United States
Environmental Protection
Agency
Office Of
The Administrator
(A101F)
171-R-92-024
August 1992
vvEPA
Affordable
Drinking Water Treatment For
Public Water Systems
Contaminated By Excess Levels
Of Natural Fluoride
-------
DISCLAIMER
This report was furnished to the U.S. Environmental Protection
Agency by the student identified on the cover page, under a National
Network for Environmental Management Studies fellowship.
The contents are essentially as received from the author. The
opinions, findings, and conclusions expressed are those of the author
and not necessarily those of the U.S. Environmental Protection
Agency. Mention, if any, of company, process, or product names is
not to be considered as an endorsement by the U.S. Environmental
Protection Agency.
-------
NATIONAL NETWORK FOR
ENVIRONMENTAL MANAGEMENT STUDIES
RESEARCH REPORT
AFFORDABLE DRINKING WATER TREATMENT
FOR PUBLIC WATER SYSTEMS
CONTAMINATED BY EXCESS LEVELS OF NATURAL FLUORIDE
BY
CARL MASTROPAOLO
DECEMBER 1991
U 3 Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson 6ouiaydid, IZifl rioor
Chicago, !L 60604-3^90
-------
TABLE OF CONTENTS
page
STUDY APPROACH 2
HISTORICAL PERSPECTIVE 4
SOURCES OF FLUORIDE IN GROUNDWATER 11
GEOGRAPHIC AREAS WITH EXCESS FLUORIDE 13
COMPLIANCE OPTIONS 15
AVAILABLE TREATMENT TECHNOLOGIES 17
MORE ON POINT-OF-USE AND POINT-OF-ENTRY 26
TREATMENT COSTS .31
SUMMARY OF CONSTRUCTION COSTS 39
SUMMARY OF ANNUAL COSTS 41
MISCELLANEOUS NOTES REGARDING COST CALCULATIONS 45
APPLICATION TO VIRGINIA PUBLIC WATER SYSTEMS 47
CONCLUSIONS 51
RECOMMENDATIONS 55
REFERENCES 57
APPENDIX A: COST CHARTS FOR TREATMENT TECHNOLOGIES A-l
-------
STUDY APPROACH
This study was performed during the period 10 June through
early September 1991. The author researched the published
literature regarding fluoride occurrence in groundwater,
remediation techniques associated with its removal, and the dose-
related health effects associated with exposure. Sources
explored included texts and professional journals, and technical
articles written by the United States Environmental Protection
Agency, or by individuals or companies under EPA contracts.
Professional papers and water resources investigations prepared
by the United States Geological Survey were also consulted.
In addition, over fifty telephone conversations unearthed
information deemed relevant for inclusion in this report.
Telephone calls were placed to other EPA regional offices,
headquarters EPA in Washington DC, the EPA research facility in
Cincinnati, various state level drinking or groundwater offices,
the National Institute of Health, medical and dental experts,
United States Geological Survey offices, university researchers,
public water system managers and operators, private contractors
involved in the water industry, and other individuals. A number
of these contacts submitted written or faxed information, a
considerable amount of which has been referenced in this report.
An attempt was made*to quantify various treatment technology
construction, operation and maintenance, chemical, and disposal
costs, and to determine whether a centralized treatment facility,
or a point-of-use/point-of-entry treatment is more cost effective
based upon the average flow rate of a given public water system.
Cost figures were primarily generated by updating data contained
in an extensive cost study of small water treatment facilities
performed by Culp/Wesner/Culp under EPA contract during the
period September 1981 through January 1984 (84-1).
We feel that the cost and technology information acquired in
this research could be applied to any public water system
experiencing excessive fluoride concentrations. We attempted to
specifically apply this information to a group of southeastern
Virginia public water systems regulated by the Commonwealth of
Virginia which experience excessive fluoride groundwater
concentrations.
We wish to note that there were many avenues which we could
not explore since this research project lasted for only three
months. We have tried to highlight some of these in the hope
that further studies might be performed. Specifically we hope
that further research is able to determine costs associated with
the following compliance options: drilling a new well;
connecting or hauling water from another public water system;
point-of-entry treatment; and distributing bottled water.
-------
A draft copy of this report was circulated among various
individuals both within and outside the Environmental Protection
Agency for their comments. The author of this report agreed to
review these comments and incorporate them into the final version
of this report during a one week period in December 1991. It was
determined however that a considerable amount of research time
would be required in order to properly address some of the
comments submitted. The author has attempted to address all
comments and to note within the text of this report those
comments which could not be fully addressed due to the short time
frame set aside for developing the final version.
-------
HISTORICAL PERSPECTIVE
The physiological effects of fluoride upon human health have
been studied since the early part of this century. As early as
1925 (25-1) dental fluorosis, a condition characterized by
discoloration and softening of teeth (90-4), was attributable to
drinking water.
During the 1930s and 1940s, scientific research established
that the incidence of dental fluorosis was directly related to
the fluoride content of drinking water. High fluoride exposure
was also linked to increased bone density of the spine and
pelvis, lenticular opaqueness of the eye, and brittleness and
blotching of fingernails (53-1).
In response to growing scientific evidence regarding the
negative effects of exposure to excessive fluoride, the United
States Public Health Service (PHS) established a range of
recommended concentrations for fluoride in drinking water (1.4 to
2.4 mg/L), dependent upon the maximum daily air temperature of
the location of the water supply (62-1). A temperature dependent
value was required to compensate for increased water consumption
in warmer climates.
Several reports and studies in the 1960s and 1970s established
both the risks of high fluoride dosing and the benefits of
minimal exposure. Low drinking water concentrations of fluoride
were deemed responsible for inhibiting dental caries (tooth
decay), while higher concentrations were linked to permanent
tooth fluorosis, and skeletal fluorosis, the latter responsible
for aberrations ranging from stiffness to crippling rigidity (65-
1, 70-2, 70-3, 71-1, 73-2, 73-3, 73-4, 75-2). One of these
studies (71-1), performed by the National Academy of Science,
determined that dental tissues and the skeleton accumulate
fluoride more rapidly during formation and mineralization thus
negatively impacting children up to about 10 years old more so
than the general population.
Through continued scientific research, it became apparent that
an optimal intermediate concentration of about 1 mg/L would
produce the benefit of reduced tooth decay without the risks
associated with higher doses (70-1).
1975 National Interim Primary Drinking Water Regulations
In 1975, pursuant to the 1974 Safe Drinking Water Act, the
United States Environmental Protection Agency promulgated the
National Interim Primary Drinking Water Regulations, establishing
maximum contamination levels (MCLs) for ten inorganic
contaminants, including fluoride, as well as for, turbidity,
coliform bacteria, pesticides, herbicides, total trihalomethanes,
and radionuclides. EPA adopted the Public Health Service
-------
temperature dependent standards for fluoride (1.4 to 2.4 mg/L).
As part of its on-going regulatory review process, EPA
requested the advice of the National Academy of Science
concerning MCLs for fluoride and other contaminants. In 1980 the
National Academy of Science concluded that while "dental mottling
and changes in tooth structure may develop in children when
fluoride levels exceed approximately .7 to 1.3 mg/L", consumption
of drinking water with concentrations as high as 4 mg/L provides
a beneficial reduction in osteoporosis (80-2). Scientific
evidence was therefore suggesting that concentrations between
about 1.3 and 4.0 mg/L produced both positive and negative
effects.
In 1981, the state of South Carolina petitioned that EPA
delete fluoride from the Primary Drinking Water Regulation and,
in 1984, formally sued EPA seeking faster action on the petition.
A consent decree signed by both parties on January 18, 1985 set
EPA about the task of re-evaluating the fluoride MCL.
Considerable scientific evidence had been accumulating in the
interim. Responding to an EPA request to examine "the issue of
the relationship of fluoride in drinking water and the health
aspects of dental fluorosis", former Surgeon General Koop
concurred with the conclusion reached by the PHS's ad-hoc
committee on dental fluorosis: "No sound evidence exists which
shows that drinking water with the various concentration of
fluoride found naturally in public water supplies in the United
States has any adverse effect on dental health as measured by
loss of function and tooth mortality" (82-2).
And in 1984, in response to an EPA request to review the non-
dental effects of fluoride, Koop stated (84-5):
Adverse health effects were defined by the
committee as death (poisoning), gastrointestinal
hemorrhage, gastrointestinal irritation,
arthralgias, and crippling fluorosis. No record
exists of poisoning death from fluorides consumed
in drinking water. There are no scientifically
credible reports of gastrointestinal effects at
levels found in drinking water. Clinical
experience suggests that arthralgias are not
likely to occur in patients who are on therapeutic
regimens of less than 20 milligrams per day.
Crippling fluorosis has been detected in some
people who have consumed 20 mg or more of fluoride
per day from all sources for twenty or more years.
Such a situation does not exist in the U.S. today.
Meanwhile, in 1984, the World Health Organization deemed
fluoride concentrations exceeding 1.5 mg/L as excessive on the
-------
basis of mottling of teeth (84-4).
1985 RMCL Proposal
In proposing a recommended maximum contamination level (RMCL)
for fluoride in 1985, EPA extensively reviewed literature, public
comments, and professional advice from diverse sources and
concluded (R85-1):
* exposure to fluoride levels of about 1 to 2 mg/L can
contribute to dental fluorosis, frequency and severity
increasing with that of the fluoride concentration
* exposure to levels between 1 and 4 mg/L have been
shown to contribute to reduced dental caries formation
* exposure to levels greater than 4 mg/L "can result in
asymptomatic osteosclerosis (increased bone density) in
a small percentage of individuals"
As a result of its review, EPA established an RMCL of 4.0 mg/L to
prevent crippling skeletal fluorosis, and planned to propose a
secondary maximum contamination level (SMCL) of 2.0 mq/L to cruard
against mottling of teeth.
The Office of Management and Budget, as well as medical
dental, and public health groups throughout the United States
opposed EPA's handling of the naturally occurring fluoride issue
and its assessment of negative fluoride effects in the period
between the promulgation of the National Interim Primary Drinking
Water Regulations and the establishment of the RMCL.
In October 1979 the American Dental Association House of
Delegates "resolved, that based on present knowledge, it is the
opinion of the American Dental Association that the natural
fluoride levels of drinking water in the United States do not
constitute a health hazard". The resolution was supported by the
American Association of Public Health Dentists.
In March 1980 the Association of State and Territorial Dental
Directors unanimously approved a resolution that opposed EPA's
inclusion of fluoride in the primary standards of the National
Interim Primary Drinking Water Regulations at levels of twice the
optimum and above as a contaminant and health hazard. It further
requested that "fluoride be changed from the Primary to the
Secondary Drinking Water Regulations".
In a May 7, 1982 letter to the HQ EPA Administrator Anne
Gorsuch, the American Medical Association concurred with the
previously stated American Dental Association position indicating
that AMA was "unaware of any unequivocal evidence showing that
naturally occurring fluorides have adverse effects on the public
health".
-------
In March 1985 EPA requested the Office of Management and
Budget to review the draft notice of proposed rulemaking (NPRM)
which would establish the RMCL for fluoride. In response, the
OMB questioned "the need for national health standards for
fluoride in view of the infinitesimal health risk and quite small
population exposure" (85-5). Using EPA's own NPRM figures, OMB
indicated that only 184,000 people were served by the 282 public
water systems (PWSs) which then had fluoride concentrations
exceeding 4 mg/L. "This is less than 0.1% of the total
population of 190-200 million served by the 60,000 drinking water
systems that are subject to the SDWA." In referring to the fact
that only 17,000 people were then exposed to concentrations
exceeding 6 mg/L, and 3,000 to 4,000 to those exceeding 8 mg/L,
OMB further commented that "at all these concentrations, the
Surgeon General and his "world class" panels have found no
demonstrated cases of adverse affects on the health of persons".
1986 Final Fluoride Rule
Prior to establishing the primary MCL for fluoride at 4.0
mg/L, EPA solicited comments relating to three options under
consideration for the regulation of fluoride:
Option 1: "Propose a National Revised Primary Drinking
Water Regulation to protect against moderate
and severe dental fluorosis and set the RMCL
at 1 or 2 mg/L, as appropriate" (R86-1)
Option 2: "Propose a National Revised Primary Drinking
Water Regulation finding that crippling
skeletal fluorosis (but not dental fluorosis)
is an adverse health effect and set the RMCL
at 4 mg/L; propose a National Secondary
Drinking Water Regulation to warn against
dental fluorosis (a cosmetic effect),
setting a secondary MCL at 2 mg/L." (R86-1)
Option 3: "Delete fluoride from the National Primary
Drinking Water Regulations based upon a
finding that levels of fluoride in U.S.
drinking water are not associated with any
adverse health effects; propose a Secondary
MCL of 2 mg/L to warn against the cosmetic
effects of dental fluorosis." (R86-1)
"The Agency proposed Option 2 for the regulation of fluoride. In
making this decision, the Agency concluded that 'based upon the
information available at this time, EPA believes that crippling
skeletal fluorosis is an adverse health effect that can be caused
by excessive amounts of fluoride in drinking water, and that 4
mg/L is the level below which no known or anticipated adverse
effect on health of persons occur and which allows an adequate
margin of safety. Thus, an RMCL is proposed at 4 mg/L' (50 FR
20164). The Agency stated that it now believed that
-------
objectionable (moderate and severe) dental fluorosis is not an
adverse health effect under the Safe Drinking Water Act, but
rather a cosmetic effect that would adversely affect public
welfare and it should be regulated under the NSDWRs. The Aqencv
therefore stated that at the time of proposal of the MCL for
fluoride, it planned to propose a Secondary MCL at 2 mg/L."
(R86—1).
In responding to comments relating to these three options EPA
considered various independent study results, some of which were
submitted by commenters. Some of the studies cited in the final
fluoride rule, and their impact upon the decision making process
follow: -* r ,-
A draft National Institute of Dental Health study (84-8)
submitted by a commenter, concluded that the consumption of'
drinking water containing 4 mg/L or less has no adverse effects
upon the teeth and that increasing fluoride exposure from .7 to 4
mg/L results in increasing the level of dental attrition, but
does not pose a concern of clinical importance.
Commenters also noted two independent studies which cited two
cases of crippling skeletal fluorosis suffered by individuals who
had been exposed to fluoride concentration between 2.4 and 3 5
mg/L and 4.0 and 7.8 mg/L, respectively (71-2, 65-2). it is'
estimated that these individuals consumed about 6 liters of water
per day each, and consumed large undetermined quantities of tea
a beverage with a considerably higher fluoride concentration than
most foods, 97.0 parts per million (70-2).
EPA also cited the Knox Report, an epidemiological study
performed by the British government (85-7), and concurred with
its findings that there is no evidence that naturally or
artificially occurring fluoride is capable of inducing cancer.
On April 2, 1986, after review of all pertinent scientific
data, EPA promulgated the final fluoride rule (R86-1),
establishing an enforceable MCL of 4.0 mg/L "to protect against
crippling skeletal fluorosis" and a non-enforceable SMCL of 2.0
mg/L "to protect against objectionable dental fluorosis" (R86-1).
Activated alumina adsorption and reverse osmosis were identified'
as best technologies generally available (BTGA-this is currently
known as BAT: best available technology) for purposes of fluoride
variances.
1991 Public Health Service Fluoride Report
Since the 1986 ruling the most comprehensive toxicological
analysis of the effects of fluoride was rendered by the Public
Health Service's February 1991 report entitled Review of
Fluoride. Benefits and Risks. The PHS addressed cancer, dental
fluorosis, and skeletal disorders as the significant health
-------
conditions in their evaluation of the health risks associated
with fluoride intake. A brief summary of the report's findings
related to each of these conditions follows:
Regarding cancer: Optimal fluoridation does not pose a cancer
risk to humans as evidenced by extensive human epidemiological
data. Two methodologically acceptable animal carcinogenicity
studies published in 1990 by Proctor and Gamble (90-6), and the
National Toxicology Program, Public Health Service, Department of
Health and Human Services (90-7) "taken together fail to
establish an association between fluoride and cancer".
Regarding dental fluorosis; Total exposure to fluoride has
increased since the 1940's due to the introduction of fluoride
into dental products. Accordingly the incidence of dental
fluorosis has increased even in non-fluoridated areas. The PHS
recommends that unnecessary and inappropriate fluoride exposure
be reduced.
Regarding skeletal disorders; "Crippling skeletal fluorosis is
not a public health problem in the United States, as evidenced by
the reports of only five cases in 30 years." It was established
that two of these subjects consumed between 15 and 20 mg/day for
20 years. PHS noted that, in foreign literature, Moudgil, et al
(86-2) had identified 41 children with evidence of crippling
skeletal deformities who had consumed water with fluoride
concentrations ranging from 7.3 to 29 mg/L. PHS concluded "human
crippling skeletal fluorosis is endemic in several countries in
the world, but is extremely rare in the United States".
