xvEPA
United States Water Engineering Research Laboratory
Environmental Protection Cincinnati, OH 45268
Agency
EPA/600/9-88/012
June 1988
Research and Development
Proceedings:
Conference on Point-of-Use
Treatment of Drinking Water
Cincinnati, OH
October 6-8, 1987
-------
-------
EPA/600/9-88/012
June 1988
Proceedings
CONFERENCE ON POINT-OF-USE
TREATMENT OF DRINKING WATER
CINCINNATI, OHIO, OCTOBER 6-8, 1987
Co-Sponsored by
Drinking Water Research Division
Water Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Office of Drinking Water
U.S. Environmental Protection Agency
Washington, D.C. 20460
American Water Works Association
6666 W. Quincy Avenue
Denver, Colorado 80235
In Cooperation With:
Center for Environmental Research Information
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
WATER ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
Disclaimer
The following papers have been reviewed in accordance with the U.S. Environmental Protection
Agency's peer and administrative review policies and approved for presentation and publication:
• Point-of-Entry and Point-of-Use Devices for Meeting Drinking Water Standards
• Regulatory Requirements for Point-of-Use Systems
• Microbiological Studies of Granular Activated Carbon Point-of-Use Systems
• Health Studies of Aerobic Heterotrophic Bacteria Colonizing Granular Activated Carbon
Systems
• Community Demonstration of POU Systems Removal of Arsenic and Fluoride
• POU/POE Point-of-View (Discussion by F. Bell)
The remaining papers were not prepared with U.S. EPA financial support and the contents do
not necessarily reflect the views of the Agency; therefore, no official endorsement should be
inferred.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's
land, air, and water systems. Under a mandate of national environmental laws, the Agency
strives to formulate and implement actions leading to a compatible balance between .human
activities and the ability of natural systems to support and nurture life. The Clean Water Act, the
Safe Drinking Water Act, the Resource Conservation and Recovery Act, the Federal Insecticide,
Fungicide and Rodenticide Act, and the Toxic Substances Control Act are five of the major
congressional laws that provide the framework for restoring and maintaining the integrity of our
Nation's water, for preserving and enhancing the water we drink, and for protecting the
environment from hazardous and toxic substances. These laws direct the EPA to perform
research to define our environmental problems, measure the impacts, and search for solutions.
The Water Engineering Research Laboratory is that component of EPA's Research and
Development Program concerned with preventing, treating, and managing municipal and
industrial wastewater discharges; establishing practices to control and remove contaminants
from drinking water; preventing deterioration during storage and distribution; and assessing the
nature and controllability of releases of toxic substances to the air, water, and land from
manufacturing processes and subsequent product uses. This publication is one of the products
of that research and provides a vital communication link between the researcher and the user
community.
The Conference on Point-of-Use Treatment of Drinking Water was held because of a national
interest in the application of point-of-use (POU) and point-of-entry (POE) systems to
improve drinking water quality. The role of POU/POE Systems has broadened substantially over
the past 8-10 years from use principally to improve the taste and odor of drinking water to
consideration for solving specific contaminant problems. The purpose of this conference was to
present information on various administrative and technical aspects of utilizing POU/POE
systems to solve individual and small community drinking water problems. The conference also
provided a forum for dialogue and interaction between the manufacturers, distributors, regulatory
officials, and users of point-of-use treatment technology. These proceedings document the
information presented to assist, not only those who attended the conference, but also those
who had a vital interest in the conference information but were unable to attend.
Francis T. Mayo, Director
Water Engineering Research Laboratory
in
-------
Contents
Foreword iii
Preface iv
Abstract v
Acknowledgments viii
Overview of Point-of-Use and Point-of-Entry Systems
Lee T. Rozelle, Olin Corporation 1
Point-of-Entry and Point-of-Use Devices for Meeting Drinking Water Standards
Stephen W. Clark, U.S. Environmental Protection Agency 4
Regulatory Requirements for Point-of-Use Systems
Ruth Douglas, U.S. Environmental Protection Agency 10
Control of Point-of-Use Water Treatment Devices in Canada:
Legal and Practical Considerations
Richard S. Tobin, Health and Welfare Canada 25
The Regulation of Water Treatment Devices in California
Robert F. Burns, California Department of Health Services 15
Wisconsin Regulation of Point-of-Use and Point-of-Entry Water Treatment Devices
Loretta Trapp, Wisconsin Dept. of Industry, Labor and Human Relations 18
Household Water Quality Education: The Cooperative Extension System Role
G. Morgan Powell, Kansas State University 22
Federal Trade Commission Regulation of Water Treatment Devices
Joel Winston, Federal Trade Commission 25
POU/POE Product Promotion Guidelines and Code of Ethics
Maribeth M. Robb, Water Quality Association 27
NSF's Listing Program for POU/POE DWTUs
Randy A. Dougherty, National Sanitation Foundation 31
Water Quality Association Voluntary Product Validation Program and
Voluntary Certification Program
Lucius Cole, Water Quality Association 35
Guide Standard and Protocol for Testing Microbiological Water Purifiers
Stephen A. Schaub, U.S. Army Biomedical Research and Development Laboratory
Charles P. Gerba, University of Arizona 37
Performance and Applications of Granular Activated Carbon Point-of-Use Systems
Karl Van Dyke and Roy W. Kuennen, Amway Corporation 44
Performance and Application of RO Systems
Donald T. Bray, Desalination Systems, Inc 62
VI
-------
Contents (continued)
Performance and Application of Ultraviolet Light Systems
Clyde Foust, Ideal Horizons, Inc : 69
Precoat Carbon Filters as Barriers to Incidental Microbial Contamination
P. Regunathan, W.H. Beauman, D.J. Jarog, Everpure, Inc 71
Microbiological Studies of Granular Activated Carbon Point-of-Use Systems
Donald J. Reasoner, U.S. Environmental Protection Agency 81
Health Studies of Aerobic Heterotrophic Bacteria Colonizing Granular
Activated Carbon Systems
Alfred P. Dufour, U.S. Environmental Protection Agency 84
Activated Alumina for POU/POE Removal of Fluoride and Arsenic
Robert L. Lake, Water Treatment Engineers 88
Modelling Point-of-Entry Radon Removal by GAG
Jerry D. Lowry, University of Maine, Sylvia B. Lowry, Lowry Engineering, Inc 90
Point-of-Entry Activated Carbon Treatment Lake Carmel - Putnam County
George A. Stasko, NY State Department of Health 99
Community Demonstration of POU Systems Removal of Arsenic and
Fluoride: San Ysidro, New Mexico
Karen Rogers, Leedshill-Herkenhoff, Inc 106
Florida's Funding for Contamination Correction
Glenn Dykes, Florida Dept. of Envirommental Regulation 111
Monitoring and Maintenance Programs for POU/POE •
Gorden E. Bellen, Thomas G. Stevens, National Sanitation Foundation 113
Point-of-Use and Point-of-Entry Treatment Devices Used at Superfund
Sites to Remediate Contaminated Drinking Water
Sheri L. Bianchin, U.S. Environmental Protection Agency, Region V 118
New Developments in Point-of-Use/Point-of-Entry Drinking Water Treatment
Gary L. Hatch, Ametek, Inc 129
POU/POE Point of View
Frank A. Bell, Jr., U.S. Envirommental Protection Agency 136
AWWA Viewpoint on Home Treatment Units
Jon DeBoer, American Water Works Association 138
POU/POE - Point of View - Association of State Drinking Water Administrators (ADWSA)
Barker G. Hamill, New Jersey Dept. of Environmental Protection 141
POU/POE: An Industry Perspective
Donna Cirolia, Water Quality Association 143
Point-of-Use Treatment of Drinking Water: Comments
Sue Lofgren 145
VII
-------
Acknowledgments
Numerous individuals were responsible for the success of this conference. Planning of the
program was accomplished by Tom Sorg, Kim Fox, Frank Bell, and James Smith of the U.S.
EPA and Jon DeBoer of the AWWA.
A special thanks is given to Patricia Cooke who arranged for the conference facilities and to all.
the U.S. EPA employees who helped during registration and provided other support throughout
the conference.
Sincere appreciation is extended to Lisa Moore, Carolyn McGill, and other members of the
JACA Corporation who were responsible for preregistration and preparation of the Proceedings,
and who also provided help throughout the entire Conference.
Sincere gratitude is expressed to the Water Quality Association (WQA) who sponsored and
arranged for the excellent equipment exhibit. Through the efforts of Donna Cirolia, Maribeth
Robb, and Lu Cole of the WQA, and the participating manufacturers, the equipment exhibit was
considered one of the highlights of the conference.
Last, but, not least, all the speakers and moderators are acknowledged for their excellent
presentations and the preparation of their written papers.
VIII
-------
OVERVIEW OF POINT-OF-USE AND POINT-OF-ENTRY SYSTEMS
Lee T. Rozelle
Olin Corporation
Cheshire, CT 06410
The 1986 amendments to the Safe Drinking Water
Act require that 83 contaminants must be regulated
within three years of signing the act, that is by June,
1989. EPA has set up a schedule to comply with the
amendments including the eight volatile organic
chemicals currently finalized.
The greatest regulatory burden will be on the 38,000
community systems serving less than 500 people.
When violations are incurred, modification or
installation of conventional water treatment systems
may be too costly and these communities may apply
for variances or exemptions. Although unregulated by
the Safe Drinking Water Act, the 850,000 rural
systems with two to 14 connections and an additional
9,000,000 individual rural systems, will certainly be
affected by this act.
The utilization of proven water treatment technologies
at the point-of-use/point-of-entry (POU/POE)
offers a potentially viable and cost effective method of
reduction of chemical contaminants to acceptable
levels in drinking water. In fact, in situations where
contaminants pose an unreasonable risk to health, the
option of point-of-use or bottled water is being
proposed by U.S. EPA as a condition for receiving a
variance or exemption. This would be on a temporary
basis until compliance with the regulations is
achieved. Also there are certain conditions for use
such as certification, bacterial safety, etc.
Point-of-entry is acceptable to EPA for long term
use in contaminant removal from drinking water
supplies, although it has not been given Best
Available Technology status. Again there are certain
conditions for use, similar to those of point-of-use.
A point-of-use treatment device consists of
equipment applied to selected taps used for the
purpose of reducing contaminants in water at each
tap.
A point-of-entry treatment device consists of
equipment applied to water entering a house or
building for the purpose of reducing contaminants
distributed throughout the house or building.
Point-of-use systems are commonly placed in the
following locations at the sink:
• Counter top. A counter top device normally fits
through a connection to the faucet on the sink and
rests on the counter or in the sink.
• Faucet Mounted. A faucet mounted filter is
attached directly to the end of the faucet.
• Under Sink Cold Tap. This device fits onto the cold
water line and treats all the cold water that flows
through the faucet.
• Under the Sink Line Bypass. This device taps onto
the cold water line, and after flowing through the
lines to a reservoir (in some cases), exits through a
special spigot attached to the sink.
Point-of-entry systems are placed where the
household water enters the house (but normally after
the outside outlets).
Common POU/POE technologies and their placement
are shown in Table 1. Ultraviolet radiation is also an
effective POU/POE technology for reduction of
microorganisms. Placement could include counter top
and under the sink as well as point-of-entry.
Table 1. Common POU/POE Technologies and Their
Placement
Technology
Normal Placement
Paniculate Filters
Adsorption Filters
Reverse Osmosis
Ion Exchange
Distillation
All POD Placements
POE
All POU Placements
POE
Countertop
Undersink Line Bypass
POE
Potentially All POU Placements
Countertop
Paniculate filters at the point of use normally have 3
to 60 iim (0.0001 to 0.002 in) ratings. These filters
consist of media such as spun bonded materials,
foam (molded in place), wound string, fabric,
membranes, and granular activated carbon (GAG).
The point-of-use paniculate filters are typically 25.4
-------
cm (10 in) in height and 7.6 cm (3 in) in diameter.
Point-of-entry paniculate filters resemble
household softeners with media consisting of sand,
granite, anthracite, etc.
The most common adsorption filter for contaminant
reduction in drinking water is granular activated
carbon (GAC). These filters reduce common tastes
and odors, some turbidity, residual chlorine, radon,
and many organic contaminants with varying degrees
of efficiency based on molecular structure and
equipment design. The most common design is a
cartridge containing a loose carbon bed. This bed-
type filter could also contain activated alumina as the
media for fluoride and arsenic (V) reduction. Fused
carbon filters and precoat filters also are utilized.
Precoat filters usually consist of powdered activated
carbon and/or diatomaceous earth applied to the
influent side of the filter.
Reverse osmosis (RO) is considered the "high tech"
method for reduction of dissolved solids. Currently it
is more applicable at the point-of-use than point-
of-entry. Typical RO membranes remove total
dissolved solids (TDS) with such efficiency that the
treated water may become aggressive and dissolve
metals from the water pipes; blending may be
necessary. Also, at the point entry for RO there would
be a need for a holding tank and a recirculation pump
resulting in a more expensive treatment system.
The reverse osmosis systems normally consist of a
particulate filter followed by an optional activated
carbon filter (if a chlorine sensitive membrane is
used), an RO module, a water reservoir containing a
pressurized rubber bladder (approximately 9.5 I [2-
1/2 gal] capacity), a final activated carbon filter (to
remove any residual taste and odor), and the special
spigot on the sink. Household under-the-sink units
operate efficiently at pressures between 2.8 and 4.9
kg/cm2 (40 and 70 psi) on nonbrackish raw waters
with up to 2,000 mg/I of TDS. The flow rate through
the spigot is typically between 0.03 and 0.06 l/s (0.5
and 1 gpm).
Cation exchange has been used for water softening
for over 50 years. However ion exchange can apply
to selective inorganic contaminant removal using
either cation or anion exchange resins.
Distillation has historically been known to be effective
and has been utilized for producing contaminant free
water.
The maintenance of point-of-use and point-of-
entry systems is necessary to maintain their
effectiveness for contaminant removal. The following
summarizes maintenance procedures of current
point-of-use/ontry devices utilizing proven
technologies:
• Particulate filters. Particulate filters at the point-
of-use are replaced before clogging, when slow
flow is observed. At the point-of-entry these
filters are backwashed periodically.
• Granular activated carbon. Granular activated
carbon filters must be replaced before
breakthrough df the contaminants. Many units
contain a shutoff or alarm device to indicate when
a certain volume of water has been filtered. If no
shutoff device is on the unit, the filter either
periodically replaced by a qualified dealer or, if
listed by the National Sanitation Foundation (NSF)
Standard 53, a 100 percent safety factor is used
for replacement based on volume flow. That is, if a
3,790-1 (1,000-gal) capacity is claimed, it must
be effective for 7,580-1 (2,000-gal) to pass NSF
Standard 63.
• Point-of-use RO systems require a periodic
replacement of filters and the RO modules (to
avoid loss of efficiency due to membrane fouling or
deterioration). According to NSF Standard 58, the
RO module must be replaced when the
conductivity rejection is below 75 percent or at a
value necessary to maintain drinking water
compliant with the Safe Drinking Water Act. With
proper maintenance of the prefilters, the cellulose
acetate modules are normally replaced after one
and a half to two years of service and the
polyamide modules replaced between two and four
years of service. The GAC and particulate filters
are normally replaced every six to 12 months.
• Ion Exchange. Ion exchange units are normally
regenerated with sodium chloride.
• Distillation. Distillers must be cleaned due to
scaling.
• Activated Aluminia. Activated aluminia is
regenerated by sodium hydroxide and then
acidified for adsorption.
The costs of point-of-use/point-of-entry units
and their replacement follow:
• Particulate filters typically cost between $20 and
$100.
• Granule)' activated carbon filters typically cost
between $50 and $300 with the lowest
replacement cost about $20. For the point-of-
entry the cost range of granular activated carbon
filters is $800 to '$1,000, with $200 to $400
replacement costs.
• Reverse osmosis devices vary in price. The
counter top devices range from $100 to $300.
Under-the-sink RO devices containing cellulose
acetate membranes range from $300 to $600 with
replacement cost at $50 to $60 for the CA
membrane element. Under-the-sink RO devices
containing thin film composite membranes range
from $400 to $800 with replacement cost of the
membrane element at around $100. .
-------
• Ion exchange costs vary from $300 to $2,000
depending on the resin (anion exchange resins
cost more than cation exchange resins) and type of
equipment.
• Distillation typically costs between $200 and $600.
• Ultraviolet units typically cost between $300 and
$700.
Reverse osmosis is the best technology for reduction
of inorganic contaminants considering no energy
input. It may be indispensable for lead and copper,
which may contaminate water from household pipes.
It should be remembered, however, that reverse
osmosis efficiency depends on the type of membrane
used. Cellulose acetate membranes normally do not
reject some contaminants as effectively as the newer
thin film composite membranes. For example, nitrates
are rejected up to 65 percent by cellulose acetate
membranes, but up to 94 percent by thin film
composite membranes under point-of-use
conditions.
Selective reduction of inorganic contaminants can be
carried out by cation exchange resins for ions such
as radium and barium and by anion exchange resins
for nitrates and arsenates.
Activated alumina is effective for reduction of fluoride
and arsenic. Granular activated carbon is very
effective for reduction of radon. With proper design
virtually 100 percent reduction has been observed.
Disposal and shielding remain an issue for general
use. Distillation is also effective for inorganic removal.
Granular activated carbon is known to be the most
effective and inexpensive method for removal of
organic contaminants in drinking water. It must be
remembered, however, that it is not perfect and the
adsorption capacity for various contaminants varies.
Thus, it is important to know which organic
contaminants are present and their adsorption
capacities for effective maintenance.
Reverse osmosis is not known for effective reduction
of volatile organic chemical or low molecular weight
organic contaminants. Reduction efficiency varies
based on molecular weight, charge, size, shape, and
relationship with the chemistry of the membrane.
There are indications, however, that total organic
carbon is more consistently reduced (80 to 90
percent) by reverse osmosis when compared to
granular activated carbon.
When GAC is used with RO, as in many line bypass
RO systems, organic reductions increase, specifically
for low molecular weight organics including VOCs.
This combination can result in a very effective point-
of-use device for removal of contaminants.
Several field studies have been carried out using
point-of-use/point-of-entry. In Suffolk County,
Long Island 3,000 GAC units have been used for over
four years to treat water with an average aldicarb
concentration of 87 iig/l. Based on this experience, at
a 100 pg/l influent concentration using 0.028 m3 (1 cu
ft] of GAC, the GAC filter life was calculated to be
170,325 I (45,000 gal) before breakthrough of 7 ug/l
aldicarb.
In Rockaway Township, New Jersey 12 GAC units
were used to remove concentrations from water
above 100 ng/l of TCE and 1,1,1-trichloroethane.
After 24 months of testing, no significant
concentrations were observed in the effluent (less
than 1 ug/l [1 ppb]). In Silverdale, Pennsylvania 47
GAC devices were tested using five models. With
influent concentrations of TCE above 100 pg/l (100
ppb), no significant concentrations were observed in
the effluent after 14 months of operation.
