44 0 & M Cost Versus Carbon Cost 63
45 Amortized Capital Cost Versus Carbon Cost 64
46 0 & M Cost Versus Reactivation Frequency 65
47 Amortized Capital Cost Versus Reactivation Frequency .... 66
48 0 & M Cost Versus Interaction Between Reactivation Frequency
and Capacity Factor 67
49 Amortized Capital Cost Versus Interaction Between Reactivation
Frequency and Capacity Factor 68
50 Percent Change in Plant Cost Versus Carbon Loss 70
ix
-------
FIGURES (Cont.)
Number Page
51 Total Unit Costs Versus Plant Size 75
52 Construction Cost for Carbon Reactivation System Versus
Reactivation Rate 78
53 0 & M Cost for Carbon Reactivation System Versus
Reactivation Rate 79
54 Cost of Transporting Carbon from a 1 mgd Plant to Regional
Reactivation Site Versus Distance in Miles 83
55 Cost of Transporting Carbon from a 5 mgd Plant to Regional
Reactivation Site Versus Distance in Miles 84
56 Cost of Transporting Carbon from a 10 mgd Plant to Regional
Reactivation Site Versus Distance in Miles 85
57 The Sensitivity of Reactivation Costs to Transportation
Cost Variations 86
58 Comparison of Costs Between Contactor System and Media
Replacement Versus Plant Capacity 88
x
-------
TABLES
Number Page
1 Design Parameters for the Cost of Chlorination 5
2 Chlorine Cost Assuming Standardized Design Levels 18
3 Chlorine Dioxide Costs Assuming Standardized Design Levels . . 20
4 Design Parameters for Ozonation 23
5 Ozone (Air) Costs Assuming Standardized Design Levels .... 36
6 Ozone (Oxygen) Costs Assuming Standardized Design Levels . . 41
7 Design Parameters for Aeration 42
8 Aeration Costs Assuming Standardized Design Levels 46
9 Design Parameters for Granular Activated Carbon 49
10 Design Parameters Affecting 0 & M Costs (100 mgd) 71
11 Design Parameters Affecting Capital Costs (100 mgd) 72
12 New Effect for Design Parameters at High and Low Levels
(100 mgd) 73
13 Carbon Costs and Reactivation for Regional Reactivation Systems 77
14 Amortized Capital and Operating Costs for Off-Site
Reactivation Systems 81
15 Reactivation Systems Cost for an Individual Plant 82
16 Assumptions for Separate Contactor Systems 87
17 Amortized Capital and 0 & M Costs for Contactor Versus Filter
Media Replacement - c/1000 gal 89
xi
-------
TABLES (Cont.)
Number Page
18 Investment Costs for Granular Activated Carbon Treatment:
Replacement of Media and Separate Contactors (Thousands
of Dollars) 91
19 Estimated Construction Cost of Granular Activated Carbon
Reactivation Furnaces 90
20 Labor Costs for 1, 10, and 100 mgd GAG Systems Reactivating
On-Site (Filter Shell Replacement) 93
21 Comparison Among Systems (c/1000 gal) 94
xii
-------
METRIC CONVERSION TABLE
English Units
1 foot
1 mile
1 square mile
1 million gallons
1 $/million gallons
1 c/1000 gallons
Metric Equivalents
0.305 meters
1.61 kilometers
2.59 square kilometers
3.79 thousand cubic meters
0.26 $/thousand cubic meters
0.26 c/cubic meter
xiii
-------
ACKNOWLEDGMENTS
The authors gratefully acknowledge the assistance of the
following individuals in the preparation of this report:
Mr. Gordon G. Robeck, Dr. James M. Symons, Dr. 0. Thomas Love,
and Mr. J. Keith Carswell of the Water Supply Research Division,
and Mr. Richard Eilers of the Wastewater Research Division.
xiv
-------
THE COST OF REMOVING CHLOROFORM AND OTHER TRIHALOMETHANES
FROM DRINKING WATER SUPPLIES
BY
* + +
Robert M. Clark , Daniel L. Guttman , John L. Crawford ,
and John A. Machisko
INTRODUCTION
The Safe Drinking Water Act of 1974 will change the way water is
handled before it is distributed to the consumer. ^ The Act contains two
types of provisions for drinking water as delivered to the consumer's tap:
a set of enforceable regulations that are health-related, and a set of
non-enforceable guidelines that are related to the esthetics of drinking
water. Each health-related standard has an associated Maximum Contaminant
Level (MCL) that must not be exceeded. The Act also contains provisions
for the Administrator, U. S. Environmental Protection Agency (EPA) to take
various courses of action when a contaminant is detected for which no MCLs
have been established.1 Trihalomethanes in general and chloroform in partic-
ular, recently determined to be a carcinogen in drinking water, are examples
of such contaminants.