Current Opinions Regarding EPA Policy
Some medical and toxicological experts still oppose EPA
regulatory policy regarding fluoride. While the primary MCL for
fluoride has been established to guard against crippling skeletal
fluorosis, those in opposition refer to studies which indicate
that the probability of occurrence of the disease in the United
States is virtually zero.
Other opposition centers upon views regarding the effects of
dental fluorosis, with some experts favoring the reestablishment
of a temperature-dependent standard, since water consumption is
known to increase with rising temperatures. This is the view
still held, for example, by the National Institute of Health in
considering the optimal fluoride exposure to reduce dental
caries. Other dissenters still view dental fluorosis as a
cosmetic problem and not a health problem, and cite studies
indicating greater incidence of dental fluorosis in non-
fluoridating areas, the result of consumer use of increasing
numbers of fluoridated products, including dental care items.
Further, they note, dental fluorosis only adversely effects teeth
of young children, a known factor which can be readily
-------
compensated for by means other than to treat an entire
community's water supply.
10
-------
SOURCES OF FLUORIDE IN GROUNDWATER
Fluoride is the most electronegative element having a valence
of negative 1 and is the smallest atom of the halogen family. It
is not found in nature in the elemental state but rather it
combines with other elements to form fluoride compounds.
Fluoride is the anion (negatively charged ion) of the fluorine
atom. Under the proper conditions in a water environment,
fluoride compounds may dissociate to produce fluoride ions in
solution.
Fluoride occurrences in groundwater have two sources:
anthropogenic and natural.
Anthropogenic Sources
Anthropogenic fluoride sources originate with the use of
various forms of fluoride-bearing minerals or reagents in
industrial processes. Effluent from electroplating operations
may contain fluoride due to surface preparation or electroplating
with metal fluoborate salts. Wastewater from steel production
facilities contain fluoride, attributable to pickling of
stainless steel with hydrofluoric acid. Wet scrubbing of
emissions from furnaces in which aluminum is separated from
bauxite ore in the presence of the fluoride-bearing mineral
cryolite results in fluoride presence in the waste stream, as
does the removal of phosphate from fluorophosphate (fluorapatite-
Ca5F(PO4)3) during the production of fertilizers. Glass
manufacturing also produces fluoride-rich effluent (75-1).
Fluorine and fluoride compounds are also used in the electronics
industries for production of semiconductors or internal washing
of cathode ray tubes. In the wastewater discharged from some of
these industries, fluoride concentrations can sometimes reach
several hundred mg/L (88-6).
Natural Sources
Geochemical dissolution of fluoride-bearing minerals is
responsible for the occurrence of natural fluoride in water
supplies of the United States (85-1) . Fluorite (CaF2) i's the
most common fluoride-bearing mineral and is usually present in
sandstone, limestone, dolomite, and granite, all of which are
common in the United States. Other less common fluoride-bearing
minerals include fluorapatite, cryolite, rhyolite and hornblends.
Fluoride also may displace other ions in the structures of micas
and clays (85-1, 89-4, 59-1). Water with high naturally
occurring fluoride is usually found at the foot of mountain
ranges and in areas with certain geological formations,
particularly those of marine origin (88-3). Fluorite in
sedimentary rocks has been identified in marine carbonates and
related evaporites ranging in geologic age from Cambrian to
11
-------
Cretaceous. Experiments have shown (51-1) that fluorite could
precipitate from seawater brines. Subsequent geochemical
dissolution could reintroduce fluoride into fresh water aquifers
upon geologic time-scale marine regressions. Groundwater
fluoride concentrations may sometimes exceed 100 mq/L as the
result of volcanic activity.
12
-------
GEOGRAPHIC AREAS WITH EXCESS FLUORIDE
Fluoride concentrations in excess of the EPA MCL effect
groundwater-source public water systems throughout the
continental United States. All EPA regions were contacted to
determine the degree to which excess fluoride in groundwater
poses a problem today. While EPA regions I and II could identify
no problem areas, all other regions regulate at least one public
water system with fluoride concentrations exceeding the MCL.
The number of PWSs with high fluoride groundwater is
especially large in both Texas and Virginia. A 1982 national
survey performed by the American Water Works Association (83-1)
indicated, for example, that 328 Texas and 104 Virginia community
water systems (CWSs) were exceeding the then current fluoride MCL
(1.4 to 2.4 mg/L). At the time, the total number of CWSs
exceeding the MCL nationally was 907.
Recent conversations with EPA Region VI (Dallas) personnel
and the Virginia Department of Health, as well as reference to
the Federal Reporting Data System (FRDS II), indicate that Texas
and Virginia PWSs still currently experience excessive fluoride
concentrations. For example, 51 Virginia public water systems,
34 in Suffolk City County, have groundwater fluoride
concentrations exceeding the EPA MCL of 4.0 mg/L. Twenty-nine of
these have concentrations which also exceed the unreasonable risk
to health (URTH) level of 5.0 mg/L, the highest being 6.60 mg/L.
A number of PWSs in Oklahoma, Arizona, New Mexico and South
Carolina also currently experience excessive fluoride levels as
do a smaller number of PWSs in Indiana, Illinois, Ohio, Missouri,
Wyoming, Nevada, North Dakota and Idaho. In most of these cases
personnel from the various EPA regions indicate concentrations
are in the 5 mg/L range. Data from the 1982 AWWA survey, for
example, indicated that 177 CWSs had fluoride concentrations
exceeding 4.0 mg/L while only 85 exceeded 5.0 mg/L.
Since most sources of fluoride are natural, and naturally
occurring fluoride concentrations change very slowly (on a
geologic time scale), AWWA's 1982 data should reasonably reflect
the current fluoride concentration distribution among CWSs
nationally. It should be noted however that the number of CWSs
nationwide with excessive fluoride concentrations has decreased
since the early 1980s. Contacts at several EPA regional offices
have indicated, for example, that a considerable number of
smaller water systems have chosen to either locate another
drinking water source, connect to a large local water system, or
pursue other options in order to avoid costly defluoridation
treatment. As a result the number of CWSs in each concentration
category below probably overstates the current situation:
13
-------
Distribution of Fluoride Concentrations Nationally
(source: 1982 AWWA survey:83-1)
concentrations number of
exceeding fma/L) CWSs
4.0 177
5.0 85
6.0 34
7-0 16
8.0 9
9.0 8
Fluoride-rich groundwater apparently occurs in geographic
pockets" and normally does not effect groundwater on a regional-
scale. The North Carolina state water agency's regional office
in the eastern part of the state, for example, reports that no
PWSs in the area are in violation of the fluoride MCL, despite
close proximity to the Suffolk Virginia public water systems
experiencing excessive fluoride. Concentrations are in the 2 0
to 3.5 mg/L range. A Gates County North Carolina water supply
operator whose wells are about 20 to 25 miles from Suffolk
Virginia, indicates that fluoride levels in his wells are about
3.1 mg/L.
14
-------
COMPLIANCE OPTIONS
The Safe Drinking Water Act (SDWA) affects approximately
230,000 public water systems (PWSs). Of these, approximately
60,000 are community water systems (CWSs) which, by definition,
serve at least 15 service connections used by year-round
residents, or regularly serve at least 25 year-round residents.
About 88% of these systems serve 10,000 residents or less,
accounting for 22% of the total population served by CWSs (91-7).
In addition approximately 25,000 non-transient, non-community
water systems (NTNCWSs) and 145,000 non-community water systems
(NCWSs) are effected by the SDWA.
PWSs which produce water with excess fluoride may exercise a
variety of options in order to provide water which meets
regulatory standards. These options can broadly be characterized
as treatment versus non-treatment.
Treatment Non-treatment
central develop a new well which produces
water which meets the MCL
point of use (POU) connect to or haul water from a
PWS which produces water which
meets the MCL
point of entry (POE) purchase and distribute bottled
water which meets the MCL
While each of these options is technically capable of reducing
fluoride concentrations, current Safe Drinking Water Act
regulations do not permit a PWS to employ POU treatment, or
purchase and distribution of bottled water, as compliance options
to attain the MCL. Rather, POU treatment and bottled water are
only permissible as short-term options. It should be noted that
the January 30, 1991 final rule for volatile organic compounds
(R91-1) allows POE devices to be used to meet regulatory
requirements, although such devices are not considered BAT.
In most cases the first three compliance options which are
considered acceptable from a regulatory standpoint are relatively
expensive operations which can be pursued cost-effectively only
by larger PWSs which can spread capital, and operation and
maintenance costs over a larger customer base. For many of these
options, fixed capital costs, which are independent of the number
of customers serviced by the PWS, are quite large. For example,
sinking a new well to develop a new water source may cost in the
order of ten thousand dollars. To a PWS which services only 50
households, this represents a one time cost on the order of $200
per household. To a PWS servicing 500 households, this cost is
only about $20 per household. In addition, some marginal costs,
15
-------
which are dependent upon the number of customers serviced by the
PWS, are also subject to an economy of scale, in effect
financially penalizing smaller systems.
A PWS can also comply with SDWA regulations by applying for
and receiving a variance or exemption (SDWA, Sections 1415 and
1416). While EPA does not consider POU treatment, or purchase
and distribution of bottled water as viable long-term options to
attain regulatory compliance, it does permit the regulatory
agency (usually the State) to require a PWS to pursue such
options as a condition of granting an exemption or variance.
The feasibility of pursuing any one of these treatment or non-
treatment compliance options is often contingent upon a multitude
of factors including those of geography, geology, and economics.
Many small South Carolina PWSs in the North Atlantic Coastal
Plain, for example, which had experienced problems with excess
fluoride, opted to connect to larger water systems instead of
adding treatment. This was technically feasible due to the
relatively dense population along the coast and the subsequent
proximity of larger water systems which were already treating for
fluoride. By contrast, about 50 PWSs scattered throughout the
Arizona desert, some servicing as few as 25 to 30 people, have
been required to install POE devices as a condition for receiving
a variance or exemption. Due to their isolation these systems
did not have an option to connect to a larger system.
16
-------
AVAILABLE TREATMENT TECHNOLOGIES
The treatment technologies to remove fluoride are, for the
most part, well documented. They include:
adsorption by activated alumina
adsorption by bone char
reverse osmosis
electrodialysis
alum coagulation
lime softening
anion exchange
nanofiltration
All of these technologies can be applied as central treatment.
Only activated alumina, reverse osmosis and anion exchange
POU/POE treatment are currently commercially available, although
at least one commercial producer of bone char is attempting to
market his product as a POU/POE device ingredient.
In addition to the treatment technologies listed above,
research for this report revealed a recently developed and
patented defluoridation technique involving adsorption by rare
earth compounds.
A considerable amount of research has been conducted regarding
operation of, and costs associated with, most of these treatment
technologies. Gumerman, et al (84-1) provides a detailed cost
analysis for small centralized systems. Bellen and Anderson (85-
1) specifically evaluate the comparative efficiencies and costs
of defluoridation based upon evaluation of a number of
centralized and POU facilities. Proceedings of a 1987 EPA
sponsored POU conference (88-5) provides discussion of a broad
range of POU related topics as well as unit costs. Rogers (90-5,
88-4) relates the development of a POU reverse osmosis pilot
study in New Mexico from prior to request-for-proposal
preparation through cost and efficiency analysis.
A brief description of each treatment technology follows.
Activated Alumina
Activated alumina is manufactured by low temperature
dehydration of hydrous aluminum oxide, the process creating a
porous adsorbent of moderately high surface area (81-2). The
mechanism of fluoride adsorption involves the attachment of
fluoride (F~) to the activated alumina and the loss of hydroxyl
ion (OH") from activated alumina to the water (86-1). The
optimal pH range for this anion exchange is pH 5 to 6. Below
this range activated alumina dissolves in the acidic environment
producing loss of adsorbing media. Above pH 6, the predominant
hydroxyl ions are preferred over fluoride for adsorption. One
17
-------
gram of activated alumina is reported to adsorb between 2 3 and
10.1 mg of fluoride, dependent upon other raw water conditions
(81-2).
In central treatment, raw water is adjusted to the optimal pH
range and fluoride is adsorbed until the alumina bed is
saturated. Fluoride is then removed by adding a caustic (usually
sodium hydroxide) to produce a high pH water which removes
fluoride in favor of hydroxyl (regenerative step). After
regeneration the pH is reduced to the optimal range and the cycle
begins anew. J
POU/POE activated alumina treatment uses canisters containing
approximately .5 to 1.0 cubic feet of activated alumina and does
not involve pH adjustment to optimize the canister's adsorption
capacity. Once the fluoride adsorption capacity of the medium is
exhausted, the canister is replaced. The contents of the spent
canister can be regenerated by exposure to high alkali water as
is done in central treatment, and reused. This is usually done
off-site.
While activated alumina readily adsorbs fluoride, other raw
water parameters have an impact upon adsorptive capacity High
concentrations of total dissolved solids (TDS), for example can
result in fouling of the alumina bed and high concentrations of
sulfates and carbonates can result in ionic competition.
Application of activated alumina treatment is most favorable when
fluoride alone must be removed from drinking water (81-2).
Bone Char
Bone char is crushed and burned animal bone, having an
approximate chemical composition of hydroxyapatite, Cain(PO ) (OH)
(91-1). This treatment process normally also includes carbon 2
adsorption to remove color and odor.
The bone char process for fluoride removal is similar to that
for activated alumina. Regeneration of the media, by removing
fluoride, can be accomplished by exposure to dilute caustic soda
(84-6), or to a regenerating solution using bone char particles
too small for use in the bone char bed (91-1).
POU/POE bone char treatment has been proposed for a number of
years (see for example 68-1) and has been practiced in Third
World villages (88-3), in which case media is not regenerated
the village being supplied instead with replacement filter bags
Researchers have expressed the intent of providing villagers with
high-temperature furnaces to prepare filter ingredients (88-3).
A New Hampshire vendor who has been using a POE bone char
device to eliminate excessive arsenic, is currently researching
the device's ability to reduce groundwater fluoride levels. As
18
-------
of the writing of this report no results had been reported.
Bone char's capacity to adsorb fluoride is reportedly far less
than that attained by activated alumina (84-6, for example).
However, while a considerable amount of research has been
performed to determine optimal processing parameters for
activated alumina, such as pH, influent flow rate, and
contaminant loading concentration (see for example 79-1), little
in this regard had been done to optimize processing with bone
char. A recently published study supported in part by the World
Health Organization, however, claims bone char adsorption
capacity of about 9.8 mg fluoride per g of bone char (91-1).
This is comparable with activated alumina capacity. Furthermore,
after regeneration of spent columns with calcium phosphate at pH
3, the column is reportedly at least as effective as it was
originally.
Reverse Osmosis
Reverse osmosis (RO) utilizes a pressure gradient and semi-
permeable membranes to remove a high percentage of almost all
inorganic contaminants, as well as turbidity, bacteria and
viruses. The membrane essentially operates as a molecular sieve.
Fluoride removal efficiencies greater than 80%, and sometimes as
high as 98% have been documented for both central and POU
treatment techniques (90-5, 88-4, 86-1, 85-1, 85-4).
RO membranes are susceptible to scaling or fouling. Chlorine
also damages certain membrane materials. As a result chemical
pre-treatment or granular activated carbon adsorption is often
required in the process line prior to the reverse osmosis unit.
Pre-carbon adsorbers are apparently now standard for all POU/POE
units.
Reverse osmosis units incur substantial process electricity
costs. Low pressure units require less electricity than high
pressure units but remove a smaller percentage of contaminants.
Reverse osmosis units tend to produce a relatively small
volume of water with respect to the waste stream generated. A
1983 POU reverse osmosis pilot study resulted in an average waste
stream of 89% for 47 devices (85-1), that is, only 11% of raw
water influent was eventually processed into product drinking
water.
Electrodialvsis
In the electrodialysis (ED) process, water flows between
alternate cation permeable and anion permeable membranes, with
direct electric current providing the driving force to cause ions
to migrate through the membranes. Electrodialysis reversal (EDR)
is essentially the same process as ED except that the electric
19
-------
charge on each membrane is occasionally automatically reversed to
expel any ions which may have collected. Virtually all
electrodialysis plants constructed since 1975 are EDR.
Fluoride removal efficiencies greater than 80% have been
reported (84-3) for these processes.
EDR plants operate with a relatively high water recovery rate
especially when compared with RO. An EDR defluoridation plant '
processing 3 million gallons per day and operated by the City of
Suffolk, Virginia, for example, typically recovers about 94% of
influent as product water.
Apparently POU/POE EDR is not a viable option. Neither the
proceedings of the 1987 EPA conference on point-of-use (88-5) nor
Gumerman, et al's extensive review of available small systems
completed in 1984 (84-1) mention an EDR POU/POE option.
A contact from the Office of Ground Water and Drinking Water,
EPA reports that pending regulations may include ED as BAT for
fluoride.