In Emmington, Illinois 63 reverse osmosis devices
were tested for removal of fluoride and high total
dissolved solids. In this one year test, the fluoride was
reduced by 86 percent (from raw water concentration
of 4.5 mg/l [4.5 ppm]) and the TDS reduced 79
percent (from a raw water concentration of 2,620
mg/l).
The actual costs of the point-of-use purchase,
operation, and maintenance in Rockaway and
Silverdale varied from $5.98 a month jn Silverdale to
$4.23 a month in Rockaway. An estimate of an
additional $1.23 per month was made for
administrative costs if used in a community of 650
customers.
In Emmington, Illinois the actual point-of-use
reverse osmosis costs were $12.48 a month. It was
estimated that if reverse osmosis was used for central
treatment, the cost to the homeowner would be
$28.50 a month.
As a comparison, bottled water used in a family of 2.8
people at 3.8 I per day (1 gpd) per person at a price
of $0.22 per liter ($0.85 per gal) would cost $67.00
per month.
The following conclusions result from the study:
• Use of proven technologies at the point-of-use
and point-of-entry is effective for reducing
contaminants from drinking water supplies.
• Reverse osmosis and distillation are most
universally effective for inorganic reduction.
• Granular activated carbon is most universally
effective for organic contaminant reduction.
• Costs for small communities appear to be attractive
particularly if these devices can be leased to the
community avoiding up front costs.
-------
Point-of-Entry and Point-of-Use Devices for Meeting Drinking Water Standards
Stephen W. Clark
U.S. Environmental Protection Agency
Washington, DC 20460
BACKGROUND
The Safe Drinking Water Act Amendments of 1986
require the Environmental Protection Agency (EPA) to
set standards for 83 contaminants by June of 1989.
These contaminants include inorganic chemicals,
radionuclides, and organic chemicals. The microbial
contaminants listed by Congress will be regulated by
requiring filtration as a treatment technique for
surface waters, and disinfection of all waters. The
chemical contaminants will be regulated by setting
maximum contaminant levels (MCLs). MCLs generally
apply at the tap and represent an achievable, safe
level of contaminants in public water supplies.
The public water supplies regulated under the Safe
Drinking Water Act include all systems serving at
least 25 people or 15 service connections. The EPA
has by regulation created three subcategories of
public water systems. Community water systems
service fifteen or more connections. There are
approximately 65,000 community water systems in
the U.S. They range in size from very small
communities to large cities like New York and
Chicago. The other major category is noncommunity
water systems, which serve at least 25 persons.
There are over 200,000 noncommunity water
systems, a category which includes restaurants,
parks, factories, and other places frequented by the
public. Nontransient, noncommunity water systems
such as schools or workplaces serve the same 25 or
more people at least six months of the year. There
are approximately 20,000 nontransient, noncommunity
water systems that will have to meet the same
standards as community water systems. The
remaining noncommunity water systems will have to
meet standards for microbial contaminants and some
acutely toxic chemicals like nitrate. The reason for the
difference is that some toxicants (e.g., fluoride)
require lifetime exposure to increase risk of diseases,
whereas acute toxicants can theoretically cause
diseases like hepatitis after one drink of contaminated
water. The majority of all kinds of water systems are
small, that is they serve less than 3,300 people or
600 service connections. Compliance is good among
large, metropolitan systems, but small systems have
historically lacked the money and the technical skill to
operate complex water treatment plants.
Recognizing the difficulty that small systems would
have complying with the many new drinking water
standards, EPA considered allowing a variety of
decentralized approaches. These approaches
included point-of-entry devices, point-of-use
devices, and bottled water.
Public comment was first sought on these
decentralized approaches in the Federal Register of
November 1985 (1). This notice proposed MCLs for
eight volatile organic chemicals as well as criteria for
the use of decentralized approaches in public drinking
water systems.
MAJOR ISSUES
Although the Federal Register notice sought comment
on these along with other issues, the U.S. EPA
decided to conduct a public hearing on decentralized
approaches in June of 1986. The three major issues
discussed at this meeting were;
• Should point-of-entry (POE) devices treating all
the water entering buildings connected to a public
water system be considered a suitable means of
compliance?
• Should point-of-use (POU) and bottled water in
addition to POE be considered suitable means of
compliance?
• Should POE, POU, or bottled water be considered
Best Available Technology (BAT) for small systems
(less than 600 persons)?
Three options were presented at this meeting. They
are discussed below.
Option 1
Consider POE to be an acceptable means of
compliance.
Explanation
Allow the application of POE devices to treat all the
water in every building for compliance purposes.
-------
Bottled water and POU could be considered as
interim means of reducing excessive risks during
emergency situations. However, POE would not be
considered BAT under this option. EPA was leaning
toward this option, at that time.
Discussion
1) From a human exposure standpoint POE could be
considered equivalent to central treatment.
2) POE treatment methods are similar to the central
treatment options that would be used for small
systems.
3) From a practical perspective there are some
differences:
a) Monitoring would have to be increased to assure
that each device is functioning properly (i.e.,
producing water meeting the MCL).
b) Operation and maintenance is much more
difficult than for a central treatment system.
c) POE is less likely to be suitable for compliance
with the microbiological standards because:
• Microbiological safety is assured by
maintaining good source waters, the
application of filtration and disinfection
technologies as appropriate (including the
maintenance of a disinfectant residual
throughout the distribution system), and
maintenance of the integrity of the distribution
system.
• Protection from acutely hazardous
contaminants (such as microbes) is critical,
and more difficult to assure in a decentralized
operation (which would naturally have less
supervisory control).
• POE devices might not be able to provide
protection equivalent to central treatment
because of these considerations.
d) Compliance for some contaminants would be
determined by multiple in-building samples for
the POE mode.
e) There might not be any cost advantage for the
POE option over for central treatment, especially
as the hydraulic capacity of the system
increases.
f) Tradition in the industry and some of the
legislative history of the Safe Drinking Water Act
suggest that the trend toward regionalization
versus decentralization.
g) Compatibility of POE devices with the central
treatment technologies currently in place or
required in the future needs to be considered.
• Without post-disinfection, GAC adsorption
POE devices would contribute
microorganisms to the water supply (as with
POU).
• In addition, this could result in exposure to
microbes via inhalation as well as by drinking.
4) Because POE would not be considered to be BAT,
EPA would not require its installation before a
variance could be granted to a water system. If a
system could install POU to gain near term benefit,
it would be allowed to do so. However, if the
system desired a variance, it would have to install
central treatment (BAT) to fulfill the statutory
conditions for variances.
5) Concern was expressed that persons may still
consume water from untreated taps of systems are
allowed to use bottled water or POU devices for
long-term compliance purposes. This is one
reason why EPA was leaning toward requiring that
all water provided to the consumer be treated.
Conditions for Choosing Option 1
1) Public water systems would have to maintain
control and responsibility for the operation and
maintenance of the POE devices.
2) A monitoring and maintenance program that
assures protection of all consumers equal to that
provided by the central treatment option.
3) Effective technology must be properly applied
including provisions for microbiological safety.
4) All consumers in every building must be protected
(i.e., have a device installed, maintained, and
adequately monitored by the responsible party).
5) The POE mode of compliance must provide
protection equivalent to that provided by central
water treatment.
Option 2
Allow POE, POU, and bottled water as acceptable
means of compliance.
Explanation
Allow POU and bottled water in addition to POE as
suitable means of compliance under defined
circumstances and criteria. None of these would be
considered BAT under this option.
Discussion
1) Since respiratory and dermal exposure have been
identified as concerns for volatile chemicals and
microbiological contaminants, then all but central
treatment or POE would be ruled out for these
substances. Under certain circumstances POE
-------
might not be acceptable for biological contaminants
(see Option 1 discussion).
2) To allow both bottled water and POU:
a) Bottled water, meeting all standards, should be
delivered to the consumers in order to be similar
to POU.
b) A special monitoring scheme for bottled water
would need to be developed.
c) Because POE would not be considered to be
BAT, EPA would not require its installation
before a variance could be granted to a water
system. If a system could install POU to gain
near-term benefits, it would be allowed to do
so. However, if the system desired a variance, it
would have to install central treatment (BAT) to
fulfill the statutory conditions for variances.
Conditions for Choosing Option 2
1) The public water system would have to maintain
control and responsibility for the operation and
maintenance of the POU or quality control over the
contents and delivery of the bottled water.
2) A special monitoring program that assures
protection of all consumers equal to that provided
by the central treatment would be required. It could
consist of application of the Part 141 monitoring
requirements.
3) Effective technology must be properly applied
including provisions for microbiological safety -
bottled water must meet the microbiological safety
standards, too.
4) All consumers in every building must be protected
(i.e., have a POU or bottled water device installed,
maintained, and* adequately monitored by the
responsible party).
S) Bottled water is not "piped water" for human
consumption, thus arguably excluding these
systems from the definition of a public water
system. To allow bottled water for drinking water
systems EPA would therefore, have to determine
that provision of bottled water, under certain
conditions, is equivalent to provision of "piped
water" by a public water system.
Option 3
Consider POE, POU, and bottled water to be Best
Available Technology (BAT) for small systems (less
than 500 persons).
Explanation
On a compound-by-compound basis, criteria would
be set by which small systems could be required to
use POE, POU, or bottled water in lieu of central
treatment prior to being granted a variance. That is,
for purposes of receiving variances to specific MCLs,
POE, POU, or bottled water would have to be
installed by small systems.
Discussion/Conditions for Choosing Option 3
1) The criteria used to determine BAT for
decentralized treatment would differ from that for
the central treatment option, and may vary by
contaminant.
2) As an example, consider criteria for the
determination of BAT for fluoride. The POU, POE,
or bottled water must be:
a) Commercially available, and capable of
satisfactorily removing fluoride from drinking
water.
b) Affordable by large metropolitan public water
systems.
c) "Best" based upon the following factors:
• Wide applicability,
• High cost efficiency,
• High degree of compatibility with other water
treatments in use or needed for the system,
and
• The ability to achieve compliance for all water
in a public water system.
3) Affordability criteria would be different for large and
small systems.
4) Central treatment would still be available.
5) The amount of space required for installation of
POE could limit the applicability of POE throughout
a system, and hence, its designation as BAT.
6) The high degree of compatibility criterion would be
considered on a compound- and technology-
specific basis.
7) EPA would need evidence that costs are
reasonable.
SUMMARY OF EPA DECISION
After considering all public comments and through a
variety of discussions at all levels of management,
EPA decided that point-of-entry devices were
suitable for compliance, but they were not BAT. It
was also decided that POU and bottled water could
be used as interim measures, but were not to be
considered BAT or a mearis of compliance.
The decentralized approaches cannot be considered
BAT because of difficulties associated with monitoring
-------
compliance and assuring effective treatment
performance in a manner comparable to central
treatment. Most of the public comments received by
EPA were against considering the decentralized
approaches BAT. The commenters cited difficulties in
controlling installation, maintenance, operation, repair,
and potential exposure through untreated taps.
However, other commenters felt that decentralized
technologies were BAT for very small systems, as
these methods were often more cost effective for
some small systems than central treatment.
In the final rule, POE and POU were not designated
as BAT because: 1) it is more difficult to monitor the
reliability of treatment performance and to control
POU and POE than for central treatment; 2) these
devices are generally not affordable by large
metropolitan water systems; and 3) in the case of
POU, not all the water is treated. In addition, POU
and bottled water are not considered acceptable
means of compliance with MCLs. Neither these
devices nor bottled water treat all the water in the
home and could result in health risks due to exposure
to untreated water. Consequently, POU and bottled
water are only considered acceptable for use as
interim measures. That is, they may be required by a
state primacy agent as a condition of obtaining a
variance or exemption, if necessary to avoid an
unreasonable risk to health before full compliance
could be achieved. Under this rule, however, POE
devices are acceptable means of compliance,
because POE provides drinking water meeting
standards at all taps in the house. Furthermore, these
devices might be cost effective for small public water
systems or nontransient, noncommunity water
systems. The ultimate goal for the EPA drinking water
program is to have water meeting all standards
through regionalization, pure source water, or central
water treatment.
FINAL REGULATIONS ALLOWING POE
FOR COMPLIANCE
Introduction
The EPA promulgated a final rule in July 1987 that
allowed POE devices as a means of compliance with
the final MCLs for volatile organic chemicals. POU or
bottled water could only be used to alleviate
unreasonable risks to health during a variance or
exemption period - that is, while the public water
systems were attempting to come into full compliance
with the MCL using POE or central treatment. A more
detailed discussion of the criteria for the use of POE
as a means of compliance follows.
Definitions
The final rule (Code of Federal Regulations, Part
141.2) defines POU and POE. These definitions are
worth repeating here for clarity (2):
"'Point-of-entry treatment device' is a treatment
device applied to drinking water entering a house
or building for the purpose of reducing
contaminants in the drinking water distributed
throughout the house or building."
"'Point-of-use treatment device' is a treatment
device applied to a single tap used for the purpose
of reducing contaminants in drinking water at that
one tap."
Criteria and Procedures
EPA is required to establish conditions for treatment
and control of public water supplies that assure
protection of public health. Specifically, EPA's primary
drinking water regulations are to contain criteria and
procedures to assure a supply of drinking water that
dependably complies with MCLs, including quality
control and testing procedures to insure compliance
with such levels and to insure proper operation and
maintenance of the system. It is under this authority
that EPA promulgated criteria and procedures
allowing the use of POE for compliance with the
volatile organic chemical MCLs. As was mentioned
earlier, EPA feels that philosophically the Safe
Drinking Water Act, including the legislative history,
emphasizes the goal of providing pure water
throughout a centrally controlled facility. Realizing that
this philosophical goal is not always attainable,
especially for small systems that lack the necessary
financial and technical resources, EPA is seeking to
allow innovation in order to gain increased compliance
among these systems. Historically, the largest
number of violations have occurred among the very
small systems. It is hoped that this rule will allow
them a more accessible means of compliance, and
would, in turn, increase their compliance rate. The
rule specifies criteria and procedures that will
hopefully assure quality and safety when POE is
applied for compliance purposes.
The five criteria necessary for compliance using POE
are summarized below.
Central Control
Originally, in the November 13, 1985 rule, central
ownership and control were required. It would be the
responsibility of the public water system (PWS) to
own, operate, and maintain all parts of the treatment
system. This appeared appropriate and necessary to
ensure adequate control of the treatment devices so
that they were working properly.
Public comments noted that while central control and
responsibility were necessary, ownership of the
devices was not. The PWS, while maintaining
responsibility and control, could lease the treatment
devices and also possibly have them operated and
maintained by a service company. The major concern
-------
of EPA was that the property owners would not
individually become responsible for these devices.
The final rule requires the public water system to be
responsible for operating and maintaining all parts of
the treatment system including each POE device.
Central ownership is not necessary so long as the
public water system maintains control of the operation
and maintenance of the device. This includes being
responsible for and supervising any service contractor
acting on behalf of the public water system. Central
control is appropriate and necessary to ensure that
the treatment device is always functioning properly.
Effective Monitoring
The public water system must develop an effective
monitoring plan and obtain state primacy agent
approval before POE devices are installed for
compliance with drinking water standards. POE
devices are fundamentally different from central
treatment in that many more devices are installed at
different locations. All mechanical devices have some
theoretical or empirical failure rate. As the numbers of
devices applied at one public water system increase
so does the probability of some devices failing.
Typically, at the central water plant, the operator
makes observations and measurements. This clearly
becomes more difficult with POE devices since they
are numerous and are generally located on private
property.
A monitoring program would include some proportion
of the devices, for example, 10 percent. Monitoring
might rotate throughout the population on a quarterly
basis. In some cases, physical inspections and flow
measurements could be made on the entire
population, with the proportionate sampling being
done for more expensive chemical analyses. The cost
of analyses for volatile organic chemicals is so high
and probability of failure in a well-designed, well-
operated treatment device so low that a small number
of samples should be adequate. The details of this
requirement remain to be determined by the state.
The state and the utility should work together to
formulate an adequate, yet affordable program.
Application of Effective Technology
Design review of plans and specifications for
modifications or additions to water works are
generally required under state authority. Almost every
state requires reviews and permits for this kind of
activity. EPA recognized this and mandated a similar
kind of review. This review should include certification
that the device will perform adequately to protect
public health in the individual application
contemplated. Most states are likely, and it is
certainly appropriate, to accept recognized third-
party certification of these devices. Third-party
certification should not preempt a review to assure
that the application of certified device is appropriate.
There are, then, two responsibilities: 1) certification of
the device for various contaminant removal situations,
and 2) review to assure that the device is being
applied to an appropriate situation.
Maintenance of Microbial Safety
The design and application of POE devices must
consider the tendency for increases in bacterial
concentrations in water treated with granular activated
carbon and possibly some other technologies. At a
central treatment plant, provisions can be made for
granular activated carbon adsorbers to be
backwashed and post-disinfection is generally
practiced. In a POE situation, the disinfectant, if
present, is in the incoming water. GAC is an effective
media for removal of chlorine from water. It also has
been shown to provide a surface for the attachment
and growth of heterotrophic bacteria. Heterotrophic
bacteria are not usually harmful to health, but do
present two concerns. The first is that they may infect
people who are sickly and have a low resistance to
bacterial infection, especially of the respiratory tract.
Secondly, high concentrations of heterotrophic
bacteria (greater than 500 per ml), can interfere with
the examination of the water for coliform bacteria.
The state might require additional monitoring for
heterotrophic bacteria to evaluate for possible
interference with the required coliform bacterial tests.
If interference is suspected or counts are high
enough to be of concern via respiratory exposure,
then the state might require post disinfection.
Post disinfection after a GAC adsorption unit would
require an ultraviolet device, or a chlorinator with an
adequate contact tank. The contact tank and
chlorinator can be designed in accordance with
standard procedures used for providing disinfection of
single buildings, using noncommunity water wells.
Post disinfection would clearly increase the cost of
POU treatment and might bias the economics toward
central treatment.
Protection of All Consumers
Every building connected to a public water system
must have a POE device installed, maintained, and
adequately monitored. The device should provide
treated water to every potable water tap within each
building. It is up to the state to determine if some taps
within or outside certain buildings may remain
untreated. For example, if the state allows nonpotable
water to be used for car washing, then this portion of
the water can be untreated. Other nonpotable uses
might include fawn watering devices, aesthetic
fountains, industrial cooling, and fire protection. As
previously mentioned there is concern, especially with
volatile chemicals, for respiratory exposure.
Therefore, the definition of nonpotable water should
never extend to living or nonindustrial working spaces
where these kinds of exposure are possible.
SUMMARY
The goal of these five criteria is to assure that when
POE devices are applied by public water systems for
-------
compliance with drinking water standards, the water is
as safe as the time-tested methods of central water
treatment. These criteria provide for adequate public
health protection, while at the same time allowing for
an innovative, decentralized approach (i.e., POE).