Trihalomethanes are found in drinking water as a direct consequence of
the practice of chlorination, a long established public health practice for
the disinfection of drinking water. It is probable that chlorine has been
reacting with certain organic materials to produce chloroform and related
organic byproducts since chlorination was initiated. These compounds in drink-
ing water escaped detection due to their low concentrations and because of
their low boiling points, which allowed them to be lost during procedures used
for performing typical organic analyses in water. ^-^
Systems Analyst, Water Supply Research Division, Municipal Environmental
Research Laboratory, U. S. Environmental Protection Agency, Cincinnati,
Ohio 45268.
Research Assistants, Water Supply Research Division, Municipal Environ-
mental Research Laboratory, U. S. Environmental Protection Agency,
Cincinnati, Ohio 45268
-------
Recently, investigators have developed new, more sensitive analytical
procedures which allow for more precise measurement of trihalomethanes. The
newly developed procedures have been used to demonstrate that the concentra-
tion of chloroform and related compounds is generally higher in finished than
in raw water, leading to the conclusion that they are being produced during
the chlorination process.
Acting on these findings, Russell Train, Administration of the U. S.
Environmental Protection Agency, on March 29, 1976, released a statement that
said in part, "... The recent test results and the fact that chloroform is
prevalent in the environment have convinced me that the prudent course of
action at this time is to minimize exposure to this chemical wherever it is
feasible to do so." He also said, "EPA will work with cities and states to
evaluate certain modifications to current treatment practices that can reduce
the formation of chloroform during the water treatment process, without
lessening the effectiveness of disinfection. EPA research has shown that
changes in chlorination procedures practiced by some water systems can result
in reduction in the levels of chloroform produced. EPA plans to share these
initial findings on chloroform reduction with the states and some cities
encountering high chloroform levels, in an effort to reduce human exposure as
quickly as possible. This will also allow EPA to gain added information to
support the development of national regulations to limit chloroform levels in
water supplies."
An interim guide, entitled "Interim Treatment Guide for the Control of
Chloroform and Other Trihalomethanes in Drinking Water," has been prepared
in an attempt to present EPA's knowledge concerning the removal and control
of chloroform and trihalomethanes in drinking water. The "Guide" covers such
items as: changing the point of chlorine application to reduce chloroform
concentrations; the use of alternative disinfectants, such as ozone or
chlorine dioxide; and the use of granular activated carbon as a medium for
the adsorption of organic compounds.
The "Guide" summarizes EPA's current knowledge and recent research
results related to the removal and control of chloroform and other trihalo-
methanes. It also presents cost information with respect to the use of
granular activated carbon, ozonation, aeration, and chlorine dioxide for
trihalomethane removal. This report has been developed as a support document
for the "Guide" and provides an in-depth examination of the costs related to
the above-mentioned techniques. -^
COST DETERMINATION
The costs presented in this report are intended for the development of
planning estimates only and not for the preparation of bid documents or
detailed cost estimates. Exact capital and operating costs are highly
variable from location to location within the United States, even for plants
of the same size and design. Variables — such as local costs of land,
materials, and labor; state or regional differences in building codes; and
existing facilities suitable for modification — may accentuate the
differences in treatment costs for similar plants to reduce chloroform and
other trihalomethanes in drinking water.
-------
These cost data are presented in such a way as to enable the planner to
make adjustments to the reported costs when local information is available.
For example, operation and maintenance costs can be reduced if the delivered
cost of chemicals is less than the costs upon which the estimates are "based.
Costs indices are used to provide a baseline for projecting costs and for
estimating escalation due to inflation. The indices used in this report are
national indices, but other indices are often available for major U. S.
cities or on a regional basis and may be substituted if desired.
Basis of Cost Estimates
The cost indices used in this report are:
a. EPA's Sewer and Sewage Treatment Plant Construction Cost Index:
This index was used because most of the basic information utilized
in the report was obtained from the Systems and Economic Evaluation
Section of EPA's Wastewater Research Division.9»3 For example,
computations for granular activated carbon were performed using
a computer program developed by the Systems and Economic Evalua-
tion Section, but operational modifications were assumed in the
analysis to reflect conditions unique to water supply.H The
index should reflect similar costs for water treatment plant con-
struction and for January 1976 is 256.7.
2
b. Wholesale Price Index: The Wholesale Price Index (WPI) is the
oldest continuous statistical series published by the Bureau of
Labor Statistics (BLS). It is a measure of the price changes for
goods sold in primary markets in the United States. "Wholesale,"
as used in the title of the index, refers to sales in large quanti-
ties, not prices received by wholesalers, jobbers, or distributors,
and for January 1976 is 175.1.
c. Bureau of Labor Statistics Labor Cost Index (Direct Hourly Wage
Rate):5 The index used in this report is, for personnel in
Standard Industrial Category (SIC), 494.7 for Water, Steam and
Sanitary Systems Non-Supervisory Workers. The base BLS Labor Cost
Index for February 1976 is 5.19.