Alum Coagulation
Upon addition to water, alum (A12 (SO,), 18H,O) dissociates to
form an aluminum hydroxide complex which removes fluoride either
by floe enmeshment or sorption. The insoluble complex
precipitates out of solution.
EPA has reported that this treatment technique tends to be
expensive for small system applications (84-6), requiring
addition of large quantities of alum. More recently, however
Nawlakhe & Bulusu (89-4) have developed a combination alum/lime
defluoridation treatment (referred to as the Nalgonda technique)
which they contend is inexpensive, effective, and adaptable to
both the domestic and community levels. The technique is
specifically designed to be operated by illiterate people in
rural communities in India.
Lime softening
Fluoride is removed by treatment systems which utilize lime
softening (lime is calcium hydroxide) to remove calcium and
magnesium hardness. Since fluoride is coprecipitated in this
process with magnesium hydroxide (86-1, 84-6), this method
requires the raw water magnesium concentration to be sufficient
to enmesh enough fluoride to produce residual concentrations
below the MCL. Substantial quantities of magnesium are required
to produce this effect, as well as the addition of large
quantities of lime.
Lime softening treatment can be performed either as a single
20
-------
stage or two stage process. As indicated in Gumerman, et al (84-
1), the majority of lime softening systems utilize single stage
processing. Two stage processing is required to remove
significantly high raw water magnesium concentrations.
While lime softening is primarily used to remove hardness,
considerable fluoride reduction is a by-product. For example,
the Yuma Bureau of Reclamations Desalting Plant typically removes
about 45% of raw Colorado River water fluoride concentration in
its lime softening operation.
As indicated in the discussion on alum coagulation, a small
scale treatment technique which employs alum and lime has
reportedly been developed which is effective and inexpensive at
the domestic and community levels.
Anion Exchange
Ion exchange is a process in which ions held by electrostatic
forces to charged functional groups on the surface of a solid are
exchanged for ions of similar charge in a solution in which the
solid is immersed (72-1).
Though this technique theoretically has potential application
for defluoridation, little information regarding its use could be
obtained either through personal contacts or literature review.
EPA pilot studies in Oregon and Alaska utilized anion exchange as
a POU arsenic removal technique (87-3) but apparently did not
pursue the device's ability to remove fluoride. EPA-sponsored
research conducted by the University of Houston (84-3) indicates
that anion exchange resins prefer exchange with chloride rather
than fluoride. Since chloride is a common major anion in
groundwater sources, especially in coastal regions experiencing
varying degrees of salt water intrusion, its competitive
advantage over fluoride in an anion exchange process would tend
to minimize this technology's effectiveness for fluoride removal.
Nanofiltration
Nanofiltration is a relatively new water treatment technique
which has been commercially available since 1986. The^nanofilter
operates on a molecular scale, removing molecular species with
diameters greater than or equal to 10 angstroms (a nanometer).
The technology has the ability of removing smaller materials than
ultrafiltration, but is incapable of removing some of the smaller
materials removable by reverse osmosis. Because of the coarser
membrane with respect to reverse osmosis, nanofiltration requires
lower operating pressures (typically between 80 to 100 psi) than
RO to force water through the membrane and therefore
substantially less energy (89-5, 89-6, 87-2).
The coarser membrane also produces a substantially higher
21
-------
water recovery rate than RO resulting in a much smaller waste
stream. According to a nanofiltration researcher from the
University of Arizona, the process experiences a 95% water
recovery rate. The researcher reports that several Florida
locations are using this technology, and the city of Fort Myers
is presently constructing a 20 million gallon per day plant which
utilizes a groundwater source.
While the researcher's bench scale tests have been designed to
determine the effects of the treatment on salinity, hardness
bacteria, virus, and organic precursors, tests with recharged
municipal effluent in the city of Phoenix indicate that
nanoflltration was able to reduce fluoride levels from 7 to 2
mg/L (89-5).
The nanofiltration researcher is currently in the process of
developing a nanofiltration device for home use.
Adsorption Involving Rare Earth Compounds
Nomura et al (90-3, 88-6) have developed and patented a high
efficiency method of selectively removing fluoride and fluoride
compounds at low concentrations in water by adsorption onto
hydrated rare earth oxides or insoluble hydrated rare earth
salts, particularly hydrous rare earth phosphates or hydrous rare
earth fluorides, effective within the pH range of 2 to 7. The
adsorbent can be prepared as a filtration cake, a dried powder
or fabricated as a pre-formed mold upon a carrier porous
material, preferably a polymeric organic (88-6).
The mechanism of adsorption, similar to that by activated
alumina, is by anion exchange between the hydroxyl groups on the
adsorbent and fluoride or fluoride compounds in solution The
percentage of removal of fluoride increases abruptly at pH 7 or
lower, approaching 100% rapidly as pH decreases. The percentage
removal of other common anions, such as chloride, nitrate, and
sulfate, is considerably lower and the pH at which these anions
are optimally adsorbed is at pH ranges less than 4. As a result
fluoride is selectively adsorbed. Below pH 2, the adsorbent
itself is highly soluble, while above pH 7 adsorption capacity
decreases significantly.
The selectivity for fluoride is reflected in data published in
Nomura, et al's patent application (88-6). In a pH 5 solution
containing equal concentrations of fluoride, chloride, nitrate
and sulfate, fluoride was selectively adsorbed by a ratio range
of 100 to 1000 over chloride, 200 to 5000 over nitrate, and 30 to
200 over sulfate. Here selectivity is defined (using chloride as
an example) by the following:
22
-------
IFlads ICllaq
where: [X] -= concentration of species X in solution (irunol/L)
[X]ads = concentration of species X in adsorbent
(meq/g of adsorbent)
Desorption of fluoride and subsequent regeneration of the
adsorbing media is accomplished by exposure of the media to high
alkalinity water (preferably pH 12 or above) to permit
replacement of fluoride ions by hydroxyl ions on the adsorbing
media. The desorption and regeneration process is similar to
that for activated alumina treatment.
Since their initial patent application, Nomura, et al have
maximized adsorption capacity by developing a carrier bead of
polyolef inic resin bearing an adsorbent of hydrous cerium oxide
powder. Optimal adsorption occurs in the 2 to 5 pH range.
Sodium hydroxide is used to regenerate the adsorbent media, with
calcium fluoride precipitating from solution upon addition of
calcium chloride (90-3) .
Nomura and his co-workers compared the adsorptive capacities
of activated alumina and hydrous cerium oxide by running tests
with influent containing a fluoride concentration of 100 mg/L.
The water treated with activated alumina was adjusted to pH 6
(optimal for activated alumina adsorption) , while water treated
by hydrous cerium oxide was adjusted to pH 3. Results indicate
that hydrous cerium oxide is nearly six times as effective as
activated alumina in removing fluoride when compared on the basis
of mole of fluoride adsorbed per mole of metal in the adsorbent:
Adsorption Capacity
mg fluoride/ mole fluoride/
g adsorbent mole metal
Ce02 1.6 H2O 105 1.13
A1203 69 .20
It should be noted however that Nomura, et al have pursued
this technique primarily to treat wastewater and that the
influent fluoride concentration (100 mg/L) of the water used for
testing comparison far exceeds fluoride concentrations normally
encountered in the United States.
While their patent application indicates their process is
economical, no cost data is presented in their published
material. Attempts to contact the two editors of the text
containing Nomura, et al's publication were unsuccessful. Nomura
and his co-workers reside in Japan.
23
-------
Disposal
Each of these treatment technologies generates a waste stream
of sludge or brine which requires an appropriate disposal
strategy which meets regulatory requirements since each of these
waste streams contains concentrated quantities of removed
contaminants. Several disposal strategies exist, two or more
applicable to a given treatment technology (following modified
from Gumerman, et al, 84-1):
Disposal strategy
sludge
laaoon
Treatment technology
activated alumina
bone char
reverse osmosis
electrodialysis reversal
alum coagulation x
lime softening x
nanofiltration
sanitary sand evaporation
sewer beds ponds
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Disposal into a sanitary sewer system is a realistic strategy
only if a sewer line is available and a centralized treatment
facility can operationally handle the increase in sludge. Slowly
releasing the sludge over a period of time (equalizing
discharge), would enable the centralized treatment facility to
more efficiently respond to the added sludge volume, and reduce
the probability of producing sudden changes in pH, temperature,
dissolved oxygen concentration, or other physicochemical
characteristics of the system.
The State may also approve sludge discharges to surface
waters, or sludge injections into deep wells when contaminant
concentrations are low. In either case, equalizing the discharge
is recommended.
Sludge lagoons and evaporation ponds are frequently designed
to accommodate several years of waste before removal of
precipitated solids to a permanent disposal location is required
(84-1). The economic feasibility of sludge lagoons and
evaporation ponds is dependent upon the availability and cost of
land in the vicinity of the treatment facility, as well as the
24
-------
evapotranspiration rate of the locale. In southeastern Virginia,
where the average annual precipitation rate, 43 inches per year
(90-2), exceeds the average evapotranspiration rate, 21.5 inches
per year (90-2), these strategies are not applicable. By
contrast, these options are favorable in desert locations of
Arizona where annual precipitation is low, evapotranspiration is
high, and land is abundant and relatively inexpensive.
25
-------
MORE ON POINT-OF-USE AND POINT-QF-ENTRY
POU treatment techniques, and purchasing and distributing
bottled water are both capable of reducing groundwater fluoride
levels below the MCL. Currently, EPA does not accept these
options as methods of attaining long-term compliance with the
federal fluoride regulations but does permit the regulatory
agency (usually the State) to require a public water system to
use these options as a condition of granting an exemption or
variance from the fluoride MCL.
The following discussion intends to provide some insight into
the POU and POE options based upon two reports published in 1990
and recent discussions with several contacts, including personnel
at various EPA offices. The short time frame of this current
research prevented the gathering of detailed information
regarding the bottled water option.
The final volatile organic compound rule (R91-1) defines
specific criteria which a public water system must meet in order
to receive an exemption or variance using these options. This
rule was promulgated on January 30, 1991, to become effective
July 30, 1992. A synopsis of these criteria follows:
Public water systems that use POU/POE devices as a
condition for obtaining an exemption or variance must:
1-maintain the POU/POE devices
2-obtain the approval of a monitoring plan which
ensures the devices provide health protection
equivalent to that provided by central treatment
3-apply effective technology and maintain micro-
biological safety
4-provide the State with certification of performance
5-consider the potential for heterotrophic
bacteria and provide a means to assure that the
microbiological safety of the water is not
compromised
6-assure the State that the devices are sufficient
in number, and that they are properly installed,
maintained, and monitored
Public water systems that use bottled water as a
condition for receiving an exemption or variance must:
1-develop a monitoring program for State approval
2-receive certification from the bottled water
company that the supply is taken from an
approved source
3-assume full responsibility for providing a
sufficient supply of bottled water to every person
supplied by the public water system via door-to-door
delivery
26
-------
POU/POE devices both require a relatively small capital
investment when compared with central treatment. While operation
and maintenance costs on a per gallon basis are usually higher
for these devices than for central treatment, POU devices treat
only drinking (and possible cooking) water which is usually about
10% of the total household water usage. As a result overall POU
operation and maintenance costs can be smaller than that for
central treatment, especially for small systems.
Some of the disadvantages of using POU/POE devices include:
the potential for bacterial growth on the surface of the media
within the treatment units; inability to optimize process
parameters; and difficulties associated with monitoring, testing,
and servicing.
Two recently published studies, one sponsored by EPA, the
other performed by the National Sanitation Foundation, address
the problems associated with POU/POE drinking water treatment.
Seventy-two under-the-sink model reverse osmosis POU units
were installed and monitored in the village of San Ysidro, New
Mexico in an EPA-sponsored study (90-5). The local groundwater
source contains leachate from geothermal activity and therefore
is high in mineral content. At the time of the study the
groundwater exceeded EPA primary MCLs for arsenic and fluoride,
and secondary MCLs for iron, manganese, chloride, and total
dissolved solids. Eighteen months of operational and_maintenance
data were collected and evaluated. Study results indicate:
* POU drinking water treatment is an effective, economical,
reliable, and viable alternative to central treatment
for the community
* A POU treatment system requires a more time-intensive
monitoring program than that for central treatment
* A POU treatment system requires special municipal
regulations regarding customer responsibilities, and
water utility responsibilities
* POU systems require special considerations from regulatory
agencies to determine appropriate methods for record
keeping, monitoring, and testing frequencies
* A POU treatment system requires customers to permit access
for monitoring, testing, and maintenance of equipment
* A POU reverse osmosis unit can reduce raw water arsenic and
fluoride levels which exceed the EPA MCL to below regulatory
limits
* A POU reverse osmosis unit can reduce raw water iron,
manganese, chloride, and total dissolved solids levels
which exceed the EPA secondary MCL to below those regulatory
limits
* While the monthly treatment cost on a per gallon basis is
about 3 times that of a central treatment facility, the
actual household treatment cost is less than half since POU
27
-------
*
*
devices treat only that small portion of household water
demand used for drinking and cooking
During the testing a few RO-treated water samples tested
positive for coliforms. An investigation revealed that all RO
units had been installed with the RO drain connected directly to
the kitchen sink drain without an air gap. This was strongly
suspected to be causing the positive coliform test results.
After the air gap problem was corrected, all coliform test
results were negative.
The National Sanitation Foundation study pursued the viability
of using POE treatment to reduce radium levels in the town of
Bellevue, Wisconsin (90-8). Results from this study indicate
that:
* Residents did not perceive POE devices as a treatment
technology
* Residents were unwilling to permit equipment inspections
more than twice per year
The legal issues that will be most difficult to address
involve access and entry to private property for
maintenance, inspection, and monitoring
Although the capital costs may be favorable when compared
with central treatment or connection to an existing system,
the monitoring, maintenance and administrative costs
associated with POE treatment would be much greater than
other possible options
Due to its proximity to a central treatment facility in Green
Bay, the city of Bellevue is currently deciding whether to
connect to the Green Bay system. The study reveals a general
public uneasiness regarding decentralized treatment.
People contacted during this study had varying opinions
regarding decentralized treatment. Many people concurred with
some of the findings of the two previously mentioned studies
specifically that decentralized treatment is effective but
presents monitoring, testing and maintenance problems since
entrance to the consumer's property is always required for POU
devices, and usually required for POE devices. Many contacts
were also concerned about contaminant breakthrough once the
device's media had become saturated. Should breakthrough occur
undetected, the consumer is potentially susceptible to larger
contaminant doses than is contained in the raw water due to
contaminant concentration in the device's media. A contact from
the state of Arizona's drinking water office pointed out that
monitoring a few devices is not effective since it assumes that
devices at each residence respond identically. This may not be
the case.
Other contacts, however argued that, while POU/POE devices may
28
-------
not be preferred over centralized treatment, they are sometimes
the only cost effective alternative for small systems. A contact
from the Rural Water Association, for example, had visited a
number of sites using POE devices managed by a contracted
"circuit-rider", or system monitor, in the Arizona desert and
felt the monitoring system worked well. He indicated that, for a
POE treatment system to be viable, the state regulatory agency
should specify that a contract must exist between the home owner
and the PWS for reasons of monitoring, testing, and maintenance
so that access to the customer's property is assured. He feels
in general that management must be more creative and evaluate the
trade offs between centralized and decentralized treatment. When
this contact was informed regarding the natural fluoride
contamination currently being experienced by small PWSs in
southeastern Virginia, he indicated that the Virginia office of
the Rural Water Association would be available for assistance
should EPA be interested in pursuing a pilot decentralized
program in the state.
One contact from the National Sanitation Foundation maintains
that EPA should be working with the small public water systems in
determining optimal monitoring procedures and in sharing
decentralized system cost information. The State of California,
according to a contact in the State's Department of Health
Services, has wholeheartedly adopted the philosophy of the
National Sanitation Foundation. The State of California contact,
however indicated that California, Iowa, and Wisconsin are the
only states with laws regarding regulation of POU/POE devices.
The King's Point Subdivision in the City of Suffolk, Virginia
is expected to be a test site for POU/POE fluoride removal. When
contacted in late July 1991 the HQ EPA coordinator for the Low-
cost Innovative Small System Initiative indicated that requests
for proposal (RFPs) should have been disseminated by early August
1991. He expected that a mix of POU and POE devices would
probably be installed. The coordinator further indicated that
POU/POE technology is progressing. Some systems, for example are
currently capable of on-line real-time assessment of treatment
plant operations. A New Hampshire vendor, for example, indicates
that his company has developed a bone char POE device for arsenic
removal which is automatically backwashed.