Hopefully, this approach will allow a cost effective
means of compliance for smalj systems that have had
the most violations of EPA's drinking water standards.
The criteria developed by EPA will be adopted and
implemented by the states with the goal of providing
safe drinking water to all communities.
REFERENCES
1. National Primary Drinking Water Regulations,
Volatile Synthetic Organic Chemicals, Final Rule
and Proposed Rule. Fed. Reg. 50:219:46880-
46933. November 13, 1985.
2. National Primary Drinking Water Regulations,
Synthetic Organic Chemicals, Monitoring for
Unregulated Contaminants, Final Rule. Fed. Reg.
52:130:25690-25717. July 8, 1987.
-------
REGULATORY REQUIREMENTS FOR POINT-OF-USE SYSTEMS
Ruth Douglas
Registration Division
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Washington, DC
There are three general categories of water treatment
units: 1) units not intended to prevent, destroy, repel,
or mitigate any microorganisms or other pests (e.g.,
carbon or some other coarse filtering material); 2)
units that consist of only a physical or mechanical
means of preventing, destroying, repelling, or
mitigating any microorganisms or pests (e.g.,
devices); and 3) units that incorporate a chemical
antimicrobial agent or units that consist of a
combination of physical and chemical treatment
intended to prevent, destroy, or mitigate
microorganisms or pests (e.g., pesticides).
Products in the first category are subject to neither
registration nor regulation under the Federal
insecticide, Fungicide, and Rodenticide Act (FIFRA).
Products in the second category are only subject to
regulation under FIFRA. Products in the third category
are subject to both registration requirements and
regulation under FIFRA.
There are approximately 147 registered water
treatment products. The first registration was issued
in 1965 by the U.S. Department of Agriculture. The
remaining products were registered by the U.S.
Environmental Protection Agency beginning in 1975.
The majority of the water treatment products are
registered for use in conjunction with municipally
treated or microbiologically potable water. Five of
these products are registered for use on untreated or
raw water (i.e., water of unknown quality or source).
Types of registered water treatment products are as
follows:
• Water filters 118
• Filtering media 13
• Replacement cartridges 11
* Water purifiers 5
Prior to 1979, data requirements for bacteriostatic
water filters consisted of bacteriological and chemical
data. These data requirements were published in the
Federal Register (1) as the Interim Requirements for
Registration of Bacteriostatic Water Treatment Units
for Home Use. Since the promulgation of the
conditional registration regulations in 1979, we have
only required chemical data demonstrating that no
more than 50 yg/l (50 ppb) silver are re/eased into
the effluent water. This is because with the
promulgation of the conditional registration
regulations, microbiological data are no longer
required for pesticide products with non-public
health related uses. Bacteriostatic water filters are in
this category because they can only be
recommended for use in conjunction with municipally
treated water or water that is already microbiologically
potable. The only pesticidal claim allowed for this type
of product is that it "inhibits (slows down) the growth
of bacteria with the filter medium." Other acceptable
claims for bacteriostatic water filters are of an
aesthetic nature, such as, "removes chlorine, makes
the water taste better, clarifies the water, and filters
out suspended particles."
On the other hand, water purifiers fall in the category
of pesticide products with public health-related uses
because they are recommended for use on
raw/untreated water or water of unknown source or
quality. Therefore, bacteriological and chemical data
are still required for those products. The products
currently registered as water purifiers are only for
emergency use. They are not registered for use on a
continuous basis.
In 1984, EPA formed a task force for the specific
purpose of developing definitive guidance and specific
test parameters for demonstrating effectiveness of
water treatment units claiming to microbiologically
purify water under conditions that simulated actual
use. This task force, which consisted of 17 people,
was chaired by Dr. Stephen Shaub of the U.S. Army
Medical Bioengineering R&D Laboratory in Frederick,
Maryland. The culmination of the efforts of this task
force resulted in the Guide Standard and Protocol
dated April 1986 and revised in April 1987.
Our current requirements for microbiological water
purifiers consist of data showing effectiveness of the
10
-------
product against bacteria, viruses, and protozoan
cysts.
In summary, the requirements for registration of
bacteriostatic water filters have not changed since
1979. On the other hand, we now have more
definitive guidance and specific testing parameters for
products claiming effectiveness as microbiological
water purifiers - the Guide Standard and Protocol for
Testing Microbiological Water Purifiers.
REFERENCE
1. Federal Register, Volume 41, No. 152, August 5,
1976.
11
-------
CONTROL OFPOINT-OF-USE WATER TREATMENT DEVICES IN CANADA:
LEGAL AND PRACTICAL CONSIDERATIONS
Richards. Tobin
Environmental Health Directorate
Health and Welfare Canada
Ottawa, Ontario K1A OL2
Canada
Introduction
With a natural resource of about 25 percent of the
world's fresh water, Canada can be considered a
water-rich country. Over 7.6 percent of Canada's
surface is covered by water, although it is not always
located where water demand is highest. With this
tremendous supply of water, about 2,500
communities are served by a water distribution
system covering 87 percent of the population (1).
Most of the remaining 13 percent of the population
are served by private wells or other small private
sources. It has been estimated that there are as
many as 1.38 million private wells in Canada, serving
about 4 million people.
Unfortunately, this seeming abundance has resulted
in reckless use of water. While Canadians consume
only 1.3 L of water per person per day (0.3 gpd) (2)
the average rural household uses about 150
L/person/day (40 gpd), and the average demand for
municipal systems averages about 500 L/person/day
(132 gpd).
Water Jurisdictions
In Canada, the legislative base is derived from the
Constitution Act of 1981 (including previous
Constitution Acts, referred to in this act). Although the
acts do not specifically address water, the ownership
of natural resources, including water, is vested with
the provinces. The provinces, therefore, have the
right to enact legislation with regard to water and to
have exclusive Jurisdiction over municipal institutions,
local works and undertakings, and other matters
within the province. Under the Department of National
Health and Welfare Act, this department has a
responsibility to investigate and conduct programs
related to public health. In conducting this program,
under Section 5 of the act, the department must
coordinate its efforts with those of the provinces. For
example, although there is no national safe drinking
water act, the Guidelines for Canadian Drinking Water
Quality are developed by a Federal-Provincial
Subcommittee that reports to a Federal-Provincial
Advisory Committee on Environmental and
Occupational Health (in turn reporting to the
Conference of Deputy Ministers of Health). Thus, the
provinces assume the lead role in ensuring a safe
supply of drinking water whereas the Federal
government provides leadership in ensuring
guidelines for drinking water quality.
In some circumstances, the Federal government is
entirely responsible for the provision and quality of
drinking water. These include administering potable
water regulations for all common carriers
(transportation crossing Canadian Interprovincial and
International borders), and on Canadian coastal
shipping vessels, and the provision of potable water
in the Territories, Indian reservations, national parks,
and military bases.
Under the Food and Drugs Act, administered by
Health and Welfare Canada, a "food" is defined (in
Section 2) as including "any article manufactured,
sold or represented for use as a food or drink for
man, ..., and any ingredient that may be mixed with
food for any purpose whatever." It is an offense
under this act to sell a food that contains harmful or
poisonous substances, that is unfit for human
consumption, or that has been prepared under
unsanitary conditions. The Minister, therefore, has the
authority to prescribe regulations for water, although
this has only been done for bottled and spring waters
(Division 12, Food and Drug Regulations), due to the
primary role generally assumed by the provinces.
Point-of-Use Devices
Point-of-use devices are becoming common
household appliances in Canada, as they are in the
U.S. Total sales of all types of devices have been
estimated at about 100,000 per year. By means of a
telephone survey of over 16,000 homes in cities
across Canada, we learned that the use of activated
12
-------
carbon filters alone ranged from about 0.3 percent to
16 percent (3), largely depending upon the perception
of the water quality. Interestingly, the highest value
was found in a city with a notorious taste and odor
problem, but with no particular chemical problem.
In Canada, there is no specific legislation controlling
the sale, use, or performance of point-of-use water
treatment devices. A number of acts, however,.do
have provisions that could be related to these
devices. Some of these are briefly described below.
The Pest Control Products Act (1968-69, c. 50,
S.1), administered by Agriculture Canada, regulates
products that are intended to control any kind of pest.
A pest is "any injurious, noxious, or troublesome
insect, fungus, bacterial organism, virus, weed, rodent
or other plant or animal pest, and includes any
injurious, noxious or troublesome organic function of
a plant or animal" (Section 2 of the act).
An exemption (Regulations, Section 3(a)) to the act
stipulates that it may not be applied when the product
is a food. Since "drink" is defined as a food in the
Food and Drugs Act, an administrative agreement
between Agriculture Canada and Health and Welfare
Canada has exempted water treatment chemicals and
devices from consideration under the provisions of
the act (4). The agreement states that for "chemicals
or devices for water purification," the Environmental
Health Directorate of' Health and Welfare Canada
should be consulted. In fact, water treatment
chemicals are not currently covered by legislation at
the Federal level, except for emergency disinfection
chemicals which may be considered as food additives
or drugs (depending on claims made) under
Regulations of the Food and Drugs Act. Thus, it is
conceivable that such devices (e.g., ozonators,
chemical feeders) may be covered by the Act when
they are used for nonpotable water (e.g., swimming
pools, spas) but not for potable water.
The Medical Devices Regulations (established by PC
1976-2031 and revised periodically) of the Food and
Drugs Act are used to control devices that are
"manufactured, sold or represented for use in ...
prevention of disease ...." Thus, depending on the
exact claims made, devices may come under the
provisions of these regulations. If the device claims to
disinfect water and prevent enteric disease, for
example, it could well be interpreted to fall under
these provisions. The regulations require the
manufacturer to notify the department when a device
is put on the market and to furnish certain information
including a statement of purpose of the device, and a
copy of instructions. The department may also require
evidence of the safety and effectiveness of the device
(Regulations, Section 27(1)). Thus far, these
provisions have not been used for point-of-use
devices.
The Hazardous Products Act (1968-69, c. 42, S.1),
administered by Consumer and Corporate Affairs
Canada (and naming Health and Welfare Canada in
certain sections) authorizes the Minister to carry out
investigations and demand information regarding
consumer products, to determine whether such
products are likely to be a danger to the health or
safety of the public. Where it is considered necessary
to remove a product from the market entirely, it is
included in Part I of the Schedule to the Act. Where it
is considered that specific regulations can be
prescribed to which the products must comply in
order not to present a hazard, then the products are
listed in Part II of the Schedule.
In 1981, it was proposed to prohibit the sale of
activated carbon water filters because of the problem
of bacterial growth on the filters (5). As a result of
negotiations with industry, an agreement was reached
whereby such filters would be labeled to prevent their
use on microbiologically unsafe waters. Subsequently,
the proposal to ban these devices was discontinued
(6).
Another piece of legislation administered by
Consumer and Corporate Affairs Canada is the
Competition Act (R.S., c. C-23, S.1; 1986, C-26,
S.19) which supersedes the Combined Investigation
Act. Parts of this act have been used where
misrepresentation of devices has been alleged.
Section 36(1) states that
"No person shall ... (a) make a representation to
the public that is false or misleading in a material
respect; (b) make a representation to the public in
the form of a statement, warranty or guarantee of
the performance, efficacy or length of life of a
product that is not based on an adequate and
proper test thereof, the proof of which lies upon
the person making the representation; ... the
general impression conveyed by a representation
as well as the literal meaning thereof shall be taken
into account in determining whether or not the
representation is false or misleading in any material
respect."
Anyone found guilty of such an offense on conviction
or indictment is subject to a fine in the discretion of
the court or to imprisonment for five years, or both.
Since these provisions require the representations
made for a device to be true and backed by adequate
and proper proof, it is clear that these are powerful
tools against misleading advertising for all types of
devices. The Department has worked closely with
officials of Consumer and Corporate Affairs Canada
and lawyers in the Justice Department by advising
them on technical matters and suggesting test
protocols for testing of devices. In many cases we
have been asked to be prepared to serve as expert
13
-------
witnesses in the event that the case went to court,
which in most cases was not required. Often a guilty
plea was entered by the defendants, obviating the
need for a trial.
Nonlegislative Activities
Although these legislative tools are at our disposal for
more serious problems, the departmental policy has
been to avert problems before they occur. Thus, our
program on point-of-use devices includes the
following aspects: testing and evaluation of devices;
provision of advice and educational materials; and
cooperation with industry and nonprofit organizations.
In our program of testing devices, summarized
elsewhere (7), we have elucidated a basic test
protocol and have applied it on a number of water-
disinfectant devices, including ultraviolet (UV),
filtration, and iodine-releasing devices. Bacterial
growth studies have been conducted on several types
of activated carbon water filters: normal granular
activated carbon, silver containing, compressed
carbon, precoated carbon, etc. Results from these
studies were qualitatively similar; all types of carbon
filters supported bacterial growth, and these bacteria
contaminated the finished water.
Evaluation of devices is conducted on an ongoing
basis at the request of the public, governmental and
nongovernmental agencies, and industry. Normally an
evaluation involves the review of data and claims
made for a device and provision of an opinion on the
suitability of the device for a given purpose, the
validity of claims, adequacy of the test protocols, etc.
Often we find that there is not sufficient good
evidence on which an evaluation can be based.
Advice on the choice and use of devices is often
sought by members of the public. Sometimes
individual advice is required, but often the educational
Tearsheets, Dispatches, Information Letters,
Environmental Health Directorate Reports, and
scientific articles can be sent to the person for in-
depth study.
A number of cooperative efforts have been made with
industry associations and nonprofit organizations. For
example, numerous discussions have been held with
the Canadian Water Quality Association to discuss
perceived problems with the advertising and
promotion of water treatment devices. They have
published voluntary guidelines for use by the industry
for the advertising and promotion of carbon water
filters (8) and all products (9). In another area, we
have worked closely with the National Sanitation
Foundation (NSF), who has assumed a lead role in
development of performance standards and the
testing and listing of devices. Departmental
representatives have served on NSF Working Groups
during preparation of the draft UV standard, on the
revision of the standards on health effects and
easthetic effects devices, and on the Joint Committee
on water treatment units and on the Council of Public
Health Consultants. It is considered that the NSF
listing of devices provides the consumer with an
easily identifiable proof of performance for removal of
specific contaminants. .Ultimately, compliance with
these voluntary performance standards should make
the selection of an appropriate device more
straightforward for the consumer.
Conclusion
Although there is no specific Canadian legislation
respecting point-of-use water treatment devices,
there are a few pieces of legislation that have been or
could be used for particular problems. At the present
time less formal methods are generally used to
provide information on these devices and to ensure
their safety and efficacy in use.
References
1. Anon. National inventory of municipal waterworks
and wastewater systems in Canada 1981. Supply
and Services Canada, Ottawa, 1981.
2. Environmental Health Directorate. Tapwater
consumption in Canada. 82-EHD-80, Health and
Welfare Canada, Ottawa, 1981.
3. Tobin, R.S., Junkins, E.A. 'and Eaton, F.E. Survey
of the use of activated carbon water filters in
Canadian homes. Can J. Public Health. 76:384-
387, 1984.
4. Health and Welfare Canada. Antimicrobial products
subject to the Pest Control Products Act and Food
and Drug Act. Information Letter No. 536. Health
Protection Branch, 1978.
5. Health and Welfare Canada. Point-of-use water
treatment devices. Information Letter No. 601.
Health Protection Branch, 1981.
6. Health and Welfare Canada. Activated carbon
water treatment devices. Information Letter No.
635. Health Protection Branch, 1982.
7. Tobin, R.S. Testing and evaluating point-of-use
treatment devices in Canada. JAWWA (In press),
1987.
8. Canadian Water Quality Association. Canadian
water filter industry voluntary guidelines for carbon
water filter advertising and promotional claims.
CWQA, Waterloo, Canada, 1982.
9. Canadian Water Quality Association. Voluntary
water quality industry product promotion guidelines.
CWQA, Waterloo, Canada, 1984.
14
-------
THE REGULATION OF WATER TREATMENT DEVICES IN CALIFORNIA
Robert F. Bums
Sacramento, CA 95814
During 1986, the California legislature introduced two
legislative measures regarding point-of-use (POU)
and point-of-entry (POE) water treatment devices.
One measure, Senate Bill SB 2119(1) by Senator
Torres, addressed the performance of POU/POE
water treatment devices for which a claim relative to
the health or safety of drinking water is made. The
other measure, SB 2361(2) by Senator
McCorquodale, addressed advertising claims made in
the sale of POU/POE water treatment devices. Both
bills were passed by the legislature, signed by the
Governor, and became effective on January 1, 1987.
In general, the two bills were considered to be tough
pieces of legislation. A frequently asked question is
why the legislature decided to regulate the water
treatment device industry. Since the California
legislature does not maintain a record of their
committee proceedings, one can only speculate why
these bills were passed. What is known is that a
number of events, preceding the introduction of the
bills, had come to the attention of the legislature.
Since the late 1970s, Californians have realized that
their ground water sources of drinking water were
potentially vulnerable to chemical contamination. In
1978, a large number of wells in the San Joaquin
Valley were found to be contaminated with the
agricultural fumigant, dibromochloropropane (DBCP).
Many of these wells were found to exceed the state's
action level for DBCP of 1 ug/l (1 ppb).
During 1980, wells in the heavily populated San
Fernando Valley and San Gabriel Valley in Los
Angeles County, were found to be contaminated by
industrial solvents such as trichloroethane (TCE) and
perchloroethylene (POE). These same industrial
solvents were also detected in wells in the Santa
Clara Valley, which is often referred to as Silicon
Valley.
In 1985, the California Department of Health Services
(DHS) sampled over 3,000 wells used by large public
water systems (over 200 connections) for organic
chemical contaminants (3). A significant number (18.3
percent) of the wells sampled had measurable
concentrations of one or more organic chemicals.
One hundred and sixty five of these wells (5.6
percent) had concentrations of chemicals that exceed
the State Maximum Contaminant Level (MCL) or a
State Action Level (4). When contamination levels
were found to exceed an MCL or State Action Level,
public notification was initiated by means of a public
news release.
During 1985 and 1986, the newspapers and television
stations, particularly in Southern California, frequently
reported on drinking water contamination problems.
The public was alarmed by these news stories and
became very concerned about the quality- of their
drinking water.
As a result of the increased public concerns about
drinking water quality, the bottled water industry
recorded a significant increase in sales in 1985 and
1986. The water treatment device industry appears to
have experienced a similar increase in sales during
the same period. Unfortunately, there were a number
of cases of consumer abuse and fraud as a result of
the overly aggressive mart^ting efforts by a few
companies selling water treatment devices.
The introduction of SB 2361 by Senator
McCorquodale has not been tied to any specific
consumer problems. However, the Senator
represents the Santa Clara Valley and his office had
been contacted by constituents about the marketing
techniques used by the water treatment industry in
that area.