Costs reported as Capital Costs include:
a. construction for site preparation,
b. plant construction,
c. legal, fiscal, and administrative services,
d. interest during construction, and
e. start-up costs.
-------
Costs reported as Operating and Maintenance Costs include:
a. chemical costs,
b. labor costs, and
c. operation and maintenance costs, such as utilities, annual replace-
ment of expendable items, etc.
Because the formation of trihalomethanes is associated with chlorination,
several options will be considered in lieu of chlorination, all more expensive
than the basic chlorination process itself. The benefits as well as the
problems associated with changing to some alternative must be weighed when
choosing another form of disinfection. For example, ozonation, while not
creating chloroform, also does not provide any kind of residual disinfectant
in the distribution system. Moreover, it is conceivable that the byproducts
resulting from disinfecting with chlorine dioxide or ozone might be more harm-
ful than chloroform itself. Granular activated carbon (GAC) is not a direct
alternative to chlorination, but provides a means of removing organics from
water supplies. The cost of the processes to be considered - chlorine dioxide,
ozonation, and granular activated carbon - must be weighed against the basic
cost of chlorine disinfection. Initially, the costs associated with chlori-
nation will be presented. Next, the directly competitive disinfection
alternatives of chlorine dioxide and ozonation will be discussed. The cost
of aeration as a means of removing chloroform and other trihalomethanes once
they have been formed and, finally, a cost analysis for granular activated
carbon as a general means of removing organics will be presented.
COST OF CHLORINATION
Chlorine was first used as a disinfectant for municipal water supplies
in the United States in 1908 to disinfect continuously the water supply of
Jersey City, New Jersey.^ In many water supplies, chlorination is often
the only treatment process used. When other treatment methods are used,
disinfecting chlorine may be added to the raw water (prechlorination), the
partially treated water, or the finished water (postchlorination). Where the
distribution system contains open reservoirs, the treated water may be re-
chlorinated in the distribution system.
In this analysis, a number of variables have been evaluated as to their
effect on the cost of chlorination. Baseline or standardized design values
for a set of design parameters were assumed and the cost of chlorination,
including chlorine feeding equipment and contact chambers for 1, 5, 10, 100,
and 150 million gallons per day (mgd) systems, was calculated. The para-
meters and their associated "standardized" levels are shown in Table 1.
Figure 1 shows the effect of economies of scale, by displaying the total
amortized unit cost of a chlorination system in C/1000 gal versus plant
capacity, based on the values in Table 1. It can be seen that cost varies
from 3.6C/1000 gal at. 1 mgd to 0.6C/1000 gal at 150 mgd. Figure 2 separates
the total unit costs shown in Figure 1 into unit Amortized Capital and 0 & M
costs.
-------
TABLE 1. DESIGN PARAMETERS FOR THE COST OF CHLORINATION
Design Parameter (Variable)
Chlorine Dose
Chlorine Contact Time
Cost of Chlorine
Construction Cost Index
Wholesale Price Index
Direct Hourly Wage Rate
Amortization Interest Rate
Amortization Period
Electric Power Cost
Level
2 mg/1
20 min
300 $/ton
256.7
178.1
5.19 $/hr
7 percent
20 yr
$0.01/KWH
Design Parameter (Fixed)
Capacity Factor
70 percent
-------
O 4
o
o
o
- 3
K
(fl
O
o
t 2
5 1
°1510 20 40
60 80 100 120 140 160 18O 200
PLANT SIZE (MGD)
FIGURE 1. UNIT COST FOR CHLORINATION VERSUS PLANT SIZE
-------
AMORTIZED CAPITAL COST
'1 510 29
40
60
80 100 120
PLANT SIZE (MGO)
140
160
180
200
FIGURE 2. CAPITAL AND O&IUI COSTS FOR CHLORINATION VERSUS PLANT SIZE
-------
Figures 3 through 11 show the sensitivity of the cost of chlorination
to variations in all of the design variables listed in Table 1, with the
exception of capacity factor. Capacity factor has been fixed at 70 percent
to reflect the fact that water treatment systems generally operate at less
than design capacity. Costs presented in this report, therefore, reflect the
average costs to be expected over a year's operating period. The sensitivity
analysis assumes that one parameter is varied while the others are fixed at
the standardized levels, and is an attempt to show how costs might vary due
to local conditions.