Subsequent comments submitted by the coordinator for the Low-
cost Innovative Small System Initiative (91-12) indicate that
award announcements for the King's Point Subdivision project are
expected sometime in December 1991. A December 1991 telephone
call to the consulting firm hired to coordinate the project
indicates that the State of Virginia and the Subdivision are
currently reviewing the proposals submitted by four bidders. The
firm is encouraging the acceptance of all four proposals.
An EPA Cincinnati contact indicated that he, in conjunction
29
-------
with the Water Quality Association, recently polled between 250
and 300 POU/POE manufacturers nationwide to assess currently
available products. While many manufacturers were reluctant to
furnish effectiveness data, some significant information was
collected.
Another HQ EPA contact ventured to predict that POU/POE
devices will one day be considered BAT but probably not for many
years, citing the legal and managerial problems previously
mentioned in this report.
It should be noted that in extreme cases POU/POE may be the
only option available to small public water systems. A case in
point is an Indiana public water system servicing about 6500
people. An EPA Region V contact indicated that the system has
been producing groundwater source drinking water with fluoride
levels consistently in the upper 4 mg/L range for the past 13
years. A contractor hired by the PWS to assess treatment options
determined that he could provide neither centralized activated
alumina nor centralized reverse osmosis (the two BAT choices) at
any reasonable cost. When the beleaguered system pursued
connecting to another water system it discovered that such action
required approval by Canada and other states bordering the Great
Lakes watershed. The case has yet to be resolved and is awaiting
action by the US Department of Justice.
30
-------
TREATMENT COSTS
While nine available treatment techniques have been
identified, we propose to explore related costs for only six of
these as central treatments (activated alumina, reverse osmosis,
electrodialysis, alum coagulation, lime softening, and
nanofiltration), and two as POU treatments (activated alumina and
reverse osmosis). Bone char and rare earth adsorption costs are
not available in the literature. Anion exchange costs were not
explored due to the uncertain nature of its application for
defluoridation. Activated alumina and reverse osmosis are the
only techniques for which POU/POE defluoridation treatment is
currently available. However only detailed POU costs are
available in the literature. POE costs are therefore not
included below.
The following discussion and development of treatment
technology costs is largely based upon cost analysis performed by
Culp/Wesner/Culp Consulting Engineers (Gumerman, et al, 84-1).
Under EPA contract, Culp/Wesner/Culp undertook an extensive two
and a half year study in 1981 to determine costs associated with
unit processes capable of removing contaminants included in the
National Interim Primary Drinking Water Regulations. The study
(84-1), performed during the period September 1981 to January
1984, considered cost data for 45 centralized and five POU/POE
treatment techniques.
The discussion and development of disposal strategy costs,
while partially based upon the Culp/Wesner/Culp study, draws also
upon a 1989 feasibility study performed by Engineering-Science,
Inc. (89-7) which studied four west Texas public water systems
experiencing excess fluoride levels and developed cost
projections for remediation using centralized activated alumina
and reverse osmosis at each location.
Costs related to nanofiltration are derived from a 1989 journal
article (89-8) and telephone conversations with a researcher from
the University of Arizona and his marketing representative.
Comments submitted by the Office of Ground Water and Drinking
Water (OGWDW), HQ EPA, (91-12), indicate reluctance on the part
of the OGWDW to rely heavily upon the Culp/Wesner/Culp model for
small system costs. According to OGWDW, the model, which was
developed under contract to the EPA Office of Research and
Development, does not necessarily accurately reflect the method
small systems employ to solve water treatment issues. Instead of
hiring consulting engineers to design and build treatment
facilities, as the Culp/Wesner/Culp model suggests, smaller
systems tend to rely upon pre-engineered complete systems which
are not custom designed.
31
-------
Cost Categories
Centralized treatment facilities incur costs which can be
separated into four categories: Construction, operation and
maintenance, chemical, and disposal. Construction costs demand a
one-time consumer expenditure whereas the other three cost
categories require annual consumer expenditures. POU/POE
technologies incur only construction, and operation and
maintenance costs.
Culp/Wesner/Culp developed construction cost data by
segregating costs into eight principal components which were
selected to aid in subsequent cost updating: excavation and site
work, manufactured equipment, concrete, steel, labor , pipe and
valves, electrical equipment and instrumentation, and housing.
The subtotal of the costs of these components includes the cost
of material and equipment purchase and installation, and a
subcontractor's overhead and profit. A 15% contingency allowance
is then added to the subtotal to determine the total process
cost.
It should be noted that Culp/Wesner/Culp construction cost
data do not represent capital cost. To convert the construction
cost data into capital cost data requires adjustments for the
following, where appropriate:
' j
- overall site work, interface piping, roads
- general contractor's overhead and profit
- engineering
- land
- legal, fiscal, and administrative costs
- interest during construction
It is also important to realize that Culp/Wesner/Culp
construction cost data do not include costs associated with
constructing disposal facilities or installing pumps. While it
is extremely likely that an existing public water system already
has pumps in place (an exception might be a PWS with an artesian
well water supply), it is extremely unlikely that a PWS which is
not currently treating for a contaminant would already have a
sanitary sewer, sand bed, evaporation pond, or sludge lagoon in
place.
Culp/Wesner/Culp developed operation and maintenance costs
for three separate components: energy, maintenance material, and
labor. The energy component includes process and building
electrical energy, and diesel fuel where appropriate.
Maintenance material costs include those for replacement parts
but does not include chemical costs. Labor requirements include
both operational and maintenance costs.
Chemical costs are incurred by all centralized treatment
32
-------
facilities. Included are costs for coagulants, such as alum and
lime, adsorbers, such as activated alumina, and acids and
alkalies, typically sulfuric acid and sodium hydroxide,
respectively.
Disposal costs are associated with the various fluoride
treatment disposal strategies, including sludge lagoons, sanitary
sewers, sand beds, and evaporation ponds. All centralized
treatment facilities incur such costs. While technically spent
POU/POE activated alumina canisters require disposition, canister
contents are regenerated and/or disposed off-site. Labor costs
associated with delivery of spent canisters to off-site
regeneration/disposal facilities are included in operation and
maintenance costs.
Cost Standardization
In nearly all categories, Culp/Wesner/Culp based costs upon
standardized Bureau of Labor Statistics (BLS) or Engineering
News-Record (ENR) indices, or, in the case of electricity, upon
the current cost per kilowatt hour (kWh).
While all costs are in December 1983 dollars, the
Culp/Wesner/Culp report provides the means by which cost figures
can be updated.
Each of the eight construction cost components is associated
with either a Bureau of Labor Statistics (BLS) or Engineering
News-Record (ENR) standardized cost index.
Operation and maintenance costs are segregated into the three
individual components: energy, maintenance material, and labor.
Since the report provides energy requirements in Kwh/yr, applying
current cost of electricity quantifies energy costs. Maintenance
material costs are based upon the Producer Price Index for
Finished Goods. Operation and maintenance labor costs can be
updated using the ENR skilled labor index. Culp/Wesner/Culp
applied labor costs of $11.00 per hour for 1983 operation and
maintenance costs.
The Culp/Wesner/Culp report does not explicitly identify
specific BLS or ENR indices for chemical costs. However a series
of BLS chemical indices were selected by this researcher for cost
updating purposes.
33
-------
Construction Cost Component
Excavation and site work
Manufactured equipment
Concrete
Steel
Labor
Pipe & Valves
Electrical
Housing
Index
ENR Skilled Labor index
BLS General Purpose
Machinery and Equipment
(Code 114)
BLS Concrete Ingredients
and Related Products
(Code 132)
BLS Steel Mill Products
(Code 1017)
ENR Skilled Labor index
BLS Miscellaneous
Gen'l Purpose Equipment
(Code 1149)
BLS Electrical Machinery
& Equipment
(Code 117)
ENR Building Cost index
Operation and Maintenance
Cost Component
Energy
Maintenance materials
Labor
Index
Current electricity rate
in $/Kwh
BLS Producer Price index
for Finished Goods
ENR skilled labor index
34
-------
Chemical Cost Component
sodium hydroxide
carbon dioxide
sulfuric acid
alum
soda ash
sodium hexametaphosphate
lime
Index
BLS Sodium Hydroxide
(Code 2812-3)
BLS Carbon Dioxide
(Code 2813-3)
BLS Sulfuric Acid
(Code 2819-3)
BLS Other Aluminum
Cmpnds (Code 2819-671)
BLS Sodium Cmpnds
(Code 2819-7A)
BLS Other Sodium Phos-
phates (Code 2819-739)
BLS Alkali Earth Metal
Cmpnds (Code 2819-9A)
Disposal costs present a much more complex problem. For the
purposes of developing comparative costs, this researcher has
decided to update only those disposal costs associated with
disposal to an existing sanitary sewer. While this is a gross
generalization, several factors influence this decision:
1-The sanitary sewer disposal strategy is the only strategy
common to all of the centralized treatment technologies
under consideration. It was therefore felt that updating
sewage disposal costs would be most useful in general.
2-Since two or more disposal strategies apply to each
treatment technology, quantifying disposal costs
would require updating a large volume of data.
3-Culp/Wesner/Culp construction costs for disposal to
sanitary sewers is based upon sludge flow rate in
gallons per day. Knowing the ratio of average sludge
flow rate (in gallons per day) to the average plant
flow rate (in gallons per day) would permit a
relatively straight forward computation of sanitary
sewer disposal costs. It is felt that these ratios
can reasonably be estimated. For activated alumina
it is well established that the waste flow is
between 1 and 4% of plant flow (89-7, 85-1, 84-1, 80-4).
The average ratio for RO for the four PWSs researched
in the west Texas study (89-7) was 13%. Considering
the relatively low water recovery of RO when compared
with other treatment technologies, this figure should
35
-------
be an upper bound and can be conservatively applied
to the other treatment technologies.
4-By contrast, Culp/Wesner/Culp costs for disposal to sludge
lagoons is based upon lagoon volume, and to sand beds and
evaporation ponds upon surface area. Determining the
size of these disposal facilities based upon average plant
flow rate is a complex site-specific computation. Costing
these disposal strategies is therefore complex.
5-The short three month duration of this current
research precludes exploring all possible options.
Culp/Wesner/Culp determined costs for disposal to sanitary sewers
based upon a 1982 survey of 40 wastewater treatment plants (82-
3). No standardized indices were used. This researcher selected
the BLS index for General Purpose Machinery and Equipment for
cost updating purposes:
Disposal Cost Component Index
disposal BLS General Purpose
Machinery & Equipment
(Code 114)
The Culp/Wesner/Culp costs for disposal to sanitary sewers is
given for a wide range of groundwater suspended solids
concentration, with costs increasing as suspended solids
concentration increases. Costs were updated assuming a suspended
solids concentration of 2,000 mg/L as a conservative average.
It should be noted that disposal to an existing sanitary sewer
is the least costly of all disposal strategies. Comments
received from the Office of Ground Water and Drinking Water
(OGWDW) HQ EPA, (91-11) indicate that OGWDW has updated cost
figures (1990) for several disposal options. The author of this
current report was unaware of this data at the time the draft
report was written. The short time frame set aside for the final
version of this report precluded investigation into disposal cost
data available from OGWDW.
Updating Costs
The intent of this current report is to update these cost
figures based upon current dollars.
Culp/Wesner/Culp based costs associated with ENR indices upon
the 8 December 1983 index values. To develop current costs for
ENR categories, it is necessary to divide the current skilled
labor cost index, and current building cost index (values
contained in 91-8) by their respective 8 December 1983
36
-------
counterparts (values contained in 83-2). The resulting factors
are then multiplied by the Culp/Wesner/Culp 1983 costs to
generate 1991 costs.
Culp/Wesner/Culp based costs associated with BLS indices upon
the December 1983 index values. To develop current costs for BLS
categories, it is necessary to divide the current index values
for each of the various BLS indices by their respective December
1983 counterparts (both sets of values supplied by Philadelphia
BLS office personnel). The resulting factors are then multiplied
by the Culp/Wesner/Culp 1983 costs to generate 1991 costs.
The only remaining cost category is for electricity. Per
Philadelphia area BLS personnel, the June 1991 electricity cost
is $.134 per kwh. Applying this rate to Culp/Wesner/Culp
electricity requirements generates 1991 electricity costs.
Computation of the cost factors follows:
37
-------
ENR Categories
Skilled labor index
Building cost index
Genl Purpose Machinery
& Equip (Code 114)
Concrete Ingredients &
Related Prods (Code 132)
Steel Mill Products
(Code 1017)
Misc Genl Purpose
Equipment (Code 1149)
Electrical Machinery °
& Equip (Code 117)
Producer Price Index for
Finished Products
Sodium Hydroxide
(Code 2812-3)
Carbon Dioxide
(Code 2813-3)
Sulfuric Acid
(Code 2819-3)
Other Aluminum Compounds
(Code 2819-671)
Sodium Compounds
(Code 2819-7A)
Other Sodium Phosphates
(Code 2819-739)
Alkali Earth Metal Cmpnds
(Code 2819-9A)
!2 Dec 83
3674
2405
2)
r
5
Is
.72
.56
BLS
Dec 83
101.9
101.5
103.0
101.1
105.3
102.3
86.3
124.5
94.2
100.0
102.8
101.4
101.5
38
29 Jul 91
4475.35
2757.18
Categories
Jun 91
128.1
119.0
109.8
136.7
120.9
121.9
167.5
118.4
104.8
131.0
112.4
113.2
114.3
cost facto
1.218
1.146
cost factor
1.257
1.172
1.066
1.352
1.148
1.192
1.941
.951
1.113
1.310
1.093
1.116
1.126
-------
SUMMARY OF CONSTRUCTION COSTS
A summary of construction costs for both central and POU
defluoridation treatment follows. Total household cost (THC) is
based upon an average of 3 people per household (HH) using 150
gallons per day (gpd) each.
Most costs have been derived by updating costs developed by
Culp/Wesner/Culp (84-1) by using cost factors generated in this
current report. Cost estimates for the nanofiltration technology
were provided by a nanofiltration researcher and his marketing
representative and may or may not contain all of the cost
components considered by Culp/Wesner/Culp for the other
technologies. The smaller of the two POU RO costs is derived
from the EPA-sponsored San Ysidro, New Mexico experience.
Costs associated with constructing disposal facilities were
not explored due to the short duration of this study, and are
therefore not included in the following construction costs.
Average plant flow rates are in units of thousands of gallons
per day.
Central Treatment
THC f$ one-time cost per household)
Avg Plant
Flow Rate
1.5
2.5
7
8
10
15
30
40
46
50
63
72
100
150
180
216
300
406
430
500
560
720
>750
Number
of HH
3
6
16
18
22
33
67
89
100
110
140
160
220
330
400
480
670
900
960
1,100
1,250
1,600
>1,700
AA
562
384
293
168
158
142
130
RO ED
12,100
9,000
3,100
Alum
Lime Nano
640
4,400
2,400
3,100
1,246
1,000 1,100
730
700
680
560
650
720
600
280
180
160
130
490
230
170
430
340
39
-------
POU Treatment
•—THC fS one-time cost per household'
Activated Alumina Reverse Osmosis
407-597 417_ 547
The updated construction costs indicate that POU treatment is
more cost effective than central treatment for PWSs servicinq
less than approximately 100 to 200 households (300 to 600
consumers). For systems servicing greater than about 200
households (600 consumers) which have a realistic option of
disposing to an existing sanitary sewer, central defluoridation
treatment becomes increasingly more cost effective than POU
Should disposal to an existing sanitary sewer not be a realistic
option, a greater number of households is required for central
defluoridation treatment to be more cost effective than POU
40
-------
SUMMARY OF ANNUAL COSTS
A summary of total annual costs for both central and POU
defluoridation treatment follows. Costs represent the sum of
operation and maintenance, chemical, and disposal cost
components. Total household cost is based upon an average of 3
people per household (HH) using 150 gallons per day (gpd) each.
Most costs have been derived by updating costs developed by
Culp/Wesner/Culp (84-1) by using the cost factors generated in
this current report. Cost estimates for the nanofiltration
technology were provided by a nanofiltration researcher and his
marketing representative and may or may not contain all of the
cost components considered by Culp/Wesner/Culp for the other
technologies. The smaller of the two POU RO costs is derived
from the EPA-sponsored San Ysidro, New Mexico experience.
Comments from the Office of Ground Water and Drinking Water
(OGWDW), HQ EPA, (91-11) indicates that strict reliance upon the
Culp/Wesner/Culp report (84-1) for approximating costs based upon
average and design—flow rates will not reflect EPA's best
estimates, contending that OGWDW is developing a new flow regime
which will result in sharply decreased estimates for small system
treatment costs. OGWDW has been in the process of revising the
estimation of household usage as a function Of flow rate for the
past two years, using not only the Culp/Wesner/Culp report, but
also a document which is referred to as the "Fluoride Design
Manual" (84-6). A December 1991 conversation with an employee
from the OGWDW, however, indicated that OGWDW's revisions are
still in progress, and that the average daily water usage per
person used by this researcher in this current study (150 gallons
per day) is reasonable. Since the OGWDW anticipates updating
fluoride technology and cost data, contacting that office at some
future time and applying their cost estimates to this current
study may result in more refined cost estimate for smaller
systems.