SB 2361 enacted a statute which provides "truth-
in-advertising" as it relates to water treatment
devices. The statute addresses false or misleading
advertising with key provisions which make it unlawful
to:
• Make false claims or statements about the quality
of water provided by a public water system.
• Make false claims about the health benefits
provided by the use of a POU/POE water treatment
device.
• Make any product performance claims unless such
claims are based on actual, existing factual data.
15
-------
• Make any other attempts to mislead the consumer
or misrepresent the product.
This statute is expected to deter unscrupulous
salespersons and reduce the number of complaints
relative to fraudulent sales. The statute will also assist
the consumers who are victims of fraudulent sales by
providing them with a means to file a criminal
misdemeanor action and recover damages. The
statute does not assign enforcement responsibility to
any specific agency. It is expected that local district
attorneys and the State Attorney General will take
legal action against companies acting in violation of
this new law.
The introduction of SB 2119 by Senator Torres is
often described as a response to a consumer abuse
problem in McFarland, California. A water treatment
device company was reported to have advertised and
convinced customers in McFarland that their water
treatment device could remove all cancer-causing
chemicals. This marketing effort occurred at a time
when this community was very concerned about a
cluster of childhood cancer cases that were being
investigated by state and local health agencies. The
company was successfully prosecuted by the State
Attorney General and the settlement allowed the
consumers to rescind their sales contracts.
SB 2119 enacted a statute which requires that any
water treatment device for which a health benefit
claim is made, cannot be sold in California unless the
device had performance testing that has been
certified by the Department of Health Services (DHS).
This law further requires DHS to adopt regulations
setting forth the criteria and procedures for
certification of water treatment devices. These
regulations must include appropriate testing protocols
and procedures to determine the performance of
these devices. The cost of this new program is to be
paid for through fees imposed on the applicants. The
law also assigned responsibility for enforcement to
DHS or local health departments.
This statute outlines a very specific Plan for the
regulation of POU/POE water treatment devices
(WTDs). The general provisions of this statute include
the following:
• The DHS is required to adopt regulations which set
forth the criteria and procedures for the certification
of WTDs that are claimed to affect the health and
safety of drinking water.
• A "water treatment device" (WTD) is defined to
mean any point-of-use or point-of-entry
instrument or contrivance sold or offered for rental
or lease for residential, commercial, or institutional
use, without being connected to the plumbing of a
water supply intended for human consumption in
order to improve the water supply by any means,
including, but not limited to, filtration, distillation,
adsorption, ion exchange, reverse osmosis, or
other treatment.
• No WTD which makes product performance claims
or product benefit claims that the device affects
health or the safety of drinking water, shall be sold
or otherwise distributed unless the device has
been certified.
• WTDs which are not offered for sale or distribution
based on claims of improvement in the
healthfulness of drinking water need not be
certified.
• A WTD initially installed prior to the operative date
of the statute is not required to be certified.
• The requirement that a WTD be certified does not
become operative until one year after the effective
date of the regulations.
• The DHS or any testing organization designated by
the DHS may agree to evaluate test data in a WTD
offered by the manufacturer, in lieu of the
requirements of the statute, if the DHS or the
testing organization determines that the testing
procedures and standards used to develop the
data are adequate to meet the requirements of the
statute.
• The DHS may accept a WTD certification issued
by an agency of another state, by an independent
testing organization, or by the Federal government
in lieu of its own if the DHS determines that
certification program meets the requirements of the
statute.
The provisions that are to be included in the DHS
regulations were defined in the statute with
considerable detail. The provisions that are required
or allowed as part of the regulations are as follows:
• The regulations shall include appropriate testing
protocols and procedures to determine the
performances of WTDs in reducing specific
contaminants from public or private water supplies.
• The regulations may adopt, by reference, the
testing procedures and standards of one or more
independent testing organizations if the DHS
determines that they are adequate to meet the
requirements of the statute.
• The regulations may specify any testing
organization that the DHS has designated to
conduct the testing of WTDs.
• The regulations are required to include minimum
standards for (a) performance requirements, (b)
types of tests to be performed, (c) types of
allowable material, and (d) design and construction.
16
-------
• The regulations are required to include
requirements relative to product instructions and
information, including product operation,
maintenance, replacement, and the estimated cost
of these items.
• The regulations may include any additional
requirements, not inconsistent with the statute, as
may be necessary to carry out the intent of the
statute.
Finally, the statute specifies procedures for the
enforcement of the act. Key enforcement provisions
include the following:
• The DHS, or any local health officer with the
concurrence of the DHS, is responsible for the
enforcement of the act.
• The DHS may suspend, revoke, or deny a
certificate upon its determination that either (a) the
WTD does not perform in accordance with the
claims for which certification is based, or (b) the
manufacturer, or any employee or agent thereof,
has violated the statute.
• The act provides that any person, corporation, firm,
partnership, joint stock company, or any other
association that violates any provision of the act, is
liable for a civil penalty not to exceed $5,000 for
each violation.
The DHS Public Water Supply Branch (PWSB) has
been given the responsibility for the implementation of
SB 2119 and is currently developing policy and
regulations for the implementation of the certification
program. The PWSB has established an informal
advisory committee consisting of representatives for
industry, water utilities, and consumers to assist in
identifying and addressing issues. The following are
some of the elements of the program and regulations
that are being considered:
• The DHS plans to adopt existing protocols and
standards such as those established by the
National Sanitation Foundation (NSF).
• A "health or safety claim" will be defined in terms
of the Primary Drinking Water Standards adopted
by the DHS or the U.S. Environmental Protection
Agency.
• Certification of a WTD will be based on specific
contaminants for which the manufacture has made
a health or safety claim.
• The DHS will not establish a state laboratory to
conduct the testing required for certification. The
DHS plans to contract with outside laboratories or
testing organizations for the testing and other
administrative tasks relative to certification.
• The DHS may choose not to accept any
manufacturer's data relative to the performance
testing that will be required for state certification.
The water treatment device industry is very
concerned as to how the WTD certification program
will impact the marketing and sales of their product in
California. The advisory committee has been very
helpful in bringing the industries concerns to the
attention of the PWSB. Some of the concerns that
have been identified are as follows:
• The failure to accept manufacturer's data would
impose a substantial cost on the industry.
• If retesting by a third-party laboratory or testing
organization is required, the manufacturers will
have to pass on the added expense to the
consumers.
• The cost of testing under NSF or equivalent
standards will be very expensive.
• The one year grace period in which all testing must
be completed may exceed the capacity of State
contract laboratory or laboratories designated to
conduct WTD performance testing.
• In order to reduce the costs associated with
performance testing, consideration must be given
to testing approaches such as the use of
surrogates and the extrapolation of data whenever
possible.
It is evident that the California legislature has given
the DHS a difficult assignment. However, the
Department is committed to the establishment of a
WTD certification program that will serve the needs of
the California consumers and still be responsive to
some of the unique problems of the water treatment
device industry. The DHS is also confident that the
California program will not be in conflict with any
efforts to establish a national certification program.
REFERENCES
1. SB 2119 (Chapter 1247, Statutes of 1986).
2. SB 2361 (Chapter 1278, Statutes of 1986).
3. Organic Chemical Contamination Of Large Water
Systems In California. California Department of
Health Services, April 1986, page ii.
4. Drinking Water Action Levels set by the
Department of Health Services.
17
-------
WISCONSIN REGULATION OFPOINT-OF-USE AND POINT-OF-ENTRY
WATER TREATMENT DEVICES
Loretta Trapp
Department of Industry, Labor and Human Relations
State of Wisconsin
Madison, Wl 53707
Wisconsin's involvement in regulating point-of-use
and point-of-entry water treatment devices involves
five different state agencies. Two of those state
agencies, the Department of Justice (DOJ) and the
Department of Agriculture, Trade and Consumer
Protection (DATCP), have consumer protection
sections. Water treatment device manufacturers and
dealers only become involved with DOJ and DATCP if
their advertising literature or sales practices appear to
bo false or misleading.
A third state agency, the Department of Health and
Social Services (DH&SS), is responsible for
recommending enforcement standards for ground
water contaminants of public health concern. The
enforcement standard may be the actual maximum
contaminant level (MCL) set by the United States
Environmental Protection Agency or may be below
the MCL if scientific evidence for the lower number is
presently available but was not considered when the
MCL was established. When an MCL has not been
set for a contaminant of public health concern, the
enforcement standard will establish the upper limit
concentration for the contaminant in ground water.
Water treatment device manufacturers and dealers
rarely become involved with any DH&SS activities.
However, the Department of Natural Resources
(DNR), the fourth state agency, uses the enforcement
standards to establish whether or not a water supply
is contaminated. After a public hearing process, the
DNR usually adopts the recommended enforcement
standard into its regulations. If a water supply
contains a contaminant of public health concern in
excess of an enforcement standard, the water supply
is deemed contaminated. The DNR develops
regulations for methods to be pursued in obtaining
pure or noncontaminated drinking water for human
consumption.
If a water supply contains a contaminant in excess of
an enforcement standard, the DNR requires the
owner of that water supply to first seek a naturally
safe water supply which can involve:
• Extending a well casing;
• Drilling a new well; or
• Connecting to a public water supply or other
noncontaminated well.
The DNR requires all water for human consumption
to be noncontaminated. [Department of Industry,
Labor and Human Relations (DILHR) regulations
essentially require all water going to plumbing fixtures
to be noncontaminated.] Point-of-use or point-
of-entry water treatment devices used to reduce the
concentration of contaminants below the enforcement
standard may only be installed upon approval of the
DNR. The DNR also has the authority to require
sampling and maintenance for these water treatment
devices. Point-of-use devices are usually not
designed to produce the volume or flow rate of
noncontaminated water needed and so at this time
are not allowed for use on contaminated water
supplies. The DNR also considers point-of-entry
water treatment devices at this time to be the last
resort or at best an interim solution until a naturally
safe water supply can be obtained.
Water treatment device manufacturers and dealers
may become involved with DNR regulations if they
want their devices to be used to reduce the
concentration of a contaminant below the
enforcement standard.
The fifth state agency is the Department of Industry,
Labor and Human Relations (DILHR), which reviews
all point-of-use and point-of-entry water
treatment devices for the following:
• Rendering inactive or removing aesthetic and
health related contaminants;
• Suitability of construction materials for use with
potable water;
« Ability of the device to withstand the pressures to
which it will be subjected; and
• Proper installation instructions.
18
-------
The information that DILHR requires for review is
contained in Appendix A.
Presently DILHR has adopted into regulation only one
nationally recognized standard, the Water Quality
Association Standard S-100. DILHR has proposed
additional regulations, a copy of which is contained in
Appendix B.
In conclusion, water treatment device manufacturers
and dealers will most often become involved with
DILHR regulations, and to a lesser extent DNR
regulations. Appendix C contains a list of contact
people for each of the five state agencies.
Appendix A
APPLICATION FOR PLUMBING PRODUCT
REVIEW - REQUIRED INFORMATION FOR
WATER TREATMENT DEVICES
\
A letter, requesting approval, must be submitted by
the device manufacturer or distributor. Each letter
may contain only one product review request. The
following information shall be submitted with each
request for product review:
1. Product trade name and model number.
2. Manufacturer's name, address and telephone
number.
3. Product engineer's name, address and telephone
number.
4. Two copies of sales brochure, catalog and other
promotional literature.
5. Written detailed description of the composition
and function of device.
6. Detailed assembly drawings.
7. Information regarding marking of device:
a. Method of marking.
b. List of marking information on device.
c. Location of markings on device.
8. Complete installation instructions, including
detailed installation drawings indicating all
connections between the device and plumbing
system.
9. A list and copy of all national standards to which
the device, or the device's construction materials,
conforms.
10. A list of material specifications if other than
construction materials in referenced national
standards. Documentation shall also be provided
indicating that the construction material is
accepted for use with potable water, by the
National Sanitation Foundation (NSF) or other
national agency.
11. The trade name, scientific name and chemical
formula of any chemical used in the device that
may be added or leached into the water. A
toxicity rating and the source of the toxicity rating
must also be provided. Documentation shall be
included showing that these chemicals are
accepted for use with potable water, by the U.S.
Environmental Protection Agency (EPA), U.S.
Food and Drug Administration (FDA), National
Sanitation Foundation (NSF) or other national
agency.
12. A signed report, by an approved testing laboratory
or the manufacturer, which concludes that the
device functions and performs in accordance with
assertions submitted to the department. This
report must include but is not limited to the
following:
a. A detailed explanation of the test method(s).
b. The influent temperature, pH, hardness, total
dissolved solids and concentration of
contaminants.
c. The effluent temperature, pH, hardness, total
dissolved solids and concentration of
contaminants.
d. The minimum detection concentration of the
contaminants that may be achieved by the test
method.
e. Test results of at least one duplicate sample.
f. Test results of a reagent blank.
g. Test results of a spiked sample.
h. The percentage of influent disposed as waste.
i. An estimate of the error in the test results.
j. A sample calculation.
k. Test results proving conformance to
referenced national standards
I. Test results indicating the burst pressure.
m.Test results indicating the working water
pressure and temperature range.
19
-------
n. The name, address and telephone number of
the laboratory.
o. The name of the individual(s) performing the
test(s).
13. Disposal requirements of any wastewater,
backwash fluid, filter, membranes or other
replaceable device components.
14. A graph indicating the pressure loss, in psig,
through the device, over the entire flow rate
range, in U.S. gallons per minute.
15. An operation and maintenance manual or
instructions, including but not limited to the
following:
a. Maintenance cycle under given influent
conditions.
b. Maintenance procedures.
c. Operating pH range.
d. Operating pressure range.
e. Operating temperature range.
f. Operating flow rate range.
g. Operating total dissolved solids range.
h. Any influent conditions that will adversely affect
the stated performance of this device.
Appendix B
PROPOSED DRAFT REGULATIONS
Section ILHR 82.11 (181) "Water treatment device"
means a device which:
Renders inactive or removes microbiological,
particulate, inorganic, organic or radioactive
contaminants from water which passes through
the device or the water supply system
downstream of the device.
Section ILHR 84.20 (6) (o) Water treatment devices.
1. Water softeners shall conform to WQA S-100.
2a. Except as provided in subpar. b., water treatment
devices shall function and perform in accordance
with the assertions submitted to the department
under s. ILHR 84.10, relating to rendering inactive
or removing contaminants.
2b. A water treatment device which injects a water
treatment compound into a water supply system
shall maintain the compound concentration in the
system over the working flow rate range and
pressure range of the device.
3. Except as specified in subd. 4., water treatment
compounds introduced into the water supply
system by a water treatment device shall be listed
as an acceptable drinking water additive by a
listing agency approved by the department.
Listing agencies approved by the department
shall include:
a. United States Environmental Protection
Agency;
b. United States Food and Drug Administration;
and
c. National Sanitation Foundation.
4. A water supply system shall be protected from
backflow when unlisted water treatment
compounds, which may affect the potability of the
water, are introduced into the system. The
department shall determine the method of
backflow protection. Water supply outlets for
human use or consumption may not be installed
downstream of the introduction of an unlisted
water treatment compound.
6. Water treatment devices designed for
contaminated water supplies shall be labeled to
identify the following information:
a. The name of the manufacturer of the device;
b. The device's trade name; and
c. The device's model number.
Appendix C
PEOPLE TO CONTACT
Department of Agriculture, Trade & Consumer
Protection
Trade & Consumer Protection Division
Consumer Protection Bureau
801 W. Badger Road
Madison, WI53713
Jane Jansen, Director
(608) 266-8512
Department of Health & Social Services
Health Division
Community Health & Prevention Bureau
Section of Environmental and Chronic Disease
Epidemiology
1 West Wilson, Rm. 318
P.O. Box 309
Madison, Wl 53701-0309
20
-------
Henry Anderson, M.D., Section Chief
(608) 266-1253
Department of Industry, Labor and Human Relations
Safety and Buildings Division
Office of Division Codes and Applications
201 E. Washington Avenue
P.O. Box 7969
Madison, Wl 53707
Loretta Trapp, Plumbing Product Review
(608) 266-2990
Department of Justice
Legal Services Division
Consumer Protection Bureau
123 W. Washington Avenue
Madison, Wl 53703
Kevin O'Conner, Assistant Attorney General
(608) 266-2426
Department of Natural Resources
Environmental Standards Division
Water Supply Bureau
Private Water Supply Section
101 S.Webster
P.O. Box 7921
Madison, Wl 53707
William Rock, Section Chief
(608) 267-7649
21
-------
HOUSEHOLD WATER QUALITY EDUCATION:
THE COOPERATIVE EXTENSION SYSTEM ROLE
G. Morgan Powell
Cooperative Extension Service
Kansas State University
Manhattan, KS 66506
Water quality is one of the national priority initiatives
for the Cooperative Extension Service System. This
paper addresses household water quality education
through the Kansas Cooperative Extension Service. It
focuses on education relating to one portion of a
broader statewide water quality subject. To help you
understand why Kansas household water quality
programs may not be transferred directly to other
states, I will begin with a discussion of the extension
system organization using some specific examples
from Kansas.
THE COOPERATIVE EXTENSION
ORGANIZATION
The extension system in this country consists of the
50 state Cooperative Extension Services and
Extension Service of the U.S. Department of
Agriculture (USDA). The 1862 land-grant
universities, the 1890 land-grant universities, and
Tuskegee Institute operate the state Extension
Services. These total 67 State Cooperative Extension
Service organizations. These extensions are generally
funded jointly by Federal, state, and local sources,
which explains the name Cooperative Extension
Service (two states use the term "Agricultural
Extension Service").
Federal funds come through the Extension Service
USDA, while state funds come through the respective
university systems. Typically, Federal and state funds
support the state and area offices consisting of
directors, administrators, subject matter specialists,
technicians, and support staff. State and local funds
support county offices, which consist of county
agents, paraprofessionals, and support staff. Local
funds come from local governments, usually counties,
but sometimes including cities/counties.
The Cooperative Extension Service is the informal,
noncredit education arm of the land grant universities.
Much of this education occurs at the county level.
Subject matter specialists at state and area levels
provide support to county extension agents,
paraprofessionals, and volunteer teachers and
leaders. Because education programs cover broad
areas including agriculture, home economics, 4-H,
horticulture, and community development, the support
must also be broad based. The Cooperative
Extension Service includes a very broad range of
disciplines to support county education programs.
Table 1 shows the number of full time specialists by
program area and discipline for Kansas.
Table 1. Extension Subject Matter Specialists in Kansas
Agriculture
Agriculture Economics
Agriculture Engineering
Agronomy
Animal Science
Entomology
Grain Science
Horticulture
Plant Pathology
Veterinary Medicine
Sub Total
Forestry
Home Economics
Community Development
Energy
4-H and Youth
Information (writers,
editors, radio, TV, etc.)