The parameters which affect only 0 & M cost are shown in Figures 3, 4,
and 5 as follows: cost of chlorine, direct hourly wage rate, and wholesale
price index. It can be seen that all of the parameters have a significant
impact on 0 & M cost. For example, for a 100 mgd plant, increasing the
cost of chlorine from 300 $/ton to 400 $/ton increases the 0 & M cost from
0.36C/1000 gal to 0.46C/1000 gal (28 percent). For a 1 mgd plant, increasing
the Direct Hourly Wage Rate from 5.19 $/hr to 6 $/hr raises the 0 & M cost
from 1.40C/1000 gal to 1.56<:/1000 gal (11 percent).
The parameters which affect Amortized Capital costs are as follows:
chlorine contact time, interest rate, construction cost index, and amortiza-
tion period. The effects of these variables are shown in Figures 6, 7, 8.,
and 9.
The parameter (aside from the load factor) which affects both Amortized
Capital and 0 & M costs is chlorine dosage and is shown in Figures 10 and 11.
It can be seen that an increase of 1 mg/1 dosage of chlorine increases the
0 & M cost alone by 0.15C/1000 gal for a 100 mgd plant. Table 2 summarizes
the cost of chlorination for 1, 5, 10, 100, and 150 mgd systems in C/1000 gal
assuming the levels of the design variables as shown in Table 1.
As can be seen from the previous analysis, chlorination costs are
relatively stable. Another form of disinfection is chlorine dioxide dis-
infection which will be discussed in the following section.
COST OF CHLORINE DIOXIDE
An alternative to chlorine is chlorine dioxide, which does not produce
measurable quantities of trihalomethane, however, there is a possibility of
toxic organic byproducts resulting from the reaction of chlorine dioxide
with organic matter in water.? In this cost analysis, it was assumed that:
half the dosage of chlorine dioxide as compared to chlorine is required to
achieve equivalent disinfection results. Therefore, it was assumed that
1 mg/1 of chlorine dioxide would achieve disinfection results equivalent to
those achieved by 2 mg/1 of chlorination. If 0.5 iag/1 of chlorine is combined
with 1.6 mg/1 of technical grade sodium chlorite, a 1 mg/1 dosage of chlorine
dioxide will result. The equation below shows this relationship (assuming
80 percent pure NaC102 yields 1.3 mg/1 of reactive material):
-------
o
o
o
(0
O
u
S 1
«0
O
1 MGD
5 MGD
£_MGJL
100 MGD
_ "
150 MGD
100 200 300
COST OF CHLORINE (S/TON)
400
FIGURE 3. O&M COST FOR CHLORINATION SYSTEMS VERSUS COST OF CHLORINE
-------
23456
DIRECT HOURLY WAGE RATE (S/HR)
10
FIGURE 4. O&M COST FOR CHLORtNATION SYSTEMS VERSUS DIRECT HOURLY WAGE RATE
10
-------
O
O
O
O
<•> 1
5
ofl
O
1.2
1.4
1 MGD
5 MGD
10 MGD
100 MGD
150 MGD
1.6 1.8 2.0
WHOLESALE PRICE INDEX (100)
2.2
FIGURE 5. O&M COST FOR CHLORINATION SYSTEM VERSUS WHOLESALE PRICE INDEX
11
-------
O
o
o
o
V)
82
a
UJ
N
1-1
cc
o
5
5 MOD
150 MGD
10 20 30
CHLORINE CONTACT TIME (MINUTES)
40
50
FIGURE 6. AMORTIZED CAPITAL COST FOR CHLORINATION SYSTEMS VERSUS CHLORINE
CONTACT TIME
12
-------
6 7
INTEREST RATES (%)
MGO
FIGURE 7. AMORTIZED CAPITAL COST FOR CHLORINATION SYSTEMS VERSUS INTEREST RATE
13
-------
<
o
O 2'
O
a
<
o
o
UJ
N
P
DC
O
2
5 MOD
100MGD
150 MGD
2.0
2.2
2.4 2.6
CONSTRUCTION COST INDEX (100)