It should be noted that the raw water fluoride concentrations
of many PWSs exceed the primary MCL by only a few mg/L (see chart
on page 14 of this report). For these systems the blending of
treated and untreated water can result in a product which meets
the primary MCL at a lower cost than treating all of the raw
water. While this is a valid option for many PWSs, the short
duration of this study precluded developing costs related to this
option.
Disposal costs assume disposal to an existing sanitary sewer.
The short time frame of this report and other considerations
prevented the development of costs for other disposal options
(see the discussion on pages 34 and 35 of this report). It
should be noted that other disposal options would be more costly
than disposal to an existing sanitary sewer.
41
-------
AA
Total H
RO
1553-1780
603-738
OUSGrlol
ED
1395
509
d Cost (S/vrl
Alum Lime
Nano
Average plant flow rates in the following table are in units
of thousands of gallons per day.
Central Treatment
Avg Plant Number
Flow Rate of HH
1.5 3
2.5 6
8 18
10 22
15 33 599-842
30 67 311
40 89 216
46 100 194-297
50 110 374-482
72 160 162
100 220 153-198 315-414 243
150 330 113-158
216 480 153
500 1,100 158
POU Treatment
Total Household Cost ($/yr)
Activated Alumina Reverse Osmosis
252 H2 - 327
The annual central treatment costs presented above are
segregated by operation and maintenance, chemical, and disposal
costs for each treatment technology below. Total household cost
(THC) is based upon an average of 3 people per household using
150 gallons per day (gpd) each.
42
-------
Central Treatment
Activated Alumina
Avg Plant
Flow Rate
fgpd)
45,000
100,000
Avg Plant
Flow Rate
fgpd)
2,500
10,000
50,000
100,000
Avg Plant
Flow Rate
fgpd)
1,500
8,000
30,000
100,000
Avg Plant
Flow Rate
fgpdl
72,000
216,000
Avg Plant
Flow Rate
(gpd)
15,000
150,000
Avg Plant
Flow Rate
fgpdl
40,000
500,000
Number
of
H-holds
100
220
O & M
f$/qpd)
.19-. 42
.10-. 20
Chem Disp
.21 .03
.21 .03
Total
f$/gpd)
.43-. 66
.34-. 44
THC
f$/vr)
194-297
153-198
Reverse Osmosis
Number
of
H-holds
6
22
110
220
Number
of
H-holds
3
18
67
220
Number
of
H-holds
160
480
Number
of
H-holds
33
330
Number
of
H-holds
89
1,100
0 & M
(S/gpd)
2.74-3.18
1.08-1.38
.60- .84
.50- .72
0 & M
(S/gpd)
2.98
1.02
.58
.43
0 & M
f$/gpd)
.04
.02
0 & M
($/qpd)
1.21-1.71
.13- .19
Chem Disp
($/gpd) ($/gpd)
.15 .12
.15 .11
.12 .11
.09 .11
Electrodialysis
Chem Disp
0 .12
0 .11
0 .11
0 .11
Alum Coagulation
Chem Disp
f$/gpd) (S/gpd)
.21 .11
.21 .11
Lime Softening
Chem Disp
.01-. 05 .11
.01-. 05 .11
Nanofiltration
0 & M & Chem Disp
f$/qpd) f$/qpd)
.37
.24
.11
.11
Total
f$/gpd)
3.01-3.45
1.34-1.64
.83-1.07
.70- .92
Total
($/gpd)
3.10
1.13
.69
.54
Total
($/gpd)
.36
.34
Total
1.33-1.87
.25- .35
Total
f$/gpd)
.48
.35
THC
f$/vr>
1553-1780
603- 738
374- 482
315- 414
THC
($/vr)
1395
509
311
243
THC
f$/vr)
162
153
THC
($/vr)
599-842
113-158
THC
($/vr)
216
158
43
-------
™ uPdated annual costs indicate that POU treatment is more
cost effective than central treatment for PWSs servicinq less
than approximately 100 to 200 households (300 to 600 consumers)
For systems servicing greater than about 200 households (600
consumers), which have a realistic option of disposing to an
existing sanitary sewer, central defluoridation treatment becomes
increasingly more cost effective than POU. Should disposal to an
existing sanitary sewer not be a realistic option, a greater
number of households would be required for central defluoridation
treatment to be more cost effective than POU.
44
-------
MISCELLANEOUS NOTES REGARDING COST CALCULATIONS
BLS Indices
The value of Bureau of Labor Statistics index 2812-3, Sodium
Hydroxide, is not available for the period January 1981 through
November 1985. In December 1980 the value of the index was 100.0
and in December 1985 it was 86.3 with the value falling
throughout 1986. For the purposes of updating costs we have
conservatively assumed the December 1983 value to be 86.3.
The value of BLS index 2819-671, Other Aluminum Compounds, is
not available prior to June 1987, when it was 100.0 and remained
100.0 throughout most of 1988. For the purposes of updating
costs we have assumed the December 1983 value to be 100.0.
Activated Alumina
Regarding chemical costs, an economy of scale is apparent in
these figures. Annual sulfuric acid cost requirements for a
473,000 gallon per day facility may actually be cheaper than that
for an 11,900 gpd operation since the larger facility can buy
sulfuric acid in quantities of tons in lieu of gallons.
Alum Coagulation
Regarding construction costs, Culp/Wesner/Culp provides the
following conversion between settling surface area and average
plant flow rate (84-1, p 395):
surface area x rise rate = avg plant
flow rate
(ft x ft ) fcrod/ft x ft) fgpdl
70 1,030 72,100
140 1,540 216,000
250 1,730 432,000
370 1,560 576,000
740 1,560 1,152,000
Regarding chemical costs, researchers have determined that 250
mg/L of alum are required to reduce fluoride concentrations from
3.5 mg/L to 1.5 mg/L, and 350 mg/L to reduce concentrations from
3.5 mg/L to 1.0 mg/L (78-2). For cost calculations we will
conservatively assume application of 350 mg/L.
The following sample calculation will indicate how to
determine annual chemical costs for a 72,000 gal per day
facility:
45
-------
350 mg 72,000 gal 365 davs 3.785 L 2.205xl(r6lb = 76.800 Ib alum
L day year gal mg yr
Multiplying this figure by current alum costs per Ib yields
annual chemical costs. Culp/Wesner/Culp used a figure of $.15
per Ib of alum in Dec 1983 dollars.
Reverse Osmosis
Costs associated with the San Ysidro, New Mexico POU RO study
have been updated using appropriate BLS and ENR indices and
included in the construction cost summary and operation and
maintenance cost summary. Since the San Ysidro report was
completed in November 1988, we are assuming all costs were in
November 1988 dollars.
Construction costs include manufactured equipment (BLS General
Purpose Machinery and Equipment, Code 114) and labor (ENR Skilled
Labor Index). Development of construction cost factors follows:
Nov 1988 Jun 1991 29 July 1991 cost factor
BLS code 114 115.1 128.1 1.113
ENR skilled labor 4133.54 4475.35 l!o83
Applying these factors to San Ysidro cost of $290 for equipment
(RO unit) and $36 for labor (installation), yields summer 1991
values of $323 and $40, respectively. It should be noted that
the San Ysidro costs for RO units resulted from receiving a
sizeable manufacturer's discount since 80 units were bought at
once.
About 67% of the operation and maintenance costs associated
with the San Ysidro experience were for labor. To update San
Ysidro's 1988 costs we have chosen therefore to simply apply the
ENR Skilled Labor Index cost factor of 1.083 as developed above
to San Ysidro's actual monthly operation and maintenance costs of
$8.60 per household. This yields a summer 1991 value of $9.31
per household per month, or $112 per household per year in summer
1991 dollars.
46
-------
APPLICATION TO VIRGINIA PUBLIC WATER SYSTEMS
Areas With Excessive Fluoride
Information contained in the Federal Reporting Data System
(FRDS II) was used to determine geographic areas in Virginia
which experience groundwater fluoride concentrations which exceed
the EPA MCL. Fifty-one Virginia PWSs were so identified.
Twenty-nine of these had concentrations which also exceeded the
unreasonable risk to health (URTH) level of 5.0 mg/L, the highest
being 6.60 mg/L.
Virginia Public Water Systems with Excess Fluoride
# of PWSs # of PWSs # of PWSs
exceeding exceeding exceeding
County/City 4.0 ma/L 5.0 mq/L 6.0 mg/L Highest
Isle of Wight Cnty 10 2 0 5.13
James City Cnty 1 0 0 4.41
Southampton Cnty 4 2 2 6.28
Chesapeake City 2 0 0 4.33
*Suffolk City 34. 25 2. 6.60
51 29 4
source: FRDS II: Summary Violations and Related Enforcement
* while technically a city, Suffolk has incorporated what
was formerly Nansemond County
With the exception of the one PWS in James City County which is
just north of the James River on the York-James Peninsula, the
other PWSs all are located in the southeastern corner of
Virginia, in an area bounded by the James River to the north, the
North Carolina border to the south, and Virginia Beach to the
east.
Fluoride-rich groundwater apparently occurs in geographic
"pockets" and normally does not effect groundwater on a regional-
scale. For example the North Carolina state water agency's
regional office in the eastern part of the state reports that no
PWSs in the area are in violation of the fluoride MCL,
concentrations being in the 2.0 to 3.5 mg/L range. A Gates
County North Carolina water supply operator whose wells are about
20 to 25 miles from Suffolk Virginia, for example, indicates that
fluoride levels in his wells are about 3.1 mg/L.
Hydrogeoloaic Research of Southeastern Virginia
The reasons for the high fluoride groundwater concentration in
southeastern Virginia groundwater supplies are unclear. Over the
47
-------
past few years, however, the United States Geological Survey, has
been performing a series of hydrogeologic studies of the Virginia
Coastal Plain, of which southeastern Virginia is a part (see for
example, 90-2, 89-1, 89-2, 88-1, 81-1). Some of this research
relates to major ion transport, including fluoride.
Recently Published USGS Research
The most comprehensive of these USGS studies is a 1988 report
(88-1) which defined hydrogeologic characteristics for the water-
table aquifer and seven confined aquifers and intervening
confining units in southeastern Virginia. The groundwater system
was modeled based upon known prepumping water levels, that is,
prior to 1891. Using the model, pumping conditions were
simulated to reflect current pumpage rates. Model results were
consistent with current known water level data, therefore
validating the model. USGS then tested the response of the
groundwater flow system to increased pumping and determined that
water level decline would be substantial, resulting in
considerable well interference and degradation of water quality.
By testing system response with various pumping scenarios, USGS
determined that year-round pumping at a constant rate would
prevent extreme water level declines normally experienced during
the seasonally dry three month summer period.
This USGS study divides each of the eight aquifers of the
model area into 4,784 geographic squares, each measuring 1.75
miles on a side (3 square miles per square), and assigns
hydraulic properties, such as transmissivity, hydraulic
conductivity, and specific yield, to each square, based upon
values quantified in the field, the laboratory, or published
literature. Isocline maps are included in the report, as well as
maps which demonstrate the effects of increased pumping
simulation, and the resulting creation of, or increase in, cones
of depression.
Such analysis is included for the shallower aquifers which are
considered lacking in sufficient water to support a PWS, but
provide good quality potable drinking water with fluoride levels
below the MCL. As a result, this USGS report represents a
potentially valuable water resources management tool which could
be used to optimize pumping of good quality water from shallower
aquifers by small PWSs.
Current Research
A contact from the USGS Water Resources Division, Richmond
office, is currently in the early stages of a review process in
which he is mapping the major ions in the major water producing
aquifers of the area. This review may take several months to
complete.
48
-------
Another contact from the USGS Richmond office is currently
writing two reports regarding the hydrogeologic characteristics
of the aquifers of the Virginia Coastal Plain. One of these
reports discusses the results of water flow and ion transport
(including fluoride) modeling which USGS has performed on the
specific area in southeastern Virginia which is experiencing the
excess fluoride problem.
The second report is concerned with the decrease in hydraulic
conductivity of the Virginia Coastal Plain from the fall line in
the west to the ocean in the east. Hydraulic conductivity is a
property of the porous media consistency of a particular aquifer.
When multiplied by aquifer thickness, the resulting quantity,
transmissivity, provides a measure of an aquifer's water yield.
This report will provide transmissivity values for all aquifers
in the Virginia Coastal Plain area, including those shallower
aquifers which have excellent water quality in general, including
low fluoride levels, but are often considered to have
insufficient yields to support PWSs in the area.
These reports may be available for public review starting in
the fall 1991.
A graduate researcher at Old Dominion University in Norfolk
Virginia, is exploring the mineralogy of the coastal region to
specifically determine the source of groundwater fluoride. His
research has indicated that sodium bicarbonate water worldwide
tends to be fluoride-rich, a correlation existing between
concentration of sodium bicarbonate and fluoride solubility. The
southeastern Virginia fluoride-rich PWSs are in fact in a
hydrogeologic band of sodium bicarbonate-rich water.
USGS personnel indicated there are no anthropogenic fluoride
sources in southeastern Virginia.
Defluoridation Costs
The number of consumers serviced by the 51 Virginia PWSs with
fluoride concentrations exceeding the EPA MCL (4.0 mg/L) is
highlighted below:
# of # of % of total
consumers PWSs # of PWSs
< 300 33 65
300 - 600 15 29
601 -1080 _3_ 6
51
Before we discuss the means by which these PWSs can most cost
effectively defluoridate, it should be noted that, of the four
disposal strategies, only sanitary sewers and sand beds can be
49
-------
considered as realistic options for southeastern Virginia since
sludge lagoons and evaporation ponds are viable options only in
relatively dry locales with high evapotranspiration rates, in
southeastern Virginia, where the average annual precipitation
rate, 43 inches per year (90-2), exceeds the average
evapotranspiration rate, 21.5 inches per year (90-2), these
disposal strategies are unrealistic.
As indicated earlier in this report, POU is more cost
effective than central defluoridation treatment for PWSs
servicing less than approximately 100 to 200 households (300 to
600 consumers). POU defluoridation is therefore more cost
effective for at least 65% and possibly as high as 94% of the 51
Virginia PWSs with excessive fluoride groundwater concentrations.
This cost analysis is based upon central treatment disposal to
existing sanitary sewers. This current research has not
determined however whether any of the 51 PWSs are located in
close proximity to an existing sanitary sewer. Should a PWS be
required to construct either a sanitary sewer or a sand bed, a
greater number of households would be required for central
defluoridation treatment to be more cost effective than POU.
It is possible that PWSs of a given size might more cost
effectively comply with federal fluoride regulations by
exercising other compliance options such as: developing a new
well which produces water which meets the MCL; connecting to or
hauling water from a PWS which produces water which meets the
MCL; or purchasing and distributing bottled water which meets the
MCL. The short time frame of this current research, however, has
prevented this researcher from exploring costs related to these
options.
50
-------
CONCLUSIONS
1. On 2 April 1986 EPA promulgated the final fluoride rule
(R86-1), establishing an enforceable primary MCL of 4.0 mg/L
"to protect against crippling skeletal fluorosis", and a
non-enforceable secondary MCL of 2.0 mg/L "to protect
against objectionable dental fluorosis". The regulation
also identified activated alumina and reverse osmosis as
BAT. This is the currently applicable federal fluoride
regulation.
2. Some prominent national medical, dental and public health
groups did not support EPA's inclusion of fluoride as a
contaminant to be regulated by the Primary Drinking Water
Regulations. Representatives from these organizations still
maintain opposition.
3. The most recent comprehensive toxicological analysis of the
effects of fluoride was published by the Public Health
Service in February 1991. PHS concluded there is no
link between fluoride and cancer, and that crippling
skeletal fluorosis is extremely rare in the United States.
EPA review of the PHS and other documents is currently in
progress.
4. In 1985 about 185,000 people nationwide were serviced by
public water systems with fluoride concentrations exceeding
4 mg/L, 17,000 people by those exceeding 6 mg/L, and 3,000
people by those exceeding 8 mg/L. Some of these public
water systems have since pursued other drinking water
sources in order to avoid defluoridation costs. While the
total number of PWSs with excessive fluoride has decreased,
this report has not identified whether the total number of
consumers exposed to excessive fluoride in drinking water
has changed.
Comments from the Office of Ground Water and Drinking Water
(OGWDW), HQ EPA (91-12), indicate that OGWDW has estimates
regarding population exposure to fluoride at various
concentrations and that this information will be updated
concurrent with the next review of the fluoride rule.