Sub Total
Total
Discipline
Specialists
20
8
13
16
10
3
5
5
2
12
20
9
6
13
21
Program Area
Specialists
82
12
20
9
6
13
21
81
163
COUNTY EXTENSION PROGRAM
The county extension education effort helps people:
• Understand, evaluate, and solve problems;
• Learn through informal, out-of-school education
opportunities; and
• Work together to develop personally and develop
leadership skills.
22
-------
To achieve these goals, county programs include a
wide variety of efforts to reach their clientele. These
efforts include public meetings, workshops, short
courses, symposia, tours, and demonstrations.
Extension programs also reach the public through
radio, television, newspapers, and newsletters.
Extension also works closely with clubs and
organizations such as 4-H clubs, Extension
Homemaker Units, conservation tillage clubs,
agricultural commodity clubs, and marketing clubs in
Kansas. However, extension also has a history of
working closely with the Farm Bureau, soil
conservation districts, drainage districts, land
improvement contractors, rural water districts, and
many other similar groups.
In Kansas, more than 300 county agents conduct
county extension programs (Table 2). Kansas also
has thousands of volunteer teachers and leaders who
add immeasurably to the extension program. A recent
survey showed nearly 38,000 part-time volunteers
serving in extension sponsored and related
organizations, equa) to 1.5 percent of the population.
Table 2. County Professional Staff in Kansas
Agricultural Agents
4-H/Youth Agents
Home Economic Agents
Horticulture Agents
Total Agents
Paraprofessionals
113
37
127
12
289
21
Total County Staff
310
because they are man-made and the result of fairly
recent activity. This suggests that the presence and
concentration of organics in private as well as public
water.supply wells may be increasing and will likely
be of more concern in the future.
Table 3. Inorganic Contaminants in Farmstead Wells -
Parameters Abvove Maximum Contaminant Level
(MCL)
Percent Confidence"
Nitrate
Selenium
Fluoride
Lead
Total Inorganic
28
9
2
2~
37 + 9
* 95 percent confidence level.
** Lead was found to be a result of plumbing in the well water.
Table 4. Organic Contaminants in Farmstead Wells
Percent Confidence*
Atrazine
2,4-D
2,4,5-T
Tordon
Chlordane
Heptachlor Epoxide
Alaohor
Wells with Pesticides?*
1 ,2-Dichloroethane
Benzene
4
1
• 1
1
1
1
1
8 ±6
1
1
Wells with VOC
±3
95 percent confidence level.
Two pesticides were found in each of two wells.
KANSAS WATER QUALITY SITUATION
Kansas has relatively few serious water quality
problems. However, it does have conditions that are
cause for concern and that need careful monitoring.
Central water systems roughly serve 80 percent of
Kansas residents. Over half of this supply is from
ground water. For two years, the Kansas Department
of Health and Environment has checked public supply
wells for volatile organics and pesticides that are
mandated by new regulations of the Safe Drinking
Water Act to be implemented from 1987 to 1991. The
state has shut down 52 (2.9 percent) of 1,800 wells
checked because of contamination.
The Kansas Department of Health and Environment,
in cooperation with Kansas State University, randomly
surveyed 104 private farmstead wells. They found
nitrates in 28 percent of the wells, selenium in nine
percent, and fluoride in two percent, where the
inorganic contaminants exceeded the MCL (Table 3).
They found pesticides and volatile organic chemicals
(VOCs) respectively in eight and two percent of the
wells (Table 4). These organics are worrisome
According to the 1980 census, Kansas has about
125,000 private water supplies, almost all wells.
Based on 3.9 persons per well, the numbers served
by private wells in the survey, roughly 500,000 people
(about 20 percent of the state's population) depend
on private water supplies. No regulations or testing
requirements apply to these private water systems.
Users or owners are responsible for the quality of
these supplies. They are the operators and the
sanitarians. An optimistic estimate is that owners test
only a few of these wells each year.
Another concern in Kansas is the large number of
abandoned wells that remained unplugged. The
Kansas Department of Health and Environment
estimated that the state has 250,000 abandoned
wells. We believe Kansas could have 500,000 or
more abandoned wells.
At Kansas State University, we have determined that
a coordinated extension program is essential to
address the problems related to private water
supplies and agriculture's impact on water quality,
23
-------
especially that of ground water. The Kansas
Department of Health and Environment is also
concerned about, and supports, this extension
education effort.
WATER QUALITY TASK FORCE
The Kansas Cooperative Extension Service initiated a
five-member Water Quality Task Force in 1985. This
task force began as a component of the agricultural
programs portion of extension. The goals were to
address the impact of agriculture on water quality and
to initiate publications and other educational material
that would be needed. In 1986, the task force was
expanded to involve other extension program areas.
Ten persons representing agriculture, home
economics, 4-H, and youth and community
development now serve on this task force.
The task force is an efficient and effective way of
bringing persons from different disciplines and
responsibilities together for discussion, information
sharing, program planning, and task assignments.
Some persons as they became involved on the task
force were reluctant participants. However, as they
learned more about specific water quality problems,
effects on people, and how corrective action can be
taken, these people became enthusiastic participants.
Task force members are all involved in preparing
extension publications, education programs, and
water quality training.
HOUSEHOLD WATER QUALITY
PROGRAM
The Extension Household Water Quality Program at
Kansas State is coordinated through the Water
Quality Task Force. It involves preparation of a wide
range of extension publications and supporting media
(video and slide/tape) materials, agent and lay leader
training, and news stories for newspaper, magazine,
radio, and TV.
To date, 12 extension publications are completed and
30 others planned. These publications plus materials
from several other sources now make up the
Household Water Quality Resource Notebook.
We conducted 10 agent training classes for more
than 150 people in October and November 1987. We
trained agriculture, home economics, and 4-H/youth
agents as well as health services personnel in
household water quality.
These one-day training sessions familiarized people
with private household water quality and acquainted
them with the resource notebook, agency contacts,
and how to use them. Our goal was to give local
professionals a background so they can be local
resources for those with water quality problems.
SUMMARY
The extension system is a complex organization of
the 50 state Cooperative Extension Services and the
Extension Service, LJSDA. Federal, state, and local
sources fund extension programs. The extension
system is the noncredit informal education arm of the
land-grant universities. It includes substantial
professional staff but many volunteer teachers help to
make extension a dynamic and important educational
tool for adults and youth.
Kansas has water quality problems among the
125,000 private water wells that serve 20 percent of
its population. A recent survey found that 37 percent
of the wells exceeded the MCL for inorganics, and 10
percent contained organic contaminants. Although, no
data were collected on bacteriological contaminants,
based on data from some counties, they could also
be substantial. We expect that at least half of the
state's private water supplies would not meet
standards established by the Safe Drinking Water Act
and amendments.
The Kansas Cooperative Extension Service conducts
an educational program on household water quality. It
provides information on quality of water from private
water wells and in all homes. It includes training of
county agricultural, home economics, 4-H and youth
agents, and county health services personnel as local
sources of information. New extension bulletins and
leaflets address water quality, water testing, water
treatment, and water quality protection. Other
resource materials include video tape and slide/tape
sets and news programs (radio, TV, newspaper, and
newsletter).
Safe drinking water is an important issue. Until more
people have safe water, this will-likely be an
increasingly important issue. Our household water
quality program addresses the questions, helps
people evaluate their problems, and shows them how
to seek solutions.
24
-------
FEDERAL TRADE COMMISSION REGULATION OF WATER TREATMENT DEVICES
Joel Winston
Federal Trade Commission
Washington, DC 20580
The Federal Trade Commission (FTC) is a small
independent Federal agency established in 1914. It
has a broad mandate: to promote free and open
competition and protect consumers from unfair and
deceptive practices. The FTC's mission includes the
regulation of anticompetitive practices such as price
fixing and monopolization, as well as regulation of
business practices that deceive consumers, including
false advertising. The FTC works with private and
governmental consumer agencies including the Better
Business Bureaus, U.S. Postal Inspectors, state
attorneys general, and other Federal agencies.
Generally, the FTC can act only in the broad public
interest, i.e., when a problem affects large groups of
consumers or causes significant harm to smaller
groups. Advertising regulation by the FTC tends to
focus on national or regional ad campaigns.
The FTC has strong remedial powers when it finds a
practice to be unfair or deceptive. While it cannot
impose criminal penalties, it can issue broad cease
and desist orders to prevent recurrence of the
violations or similar practices. In some cases the FTC
can obtain Federal court orders enjoining violations,
imposing civil penalties, and/or requiring the
wrongdoer to make restitution to deceived
consumers.
In determining whether an advertisement is deceptive,
the FTC first looks at the ad itself to determine what
message it conveys to consumers. That message
may be explicit or implicit. The FTC then determines
whether the ad is likely to deceive reasonable
consumers. Both affirmative misrepresentations and
omissions of fact may be deceptive. In either case,
the FTC will act only when the representation or
omission is material, i.e., likely to affect consumers'
purchase decisions.
The FTC also enforces its advertising substantiation
doctrine. Advertisers making objective claims about
their products must have a reasonable basis for the
claims prior to making them. The type and amount of
substantiation that is required will depend on several
factors, including the type of claim, the product, the
consequences if the claim is false, the benefits if the
claim is true, the cost of developing substantiation for
the claim, and the amount of substantiation experts in
the field believe is reasonable.
The FTC does not have the resources to pursue
every potentially deceptive practice. The FTC
considers the following factors in deciding whether to
exercise its discretion to act in a particular case:
• The seriousness of the deception;
• The extent of consumer injury, including physical
and economic injury;
• The number of consumers misled; and
• The type of product involved, i.e., whether it is a
tow-cost, repeat-purchase or high-cost,
infrequently purchased item.
The FTC has no specific regulations governing the
advertising of water treatment devices. Like other
advertising, claims for these devices must be truthful
and substantiated. Claims for these devices are often
credence claims. This means that consumers are not
able to evaluate the truth of the claims themselves.
For example, a representation that a water treatment
device protects users from the hazards of chemical
pollutants cannot be evaluated by consumers. Under
these circumstances the FTC scrutinizes the
advertising more closely.
In recent months, the FTC has brought two formal
actions against advertisers of water treatment
devices. In the summer of 1987, New Medical
Techniques, Inc. was charged with false advertising
of its Aquaspring Home Water Distiller. The company
made a variety of claims about the capabilities of the
Aquaspring — that it produces pure water, that it
will remove all contaminants from the water, and that
it will remove all chemicals. In fact, the FTC alleged,
the devices were not capable of filtering volatile
organic chemicals (such as chloroform and benzene),
many of which may be hazardous to health. The
company agreed to sign a consent order which
25
-------
prohibits the deceptive claims and requires certain
affirmative disclosures in future advertising.
In August 1987, the FTC issued an administrative
complaint against North American Philips Corporation,
maker of the Norelco Clean Water Machine, a table-
top activated carbon filter device. The company's
advertisements allegedly represented that the Clean
Water Machine would make tap water clean or
cleaner, and would help remove organic chemicals.
According to the FTC complaint, the machine actually
introduced a potentially hazardous organic chemical,
methylene chloride, into the water it treated. The
company is charged with false advertising and failing
to disclose the alleged methylene chloride
contamination. The case is scheduled for trial before
an administrative judge in early 1988.
In addition to these cases, the FTC continues to
monitor advertising for water treatment issues. They
are assisted in this effort by the EPA, with which they
share authority in this area. EPA has, and continues
to provide, scientific expertise to enable the FTC to
evaluate and prosecute cases.
In general, the types of advertising claims that the
FTC may be concerned about in this area are as
follows:
• Claims that a device will purify the water or remove
all contaminants. These claims should be
supported by reliable evidence and appropriately
qualified.
• Claims that ordinary tap water is hazardous to
health. These claims should be carefully
substantiated and qualified.
• False claims of consumer or expert
endorsements. Consumer testimonials should be
representative or qualified clearly.
26
-------
POUlPOE PRODUCT PROMOTION GUIDELINES AND CODE OF ETHICS
Maribeth M. Robb
Water Quality Association
Lisle, IL 60532
INTRODUCTION
To gain perspective on our subject, we have to go
back about 50 years, when the point-of-use/point-
of-entry (POU/POE) water quality improvement
industry was born. It was hardly the time to be
developing a new consumer product. The country
was still in the depths of the depression, and
economic prospects in any industry were uncertain.
Founding an emerging industry in water quality, in a
country known for some of the best water in the
world, made the venture even more speculative.
But, those early entrepreneurs saw a niche they
thought needed filling: to take what many considered
perfect water and tailor it to the specific, individual
needs of the user. So they took their ideas and their
products into homes and businesses. They
demonstrated product effectiveness and, through
aggressive marketing, the POU/POE water quality
improvement industry evolved.
The success of the industry in the decades since has
validated the judgment of those pioneers. For them
and for POU/POE, pounding the pavement, talking to
customers, and marketing their products were
dominant activities for many years.
As the industry grew, so did the number of members
representing it. The need for training and education
also grew as the many purveyors of POU/POE
needed to keep pace with the rising sophistication of
the industry. Industry-sponsored organizations, like
the Water Quality Association (WQA), were born to
help fill that need. WQA continues to fulfill that
function today.
ANSWERING CONSUMER CONCERNS
The influences of the marketplace began to change.
Because new testing techniques permitted scientists
to determine very low levels of toxicity, the quality of
the nation's water supply was being reassessed on a
daily basis in our newspapers. Many Americans
became concerned about the quality of their drinking
water. So more products were added to the original
lines to meet those consumer needs as well. The
industry took on yet another dimension.
Today, experts disagree on the seriousness of the
water quality problem. Some argue that the chemical
compounds are in such minute quantities that they
pose little or no risk to health. Others are more
concerned. They worry about long-term exposure to
many of the chemicals now detectable.
Even though the issue is unresolved, the fact remains
that many individuals find their water unacceptable:
too hard, cloudy, smelly, or funny tasting. They may
have health concerns about water 'quality for their
family, their infants and young children, and during
pregnancies or times of special illnesses.
As a result, there is a growing demand for home
treatment systems and for bottled water. Due to this
demand, the water quality improvement industry has
been thrust from its traditional role of aesthetic water
treatment into the role of reducing health-related
contaminants. This, in turn, has raised new questions
about promotional claims made for various industry
products.
VOLUNTARY INDUSTRY PRODUCT
PROMOTION GUIDELINES
The promulgation of the Voluntary Industry Product
Promotion Guidelines and the creation of the Water
Quality Industry Review Panel was prompted by
concerns expressed to WQA by various agencies of
the Federal governments of the United States and
Canada, state and provincial enforcement agencies,
and members of the water quality improvement
industry. They questioned the general level of
industry advertising and promotional claims, and
expressed the view that the ads often fall below
acceptable norms of accuracy and completeness.
Although WQA was not necessarily in full agreement
with these opinions, it nevertheless believed it should
respond to them on behalf of the industry in a positive
and effective manner. It was hoped that this response
would also stimulate companies in the industry to
27
-------
undertake a thorough and comprehensive review of
their promotional material. There is plenty of evidence
that this occurred and that this activity is ongoing.
The guidelines are not intended to provide all or any
part of the wording of anyone's specific promotional
material. They are merely designed to provide a
general framework within which more accurate and
informative advertising, promotional, and sales
presentation material can be prepared in such a way
as to avoid misleading consumers about the
capabilities of water quality improvement products.
Companies in the POU/POE industry are fully aware
that they cannot, acting either through WQA or
otherwise, agree on matters relating to the form or
content of their promotional material or their policies
in these areas.
The industry recognizes that promotional material,
including advertising and sales presentation material,
is a key element in competition and must be left to
each individual company to develop for itself.
EPA SUPPORT
From the beginning, the U.S. EPA encouraged WQA
to develop a private sector program to address this
issue. In a letter dated and signed by three EPA
officials, we received much needed support.
"Over the past ten years, we have experienced a
variety of questionable advertising and sales claims
by manufacturers or salespeople of water
treatment units.
"Unfortunately, the 'wildest' claims have come to
be associated with the water quality industry in the
mind of public water supply professionals.
"Yet, this is probably not a true characterization of
the major companies operating in the water quality
field. It would appear that WQA's program is a
substantial forward step in correcting this false
image.
"Perhaps one of the greatest contributions of the
WQA Review Panel would be the preparation of
short 'state-of-nation' assessments of
advertising and sales claims at the start of the
project and periodically in such a way as to
enhance public understanding and knowledge. It
will also be worthwhile to have an industry forum to
which questionable advertising claims can be
referred.
"We realize that the WQA Panel will be trying to
cope with a difficult problem, but its importance
emphasizes the need for the work. Please be
assured of our support for your efforts."
GUIDELINES PROVISIONS
The guidelines were adopted in March of 1985 and
revised in April of 1987.
The painstaking process which led to these guidelines
assured the participants of the following:
Complaints that are based on factual data;
Product performance and benefit claims that are
verifiable;
Visuals that are clear and unambiguous;
Prohibition of untrue, misleading, deceptive,
fraudulent, or falsely disparaging claims;
Prohibition of sweeping, absolute statements;
True and accurate advertisements;
Inclusion of pertinent facts;
Avoiding confusing terminology;
Performance claims that are based on fact; and
Problem/solution scenarios that enumerate
circumstances and specifics.
Other provisions are spelled out in the handling of:
Warranties, guarantees, equivalent terms;
Layouts and illustrations;
Asterisks;
Abbreviations;
Comparisons/disparagement of competition; and
Testimonials and endorsements.
GUIDELINES REQUESTS
Two types of requests can be submitted through the
Voluntary Industry Product Promotion Guidelines
process. They are a Complaint Request, filed on an
existing advertisement or promotion which can be
voluntarily resolved or can move on the Review
Panel; and an Advisory Request, in which a company
submits its promotional material or advertising for
compliance prior to printing or releasing it.
As you can see from Figure 1, communication among
involved parties is important to this process from the
very beginning. In fact, many of the Complaint
Requests submitted are voluntarily resolved before
they leave the Staff Review Committee.
REVIEW PANEL
For those that do progress to the Review Panel, the
complaint is heard by a highly qualified and
conscientious independent panel. This panel forms
the "teeth" of the program. Their experience,
impartiality, and thorough consideration of each
request give the program credibility and consistency.
The guidelines are specific on the credentials of the
Review Panel members. It calls for:
28
-------
Figure 1. Procedures for POU/POE voluntary product promotion guidelines.
Complaint Request
Mailed to the Subject
20 days
_L
Subject Agrees
Mailed to Complaining Party
15 days
_L
Subject C
)isagrees
I No Re
sponse
No Response
Withdraw Request
Refer to Staff
Review Committee
Refer to Staff
Review Committee
Ends Request
Ends Request
Complaint is
•
Subject Agrees
Decision is Communicated
to Both Parties
I
1 5 days
Subject Disagree
I
s No Response
> Unjustified
Ends Request
Ends Request
Lj
Review Panel
n
Calendar
Panel Option
• A citizen/consumer - Currently that post is filled
by Linda Elwell, who brings her background and
experience in direct consumer marketing to her
position;
• An educational institution-connected water
chemist - This position is well served by Roger E.