2.8
3.0
FIGURE 8. AMORTIZED CAPITAL COST FOR CHLORINATION SYSTEMS VERSUS
CONSTRUCTION COST INDEX
14
-------
-J 3
<
O
o
o
o
W
82
-J
<
a
o
a
UJ
N
p 1
cc
o
S
1 MGD
100 MGD
150 MGD
0 10 15 20 25 30
AMORTIZATION PERIOD (YEARS & 7% INTEREST RATE)
FIGURE 9. AMORTIZED CAPITAL COST FOR CHLORINATION SYSTEMS VERSUS
AMORTIZATION PERIOD
15
-------
o
o
o
O
o
s
00
o
2 3
CHLORINE (MG/L)
FIGURE 10. O&M COST FOR CHLORIIMATION SYSTEMS VERSUS CHLORINE DOSAGE
16
-------
,~ 3
0
o
o
o
V)
O 2
o
a
o
UJ
H 1
t-
£C
O
5
5 MGD
100 MGD
150 MGD
1 2 3 4 £
CHLORINE DOSAGE (MG/L)
FIGURE 11. AMORTIZED CAPITAL COSTS FOR CHLORINATION SYSTEMS VERSUS
CHLORINE DOSAGE
17
-------
"O
S o
SPl i 0
C
O
s
-3
o
•3
4J
-1
a
a
<=
0
0
s
-T
0
rl
:N
vC
O
-T
•"~i
CN
«N
CN
g
Id
a
a
« rH
a
o M
M O
s o
3 O
CO
3
t/J
Cd
Z
H4
a:
3
X
O
*H
•o
so
g
o
o
1— 1
lo
£
o
t— *
0>
*
0
o
0
c,
^c
ro
>T
0
^
o^
0
Q
O
o
0
CT»
U
E
o
4_(
b^
C
X
03
33
CN
•H
U
o
m
u*>
^_j
m
O
O
«M
lA
,
U"l
m
CM
C4
m
o>
^0
-------
2 NaClCL + C10 —> 2 CIO., + 2 NaCl
L L / (1)
(181) (71) (135) (117)
1.3 mg/1 0.5 mg/1 0.9 mg/1 0.8 mg/1
Therefore, the chlorine feeding system and contact basins were estimated for
a 0.5 mg/1 dosage of chlorine with the rest of the standardized values fixed
at the levels shown in Table 1. The cost of sodium chlorite was estimated
as follows:
NaC102 cost (
-------
§
a
u
2
<
H
O
z ^
en
E-
8
X
o
u
z
s
CJ
O
O
O
13
00
g
O
in
-H
^
fO
0
H
S
i^3
0
rH
a
4-*
•H
O.
a
u
fH
M
O
0
rH
to
u
0
H
X
«3
0
iH
<3
4J
•H
a
a
rH
•o
00
O
rH
•o
00
e
in
•o
at
e
rH
a
o
H
s
*
O
rH
01
JJ
•H
CL
(Q
u
rH
CO
U
O
H
£
-3
O
u
a.
fl
u
rH
0
H
£
-3
O
rH
to
•H
a
to
o
0)
>-*
o
•
o
0
o
o
o
vO
o
rH
§
o
d
o
rH
CO
CN
0
o
o
o
00
CN
tn
10
»^
o
o
o
o
•a
tn
^y
o>
0
o
o
o
rH
^
y\
C
en
t
33
CN
CJ
0V
.
00
CO
o
rH
0
CN
M
*H
CO
CT\
o
00
O
m
M
m
5
o
in
rH
r»x
^
_,
tN
tn
CN
00
--T
rH
r^.
m
o
00
E
u
en
en
•o
"j
fc.
tN
u
tn
tn
tn
o^
O
O
o
o
m
ON
en
°1
O
o
0
d
O^
tn
S
O
0
o
d
en
tn
o
0
o
o
•n
0
u
esi
0
^_;
z
CM
rH
fN
•H
rH
CN
O
"<
^
rH
CO
m
^
^
o
rH
00
o
as
in
^
>0
O
rH
CJ>
^
CN
00
rH
^.
rH
rH
C*
r-
M
O
m
«-(
rH
CO
r-*
rH
^j
01
c
— H
fl
o
-------
o
o
t-
V)
o
o
'1510 20 40
60 80 100 120 140 160 180 200
PLANT SIZE (MGD)
FIGURE 12. TOTAL UNIT COST FOR CHLORINE DIOXIDE VERSUS PLANT SIZE
21
-------
.1
.2
.3 .4 .5 ,6 .7
SODIUM CHLORITE COST ($/LB)
.8
.9
FIGURE 13. O&M COST FOR CHLORINE DIOXIDE SYSTEM VERSUS COST OF
SODIUM CHLORITE
1.0
22
-------
TABLE 4. DESIGN PARAMETERS FOR OZONATION
Design Parameters (Variable)
Ozone Dose
Ozone Contact Time
Cost of Oxygen
Construction Cost Index
Wholesale Price Index
Direct Hourly Wage Rate
Amortization Interest Rate
Amortization Period
Electric Power Cost
Design Parameter (Fixed)
Capacity Factor
Level
1 mg/1
20 min
0.046 $/lb
256.7
178.1
5.19 $/hr
7 percent
20 yr
$0.01/KWH
70 percent
23
-------
510
100
PLANT SIZE (MGD)
150
FIGURE 14. TOTAL UNIT COSTS FOR OZOIMATION (AIR) VERSUS PLANT SIZE
24
-------
5
o
o
o
15
tf)
CAPITAL COST
O&M COST
01 510
100
PLANT SIZE (MGD)
150
FIGURE 15. AMORTIZED CAPITAL AND O&M COSTS FOR OZONATION (AIR)
VERSUS PLANT SIZE
25
-------
cost are shown in Figures 16 through 18. These variables are as follows:
direct hourly wage rate, wholesale price index, and cost of electric power in
kilowatt hours.