5. Public water systems faced with excess fluoride in their
groundwater drinking water supply currently have only
four options to permanently meet fluoride regulations:
construction of central treatment; developing a new well
which meets the fluoride regulations; connecting to or
hauling water from a public water system which is meeting
the fluoride regulations; or installing point-of-entry
(POE) devices. In most cases these compliance options are
relatively expensive operations. Small systems are, in
effect, penalized by the principle of economy of scale in
51
-------
constructing and operating central treatment facilities and
in developing new wells.
This report has not identified costs related to connecting
to or hauling water from a public water system which is
already complying with federal fluoride regulations.
Developing costs associated with drilling new wells and
connecting to compliant systems, in conjunction with costs
developed in this report for other compliance options, would
more completely determine the best compliance option which a
public water system of a given size should pursue in
order to meet federal fluoride regulations.
6. Point-of-use treatment, and purchasing and distributing
bottled water are not currently acceptable as long-term
options to meet fluoride regulations. For small systems
these are oftentimes the only compliance options which are
affordable.
7. The effectiveness of POU/POE activated alumina and reverse
osmosis -defluoridation techniques has been proven. Most
experts agree that legal and managerial problems relating
to monitoring, testing, and maintenance preclude POU/POE
as BAT for fluoride. Some of the operational, legal, and
managerial problems associated with POU/POE have been
addressed by various reports and studies, particularly,
the San Ysidro, New Mexico EPA-sponsored research, and
solutions to some of these problems have been proposed.
8. On a per gallon basis, monthly treatment costs for POU/POE
technologies are substantially higher than that for central
treatment. However, while POE technology processes
100% of household water demand, POU technology requires
treatment of only about 10% of such demand. As a result,
POU monthly household treatment costs are less than that
for both central treatment and POE devices.
9. One time construction costs for POU techniques are
substantially less than construction costs for central
treatment facilities, especially for small water systems.
10. EPA Cincinnati is currently in the process of writing an
RFP to hire a contractor to assess performance of the
San Ysidro test.
11. POU/POE technology is still in a state of development but
some systems are incorporating techniques approaching state
of the art.
12. Construction and annual treatment costs updated for this
report indicate that POU defluoridation treatment is more
cost effective than central treatment for public water
52
-------
systems servicing less than about 100 to 200 households
(300 to 600 consumers). For public water systems servicing
greater than about 200 households (600 consumers), which
have a realistic option of disposing to an existing
sanitary sewer, central defluoridation becomes increasingly
more cost effective than POU. Should disposal to an
existing sanitary sewer not be a realistic option, a greater
number of households would be required for central
defluoridation treatment to be more cost effective than POU.
13. The HQ EPA Low-cost Innovative Small System Initiative
anticipates that a mix of POU and POE technologies will be
used at the King's Point Subdivision site in Suffolk
Virginia to assess the possibilities of defluoridation by
decentralized technologies. As of the writing of this
report the State of Virginia and the Subdivision were in
the process of reviewing the bids submitted by four
contractors.
14. A New Hampshire vendor is currently exploring the
effectiveness of a POE bone char device to remove fluoride.
The device's ability to effectively remove arsenic has been
demonstrated. It is known that arsenic and fluoride are
similarly removed, in fact competitively, with arsenic
removal dominating fluoride removal. A proven arsenic
removal technique could potentially remove fluoride,
especially in a groundwater source which has small arsenic
concentrations.
15. Regarding bone char as a treatment technique, little
research has been done to determine optimal processing
parameters, though researchers have developed simple
operating techniques using bone char for use by illiterate
Third World communities. A recently published World Health
Organization study indicates that bone char adsorption
capacity is comparable to that of activated alumina.
Activated alumina is BAT for fluoride. No cost analyses
regarding bone char are currently available.
16. A combined alum/lime small-scale defluoridation treatment
known as the Nalgonda technique has been developed for use
by illiterate people in rural communities in India.
17. The nanofiltration treatment technique has been developed
since the April 2, 1986 promulgation of the final fluoride
rule. Researchers indicate that the technology requires
less energy and produces a substantially smaller waste
stream than reverse osmosis. Reverse osmosis is considered
best available technology (BAT) for fluoride. Home use
nanofiltration devices are currently being developed.
18. A defluoridation technique using hydrous cerium oxide has
53
-------
recently been developed and patented. While the technique
has been used predominantly to treat wastewater, it may
be effective for treating groundwater. No cost figures are
currently available.
19. Fluoride concentrations in excess of the EPA MCL (4.0 mg/L)
effect groundwater-source public water systems throughout
the continental United States. Fluoride-rich groundwater
occurs in geographic "pockets" and does not normally effect
groundwater on a regional-scale. Fluoride in groundwater is
predominantly a naturally occurring phenomena.
20. On-going United States Geological Survey research is
quantifying major hydrogeological and water chemistry
parameters of the Virginia Coastal Plain on an aquifer
by aquifer basis using modeling nodes separated by as
few as two miles. This data represents a potentially
valuable water resources management tool which could be used
to assist local public water systems in selecting well
sites. Some of the reports which USGS is currently
preparing may be available for public view as early as
fall 1991.
21. While 51 Virginia public water systems use groundwater
drinking water sources which exceed the EPA fluoride MCL
(4.0 mg/L), only four of these water supplies have
concentrations exceeding 6.0 mg/L, the highest being
6.60 mg/L.
22. POU defluoridation is more cost effective than central
treatment for at least 65% and possibly as high as 94% of
the 51 Virginia public water systems with fluoride
concentrations exceeding the EPA MCL. Should any of these
systems be required to construct either a sanitary sewer or
a sand bed, the number of systems for which POU is more cost
effective would increase. This report has not identified
the proximity of existing sanitary sewers to any of the 51
public water systems.
23. It is possible that some of the 51 Virginia public water
systems with excessive groundwater fluoride concentrations
might comply more cost effectively with federal fluoride
regulations by either developing a new compliant well,
connecting to or hauling water from a compliant public
water system, or purchasing and distributing compliant
bottled water. This report has not identified costs
associated with these three options.
54
-------
RECOMMENDATIONS
1. EPA could profitably gain by continuing efforts to determine
the number of consumers exposed to excessive fluoride at
various levels in drinking water in order to determine risk
factors associated with fluoride exposure in the United
States today.
(See conclusion 4)
2. EPA could profitably gain from efforts to develop costs
associated with drilling new wells, connecting to compliant
systems, and purchasing and distributing compliant bottled
water to determine the best compliance option which a public
water system of a given size should pursue in order to meet
federal fluoride regulations.
(See conclusion 5 and 23)
3. EPA could profitably gain by continuing to address the
operational, legal and managerial problems associated with
POU/POE techniques, as revealed in the San Ysidro, New
Mexico research and other POU/POE studies. Resolving these
problems could lead to eventual inclusion of POU/POE
defluoridation as BAT.
(See conclusions 6 through 12, and 22)
4. EPA could profitably gain by continuing to stimulate
industry development of POU/POE techniques by encouraging
industry to engage in the Low-cost Innovative Small System
Initiative and other creative and alternative programs. The
San Ysidro experience should assist EPA in honing RFP
requirements to achieve better industry response.
(See conclusions 10, 11 and 13)
5. EPA could profitably gain from efforts to assess the
effectiveness of bone char as both a central and POU/POE
treatment technique.
(See conclusions 14 and 15)
6. EPA could profitably gain from efforts to assess the
effectiveness of the Nalgonda technique as a small-scale
defluoridation option.
(See conclusion 16)
7. EPA could profitably gain from efforts to assess the
effectiveness of nanofiltration as both a central and
POU/POE defluoridation option.
(See conclusion 17)
8. EPA could profitably gain from efforts to assess the
effectiveness of adsorption by cerium oxide as a
defluoridation option.
(See conclusion 18)
55
-------
9. EPA could profitably gain by working in conjunction with the
United States Geological Survey in assisting public water
systems in selecting well drilling sites.
(See conclusions 19 and 20)
56
-------
REFERENCES
Regulations:
R91-1 Federal Register, Part II, Environmental
Protection Agency, 40 CFR Parts 141,142, and 143
National Primary Drinking Water Regulations;
Final Rule, January 30, 1991
R86-1 Federal Register, Part II, Environmental
Protection Agency, 40 CFR Parts 141, 142 and 143
National Primary and Secondary Drinking Water
Regulations; Fluoride; Final Rule, April 2, 1986
R85-1 Federal Register, Part II, Environmental
Protection Agency, 40 CFR Part 141 Fluoride;
National Primary Drinking Water Regulations;
Proposed Rule, May 14, 1985
Journals, Technical Articles, Texts and Correspondences:
91-1 Regeneration by Surface-Coating of Bone Char Used for
Defluoridation of Water, by J. Christoffersen, M.
Christoffersen, Larsen & Holler, Water Resources, v25, n2,
1991
91-2 Review of Fluoride Benefits and Risks. Report of the Ad Hoc
Subcommittee on Fluoride of the Committee to Coordinate
Environmental Health and Related Programs, conducted by the
National Toxicology Program of the Public Health Service, US
Department of Health and Human Services, February 1991
91-3 Completed Questionnaire for Potential Public-Private
Partnership Demonstration Projects submitted by the Public
Utilities Manager, Town of Colchester, Connecticut to EPA
Region I. Provided by George Mollineaux, P3 contact point,
EPA Region I
91-4 Fax from Chuck Botdorf, Yuma Proving Grounds, 10 July 1991
91-5 Fax from Chuck Botdorf, Yuma Proving Grounds, 15 July 1991
91-6 Information package re: HMR Black (Bone Char) submitted by
Pristine Filter Corporation, Melvindale Michigan, contact:
Mike Szczepanik, 15 July 1991
91-7 Revised Draft Workgroup Version of Variance and Exemption
Rule Preamble and Regulations, originated by Office of
Ground Water and Drinking Water, WH550E, HQ EPA, cover
letter dated 16 May 1991
57
-------
91-8 Engineering News-Record, 29 July 1991
91-9 Producer Price Indexes Data for April 1991, US Dept of
Labor, Bureau of Labor Statistics, May 1991
91-10 Fax from Andrea Jones, US Dept of Labor, Bureau of Labor
Statistics, Philadelphia, 14 August 1991
91-11 Review Comments on Draft Fluoride Report, letter from
James M. Conlon, Director, Drinking Water Standards
Division, Office of Ground Water and Drinking Water,
HQ EPA, October 4, 1991
91~12 Comments on the NNEMS Intern Report on Fluoride Control
Options for Small Water Systems, letter from David W.
Schnare, Office of Ground Water and Drinking Water, HQ
EPA, September 20, 1991
90-1 Environmental Pollution Control Alternatives; Drinking
Water Treatment for Small Communitiesf Center for
Environmental Research Info, Cincinnati Ohio,
EPA/625/5-90/025, April 1990
90-2 Conceptualization & Analysis of Ground-Water Flow
System In the Coastal Plain of Virginia & Adjacent
Parts of Maryland and North Carolinar by Harsh &
Laczniak, USGS Professional Paper 1404-F, 1990
90-3 Removal of Fluoride Ion From Wastewater bv a Hydrous
Cerium Oxide Adsorbent, by Nomura, Imai & Miyake, in
Emerging Technologies in Hazardous Waste Management,, edited
by Tedder & Pohland, 1990
90-4 Hazardous Waste Chemistry. Toxicology and Treatment by
Stanley E. Manahan, 1990 '
90~5 Point-of-Use Treatment of Drinking Water in San Ysidro. NM
by Rogers, EPA/600/S2-89/050, March 1990
90-6 Dose Determination and Carainocfenicity Studies of Sodium
Fluoride in Crl:CD-l Mice and CrlrCD (Soraaue Davley^ RT?
Rats, sponsored by Proctor & Gamble, June 28, 1990
90~7 NTP Technical Report on the Toxicology and Carcinoaenesis
Studies of Sodium Fluoride in 344/N Rats and B6C3F1 Mice
(Drinking Water Studies!. DHHS, NIH Report No. 90-2848, 1990
90~8 Feasibility Study; Use of POE Treatment for Radium Reduction
in Bellevue. Wisconsint by Stevens and Anderson, National
Sanitation Foundation, Ann Arbor, Michigan, March 1990
89-1 Hvdroaeoloaic Framework of the Virginia Coastal Plainr by
58
-------
Meng & Harsh, USGS Professional Paper 1404-C, 1989
89-2 The Occurrence & Geochemistry of Salty Ground Water in the
Northern Atlantic Coastal Plain, by Meisler, USGS
Professional Paper 1404-D, 1989
89-3 Fluoride in Drinking Water-Evaluation of Health Effects.
letter from Michael Cook, Director, Office of Drinking Water
(WH-550), HQ EPA, and Renate Kimbrough, MD,Director, Health
and Risk Capabilities, 21 February 1989
89-4 Nalgonda Technique-A Process for Removal of Excess Fluoride
from Water, by Nawlakhe & Bulusu, Water Quality Bulletin,
v!4, n4, October 1989
89-5 Test of Nanofilter Method of Treating Recharged Municipal
Effluent. Cluff, et al, February 1989
89-6 Test of Nanofilter Method of Treating Central Arizona
Project Waterf Cluff, et al, March 1989
89-7 Summary Report; Fluoride Reduction Feasibility Studies of
Four West Texas Water Systems, prepared for USEPA,
Municipal Facilities Div, prepared by Engineering-Science
Inc., Austin Texas, December 1989
89-8 Membrane Softening; A Treatment Process Comes of Age, by
Conlon & McClellan, Journal AWWA, November 1989
88-1 Hydrogeology & Analysis of the Ground-Water Flow System in
the Coastal Plain of Southeastern Virginia, by Hamilton &
Larson, USGS Water-Resources Investigations Report 87-4240,
1988
88-2 Title unknown, by Chaturvedi, Pathak & Singh, Appl. Clay
Sci., V3, p 337, 1988
88-3 A Defluoridator for Individual Households, by Phantumvanit,
Songpaisan & Moller, World Health Forum, v9, 1988
88-4 Point-of-Use Treatment of Drinking Water in San Ysidro. NM.
by Rogers, in fulfillment of Cooperative Agreement No.
CR-812499-01 by the Village of San Ysidro under the
sponsorship of the USEPA, covering the period July 1985
through January 1988, completed November 1988
88-5 Proceedings; Conference on Point-of-Use Treatment of
Drinking Water, chaired by Tom Sorg, Drinking Water Research
Division, USEPA, Cincinnati Ohio, EPA/600/9-88/012, June
1988
88-6 United States Patent Number 4,717,554, Process for
59
-------
Adsorption Treatment of Dissolved Fluoriner by Nomura,
Imai, Ishibashi & Konishi, 5 January 1988
88-7 Engineering News-Record, 24 November 1988
87-1 Removal of Arsenic From Bedrock Wells bv Activated Aluminar
by Kessler, masters' report submitted to Department of Civil
Engineering, College of Engineering, Northeastern
University, Boston Massachusetts, December 1987
87~2 City of Fort Myers Water Treatment Process Testing and
Evaluation Program, Boyle Engineering Corporation
87-3 Controlling Arsenic. Fluoride, and Uranium by Point-of-Use
Treatment, by Fox & Sorg, EPA Cincinnati, Journal AWWA
October 1987
86-1 Removing Dissolved Inorganic Contaminants from Water, by
Clifford, Subramonian & Sorg, Environmental Science &
Technology, v20, nil, 1986
86-2 Fluorosis with Crippling Skeletal Deformitiesr by
Moudgil, et al, Indian Pediatri, 23(10): 767-73, 1986
86-3 Groundwater Pollution Control, by Canter & Knox, 1986
85-1 Defluoridation of Drinking Water in Small Communities, by
Bellen & Anderson, National Sanitation Foundation, Sep 85
85~2 Authiaenic Fluorite in Dolomitic Rocks of the Floridan
Aquifer, by Cook, Randazzo & Sprinkle, Geology, v!3, 1985
85-3 Management of Point-of-Use Drinking Water Treatment Systems.
by Bellen, Anderson & Cottier, National Sanitation
Foundation, under EPA contract R809248010, post-September
1985, undated
85-4 Point-Of-Use Treatment for Removal of Inorganic
Contaminants, by Clifford, Proceedings Fourth Domestic Water
Quality Symposium, Chicago II, December 1985
85-5 Letter from Douglas Ginsburg, Administrator, Office of
Information and Regulatory Affairs, Office of Management &
Budget, to Lee Thomas, Administrator, EPA, Wash DC,
26 April 1985
85-6 Drinking Water Criteria Document on Fluoride. EPA Office of
Drinking Water, Contract 68-03-3279, 1985
85-7 Fluoridation of Water and Cancer: A Review of the
Epidemioloaical Evidence, by Knox, 1985
60
-------
84-1 Estimation of Small System Water Treatment Costs, by
Gumerman, Burris & Hansen, Culp/Wesner/Culp Consulting
Engineers, under contract no. 68-0303093, under EPA
sponsorship, March 84
84-2 Fluoride in the Ground Water of Northeastern Ohio, by
Corbett and Manner, Groundwater, v22, nl, Jan-Feb 1984
84-3 A Mobile Drinking Water Treatment Research Facility for
Inorganic Contaminants Removal: Design. Construction, and
Operation, by Clifford & Bilimoria, under contract to the
EPA Municipal Environmental Research Laboratory, Cincinnati
Ohio, EPA-600/S2-84-018, March 1984
84-4 Guidelines for Drinking-Water Quality. Volume 1.