Machmeier, Ph.D., P.E., of the University of
Minnesota;
• A water treatment equipment specialist--
nonindustry - Nina McClelland, Ph.D., brings
years of experience with regulation, validation and
testing to bear on her panel considerations;
• A water treatment equipment specialist--
industry, but no longer actively employed by
industry member - Wes McGowan now works as
a consultant and brings years of industry
experience to his panel decisions; and
• A current former Federal or state government
regulatory person who has or is working on
consumer and/or misleading advertising problems
- Ron Graham of the Better Business Bureau of
Minnesota, Inc. combines his technical prowess
with a working knowledge of small-business
operations.
COMPLAINT REQUEST CRITERIA
The Complaint Requests heard by the panel
never frivolous. The criteria includes:
are
Name, address, and telephone number of source
material company;
Written and dated entries;
Copy of all materials to be reviewed;
Details-how, when, where the material was
used;
Particular guidelines possibly violated;
Explicit request for opinion;
Full legal name, address, and telephone number of
submitting party; and
The requesting party agrees to not misrepresent
and to limit reference to opinion or results.
Figure 2 represents the first two years of the program
to date. As you can see, the total number of
Complaint Requests increased by two in the second
year of the program. However, during the same
period, those cases voluntarily resolved increased by
29
-------
four and the total number of Advisory Requests
increased by three.
Figure 2. Voluntary guidelines history.
20
10
Total CR
VRCR
Panel
CR - Complaint Requests
VRCR- Voluntarily Resolved Complaint Requests
Panol - Panel Review Cases
AR - Advisory Requests
VOLUNTARY COMPLIANCE
What we seem to be seeing is a trend toward
voluntary compliance, either after a complaint has
been filed or before the material is released. These
are encouraging figures for the industry and for
consumers as well.
There is also a question of perspective that I would
like to stress here. Of the millions of dollars spent by
the industry on advertising every year, of the
thousands of promotional pieces and the hundreds of
advertisements, fewer than 80 have been submitted
to the Voluntary Industry Promotion Guidelines
Review Program during its two years of operation. A
substantial proportion of those advertising materials
that did reach the Review Panel were submitted by
their own companies to assure compliance prior to
publication.
So, changes are being made, and they are being felt
in the marketplace.
CODE OF ETHICS
Building on the encouraging response to the
Voluntary Industry Product Promotion Guidelines,
industry members have now taken another
courageous step. Using the guidelines as a nucleus,
they have launched a Voluntary Industry Code of
Ethics program.
Although it is strictly voluntary, it is hoped that the
marketplace, which fueled this industry from the start,
will see the benefits of trading with Voluntary Industry
Code of Ethics signature companies, and will
embrace it as a stipulation for doing business.
The formal hearings procedure for the Code of Ethics
will be the same as for the Voluntary Industry Product
Promotion Guidelines.
The Voluntary Industry Code of Ethics will be
published in early 1988, with the first list of
subscribers-in-good-standing published in the
second quarter of 1988.
LOOKING AHEAD
Times have changed, and the point-of-use/point-
of-entry water quality improvement industry has
changed as well.
That change is apparent from marketing, the
Voluntary Industry Product Promotion Guidelines and
Code of Ethics, and Product Validation.
Water quality is no longer a simple issue. It demands
a complex approach to problem solving as well as a
blending of industries and specialties to assure
optimum quality for all consumers, in all regions of
the country.
The point-of-use/point-of-entry water quality
improvement industry is poised to meet that challenge
with accountability, confidence, and technical
expertise.
We have the industry products, the industry people
and the drive. By working with professionals like
yourselves .-- the regulators, the 4educators, the
resource people, and the water utility managers --
we can provide the solutions demanded and needed
by consumers today. Together, we can bring
economical, customized, quality water to every tap in
this country.
30
-------
NSF's LISTING PROGRAM FOR POUlPOE DWTUs
Randy A. Dougherty
National Sanitation Foundation
Ann Arbor, Ml 48106
The National Sanitation Foundation (NSF) was
chartered in 1944 in the State of Michigan as a
private, independent, not-for-profit organization.
The mission of NSF is to develop and administer
programs relating to public health and the
environment in areas of service, research, and
education. NSF is best known for its consensus
standards and third-party certification programs.
The subject of this article is NSF's listing program for
point-of-use and point-of-entry drinking water
treatment units (POU/POE DWTUs).
NSF standards are consensus standards and, as
such, are part of the public domain. A company can
self-certify to NSF standards, or another
organization may certify them. The standards specify
the minimum requirements for a product to satisfy
public health concerns. Additional requirements that a
company must meet to obtain and maintain
authorization for listing and use of an NSF Listing
Mark are specified by contract and administrative
policy. The key elements of NSF's listing programs
(how NSF certifies initial and continuing conformance
to NSF standards) are:
NSF standards;
Registered Listing Marks;
Public listings information;
Evaluation and testing by an independent, objective
third-party;
Monitoring;
Corrective action; and
Enforcement.
NSF STANDARDS
NSF standards are voluntary, consensus standards
developed by a Joint Committee (comprised of
regulators, users, and industry representatives),
reviewed and accepted by the Council of Public
Health Consultants (CPHC), and reviewed and
formally adopted by NSF's Board of Trustees.
The standards are developed or revised by the Joint
Committee (or task groups appointed by the Joint
Committee) with the active participation of public
health and other regulatory officials, users, and
industry.
The role of the CPHC is to assure that the
requirements of a standard satisfy public health
concerns. CPHC has 36 members from Federal,
state, and local regulatory agencies in the United
States and other countries, and academia. Industry is
not represented. The expertise of the members
includes public health, medicine, chemistry,
toxicology, epidemiology, microbiology, and
engineering. CPHC reviews and must accept a
standard or revision before it is sent to the board for
adoption; however, it does not make any changes to
a standard. If the council does not accept a standard,
it is sent back to the Joint Committee. The CPHC is
not a "rubber stamp." For example, the council
rejected proposed Standard 55 (for ultraviolet drinking
water treatment systems) in 1985 because it did not
satisfy public health concerns about cysts, turbidity,
and viruses.
The Board of Trustees reviews the standards for
business and legal consideration. As with the CPHC,
the board does not make changes in a standard. If
the board does not accept a standard, it is sent back
to the Joint Committee.
NSF has the only standards for POU/POE DWTUs
that have widespread recognition and acceptance by
public health officials. These are:
• Standard 42: Drinking Water Treatment Units -
Aesthetic Effects;
• Standard 53: Drinking Water Treatment Units -
Health Effects; and
• Standard 58: Reverse Osmosis Drinking Water
Treatment Systems.
Shortly after adoption of Standard 53 in 1981,
Canadian authorities proposed a ban on point-of-
use carbon units because of concern for
bioaccumulation - specifically, of opportunistic
pathogens. NSF organized and hosted a meeting of
31
-------
Canadian and U.S. regulatory representatives,
manufacturers, and recognized expert consultants. It
was reported at the meeting that no known illness
could be traced to carbon filters. There was general
agreement to add to the standards, and to the labels
on carbon units, the following statement: "Activated
carbon filter units covered by this standard are not
intended to be used where the water is
microbiologically unsafe or with water of unknown
quality." A revision to Standard 53, to incorporate this
statement and additional labeling requirement, was
adopted in June 1982. Standard 42 was first adopted
in 1973, but was revised in June 1982 to be a
companion document to Standard 53. Standard 42 is
consistent with Standard 53, but is for aesthetic
claims. Although these standards (42 and 53) are not
limited to specific treatment technologies, the test
protocols are appropriate for carbon or mechanical
filtration units only.
Standard 58, for reverse-osmosis systems, was
adopted in November 1986.
Them are three other standards being developed:
* Proposed Standard 44: Cation Exchange Water
Softeners;
• Proposed Standard 55: Ultraviolet Disinfection
Systems; and
• Proposed Standard 62 (for distillation systems).
The standards are for units designed to be used for
the reduction of specific contaminants from public or
private drinking water supplies. The standards have
detailed requirements and protocols for testing the
units to verify claims for the reduction of specific
chemical, particulate, or microbiological contaminants,
bacteriostasis, disinfection (Standard 55 only), or for
the addition of polyphosphates or silicates (Standard
42 only).
While the primary focus is verification of water
treatment claims, the standards include requirements
for materials, design, and construction of units to
assure that:
• Materials in contact with the drinking water do not
impart toxic substances, taste, odor, or color to the
water; and
• The units accomplish the intended purpose when
installed and operated in accordance with the
manufacturer's instructions.
For DWTUs with water treatment claims for the
reduction of contaminants that are established or
potential health hazards, the standards include
requirements for performance indicators, warnings, or
other means to alert the user when the unit is not
functioning properly. This may be by the DWTU
having a shut-off to terminate the discharge of
treated water, sounding an alarm, 50 percent
reduction in flow, or by providing a test kit. For carbon
units, one alternative is to have a 100 percent safety
factor, which is verified by testing to twice the rated
capacity. For reverse osmosis systems for nitrate
reduction, Standard 58 requires either a nitrate
monitor on the unit, or the manufacturer must provide
a test kit for nitrates with the system.
The standards also have detailed requirements for
installation and operating instructions, dataplate
information and labeling, and other information about
the function and capability of a unit, including specific
warnings for users.
NSF LISTING MARK
NSF Listing Marks are formally registered with the
U.S. Patent Office, and in Canada. NSF owns the
mark, but doesn't use it on products. The mark is for
use by other companies, on listed products and in
conjunction with listed products, as authorized by
NSF. A company applies for and contracts with NSF
for authority to use an NSF mark; and specifically
agrees to use the mark on only new products fully
complying with all NSF requirements. NSF has legal,
contractual, and ethical obligations and responsibilities
to monitor and verify that only authorized companies
use the mark, and use it properly. This is the basis of
NSF's authority relating to listed products.
To be considered listed, a DWTU must bear the NSF
Listing Mark, and must also bear a model number (or
serial number) that distinguishes it from nonlisted
units. Consumers and regulators can look for the
mark as evidence that a unit is listed by NSF.
The listing program for DWTUs differs from other
NSF listing programs in that listed units do not have
to meet the same requirements - we verify the
specific water treatment claims made by the
manufacturer for a unit. So the water treatment claims
must be directly associated with the mark. The listing
mark with an example of verified water treatment
claims is shown in Figure 1.
PUBLIC LISTINGS INFORMATION
One goal of listing services is to make current listings
information readily available and easily accessible.
This is achieved by publishing and widely distributing
seven annual listing books (see Table 1). NSF also
publishes up to nine supplements to each annual
book. These supplements include complete listing
information for new companies, and for companies
that cancel listing services; for revised listings, the
supplements have the changes only. Because listings
change daily, it is impossible to provide current
listings information by publication. Therefore,
32
-------
Figure 1. NSF listing mark on a DWTU.
Listed under NSF
Standard 53 for the
reduction of TTHMs,
Cysts, and Turbidity
only.
Listed .under NSF
Standard 42 for the
reduction of Taste, Odor,
and Chlorine only.
Caution: Do not use where the water is microbiologically unsafe
or with water of unknown quality, without adequate
disinfection before and after the unit.
EVALUATION AND TESTING BY AN
INDEPENDENT, OBJECTIVE THIRD-
PARTY
NSF evaluates and tests products as an independent,
objective third-party. As a third-party, NSF serves
the interests of regulators and users, as well as
industry. NSF has five regional offices, four in the
continental United States, and one in Brussels,
Belgium. NSF regional personnel visit each
production location (point of final assembly or
production) to evaluate products and select samples
for laboratory testing. NSF has its own modern,
state-of-the-art laboratory in Ann Arbor, Michigan,
and provides a full range of chemical,
physical/performance, and microbiological testing.
(NSF also has procedures for qualifying and
authorizing other laboratories as subcontract or
alternate testing laboratories for testing products for
listing by NSF.)
Table 1. NSF Listing Publications
Title
Distribution
Food Service Equipment and Related Products, 6,500
Components, and Materials
Plastics Piping System Components and Related 3,600
Materials
Drinking Water Treatment Units and Related 3,200
Products, Components and Materials
Swimming Pools, Spas, and Hot Tubs Circulation 3,000
System Components
Special Categories of Equipment, Products, and 2,800
Services
Wastewater Treatment Units and Related 2,400
Products and Components
Class II Biohazard Cabinetry 1,600
Total 23,100
beginning in January 1987, NSF provides for direct
electronic access by computer. Any person with a
compatible computer and modem can apply for this
service and directly access official listings information,
which is updated daily. The only cost to a user for
this service is the cost of the telephone call.
Another goal is to provide listings information that can
be useful to someone selecting a DWTU; therefore,
the listings for DWTUs include the following
information: the company name and address, a
description of the unit or system, the model number
of the DWTU and replacement element, and the
function (the verified water treatment claims). The
listings information also includes the service cycle or
capacity in gallons, flow rate, and other information.
MONITORING
Listing is on an annual basis. The listing program is
not a one-shot deal, but an ongoing program with
continuous monitoring of listed products'
conformance to NSF standards. The listing program
for DWTUs includes a requirement for annual
unannounced inspections by our regional personnel to
verify that there have been no unauthorized changes
in materials, components, design, or production of
listed units. NSF also requires periodic retesting (at
least once every five years) of listed DWTUs.
NSF investigates complaints of noncompliance of
listed units. The complaints may be from public health
or other regulatory officials, or from users. NSF also
investigates complaints from other manufacturers - a
mechanism for effective self-policing by industry. In
all cases, NSF conducts its own investigation, and
takes action with the manufacturer if, and only if, NSF
confirms that the product doesn't comply.
CORRECTIVE ACTION
The goal of the listing program is to assure that a
listed product conforms to an NSF standard. If NSF
determines that a listed unit does not conform with
the standard, NSF requires the listed company to take
appropriate corrective action, which may include the
fpllowing:
• Evaluation or testing to quality changes to
products,
• Modification of equipment,
• Destruction of product,
• Product recall, and
• Public notice.
33
-------
ENFORCEMENT
NSF zealously strives to maintain the integrity and
credibility of the mark; therefore, for repetitive or
serious noncompliance, NSF takes specific
enforcement action as follows:
Increased monitoring,
Administrative hearings,
Delisting,
Legal action, and
Cancellation of contract.
CURRENT STATUS OF THE LISTING
PROGRAM AND FUTU.RE
DEVELOPMENTS
The listing program for DWTUs is small. As shown in
Table 2, there are only 13 companies with listed
units. But, this is one of the fastest growing listing
programs. Twenty-one additional companies have
applied for listing; and two additional standards
(Standard 44 for water softeners and Standard 55 for
ultraviolet systems) are expected to be adopted by
the end of 1987. Although the program is small, it has
the third largest distribution of listing publications (see
Table 1), which is an indication of the interest and
importance of this program to regulatory officials and
consumers.
One exciting item is the development of a "model
compound" concept for testing carbon units for
volatile organic compounds (VOCs). Under the
guidance of a Standard 53 task group, NSF and a
listed company have developed isotherm data and
dynamic testing data which demonstrate that
chloroform can be used as a satisfactory model
compound for verifying the reduction of a number of
specific VOCs (regulated and nonregulated) by
carbon units. This will result in reduced testing costs
for verifying contaminant reduction claims for a large
number of organic chemicals, producing increased
participation by industry and an increased number of
listed models. But of even greater importance, as new
organic contaminants are found in drinking water
supplies, it may be possible to demonstrate that
chloroform is a satisfactory model compound for
verifying effective reduction by carbon units - which
means that there may be a large number of listed
units immediately available as a remedy.
Table 2. Listing Service Programs
Program
Food Service Equipment
Plastics Piping System Components
Swimming Pools, Spas, and Hot Tubs
Class II Biohazard Cabinetry
Drinking Water Treatment Units
Wastewater Treatment Units
Flexible Membrane Liners
Special Categories
Total
(9/1/87)
Number o
Standards
21
1
1
1
3
4
1
5
37
f Listed
; Companies
1,033
275
49
12
13
12
3
31
1,428
SUMMARY
The listing programs are voluntary. But a listing
program becomes more than voluntary when NSF
standards are referenced in regulations or codes.
Regulations or codes usually do not require that a
product be listed by NSF, but a listed product is
usually accepted by the responsible regulatory
agency. By voluntarily participating in the listing
program, with required testing, retesting, and
unannounced plant inspections by a third-party, a
company demonstrates the intent and capability to
manufacture a product conforming to an NSF
standard. The advantage to a company is wide
acceptance of its listed product(s) by regulatory
officials and consumers.
Regulatory officials and consumers have assurance
that a credible, objective third-party, widely
recognized by public health officials, has actually
tested and verified that listed products comply with
specific standards; and the cost of the program is
placed in the private sector rather than adding to the
cost of official regulation.
34
-------
WATER QUALITY ASSOCIATION VOLUNTARY PRODUCT VALIDATION PROGRAM
AND VOLUNTARY CERTIFICATION PROGRAM
Lucius Cole
Water Quality Association
Lisle, IL 60532
PRODUCT VALIDATION
One of the basic purposes of the Water Quality
Association (WQA) is to promote the acceptance and
use of point-of-use/point-of-entry industry
equipment, products, and services. One of the most
successful programs to promote this concept was the
development of voluntary industry standards.
Consistent with the goals expressed in WQA's
corporate charter, which are "to foster the further
development of equipment, products and services in
the industry for the purpose of providing a better way
of life for all mankind," the procedures set out in both
the Guide for Product Standards Development and
the Guide for Product Validation Program were
established and are followed in the development of
voluntary industry standards. The WOA developed
five voluntary industry standards:
• S-100-85 for household commercial and
portable exchange water softeners;
• S-101-80 for efficiency-rated water softeners;
• S-200-73 for household and commercial water
filters;
• S-300-84 for point-of-use low pressure
reverse osmosis drinking water systems; and
• S-400-86 for distillation drinking water systems.
Under this program, a manufacturer may voluntarily
submit a specific type of equipment or system to the
WQA laboratory where its performance will be
carefully evaluated in accordance with the appropriate
standard. When a system has successfully performed
to the specific testing protocol, it is then qualified to
receive the appropriate "gold seal." This seal alerts
the consumer that the equipment he/she is
considering purchasing complies with the
specifications of a very rigorous testing program. At
the present time, over 300 products produced by
companies in the water quality improvement industry
have been validated by the WQA laboratory. A
directory of these validated water conditioning
products is published semi-annually, and is available
to both consumers and regulatory officials.
A 10-page brochure published by the Council of
Better Business Bureaus, Inc., entitled Tips on Water
Conditioners, makes the following statement:
"When choosing a water conditioner, look for
equipment that bears the gold seal of the Water
Quality Association. This seal indicates that the
Water Quality Association has judged that the
equipment complies with the specifications of the
industry standards for water softeners (S-100)."