Figures 19 through 22 illustrate the sensitivity of capital costs to
the following variables: ozone contact time, construction cost index, and
interest rate and amortization period. Ozone dose, as can be seen from
Figures 23 and 24, affects both Amortized Capital and 0 & M costs.
Table 5 summarizes the costs for 1, 5, 10, 100, and 150 mgd plants based
on standardized levels of the design variables in Table 4.
Cost of Ozone from Oxygen
Ozone can also be generated by using oxygen. Figures 25 and 26 show
the total unit costs and the disaggregated costs (0 & M and Amortized Capital),
respectively, versus plant capacity. Figures 27 and 28 illustrate the sensi-
tivity of the cost of ozonation (0 & M and Amortized Capital costs) to the
cost of liquid oxygen.
Table 6 summarizes the costs of 1, 5, 10, 100, and 150 mgd plants using
the standardized.variables in Table 4.
COST OF AERATION
Aeration is frequently practiced for the removal of hydrogen sulfide and
reduced materials, such as ferrous iron and manganous manganese. In the
laboratory, aeration or purging is used as an analytic procedure to remove
trihalomethanes from water and it might therefore be used successfully as a
treatment technique. Experimentation has shown, however, that at typical
air-to-water ratios used in water treatment for removal of taste- and odor-
causing compounds (1:1) little removal of chloroform takes place. For this
analysis it was therefore assumed that the air-to-water ratio of 30 cu ft
of air to 1 cu ft of water would provide adequate removal of trihalomethanes,
which is consistent with laboratory results for effective chloroform removal.
Table 7 contains the standardized variables which were examined in the cost
analysis.
Figure 29 is a typical capital cost curve for an aeration basis as a
function of throughput in thousands of cubic feet. Figures 30 and 31 are
0 & M Amortized Capital cost curves, respectively, for the diffused aeration
system based on thousands of standard cubic feet per minute of air. Table 8
contains the cost per thousand gallons for a 1, 5, 10, 100, and 150 mgd
plant based on the standardized cost assumptions shown in Table 7.
COST OF GRANULAR ACTIVATED CARBON
Granular activated carbon (GAG) is not a substitute for chlorine
disinfection, but is well suited for the removal of various types of dissolved
organic materials including chloroform and other trihalomethanes. Most but
not all dissolved organics can be adsorbed, which actually removes them from
solution. Fresh, granular carbon has the following advantages for water
26
-------
-T 5
O
o
O
f- 4
^*
«
V)
83
S
•a
O
100 MOD
34567
DIRECT HOURLY WAGE RATE (S/HR)
jjSO.MGP ^
8
10
FIGURE 16. O&M COSTS FOR OZONATION (AIR) VERSUS DIRECT HOURLY WAGE RATE
27
-------
•4
.8
1.2 1.6 2.0 2.4 2.8
WHOLESALE PRICE INDEX (100)
3.2
3.6
4.0
FIGURE 17. O&M COST FOR OZONATION (AIR) VERSUS WHOLESALE PRICE INDEX
28
-------
100 MOD
150 MGD
.005 .010 .015 .020 .025 .030 .035
COST OF ELECTRIC POWER (S/KWH)
.040
.045 .050
FIGURE 18. O&M COST FOR OZONATION (AIR) VERSUS ELECTRIC POWER COST
29
-------
0
o
22
o
o
o
Q
UJ
N
ce
o
150
MOD
10 20 30
OZONE CONTACT TIME (MINUTES)
40
SO
FIGURE 19. AMORTIZED CAPITAL COSTS FOR OZONATION (AIR) VERSUS OZONE
CONTACT TIME
30
-------
o
o
o
o
o
o
a
LU
N
P
cc
o
5
2.0
5 MOD
100 MGD
150 MGD
2.2
2.4 2.6 2^8
CONSTRUCTION COST INDEX (100)
3.0
FIGURE 20. AMORTIZED CAPITAL COST FOR OZONATION (AIR) VERSUS CONSTRUCTION
COST INDEX
31
-------
3.5
3.0
O
O
O
O
O
.1.5
0.
«1.0
Q
111
N
i .5.