Recommendations. World Health Organization, Geneva, 1984
84-5 Letter from Surgeon General Koop to William D. Ruckelshaus,
23 January 1984
84-6 Design Manual:^Removal of Fluoride from Drinking Water
Supplies by Activated Alumina, by Rubel, prepared for the
Drinking Water Research Division, Municipal Environmental
Research Lab, EPA Cincinnati, August 1984
84-7 Membrane Separation Processes, by Applegate, Chemical
Engineering, 11 June 1984
84-8 Effect of Severe Dental Fluorosis on the Oral Health of
Adults, by Eklund, Ismail, Burt, and Calderone, November
27, 1984
83-1 Survey of Inorganic Contaminants, by Clifford, Sorg &
Frank, Proceedings AWWA Seminar on Control of Inorganic
Contaminants, 1983
83-2 Engineering News-Record, 22 December 1983
82-1 World Health Organization Guidelines for Drinking Water
Quality, 1982
82-2 Ad hoc Committee Report on Dental Fluorosis. Draft Report
to the Chief Dental Officer. Public Health Service, by
Albertini, Bock, Confrancesco, Driscoll, Small & Clark,
21 July 1982
82-3 Sewer Charges for Wastewater Collection and Treatment-A
Survey. Water Pollution Control Federation, 1982
81-1 Ground Water Resources of the Four Cities Area. Virginia.
by Siudyla, May & Hawthorne, Planning Bulletin 331,
Commonwealth of Virginia, State Water Control Board,
61
-------
Nov 1981
81~2 The Equilibrium Fluoride Capacity of Activated Alumina.
Determination of the Effects of PH and Competing Ionsr by
Singh & Clifford, of the University of Houston, prepared
for the EPA Municapal Environmental Research Lab,
Cincinnati, May 1981
8 °-l Recommended Dietary Allowances. National Academy of
Sciences, 1980
80-2 Drinking Water and Health. v3, National Academy of
Sciences, 1980
8°-3 Water Chemistry, by Snoeyink & Jenkins, 1980
80-4 Pilot Study of Fluoride and Arsenic Removal from Potable
Water, by Rubel, EPA-600/2-80-100, August 1980
79-1 The Removal of Excess Fluoride from Drinking Water by
Activated Aluminar by Rubel & Woosley, Water Technology/
Quality, AWWA Journal, January 1979
78-1 Removal of Excess Fluoride from Drinking Waterr by Rubel
& Woosley, for the USEPA Office of Water Supply, Technical
Report EPA 570/9^78-001, January 1978
78~2 Treatment Technology to Meet the Interim Primary Drinking
Water Regulations for Inorganicsr by Sorg, EPA Cincinnati,
Journal AWWA, v70, pp 105-112, February 1978
75-1 Defluoridation of Wastewaterr by Zabban & Helwick,
Proceedings of the 30th Industrial Waste Conference, Purdue
University, 1975
75-2 Efficacy and Safety of Fluoridation. American Medical
Association, Council on Foods and Nutrition, Chicago, 1975
74-1 Lime. CaCl.., Beat Fluoride Wastewater. by Rohrer, Water and
Wastes Engineering, Nov 1974
73-1 Geological Studies. Coastal Plain of Virginia, by Teifke &
Onuschak, Planning Bulletin 83, Parts 1, 2 & 3, Commonwealth
of Virginia, Dept of Conservation & Economic Development,
1973
73-2 A History of Water Fluoridation. by Murray, Brit. Dent. J.,
134:250-4, p 299-302, 347-350, 3 April 1973
73-3 Title unknown, by Waldbott, published by C V Mosby Co,
St. Louis, 1973
62
-------
73-4 The Chemistry of Fluorine, by O'Donnell, Pergamon Press,
Elmsford, NY, 1973
72-1 Physicochemical Processes, by Weber, 1972
71-1 Biological Effects of Atmospheric Pollutants: Fluorides.
prepared by the Committee on Biological Effects of
Atmospheric Pollutants, Division of Medical Sciences,
National Research Council, National Academy of Sciences,
Washington DC, p 209-214, 1971
71-2 Radiculomvopathy in a Southwestern Indian due to Skeletal
Fluorosis. by Goldman, Sievers, and Templin, Arizona
Medicine, 28:675-677, 1971
70-1 Water Fluoridation-The Search and the Victory, by McClure,
Human Health Services (then Health, Education & Welfare),
1970
70-2 Fluoride and Dental Health, by Adler, In Fluoride and
Human Health, p 323-354, World Health Organization, Geneva,
1970
70-3 Toxic Effects of Large Doese of Fluoride, by Bhussry, in
Fluoride and Human Health, p 230-239, World Health
Organization, Geneva, 1970
68-1 A Fluoride Filter for Domestic Use, by Roche, New Zealand
Dentistry J, v64, 1968
65-1 Effects of Fluorides on Bones and Teeth, by Hodge & Smith,
in Fluorine Chemistry. v4, p337-693, New York Academic
Press, 1965
65-2 Chronic Fluoride Intoxication with Fluorotic
Radiculomyelopathy. by Sauerbrunn, Ryan, and Shaw, Ann.
Intern. Med., 63:1074-1078, 1965
62-1 Public Health Service Drinking Water Standards-1962, USPHS,
publication 956, US Government Printing Office, Washington
DC, 1962
59-1 Dana's Manual of Mineralogy, revised by Hurlbut, 1959
53-1 Medical Aspects of Fluorosis. by Shimkin, Arnold, Hawkins &
Dean, American Association Advancement Sciences, 1953
53-2 The Problem of Providing Optimum Fluoride Intake for
Prevention of Dental Caries. by Sognnaes, a Report of the
Committee on Dental Health, publication 294, p4, November
1953
63
-------
51-1 Condition of the Formation of Fluorite in Sedimentary Rocksr
by Kazakov & Sokolova, English translation by Skitsky, USGS
TEI-386, 76 pages, 1951
42-1 Domestic Water and Dental Caries, by Dean, Arnold & Elvove,
Pub. Health Rep. 57:1155, 1942
25-1 Mottled Enamel; A Fundamental Problem in Dentistry, by
McKay, Dental Cosmos, v69, p 847, 1925
64
-------
TELEPHONE CONTACTS
Note: The following contacts furnished information deemed
pertinent to this report.
HQ EPA
Mark Parrotta, Office of Drinking Water and Ground Water
Dr. Ken Bailey, Office of Science and Technology
Al Havinga, regarding POU/POE policy
Dr. David Schnare, regarding low-cost technology initiative
EPA Cincinnati Research Facility
Tom Sorg
Jim Smith
Ben Lykins
EPA Regional Offices
John Haederlie, Region I, Boston
George Mollineaux, Region I, Boston
Steve Gould, Region II, New York
Kenneth Dean, Region IV, Atlanta
Harry Pawlowski, Region V, Chicago
Chris Urban, Region V, Chicago
Bill Davis, Region VI, Dallas
Len Pardee, Region VI, Dallas
Stan Calow, Region VII, Kansas City
Melanie Abel, Region VIII, Denver
Bob Benson, Region VIII, Denver
Bruce Macler, Region IX, San Francisco
65
-------
Alan Hemming, Region, X, Seattle
National Institute of Health, Public Health Service
John Small, fluoride expert
State Drinking Water Personnel
Dan Home, Field Office Director, Virginia Dept of Health
Shahram Mohsenin, engineer, Virginia Department of Health,
Virginia Beach Regional Office
Michael Bell, North Carolina State's Regional Drinking Water
Office in eastern part of the state, Washington NC
Marvin Murray, South Carolina Drinking Water Office
John Dahl, Arizona State's Drinking Water Office
Bob Burns, California Department of Health Services
Norm Hahn, Wisconsin Department of Natural Resources
Private Research Organizations
Gordon Bellen, National Sanitation Foundation, Ann
Arbor, Michigan
Tom Stevens, National Sanitation Foundation, Ann
Arbor, Michigan
Joe Harrison, technical director, Waste Quality
Association, Lisle, Illinois
John Trax, National Rural Water Association, Washington DC
Water Utility Managers/Operators
Chuck Botdorf, Yuma Proving Grounds
Bud Eure, Gates County North Carolina Water Supply
Chester Antonick, Sue Juan Water Co., Arizona
Tom Werner, engineer in charge of water protection,
City of Suffolk, Virginia
66
-------
Wayne Johnson, head of water lab, Yuma Bureau of
Reclamation Desalting Plant
William Harrell, Director of Public Utilities, City
of Suffolk, Virginia
University Research Facilities
Paul Maciuika, graduate student researching fluoride
in southeastern Virginia groundwater, Old Dominion
University, Norfolk, Virginia
Dr. Brent Cluff, nanofiltration researcher, University
of Arizona at Tucson
Medical Profession
Dr. Mike Morgan, Oklahoma State Health Department, State
Dental Director, former Chairman of the Association of
State and Territorial Dental Directors
Dr. Lum Young, Nebraska State Health Department, State
Dental Director, former Chairman, Fluoride Committee,
Association of State and Territorial Dental Directors
Dr. Duncan Clark, American Medical Association, New York
City
Tom Reeves, Center for Disease Control, Atlanta Georgia
Dr. Mel Ringelberg, Florida State Dental Director, Chairman
Fluoride Committee, Tallahassee, Florida
Water Industry Consulting/Marketing/Production Personnel
Ron Saylor, marketing representative, nanofiltration
technology
Frank Brigano, Culligan
Bob Lake, Water Treatment Engineers, consulting firm,
Tucson Arizona
Mike & Ann Szczepanik, Ebonex Co., producers of bone black
(bone char), Melvindale Michigan
Mark Widdison, Culligan of New Hampshire
Greg Gwaltney, ECOS, consulting firm coordinating the King's
67
-------
Point Subdivision Low-cost Innovative Small System
Initiative
United States Geological Survey
Gary Speiran, Richmond Virginia office
Mike Socazio, Richmond Virginia office
Bob Mixon, Reston Virginia office
Other
Joe Carpenter Jr., private developer in Suffolk, Virginia
area
Chris Mauro, Bureau of Labor Statistics, Philadelphia PA
Andrea Jones, Bureau of Labor Statistics, Philadelphia PA
Debra Darr, Hampton Roads Planning District Commission,
Hampton Roads, Virginia
68
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - ACTIVATED ALUMINA
plant capacity (millions of
Cost Category
Excavation and sitework
Manufactured equipment
Equipment
Activated alumina
Concrete
Labor
.046
5,725
16,090
1,834
469
1,462
.102
5,725
30,042
4,061
1,406
1,827
.180
5,725
49,149
7,074
2,110
2,436
.406
5,725
63,604
15,589
2,344
3,410
aallons per day)
.552
5,725
81,077
20,174
2,930
4,019
.720
5,725
91,635
25,676
3,750
4,141
1.080
5,725
126,957
38,514
4,805
5,116
Piping & valves 7,030 8,788 8,788 11,357 17,306 17,982 27,175
Electrical instrumentation 7,347 7,347 7,347 9,184 9,184 9,758 11,021
Housing 9,970 16,502 19,367 20,513 28,421 39,422 50,309
Subtotal 49,927 75,698 101,996 131,726 168,836 198,089 269,622
Design contingencies 7,489 11,355 15,299 19,759 25,325 29,713 40,443
TOTAL *** 57,416 87,053 117,295 151,485 194,161 227,802 310,065
*** based on labor cost of $13.40 per hour
A- 1
-------
OPERATION AND MAINTENANCE COST SUMMARY - CENTRAL TREATMENT - ACTIVATED ALUMINA
avg plant act alumina
flow rate volume
(cfpd) ( cubic feet)
Regeneration
45,000
101,000
180,000
406,000
553,000
722,000
1,083,000
frequency =
31.4
70.7
126
283
385
503
754
bldna
4.5 days
268
469
603
657
978
1,635
2,291
energy
($/yr)
process
590
1,273
2,224
2,452
2,452
2,452
2,452
total
858
1,742
2,827
3,109
3,430
4,087
4,743
maintenance
material
rS/vr)
1,073
1,907
2,980
6,079
8,106
10,370
15,138
labor
($/yr)
17,152
17,152
17,849
17,849
18,546
18,546
24,683
total
cost ***
($/vr)
19,083
20,801
23,656
27,037
30,082
33,003
44,564
Regeneration frequency = 8 days
If
45,000
101,000
180,000
406,000
553,000
722,000
083,000
31.4
70.7
126
283
385
503
754
268
469
603
657
978
1,635
2,291
1
2
2
2
2
2
590
,273
,224
,452
,452
,452
,452
1,
2,
3,
3,
4,
4,
858
742
827
109
430
087
743
834
1,311
1,907
3,576
4,768
5,960
8,702
12
12
12
12
13
13
17
,248
,248
,944
,944
,641
,641
,326
13,940
15,301
17,678
19,629
21,839
23,688
30,771
Regeneration frequency = 12 days
45,000
101,000
180,000
406,000
553,000
722,000
1,083,000
31.4
70.7
126
283
385
503
754
268
469
603
657
978
1,635
2,291
590
1,273
2,224
2,452
2,452
2,452
2,452
858
1,742
2,827
3,109
3,430
4,087
4,743
715
1,073
1,550
2,622
3,457
4,410
6,198
10,398
10,398
11,095
11,095
11,792
11,792
14,539
11,971
13,213
15,472
16,826
18,679
20,289
25,480
*** based upon labor costs of $13.40 per hour and energy costs of $.134 per kwh
A- 2
-------
OPERATION AND MAINTENANCE COST SUMMARY - CENTRAL TREATMENT - ACTIVATED ALUMINA
avg plant act alumina
flow rate volume
(crpd) (cubic feet)
Regeneration
45,000
101,000
180,000
406,000
553,000
722,000
1,083,000
frequency =
31.4
70.7
126
283
385
503
754
bldncr
19 days
268
469
603
657
978
1,635
2,291
energy
($/yr)
process
590
1,273
2,224
2,452
2,452
2,452
2,452
maintenance
material
total r$/vri
858
1,742
2,827
3,109
3,430
4,087
4,743
596
834
1,192
1,907
2,384
2,980
4,172
labor
($/yr)
8,844
8,844
9,541
9,541
10,238
10,238
12,221
total
cost ***
fS/vrl
10,298
11,420
13,560
14,557
16,052
17,305
21,136
Regeneration frequency =33 days
45,000
101,000
180,000
406,000
553,000
722,000
1,083,000
31.4
70.7
126
283
385
503
754
268
469
603
657
978
1,635
2,291
Regeneration frequency =48 days
45,000 31.4 268
101,000 70.7 469
180,000 126 603
406,000 283 657
553,000 385 978
722,000 503 1,635
1,083,000 754 2,291
*** based upon
590
1,273
2,224
2,452
2,452
2,452
2,452
590
1,273
2,224
2,452
2,452
2,452
2,452
858
1,742
2,827
3,109
3,430
4,087
4,743
858
1,742
2,827
3,109
3,430
4,087
4,743
596
715
954
1,311
1,669
2,026
2,742
596
596
834
1,192
1,430
1,669
2,146
7,772
7,772
8,469
8,469
9,166
9,166
10,613
7,316
7,316
8,013
8,013
8,710
8,710
9,916
754 2,291 2,452 4,743 2,146 9,916
labor costs of $13.40 per hour and energy costs of $.134 per kwh
A- 3
9,226
10,229
12,250
12,889
14,265
15,279
18,098
8,770
9,654
11,674
12,314
13,570
14,466
16,805
-------
CHEMICAL COST SUMMARY - CENTRAL TREATMENT - ACTIVATED ALUMINA
avg plant
flow rate
fcrpd)
11,900
473,000
sulfuric *
acid
r$/vr)
535
550
sodium *
hydroxide
r$/vr)
2,017
17,100
total
cost
f$/vr)
2,552
17,650
* see discussion in paragraph entitled MISCELLANEOUS NOTES REGARDING COST CALCULATIONS
A- 4
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - ACTIVATED ALUMINA
avg plant
flow rate
fqpdl
670
1,700
3,300
6,700
16,700
33,300
66,700
167,000
333,000
667,000
avg sludge
flow rate
(cfpd)
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
total
cost ***
f$/vr)
19
38
88
175
427
855
1,760
4,300
8,500
17,100