A great deal of effort has been made by the point-
of-use/point-of-entry industry to provide the
consumer with products that are reliable and perform
to basic standard requirements. It has also been the
industry's desire to have these systems installed in a
safe and economic fashion.
PROFESSIONAL CERTIFICATION
The WQA certification program was established by
association members to provide industry-wide
standards for evaluating the knowledge of point-of-
use/point-of-entry water treatment personnel and to
improve the knowledge of those who service the
consumer, thereby encouraging professionalism and
integrity in the industry. Since the inception of WQA's
certification program in 1977, nearly 1,800 people
have been certified in one of three categories: dealer,
specialist, or installer.
In order to provide support for industry-wide
standards for evaluating knowledge, it was vital to
develop proper study materials. The water treatment
fundamentals correspondence course was prepared
to provide comprehensive study materials for the
dealer or specialist. This course was divided into 12
lessons. Each lesson consists of a pre-lesson
questionnaire to evaluate one's knowledge about the
subject, exercises to help evaluate one's knowledge
35
-------
while preparing a lesson, and finally, questionnaires
(hat are completed and mailed to the WQA
headquarters for correction. The corrected
questionnaires are then returned to the student for his
review. This study material has been widely used by
individuals who desire to raise their level of
technological competency in the point-of-
use/point-of-entry industry, wish to participate in
the WQA certification program, and want to be able to
display the coveted certification emblem.
The necessity and function of various water
conditioning systems are well recognized. Equally
important to the consumer is the correct installation of
the equipment into the plumbing system, since the
equipment usually connects to a potable water supply
and to an appropriate drainage system.
Improper installations are not only hazardous, but also
costly when corrections must be made. In order to
prevent such problems from occurring, it is essential
that the installer be knowledgeable in acceptable
installation procedures as well as local code
requirements.
The purpose of the WQA installers home study
course is to provide installers with generally
acceptable installation procedures relating to point-
of-use/point-of entry water conditioning systems.
With such knowledge, an installer may perform the
installation procedures correctly so that a safe and
efficient installation is made. Similar lesson
procedures as previously discussed for the
fundamentals course are used to assist and evaluate
the student's knowledge.
A national directory of certified personnel is published
every two years listing the three categories of
certification - certified dealers, certified specialists,
and certified installers. Each individual listed has
completed a specific study course and demonstrated
his knowledge by successfully passing the
appropriate examination. As a certified individual, he
agrees to maintain high standards of service. The
Water Quality Association Certification and Education
Committee may revoke an individual's right to use the
seal of certification if evidence of failure to maintain
these standards is established.
The members of the Water Quality Association
continue to demonstrate their interest in serving the
consumer with both certified specialist/ installer
programs and validated products. WQA members
over the last 32 years have provided funding in
excess of $250,000 for the development and
implementation of five WQA standards and six NSF
standards to meet the association's goal "to foster
the development of equipment, products and services
by the industry for the purpose of providing a better
way of life for all mankind."
36
-------
GUIDE STANDARD AND PROTOCOL FOR TESTING MICROBIOLOGICAL WATER PURIFIERS
Stephen A. Schaub
U.S. Army Biomedkal Research and Development Laboratory
Frederick, MD 21701
Charles P. Gerba
Department of Microbiology and Immunology
University of Arizona
Tucson, AZ 85721
INTRODUCTION
Over the past several decades, a number of water
equipment manufacturers have developed technology
for the removal of chemical and microbiological
constituents from waters to be used for personal
consumption. The need for this capability arises
principally from consumer interest in improving the
quality of untreated or partially treated waters like
those used by hikers, campers, recreational home
and boat owners, and families or communities having
individual home or small system water sources.
One of the major concerns in water treatment is the
need to remove pathogenic microorganisms (bacteria,
viruses, protozoa, fungi, and helminths) from the
water before its consumption, since it is recognized
that infectious disease transmission by water is a
significant public health concern.
It is important that water treatment units or devices
designed for the protection of human health be
effective against pathogenic microorganisms in
untreated or partially treated water, and be capable of
providing this service over the designed operational
life of the equipment in waters likely to be
encountered in the United States. These
requirements are necessary for protection of the
public's health by both the water industry and the
government.
A multidisciplinary task force was formed in 1984 to
develop a Guide Standard and Protocol for Testing of
Microbiological Water Purifiers. The task force was
comprised of persons representing the interest areas
of academia, industry, and government for research
and development, product evaluation and registration,
and product regulation and enforcement. The
objective of this task force was the development of a
standard and protocol that industry, government, and
consumers could agree with and which could be
attained with current knowledge and technology. The
primary emphasis was to protect the consumer.
At this time, the guide standard and protocol has
been prepared by the task force, has been technically
reviewed (notice in Federal Register of May 29, 1986;
and U.S. EPA Science Advisory Panel), and has been
appropriately revised in consideration of these
reviews. It has been accepted on a provisional basis
by the U.S. EPA's Office of Drinking Water and Office
of Pesticide Programs, pending experimental
verification of the efficacy of the protocol under the
prescribed parameters.
The intent of this paper is to provide the major
features and considerations of the guide standard and
protocol in its current configuration. It is
recommended that persons or organizations wishing
to use the guide standard and protocol for testing
purposes, obtain the complete, detailed package from
the above U.S. EPA program offices.
REQUIREMENTS FOR A MICRO-
BIOLOGICAL WATER PURIFIER
The current definition of a microbiological water
purifier is that it must remove, kill, or inactivate all
types of disease-causing microorganisms from water
to make the product safe to drink.
Units or devices having limited claims for the
treatment or removal of a specific type of organism,
or use in a specific, limited application, can be tested
for that use in accordance with the protocol, however,
such equipment cannot be called a microbiological
water purifier. For example, a protozoan cyst removal
unit could be tested against the protocol and could
demonstrate acceptable cyst removal, but unless it
also met the required removals for the bacteria and
viruses, it could not claim to be a microbiological
water purifier.
37
-------
PRINCIPLES FOR THE
STANDARD AND PROTOCOL
GUIDE
The guide standard and protocol is to be considered
a general guide, presenting only the minimum
features and framework for testing, and may be
amended or added to for the evaluation of unique
units or specific operational problems (including
alternative organisms and procedures) as long as the
level of testing and intent of the protocol are not
diminished. It is performance-based, utilizing realistic
worst case conditions. The goal is to ensure that
microbiological requirements of the National Primary
Drinking Water Regulations are met by equipment
defined as microbiological water purifiers.
The guide standard and protocol is intended to be a
living document, subject to revision and update as
new knowledge and technology arise. The document
should be capable of addressing appropriate test
challenges for other types of purifiers when they
become available, and would consider new or
evolving pathogens of concern if they represent an
increased challenge to technology covered by the
protocol.
It was intended that the test conditions or
requirements of the protocol could be met in
reasonably well-equipped laboratories when
performed by competent scientists and engineers. It
is known that there are at least several commercial
and university laboratories in the U.S. that currently
have the capability to meet the test requirements.
The protocol does not and cannot address all
conceivable microbiological and physical/chemical
challenges that could be possible in water. Presently,
the test protocol addresses the following
technologies:
• Ceramic filtration candles or units (with or without
chemical bacteriostatic agents),
• Halogenated resins and units (with or without
filtration capabilities), and
• Ultraviolet (UV) units (with or without filtration
and/or chemical adsorption capabilities).
MICROBIOLOGICAL CHALLENGES FOR
WATER PURIFIERS
The microbiological challenges for testing were
chosen to be representative of bacterial, viral, and
protozoan pathogens of the gastrointestinal tract, and
are believed to cover the treatment requirements
presented by most other human pathogens from the
gastrointestinal tract or other origins, including fungi
and helminths. It is recognized that there are a
number of alternative organisms that could have been
selected, and which would be equally representative
for testing. A detailed rationale for the use and test
levels of the challenge organisms is provided in the
complete guide standard and protocol, which can be
obtained from the U.S. EPA offices mentioned in the
Introduction.
Table 1 provides a brief summary of the test
organisms and the culture/assay conditions required
for testing. In all cases the microbiological procedures
chosen represented well documented protocols or
standard methods, which could easily be attained in
the laboratory.
Table 2 provides the minimum microbiological
challenge levels to be used. A major point which must
be emphasized is that the challenge levels in most
instances exceed the highest concentrations that
would be found in typical source waters. It was the
task force opinion that the higher challenge levels
were less of a concern to the evaluation of
purification units than the complications arising from
the introduction of analytical errors, which could be
introduced from the effluent (product) water sample
concentration procedures. Low challenge levels would
necessitate sample concentration to quantitatively
assay the product waters if microbial removals were
significant, especially for the viruses and protozoa.
NONMICROBIOLOGICAL TEST
PARAMETERS
It was determined that, in addition to the
microbiological challenges to the various water
purifiers, there was a need to evaluate the treatment
capabilities of units in the presence of associated
physical/chemical parameters in water, which may
impact on the overall microbial removal capabilities of
each type of treatment technology. It was decided
that, for the first half of the testing procedure a
general challenge, typical of most tap waters, would
be utilized for all testing. The second half of the
testing program would use the worst case challenge
in which pH; Total Organic Carbon (TOC); Turbidity;
Total Dissolved Solids (TDS); Temperature; and for
UV light units, a UV Quenching Test Component,
would be added. Table 3 provides the test conditions
required for the various types of purifier tests and the
recommended materials or chemicals for adjusting
the water characteristics. Additionally, silver leaching
test conditions for units containing silver bacteriocide
are included. While many of the worst case
challenges appear to be on the high side of normal
conditions, they are not thought to be out of line with
conditions brought about by seasonal or
meteorological events or significant pollution events in
surface waters. The worst case challenges will be
maintained over the total duration of the second half
of the testing program with the exception of high
turbidity conditions, which would be introduced only
during sampling periods to prevent excessively rapid
clogging of units containing filtration components.
38
-------
Table 1. Microbiological Methods for Test Protocol
BACTERIA:
Organism
Culture Requirements
Assay
VIRUSES:
Organism
Culture Requirements
Assay
PROTOZOA:
Organism
Culture Production
Assay
Klebsiella terrigena
Overnight broth cultures to obtain stationary growth phase cells
Spread/pour plate or membrane filter techniques using nutrient agar, M.F.C., or M-Endo Medium*
Poliovirus Type 1 (LSc) and Rotavirus, Strain SA-11 or WA (both viruses will be tested together using equal
proportions to seed the challenge water)
Grown on tissue culture and prepared to provide monodisperse virus particles for tests
Plaque or immunofluorescent foci assays on continuous cell cultures
Giardia lamblia or Giardia muris where disinfection is principal mechanism; 4 to 6 jim spheres can be used
where occlusion filtration is the exclusive removal mechanism
Obtain and prepare cysts from feces of laboratory-infected animals
Count physical particles for filtration: determine viability of cysts (or trophozoites) for disinfectant-containing units
* Use procedures in Standard Methods for the Examination of Water and Wastewater, 16th Edition, APHA, or equivalent.
Table 2. Microbiological Challenge (testing according to NSF Standard 53 for cyst reduction
Organism Influent Challenge*
BACTERIA
Klebsiella terrigena (ATCC-33257)
VIRUS
Poliovirus 1 (LSc) (ATCC-VR-59), and
Rotavirus (WA or SA-1 1) (ATCC-VR-899 or VR-2018)
CYST (PROTOZOAN): G/ard;a~*
Girdia muris or Giardia lambia, or
As an option for units or components based on occlusion
particles or spheres, 4 to 6 ym
107/100 ml
1 x 107/l
1 x 107/l
10.6/1
filtration: 107/l
will be acceptable)
Minimum
Log
6
4
3
3
Required Reduction
Percent
99.9999
99.99**
99.9
99.9
* The influent challenge may constitute greater concentration than would be anticipated in source waters, but these are necessary to
properly test, analyze and quantitatively determine the indicated log reductions.
~ Virus types are to be mixed in roughly equal 1 x 107/I concentrations and a joint 4-log reduction will be acceptable.
*** It should be noted that new data and information with respect to cysts (i.e., Cryptosporidium or others) may in the future necessitate a
review of the organism of choice and of the challenge and reduction requirements.
Test Waters
General
Halogen Disinfection Tests
Chlorine and Others
Iodine
Ceramic Candle Tests
Ultraviolet Tests*
Silver Leaching Tests
PH*
6.5-8.5
9.0 + 0.2
5.0 ±0.2
9.0+0.2
6.5-8.5
5.0 + 0.2
TOG*", mg/l
0.1-5.0
>10
>10
£10
>10
-1.0
Turbidity*", NTU
0.1-5.0
>30
>30
>30
>30
0.1-5.0
Temperature, °C
20 + 5
4±1
4±1
4 + 1
4 + 1
20 + 5
TDSt, mg/l
50-500
1,500>150
1, 500 > 150
1,500 >. 150
1, 500 > 150
25-100
Recommended Materials for Adjusting Water Characteristics:
* Inorganic acid or base.
** Humic acids.
*** AC fine dust (part No. 1543094).
t Sea Salts (Signma Chemical Co. or equivalent).
$ p-hydroxybenzoic Acid (general purpose reagent).
Quench UV to just above alarm point. (Add color or reduce light intensity to just above point where low UV intensity alarm would be
triggered.
39
-------
PURIFIER TEST PROCEDURES
For testing of a purifier, it is recommended that three
units be set up in parallel, according to
manufacturer's instructions for normal line pressures
(~60 psig), but the flow rates through the units would
not be controlled. The required physical/chemical
characteristics for the tests would be maintained
continuously with the exception of turbidity as
mentioned above. The bacterial challenge should be
maintained continually during operation of the units,
but virus and protozoa would only be introduced
during the sampling "on" periods of the test. Samples
are to be taken in duplicate from each sampling
position (influent and product water) at the
appropriate sampling "on" periods specified in the
protocol. Sufficient void volumes will be passed
through the units before samples are taken to ensure
the uniformity of the challenge, especially for the
virus, protozoa, and turbidity, which are introduced
only during the sampling periods. An exception is
sampling after the programmed 48-hour stagnation
periods when samples are taken immediately upon
start-up. Disinfectant-containing purifier samples
(microbiological) will be immediately neutralized with
respect to that chemical.
The testing program will be conducted for 100
percent of the estimated treatment capacity of
halogenated resin-containing units and for a 10-1/2
day operating period with ceramic candle and UV
units. The operating cycles for the purifier units
should be representative of use (e.g., intermittent).
For example, a use cycle of every 15 to 40 minutes
during each operating day with an actual operating
"on" period of 10 percent for each cycle is
considered appropriate. (For example, if a 30-minute
cycle is used, the operating "on" period would be
three minutes/cycle.) If necessary, due to time or
laboratory constraints, a shorter operational day with
an extended test period can be substituted, or a daily
operating cycle of 20 percent "on" and 80 percent
"off" can be used.
A schematic of a typical test stand for the evaluation
of plumbed-in units is presented in Figure 1. This
schematic is essentially that of the National Sanitation
Foundation (NSF) Standard 53 for Drinking Water
Treatment Unit Health Effects. This set-up allows
good control of total test operations and sampling with
a minimum of variables entering the procedure.
Sampling can be performed on an automated basis.
Testing of portable or hand-held purifier units can be
set up in a batch testing procedure, which would
follow the test schematic of plumbed units as closely
as possible, although a number of the features of the
system such as flow meters, automated sampling
procedures, and delivery of virus, cyst, and turbidity
challenges would have to be modified.
Tables 4, 5, and 6 illustrate the sampling plans for the
various types of units. For units containing halogens,
the sampling of residual halogen in the product water
is conducted at the same times and frequencies as
the microbiological challenges. Additionally, at the
start of each test, the waters to be used for testing of
all type units are to be examined for U.S. EPA
primary and secondary pollutant constitutents in
accordance with standard analytical procedures. The
challenge conditions for iodine versus chlorine and
other halogen-containing units are identical except
after the 48-hour stagnation period at 75 percent of
the life of the units, wherein the pH challenge for
iodine-containing units becomes pH 5.0 rather than
9.0. The sampling for ceramic candles and UV units
is straightforward. Leaching tests for silver-
containing units are also necessary to make sure that
no dangerous levels of silver reach the product water.
MINIMUM MICROBIOLOGICAL
REMOVAL FOR ACCEPTANCE OF
PURIFIERS
In order to meet the standards of acceptable
microbiological removal, the three duplicate units
tested must continuously meet or exceed the defined
microbiological removal requirements, within allowable
tolerances as determined from paired influent and
product water samples. Not more than 10 percent of
the sample pairs from the three units can fall below
the tolerances for removal:
• Bacteria: 99.999 percent removal,
• Virus: 99.9 percent removal, and
• Protozoa: 99.5 percent removal.
If the geometric mean of all sample pairs meets or
exceeds the microbiological removal requirements,
the deficiencies of the 10 percent of the sample pairs
falling below tolerances will be allowed and the
purifier capabilities will be considered acceptable.
It is important to keep records of the test procedure
and the data if there is a claim to be made for units to
be considered microbiological water purifiers.
PRELIMINARY TEST RESULTS OF THE
FEASIBILITY OF THE PROTOCOL
Recently, studies have been conducted on cartridge
type filters, using the protocol for ceramic candle
units to help ascertain the feasibility of the protocol
for water purifier testing. The tests were conducted
specifically to evaluate the virus testing component of
the protocol. Several modifications to the protocol
were made to simplify the testing procedure. These
included elimination of.the in-line mixer and booster
pump, elimination of the pack-pressure regulator,
addition of a 380-I (100-gal) reservoir to contain
the challenge waters, and another similar reservoir to
collect the product water for disinfection prior to
discharge. The viruses were added to the proper
concentration in the reservoir in a batch mode for
40
-------
Figure 1. Test apparatus - schematic (adapted from NSF Standard 53).
Back Pressure Regulator
T
Water
Line
Pressure
Vacuum
Breaker
PressureGauge
Pressure Regulator
Flow Control or
Control Valve
Turbidity
Booster
Any Suitable
Bacterial, Chemical
and Paniculate
Delivery System
Virus and Cyst
Delivery System \I/
Effluent Sampling
Points
Electrical Supply
Flow
Control
Multiple Cam Timer
or Equivalent
NOTES:
1. Faucets are to be used in testing all units under the sink or over the sink. (Regular kitchen faucets for stationary units and faucet
attached units and smaller third faucets for by-pass units.
2. Faucet attached units and portable units are to be placed after the solenoid.
3. Whole house or similar large units need not use faucets. Flow can be regulated with valves placed on the effluent side.