5
456
INTEREST RATE (%)
10
FIGURE 21. AMORTIZED CAPITAL COST FOR OZOIMATIOIM (AIR) VERSUS INTEREST RATE
32
-------
20 25
AMORTIZATION PERIOD (YEARS)
FIGURE 22. AMORTIZED CAPITAL COST FOR OZONAT10N (AIR) VERSUS AMORTIZATION PERIOD
33
-------
o
o
o
o
o
o
1 MGD
5 MGD
10 MGD
100 MGD
15CLMGD
1.0 1.5
OZONE DOSE (MG/L)
FIGURE 23. O&M COST FOR OZONATIOIM (AIR) VERSUS OZONE DOSE
34
-------
o
o
o
°2
i-
U)
o
o
Q
HI
N
I-
cc
o
5
MOD
5 MGD
1.0
OZONE DOSE (MG/L)
1.5
FIGURE 24. AMORTIZED CAPITAL COST FOR OZONATION (AIR) VERSUS OZONE DOSE
35
-------
a
-j
z
o
D
a:
2
M
I
cn
<
cn
cn
O
CJ
•o
sc
o
m
i— <
C3
u
H
25
o
,H
*-•
a
CO
CJ
-H
00
E
O
O
CO
u
O
H
S
fafl
o
fH
ca
*J
a.
u
•o
eo
o
t-H
TI
00
S
m
,,
oc
e
fH
a
u
o
H
£
O
*
tH
a.
CO
CO
o
6-
s
"3
O
fH
4J
•H
a.
CO
a
o
.j.
>s
o
CO
u
•H
O.
CO
CJ
CO
<£
CN CJ\
•«T CN
•H C
O 0
CO
O O
CO CS
cn CM
-T CN
>H O
ON o
o o
d
in o
>T CO
CO r-*
m cn
•H O
^ o
0 0
o
0 CM
3 cn
vO CO
in >r
•H O
0 0
0
^r MS
r- co
^ -3-
in ^>
O .H
»— t O
0 O
o
cn -^
in CM
vH
00
c -
N C3 CJ O
O fc. < CJ
^
r-t
m
FH
00
o
00
•— t
tn
1— 1
tn
O
r-<
iv.
av
3
CN
^
.H
CM
iH
^
O
-1
cn
00
fH
sr
•*
CN
r>^
•a
cn'
cr.
^O
p*.
1-1
«£
a
,
O
o
o
0
o
0
0
0
o
c
0
0
o
o
o
o
o
o
o
f^.
vC
o
r^
CO
in
0
m
03
p^
cn
cn
1C
in
tO
rH
M
CM
^
a>
CTi
CM
CM
^
CM
fH
rH
cn
f— )
r"1
o
to
^
tO
Is"
cn'
vO
m
CN
U
3)
O
c_>
rH
CO
o
H
36
-------
14-,
12
o
o
o
5 8
01 5 10
100
PLANT SIZE (MGD)
150
FIGURE 25. TOTAL UNIT COST OF OZONATION (OXYGEN) VERSUS PLANT SIZE
37
-------
o
o
o
V)
t-
o
°
O
O
Q
UJ
N
o
5
AMORTIZED CAPITAL COST
5 10
100
PLANT SIZE (MOD)
150
FIGURE 26. AMORTIZED CAPITAL AND O&M COSTS FOR OZONATION
(OXYGEN) VERSUS PLANT SIZE
38
-------
1
J
<
o
o
§ 4-I 1MGD
O
O
s
«S
5MGD
10 MOD
100MGD
150 MGD
0123456789 10
LIQUID OXYGEN COST (c/LB)
FIGURE 27. O&Wl COST FOR OZONATIOIM (OXYGEN) VERSUS LIQUID OXYGEN COST
39
-------
o
o
o
1MGD
CO
O 3
O
I-
I 2
u
a
UJ
N
P 1
CC
O
5
5 MGD
10 MOD
100 MGD
150 MGD
01 23456789 10
LIQUID OXYGEN COST (e/LB)
FIGURE 28. AMORTIZED CAPITAL COST FOR OZONATION (OXYGEN) VERSUS
LIQUID OXYGEN COST
-------
!
«
g
z
5
a
a
u
r-f
9
^
a
«
o ^
KS *H
r- CO
£ 30
C/3 O
< 0
03 ~~.
E-i *>
en v*
O
CJ
r— *
U
(J
X
2
X
o
o
rJ
ca
^
_,
-a
CO
E
0
a
o
H
X
*a
o
"a
— ^
a.
r-l
0
o
H
CQ
4J
O
H
X
•4
O
in
u
Cu
0)
u
r-l
•0
OC
g
o
r-l
3
O
H
X
•&
O
t»
•H
a.
cj
00
B
in
rH
iJ
O
H
X
•d
O
W
jj
•H
O.
n
u
•o
e
^
TJ
O
H
X
fcfl
O
r-l
4j
•H
a.
to
u
s
CJ
OS
fN
—1
CN
O
CO
r*.
— ^
r>.