*** assumes a suspended solids concentration on the order of 2,000 mg/L
A- 5
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - REVERSE OSMOSIS
plant capacity (gallons per day)
Cost Category 2,500 10,000 50,000 100,000 500,000 1,000,000
Manufactured equipment 25,517 37,710 87,487 154,611 571,684 1,102,892
Labor 974 1,462 1,827 3,410 9,135 17,783
Electrical instrumentation 3,674 5,281 12-,284 21,468 52,693 71,291
Housing 13,637 15,929 18,794 21,201 44,006 60,165
Subtotal 43,802 60,382 120,392 200,690 677,518 1,252,131
Design contingencies 6,570 9,057 18,059 30,104 101,628 187,820
TOTAL *** 50,372 69,439 138,451 230,794 779,146 1,439,951
*** based on labor costs of $13.40 per hour
A- 6
-------
OPERATION AND MAINTENANCE COST SUMMARY -CENTRAL TREATMENT- REVERSE OSMOSIS (Low Pressure)
avg plant
flow rate
(qpcn
bldnq
energy
($/yr)
process
maintenance
material
total
labor
($/yr)
total
cost***
($/vr)
TDS concentration = 5,000 mg/L
2,500
10,000
50,000
100,000
500,000
1,000,000
375
442
549
657
2,090
3,926
1,327
3,524
13,413
24,174
114,329
215,204
1,702
3,966
13,962
24,831
116,419
219,130
596
2,026
9,536
17,403
79,983
140,537
4,556
4,824
6,432
8,174
11,658
15,142
*** based upon labor costs of $13.40 per hour and energy cost of $.134 per kwh
6,854
10,816
29,930
50,408
208,060
374,809
A- 7
-------
OPERATION AND MAINTENANCE COST SUMMARY -CENTRAL TREATMENT- REVERSE OSMOSIS (High Pressure)
avg plant
flow rate
(crod)
TDS concentration =
2,500
10,000
50,000
100,000
500,000
1,000,000
TDS concentration =
2,500
10,000
50,000
100,000
500,000
1,000,000
TDS concentration =
2,500
10,000
50,000
100,000
500,000
1,000,000
bldnq
5,000 mg/L
375
442
549
657
2,090
3,926
8,000 mg/L
375
442
549
657
2,090
3,926
10,000 mg/L
375
442
549
657
2,090
3,926
energy
($/yr)
process
2,412
6,459
25,607
46,150
218,286
410,844
2,412
6,459
25,607
49,982
272,851
513,555
2,412
6,459
25,607
59,992
327,429
616,266
total
2,787
6,901
26,156
46,807
220,376
414,770
2,787
6,901
26,156
50,639
274,941
517,481
2,787
6,901
26,156
60,649
329,519
620,192
maintenance
material
r$/vr)
596
2,026
9,536
17,403
79,983
140,537
596
2,026
9,536
17,761
83,678
146,497
596
2,026
9,536
18,476
87,254
152,218
labor
($/yr)
4,556
4,824
6,432
8,174
11,658
15,142
4,556
4,824
6,432
8,442
12,596
16,348
4,556
4,824
6,432
9,112
13,668
17,554
total
cost***
($/vr)
7,939
13,751
42,124
72,384
312,017
570,379
7,939
13,751
42,124
76,842
371,215
680,326
7,939
13,751
42,124
88,237
430,441
789,964
*** based upon labor costs of $13.40 per hour and energy costs of $.134 per kwh
A- 8
-------
CHEMICAL COST SUMMARY - CENTRAL TREATMENT - REVERSE OSMOSIS
avg plant sodium sulfuric sodium total
flow rate hexametaphosphate acid hydroxide cost
(gpd) ($/vr) f$/vr) ($/vr) ($/vr)
2,500 145 134 97 376
10,000 558 512 388 1,458
50,000 2,232 2,037 1,514 5,783
100,000 3,460 3,116 2,329 8,905
500,000 14,954 13,579 10,093 38,626
A- 9
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - REVERSE OSMOSIS
avg plant
flow rate
(gpd)
150
390
770
1,540
3,850
7,700
15,400
38,500
77,000
154,000
avg sludge
flow rate
(gpd)
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
total
cost ***
f$/vr)
19
38
88
175
427
855
1,760
4,300
8,500
17,100
*** assumes suspended solids concentration on the order of 2,000 mg/L
A-10
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - ELECTRODIALYSIS
plant capacity
Cost Category 1
Excavation and sitework 2
Manufactured equipment 19
Concrete
Labor
Electrical instrumentation 2
Housing 8
,500
,558
,484
352
853
,755
,939
8,000
3,654
45,378
586
1,827
4,936
12,377
30,000
4,263
103,577
1,055
2,071
9,414
20,628
(gallons per day)
100,000
5,725
161,902
1,524
3,167
11,824
25,785
300,000
6,090
349,572
2,696
7,064
22,271
36,787
1,000,000
11,327
1,229,975
5,274
24,725
67,158
48,132
Subtotal 34,941 68,758 141,008 209,927 424,480 1,386,591
Design contingencies 5,241 10,314 21,151 31,489 63,672 207,989
TOTAL *** 40,182 79,072 162,159 241,416 488,152 1,594,580
*** based upon labor costs of $13.40 per hour
A-ll
-------
OPERATION AND MAINTENANCE COST SUMMARY - CENTRAL TREATMENT - ELECTRODIALYSIS
avg plant
flow rate
(crod)
TDS concentration =
1,500
8,000
30,000
100,000
300,000
1,000,000
TDS concentration =
1,500
8,000
30,000
100,000
300,000
1,000,000
TDS concentration =
1,500
8,000
30,000
100,000
300,000
1,000,000
bldncr
1,000 mg/L
268
389
791
1,072
1,997
6,553
2 , 000 mg/L
268
389
791
1,072
1,997
6,553
3,000 mg/L
268
389
791
1,072
1,997
6,553
energy
($/yr)
process
348
1,876
7,048
23,477
70,430
234,768
496
2,667
9,983
33,259
99,776
332,588
643
3,444
12,918
43,041
129,122
430,408
total
616
2,265
7,83Q
24,549
72,427
241,321
764
3,056
10,774
34,331
101,773
339,141
911
3,833
13,709
44,113
131,119
436,961
maintenance
material
f$/vr)
238
954
3,338
10,966
27,416
108,830
238
1,073
3,934
13,112
32,661
130,524
238
1,311
4,649
15,258
38,144
152,338
labor
($/yr)
3,618
4,958
6,164
7,370
8,576
14,740
4,020
5,494
6,834
8,174
9,380
16,080
4,422
6,030
7,504
8,978
10,318
17,688
total
cost***
f$/vr)
4,472
8,177
17,341
42,885
108,419
364,891
5,022
9,623
21,542
55,617
143,814
485,745
5,571
11,174
25,862
68,349
179,581
606,987
*** based upon labor costs of $13.40 per hour and energy costs of $.134 per kwh
A-12
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - ELECTRODIALYSIS
avg plant avg sludge total
flow rate flow rate cost ***
fapd) (qpd)
150 20 19
390 50 38
770 100 88
1,540 200 175
3,850 500 427
7,700 1,000 855
15,400 2,000 1,760
38,500 5,000 4,300
77,000 10,000 8,500
154,000 20,000 17,100
*** assumes suspended solids concentration on the order of 2,000 mg/L
A-13
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - ALUM COAGULATION
average plant
Cost Category
Excavation and sitework
Manufactured equipment
Concrete
Labor - installation
Piping & valves
Electrical instrumentation
Housing
72,100
3,289
39,344
4,805
7,917
1,217
2,985
21,774
216,000
5,116
56,314
7,149
10,109
3,921
3,329
29,108
flow rate
432,000
6,577
73,157
10,079
11,693
5,949
3,559
35,984
(gallons per dav)
576,000
8,891
84,848
12,775
12,667
7,571
4,018
43,548
1,152,000
14,860
161,902
22,385
24,360
11,357
6,084
59,019
Subtotal 81,331 115,046 146,998 174,318 299,967
Design contingencies 12,200 17,257 22,050 26,148 44,995
TOTAL *** 93,531 132,303 169,048 200,466 344,962
*** based upon labor costs of $13.40 per hour
A-14
-------
OPERATION AND MAINTENANCE COST SUMMARY - CENTRAL TREATMENT - ALUM COAGULATION
settling surface
area
(square feet)
avg plant
flow rate
(cfpd) bldnq
energy
($/yr)
process
maintenance
material
total r$/vr)
labor
($/yr)
total
cost ***
f$/vr)
70
140
250
370
740
72,000
216,000
432,000
576,000
1,152,000
657
1,018
1,461
1,983
3,739
442
442
670
898
1,782
1,099
1,460
2,131
2,881
5,521
298
358
447
536
1,013
1,742
2,613
2,787
3,484
4,529
3,139
4,431
5,365
6,901
11,063
***
based upon labor costs of $13.40 per hour and energy costs of $.134 per kwh
A-15
-------
CHEMICAL COST SUMMARY - CENTRAL TREATMENT - ALUM COAGULATION
avg plant total
flow rate cost *
(gpd) ($/vr)
72,000 15,091
216,000 45,273
* see discussion in paragraph entitled MISCELLANEOUS NOTES REGARDING COST CALCULATIONS
A-16
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - ALUM COAGULATION
avg plant
flow rate
fqpdl
150
390
770
1,540
3,850
7,700
15,400
38,500
77,000
154,000
avg sludge
flow rate
fcrodl
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
total
cost ***
($/vr)
19
38
88
175
427
855
1,760
4,300
8,500
17,100
*** assumes suspended solids concentration on the order of 2,000 mg/L
A-17
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - LIME SOFTENING
plant capacity (gallons per day)
Cost Category
Single Stage
Excavation and sitework
Manufactured equipment
Concrete
Labor - installation
Piping & valves
Electrical instrumentation
Housing
Subtotal
Design contingencies
15,000
4,263
41,732
1,289
17,052
7,030
9,758
10,085
91,209
13,681
150,000
7,064
62,599
O
2,930
22,168
14,061
14,006
18,794
141,622
21,243
430,000
8,161
83,339
3,750
34,104
19,063
19,516
22,691
190,624
28,594
750,000
10,231
108,353
6,915
44,335
22,578
21,697
34,380
248,489
37,273
1,000,000
11,936
130,477
8,204
53,348
62,057
30,652
37,818
334,492
50,174
TOTAL ***
104,890
162,865
219,218
285,762
384,666
*** based upon labor costs of $13.40 per hour
A-18
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - LIME SOFTENING
plant capacity (gallons
Cost Category
Two Stage
Excavation and sitework
Manufactured equipment
Concrete
Labor - installation
Piping & valves
Electrical instrumentation
Housing
Subtotal
Design contingencies
15,000
7,064
67,250
2,110
28,501
14,061
17,015
17,878
153,969
23,095
150,000
11,327
100,937
4,922
37,027
28,257
24,682
29,796
236,948
35,542
430,000
13,520
134,499
6,446
56,881
38,126
34,210
35,984
319,666
47,950
per day)
750,000
17,052
174,849
11,720
73,933
44,616
37,884
50,424
410,478
61,572
1,000,000
19,975
210,548
13,830
88,914
124,114
53,841
55,008
566,230
84,935
TOTAL ***
177,064
272,490
367,616
472,050
651,165
*** based upon labor costs of $13.40 per hour
A-19
-------
OPERATION AND MAINTENANCE COST SUMMARY - CENTRAL TREATMENT - LIME SOFTENING
avg plant
flow rate
(qpd)
bldnq
energy
($/yr)
process
total
maintenance
material
($/vr)
labor
($/yr)
total
cost***
f$/vrl
Single Stage
15,000
150,000
430,000
750,000
1,000,000
268
549
724
1,313
1,568
1,018
1,367
2,090
4,636
5,199
1,286
1,916
2,814
5,949
6,767
2,265
3,457
4,649
5,841
7,271
14,673
14,673
19,564
19,564
24,522
18,224
20,046
27,027
31,354
38,560
Two Stage
15,
150,
430,
750,
1,000,
000
000
000
000
000
1,
1,
2,
3,
523
099
434
613
136
1,
2,
3,
7,
7,
688
278
484
048
986
2,211
3,377
4,918
9,661
11,122
3,
5,
7,
9,
14,
814
722
629
655
662
19,
19,
24,
24,
29,
564
564
522
522
346
25,589
28,663
37,069
43,838
55,130
*** based upon
labor costs of $13.40 per hour1 and energy costs of $.134 per kwh
A-20
-------
CHEMICAL COST SUMMARY - CENTRAL TREATMENT - LIME SOFTENING
avg plant
flow rate
(gpd)
Single stage
15,000
150,000
Two stage
15,000
150,000
lime
f$/vrl
195
1,950
330
3,300
soda ash
f$/vr)
0
0
390
3,900
carbon
dioxide
($/vr)
22
220
87
870
total
cost
f$/vr)
217
2,170
807
8,070
A-21
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - LIME SOFTENING
avg plant
flow rate
fqpd)
150
390
770
1,540
3,850
7,700
15,400
38,500
77,000
154,000
avg sludge
flow rate
fgpd)
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
total
cost ***
($/vr)
19
38
88
175
427
855
1,760
4,300
8,500
17,100
*** assumes suspended solids concentration on the order of 2,000 mg/L
A-22
-------
CONSTRUCTION COST SUMMARY - CENTRAL TREATMENT - NANOFILTRATION
plant capacity (gallons per day)
7,000 40,000 63,000 500,000 1,000,000 2,000,000
TOTAL 10,000 60,000 78,000 475,000* 750,000* 1,250,000
* per reference 89-8; all other data derived from written material and telephone
conversations with a nanofiltration researcher and his marketing representative
A-23
-------
OPERATION & MAINTENANCE / CHEMICAL COST SUMMARY - CENTRAL TREATMENT - NANO FILTRATION
avg plant
flow rate
total cost ($/yr)
40,000 14,600 (per nanof iltration researcher/marketing rep)
500,000 120,450 (per reference 89-8)
1,000,000 248,200 (per reference 89-8)
A-24
-------
DISPOSAL COST SUMMARY - CENTRAL TREATMENT - NANOFILTRATION
avg plant
flow rate
(qpd)
150
390
770
1,540
3,850
7,700
15,400
38,500
77,000
154,000
avg sludge
flow rate
(gpd)
20
50
100
200
500
1,000
2,000
5,000
10,000
20,000
total
cost ***
($/vr)
19
38
88
175
427
855
1,760
4,300
8,500
17,100
*** assumes suspended solids concentration on the order of 2,000 mg/L
A-25
-------
CONSTRUCTION COST SUMMARY - POU - ACTIVATED ALUMINA
construction cost
cost category POU
activated alumina filter 251
canister
plastic meter box & 10 in 0-41
PVC pipe collar
PVC piping to house 0-27
faucet & fittings 54
labor - installation 49 - 146
subtotal 354 - 519
contingency 53 - 78
total 407 - 597
A-26
-------
OPERATION AND MAINTENANCE COST SUMMARY - POU - ACTIVATED ALUMINA
labor
cost category ($/yr)
sampling/testing frequency
twice per year
4 times per year
6 times per year
media regeneration
twice per year
once per year
27
80
161
frequency
27
13
once every two years 7
repairs
low
average
high
13
27
40
materials
($/vr)
24
48
72
119
60
30
12
24
36
total cost
specific
condition
51
128
233
146
73
37
25
51
76
f$/vr) ***
mid-range
condition
128
73
51
TOTAL FOR MID-RANGE CONDITION 252
*** based upon labor cost of $13.40 per hour
A-27
-------
CONSTRUCTION COST SUMMARY - POU - REVERSE OSMOSIS
construction cost
cost category POU
manufactured equipment 323* - 415
labor - installation 40* - 61
subtotal 363* - 476
contingency 54* - 71
total 417* - 547
* these figures were derived from updating costs associated with the San Ysidro, NM
experience;
see discussion in paragraph entitled MISCELLANEOUS NOTES REGARDING COST CALCULATIONS
A-28
-------
OPERATION AND MAINTENANCE COST SUMMARY - POU - REVERSE OSMOSIS
total cost ($/vr) ***
labor materials specific mid-range
cost category ($/vr) ($/yr) condition condition
sampling/testing frequency
twice per year 27 48 75
4 times per year 54 95 149 149
6 times per year 80 143 223
pre-filter and GAG contactor replacement frequency
once per year 13 24 37
twice per year 27 48 75 75
RO membrane replacement
once per year 13 89 102
once every 2 years 7 45 52 52
once every 3 years 4 30 34
repairs
low 13 12 25
average 27 24 51 51
high 40 36 76
TOTAL FOR MID-RANGE CONDITION 112* - 327
* derived from San Ysidro, NM experience;
see discussion in paragraph entitled MISCELLANEOUS NOTES REGARDING COST CALCULATIONS
of this report
*** based upon labor costs of $13.40 per hour
A-29
------- |