CN
(
rH
o
m
CO
rH
CO
CM
,
r-(
CTv
O
r-%
en
CM
P-*
(^j
S
0
m
in
m
OS
_,
CTi
O
n
0
C
•—4
3J
3) n
c c/
01 -H
o *-•
-H
4) r-l
C -H
O 0
M 10
i-l ! 0 U.
ON
OvJ
O
O
o
o
0>
CM
CN
fO
CN
•
O
o
o
0
0
CN
f»x
.
o
o
o
o
CM
f—
vO
»
0
o
0
0
>£)
CO
<•
CM
<
"*
o
o
0
o
^
^f
CM
(^
C 0)
« D
l< >
Cu O
_
GO
o
CO
r-l
•
m
rH
0
o
*^
r*.
.
0>
fsj
*o
CM
rH
**» t
f**
CM
t
i-H
p-l
o
1-1
m
CO
vr
^
C-J
r*i
%o
en
ON
'•O
r^
t_i
Cl
ui S
3
un
"1
0
o
o
00
rH
*
in
<—*
i-H
•J
0
I-l
en
en
.
m
•-H
i— 1
^
i«4
CM
^,
'Kf
i-H
t— t
^
vO
0
"^
in
r^
O
CO
CM
in
rH
C
C
g_4
CO
a
o
VI
CN
0
CM
00
_,
CM
<"1
^
-------
TABLE 7. DESIGN PARAMETERS FOR AERATION
Design Parameters "rariable)
Air to Water Ratic
Contact Time
Construction Cost Index
Wholesale Price Iiv'ex
Direct Hourly Wage Rate
Amortization Interest Rate
Amortization Period
Electric Power Cost
Level
30 cu ft: 1 cu ft
1 • :iin
21'. 7
178.1
5.19 $/hr
7 A ercent
20 yr
$0.01/KWH
Design Parameter (Fixed)
Capacity Factor
70 percent
42
-------
10
7_
V)
cr
10
6.
O
D
V)
O
O
z
g
I io5-|
z
O
O
10
10 100
AERATION BASIN VOLUME IN CU FT (1000)
1000
FIGURE 29. CONSTRUCTION COST FOR AN AERATION BASIN
VERSUS VOLUME OF BASIN
43
-------
O
Q
CO
O
O
Z
Z
10'
10
10"
10
10 100
STANDARD CUBIC FEET PER MIN (x 1000)
1000
FIGURE 30. ANNUAL O & M COSTS FOR AIR SUPPLY VERSUS
STANDARD CUBIC FEET PER MINUTE THROUGHPUT
44
-------
10
cr
<
o
Q
V)
o
o
z
o
p
o
EC
Z
O
o
10'
10
10
10 100
STANDARD CU FT PER WIN (1000)
1000
FIGURE 31 CONSTRUCTION COST FOR AIR SUPPLY VERSUS
STANDARD CUBIC FEET PER MINUTE THROUGHPUT
45
-------
en
-J
w
>
a
j
z
o
rH
en
Ed
a
a
a
Ps]
r-t
§
<
a
§
H — .
e/3 rH
a
O 00
Z
t-t O
§ §
C.O i-t
trt "***
< 0
en
t-
en
o
u
z
o
H
1
<
go
W
_!
ffl
<
H
1 rH
i 11
O
in
rH
U
0
H
X
wO
O
p-t
B>
u
•H
a
a
u
•o
Cl
B
O
O
rH
a
jj
o
H
£
*4
O
iH
a
u
•H
a
X
-3
0
pH
a
u
•H
a
n
CO
T)
Cfl
g
in
t— t
a
u
0
H
33
-3
0
01
u
•**
a.
n
CJ
-a
0£
e
^H
rH
a
u
o
E-
s:
•o
o
*^
01
u
•H
a
a
o
E
0)
^j
M
in
O
O
O
d
rH
in
.H
O
rH
O
o
o
o
r*«
>c
rH
*T
M
f>
o
o
o
o
-3-
n
n
vO
en
-T
o
0
O
O
•J2
fn
-j
CO
in
c\
0
o
0
0
CO
m
o>
c
m
03
a
o
•H
a
•^
eg
<
xT
CO
f^l
CO
1 —
o
f>
in
CO
t*»
O
en
C7\
O
CO
00
-JO
>H
-31
m
^r
ej>
m
CO
rH
t^
-T
fM
r^
^
VO
0
vO
in
in
i^
m
*^
^
rH
>T
rH
m
o>
m
r^
rH
O
CO
•a
r-i
CM
n
rH
CN
in
CNI
r*.
C3
rH
CO
a*
m
o
rH
>.
rH
o-
o.
3
en
^
•H
<
u-i
(->
1/1
CO
r*.
O
m
in
cr\
cs
CM
n
r^
0\
CO
vO
rH
-T
|