4. All materials of construction must_be suitable for use with drinking water. !
Y Shut Off Valves (not to be used for regulating flow
/_ Solenoid Valve
(X) Check Valve
41
-------
Tablo 4. Sampling Plan Halogen-Containing Units
Tost Point
(as % of tola!
slated capacity)
0
25
SO
After 48-hr stagnation
60
75
Aftor 48-hr stagnation
30
100
Altor 48-hr stagnation
Influent
Background (all
Test Water Halogens)
General X
Challenge
pH 9.0 ±0.2
Challenge
pH 9.0 ±0.2
(chlorine & olhers) or
pH 5.0 ±0.2
(iodine)
Residual
Halogen,
Chlorine &
Others
X
X
X
X
X
X
X
X
Tests
Iodine
X
X
X
X
X
X
X
X
X
X
Microbiological
Chlorine &
Others
X
X
X
X
X
X
X
X
Iodine
X
X
X
X
X
X
X
X
X
X
Table 5. Sampling Plan - Ceramic Candles or Units and UV
Units
Tests
Tost Point*
Slart
Day 3 (middlo)
Day 6 (middlo)
After 48-hr stagnation
Day 7 (middlo)
Day 8 (near end)
After 48-hr stagnation
Day 10.5
Test Influent
Wafer Background
General X
Challenge
Micro-
biological
X
X
X
X
X
X
X
X
All days are "running days" and exclude stagnation periods.
When the units contain silver, a leaching test shall be
conducted as shown in Section 3.5.1.e and silver residual will
bo measured at each microbiological sampling point
Tablo 6. Sampling Plan - teaching Tests for Silver-
Containing Units
Tests
Tost Point
Stan
Day 2
Altor 48-hr stagnation
Influent
Background
X
Silver/
Residual
X
X
X
The test apparatus is shown in Figure 2. The results
of the virus challenge are shown in Table 7. The
results indicate that the units can remove at least 99
percent of the virus from the regular challenge
(Tucson tap water), which was used in the first half of
the test, and greater than 99.9 percent removal from
worst case challenge water used in the second half of
the tests.
CONCLUSIONS
The Guide Standard and Protocol for Testing
Microbiological Water Purifiers provides the water
industry, consumers, and government a common
approach to the evaluation of existing and
developmental products for their microbiological
removal capabilities. While the standard and testing
protocol is rigorous in terms of both microbiological
removal requirements and challenge requirements
(both microbiological and physical/chemical), it should
provide a high degree of confidence in" terms of
protection to consumers wishing to use point-of-
use microbiological water purification units or devices
to remove disease-causing organisms from their
drinking water.
introduction to the filters. When worst case challenge
waters were applied, the virus inoculum was added to
the water only at the sampling time. The testing cycle
was every 30 minutes for an eight-hour daily run
with an operating "on" period of three minutes. Virus
samples were taken when approximately 10 bed
volumes of the seeded water had passed through the
system units except for 48-hour stagnation samples.
42
-------
Figure 2. Test apparatus.
Pressure or
Bladder
Tank
Flow
Meters
Cartridge Filters
and Housings
Table 7.
Effluent
Reservoir
Virus Removal by Cartridge Filter Units Using
EPA Water Purifier Protocol
Time
(days)
1
3
6
7
8
10
10
Water Type
Tap Water
Tap Water
Worst Case Water
Worst Case Water
Worst Case Water
Worst Case Water
Worst Case Water (stagnant)
Average Percent
Removed
99.32
98.33
99.89
99.91
> 99.99
99.93
- > 99.99
43
-------
PERFORMANCE AND APPLICATIONS OF GRANULAR ACTIVATED
CARBON POINT-OF- USE SYSTEMS
Karl Van Dyke and Roy W. Kuennen
Research and Development Division
Amway Corporation
Ada, Ml 49355
Point-of-use (POU) water treatment devices based
on granular activated carbon (GAC) have been
around for many years. Most were traditionally used
to improve the aesthetic quality of drinking water:
color, turbidity, taste, and odor. As the knowledge of
trace organic chemicals in drinking water has grown,
so has the public's awareness of the problem. As the
issue of chemical contamination grows, the traditional
aesthetic claims are being supplemented by claims
for chemical removal. This requires more cjearly
defined ways to evaluate the validity of these claims.
The addition of the VOCs to the Safe Drinking Water
Act, with their low maximum contaminant levels
(MCLs), places more stringent requirements on
devices claiming to remove them to below MCL
concentrations (1,2).
It has been said one can filter anything from water if it
is filtered through enough money. This means that
the technology is available, however, it may be too
expensive to employ on a full scale. Point-of-use
water filters offer options of technology that could be
prohibitively expensive on a large scale. They offer a
means to achieve filtration if properly designed to
remove even low pg/l (ppb) levels of contaminants
that are below MCLs. POU devices containing GAC
are designed in one of three basic configurations:
GAC in single or sequential housings, GAC and PAC
in sequential housings, and pressed carbon blocks.
They may be installed in several configurations
including faucet, stationary, faucet diverter, and line
bypass.
The purpose of this paper is to summarize results of
Jab and field studies on POU devices to support the
concept of using chloroform as a surrogate
compound for making removal claims for specific
VOCs found in drinking water.
POINT-OF-USE PERFORMANCE --
TEST DATA
GULF SOUTH RESEARCH INSTITUTE
The first significant evaluation of POU devices began
with the Gulf South Research Institute (GSRI) studies
sponsored by the U.S. EPA Office of Drinking Water
and reported between May 1979 and October 1981
(3). This study consisted of three phases progressing
from lab to field. It was set up to develop basic data
and information on the performance of a variety of
small home treatment units with respect to organics
removal and bacterial/endotoxin aspects. The
philosophy was to stress the units under simulated
home use.
Phase 1
Phase 1 addressed protocol development and pilot
testing using spiked and unspiked New Orleans tap
water. The basic procedure was to run units on an
accelerated program, sampling at several points
throughout the rated life. Influent and effluent samples
from filter tests were run for trihalomethanes (THMs),
nonpurgeable total organic carbon (NPTOC), bacterial
standard plate count (SPC), endotoxin, and silver
(where appropriate). Unspiked tapwater was selected
as the main means of testing based on pilot tests.
Preliminary results on seven units in Phase 1
included two faucet filters (one bypass, one
nonbypass), one portable pour-through, one
•stationary filter, and three line bypass POU devices.
Results for trihalomethanes showed:
« Small faucet and pour-through filters removed
amounts ranging from negligible to about 25
percent of the influent THM during the
manufacturer's recommended filter life.
» Larger stationary and line bypass filters removed
greater percentages, ranging from 43 percent to
over 90 percent for one filter. The extent of THM
removal appeared to be a function of several
factors, including the quantity of carbon relative to
treated water, contact time, and design factors.
Phase 2
A total of 25 commercially available units and one
experimental unit were evaluated in Phase 2. Influent
and effluent tests were run for THMs, NPTOC, SPC,
44
-------
endotoxin, silver, and peripheral substances. Each
model was challenged with ambient New Orleans
municipal drinking water during three replicate tests.
Results for trihalomethanes showed:
• Small faucet and pour-through filters removed
from four percent to 69 percent of the influent THM
during the rated lifetime of units.
• Larger stationary and line bypass filters removed
greater percentages, ranging from 15 to 98
percent. Six line bypass units (including one unit
containing an experimental material) removed over
85 percent of the influent THM. The descriptions of
the five commercially available line bypass units
(Table 1) allows a comparison of rated capacities:
from 3,785 to 15,140 I (1,000 to 4,000 gal), carbon
weights from 300 to 1,708 grams (0.7 to 3.76 Ib),
and iodine numbers from 434 to 1,223. Units are
ranked based on carbon weight and iodine number,
a measure of capacity. Breakthrough curves
(Figure 1) show performance as expected with
three models performing well for 3,785 I (1,000
gal), and two past 7,570 I (2,000 gal). Removal
here should be noted as percent of total chloroform
applied to the filter, not influent versus effluent at
any one point.
Table 1. Line Bypass Units (Phase 2)
Unit
Culligan
SG-2
Aqualux
CB-2
Everpure
QC4-THM
Aquacell
Bacteriostatic
Seagull IV
Description
1 cartridge
w/GAC
2 cartridges
w/GAC
2 cartridges
(1 PAC, 1
GAC)
2 cartridges
w/GAC
1 cartridge
w/pressed
block
Rated
Capacity
(gal)
4,000
2,000
1,000
2,000
1,600
Carbon
Weight
(9)
1,708
1,150
765
417
300*
Iodine
Number
980
966
1,223 (GAC)
798 (PAC)
867
434*
* Carbon and binder.
Phase 3
Phase 3 included a ground water study, field study,
home study, and an addendum covering removal of
halogenated organics.
In the ground water study, 10 models were evaluated
with well water spiked with 1,1,1-trichloroethane,
carbon tetrachloride, trichloroethylene, and
tetrachloroethylene at target levels of 50, 20, 50, and
50 yg/l respectively. The five commercial line bypass
models previously detailed in Phase 2 were among
the 10 selected.
The results were:
• The selected models (from each of the four basic
configurations) removed from 76 to 99 percent of
the spiked halogenated organics in the well water.
The five line bypass models highlighted in Phase 2
removed from 93 to 99 percent of the spiked
organics.
• Generally, carbon tetrachloride and 1,1,1-
trichloroethane broke through the carbon filter first
while trichloroethylene and tetrachloroethylene did
not break through at all for some units.
The Phase 3 field study involved units tested in the
cities of Miami, Florida; Atlanta, Georgia; Pico Rivers,
California; and Detroit, Michigan.
The results were:
• Upon reviewing the THM reduction data, it appears
that the relative unit performance ranking
determined in the laboratory test is maintained
throughout the field study.
• The overall level of specific organic chemical
removal appears to be adversely affected by
increased levels of background organic material
present in the water matrix.
The conclusions drawn were:
• Filter unit percent removals of THM compounds
can be predicted with some confidence using
results based on laboratory tests. The total organic
background level, NPTOC, in water may affect
percent reduction levels of THMs slightly (to
approximately 10 to 20 percent) over the range of
background organic levels experienced in the field
study (0.6 to 7.1 mg/l NPTOC).
• The ranking of filter units appears to be maintained
regardless of the source water. The ranking could
change if one were monitoring the filter's
effectiveness in removing a different contaminant,
since many carbon adsorbents are manufactured
with some degree of selectiveness. The GSRI
experience has been that this occurs infrequently
and one can be fairly certain that units
demonstrating effective removals of THM
compounds will also demonstrate strong affinities
to most other halogenated compounds.
In the home study, three filter types were challenged
with New Orleans tap water under nonaccelerated
conditions. One of the line bypass models was
included in this test. Limited data indicated the validity
of using accelerated laboratory testing to provide an-
accurate assessment of the effectiveness of carbon
filtration in removing trace organic chemical
contaminants from drinking water. A larger data base
45
-------
OS
E
E
3
O O O O O
o o o o o
O O O O CD
CL seuBqjeuioiBqui & uoijonpay
o
0)
i_
z:
f-
£_
O
4-
tfl
•4->
d
co
10
(O
a.
o
m o
£ t-
=> a)
co a.
0)
c_
3
O1
46
-------
would be required to assure the positive correlation
between laboratory (accelerated) tests and home
(actual use) tests although preliminary data indicate
the lab tests provide valid assessment data.
The Phase 3 addendum covering removal of
halogenated organics involved testing the same 10
POD devices with additional organic spikes. The
study was flawed due to the presence of organic
solvents used to get the halogenated organics into
water. It was felt that the organic removal data
provided useful information regarding the removal
capabilities under very severe conditions.
The selected models were effective in removing the
spiked level of halogenated organics from New
Orleans tap water (49 to 99 average percent
removal). Line bypass units, generally, are more
effective in removing halogenated organics than the
other models.
Line Bypass
Faucet Mount
Stationary
Pour-through
65-99 percent
50-60 percent
49 percent
66 percent
The conclusion was that it was reassuring to observe
significant reductions of halogenated organics and
THM contaminants (49 to 99 percent and 24 to 99
percent, respectively) despite the adverse conditions
of the test water challenge.
EPA/NSF JOINT STUDY
A more in-depth study of POD devices in the home
was run using five of the units tested in the GSRI
study. An EPA-sponsored study performed by the
National Sanitation Foundation (NSF) to study point-
of-use reduction of volatile halogenated organics in
drinking water involved two communities. The project
was conducted in 1983 and 1984, in Silverdale,
Pennsylvania, and in Rockaway Township, New
Jersey to determine whether point-of-use carbon
treatment is cost effective for the control of volatile
halogenated organic chemicals in small water
systems and also to study water quality district
management techniques for point-of-use treatment
(4). Criteria for selection of devices for this study
included, among others:
• The devices must have demonstrated greater than
95 percent reduction of halogenated organic
demonstrated in the GSRI Phase 3 study or
equivalent.
• The manufacturers were required to certify that
their products met NSF Standard 53 Section 3, for
structural integrity, corrosion resistance,
nontoxicity, etc.
• Point-of-use devices were required to have a
rated capacity exceeding 2,650 I (700 gal)
(estimated one-year service life).
Only line bypass models were selected.
SILVERDALE
For the Silverdale study, the summary of influent VOC
results (Table 2) covering March 1983 through April
1984 shows trichloroethylene and tetrachloroethylene
as primary contaminants, with smaller amounts of
carbon tetrachloride and chloroform.
Table 2. Influent VOC Results - Silverdale
Compound , Mean Cone. Predevice (iig/l)
Trichloroethylene
Tetrachloroethylene
1 ,1 ,1 -trichloroethane
1,2-dichloroethane
Carbon tetrachloride
Chloroform
Bromodichloromethane
Dibromochloromethane
Bromoform
80.4
20.6
1.1
<1.0
8.0
6.7
1.5
1.4
<1.0
Breakthrough was defined as detection of the same
VOC in consecutive postdevice samples from the
same location at a concentration above the routine
detection limit of 1.0 pg/l. Breakthrough did not occur
for any of the devices tested during the 14 months of
sampling for TCE and PCE.
ROCKAWAY TOWNSHIP
For the Rockaway Township study, 12 POU devices
were installed on private wells in October 1981. The
type and concentration of VOCs was varied, with the
primary contaminants being 1,1,1-trichloroethane
and trichloroethylene. See Table 3.
Table 3. VOC Results - Rockaway Township
VOC
1,1,1 -trichloroethane
1 ,2-dichloroethane
Tetrachloroethylene
1 ,2-dichloroethane
Trichloroethylene
Trans-1 ,2-dichloroethylene
Chloroform
Trichlorofluoromethane
Range of Cone.
Found (ug/l)
1.0-240.0
6.7-20.7
1.0-12.3
< 0.4-1 0.1
0.7-240.2
0.8-5.1
1.7-2.1
<25.0
Number of
Wells wA/OCs
8
4
7
6
4
2
2
1
The local health department did sampling and
analysis from October 1981 to October 1982 with 100
percent VOC reduction. After beiqg included in the
study, four sites were monitored in October 1983, the
47
-------
24th month of operation. There were no VOCs
measured in postdevice samples after 24 months of
operation.
Data from the study indicates that point-of-use
granular activated carbon (GAG) treatment devices
effectively reduced concentrations of
trichloroethylene, tetrachloroethylene, carbon
tetrachloride, 1,1,1-trichloroethane, 1,1-
dichloroethylene, 1,1-dichloroethane, and chloroform
at influent concentrations studied. These results
confirm bench and field results from the Gulf South
Research Institute study.
NSF STANDARDS 42 AND 53
The National Sanitation Foundation has issued
several standards relating to POU devices, the two
most relevant being Standard 42 for Drinking Water
Treatment Units - Aesthetic Effects, and Standard
number 53 - Health Effects (5). The contaminants
covered in Standard 42 include such things as taste,
odor, and color and specific chemicals such as
foaming agents, hydrogen sulfide, and phenol.
Particulate reduction, while not an inherent quality of
GAG, is also covered.
The contaminants covered in Standard 53 include
total trihalomethanes, six pesticides, soluble ions -
nitrate and fluoride, eight heavy metals, plus cysts,
turbidity, and asbestos. Total trihalomethanes are run
at a 450 pg/I challenge to a 100 pg/l effluent limits.
This standard does not yet cover the newly regulated
VOCs.
DEFINITION OF END OF LIFE
Determining when a POU device is exhausted for
taste and odor can be done by consumer perception.
NSF standard determines it by chlorine reduction. For
the trihalomethanes -(MCL 100 pg/l), the NSF
Standard 53 challenges at 450 pg/l and rates capacity
based on breakthrough to 100 pg/l, with a 100
percent safety margin. End of rated life for the other
VOCs may be based on a breakthrough equal to the
MCL with an evaluated challenge of perhaps 300 pg/l.
The VOCs and MCLs are listed in Table 4 (2).
Table A. Regulated VOCs and MCLs
Compound
V«iyl Chtonde
Bonzono
Carbon Toirachkmde
1,2-dicWoroGliiane
Tnchtofoethytone
1,1-dicliforoethyIene
p-dtchtorobonzono
1,1,1-tnchloroclhano
MCL (mg/1)
0.002
0.005
0.005
0.005
0.005
0.007
0.075
0.200
To perform to these levels, a POU device will need to
be well designed and constructed.
AMWAY DATA ON A POU WATER
TREATMENT SYSTEM
Amway has developed considerable data on POU
water treatment system performance. In our
laboratories we have been working heavily on
evaluation and claims documentation for GAC-based
POU devices for the past five years (6,7). The
following data is from a unit containing a unique
pressed carbon block design. Performance is rated
on the conservative position that the only way to
prove the filter's ability to remove chemicals from
water throughout its rated life, is to test against every
chemical claimed. Testing is carried out to 150
percent of rated life to provide an extra margin of
safety. The filter is effective for removal of 116
compounds, including 100 of the EPA priority
pollutants, plus several pesticides including aldicarb,
EDB and DBCP, fuel hydrocarbons, and others.
CHEMICAL CLASSIFICATION
In order to test that many compounds, they were
placed into chemically similar classes, then into
groups, and then a group was tested together. The
classes and groups (Table 5) are based on chemical
similarity and analytical technique. Some groups were
subdivided for easier analysis.
Table 5. Chemical Classification
Class
1 . Acids
2. Base/Neutrals
3. Hydrocarbons
4. PBBs
5. PCBs
6. Pesticides
7. PNAs
8. Purgeables
Group
Phenols
a) Biphenyldiamines
b) Chlorinated Hydrocarbons
c) Cyclohexenone
d) Hatoethers
e) Nitro compounds
f) Phthalates
Gasoline/Kerosene/Diesel Fuel
PBBs
Aroclors
a) Halogenated Alkanes
b) Nitrogen/Phosphorous
c) Organochlorine
d) Organonitrogen
Polynuclear Aromatics
a) Aromatics
b) Halogenated Alkanes
c) Halogenated Alkenes
d) Trihalomethanes
Analytical
Method
EPA-625
HPLC-UVD
EPA-625
EPA-625
EPA-625
EPA-625
EPA-625
GC-FlD
HPLC-UVD
EPA-608
GC/ECD
GC/NPD
EPA-608
HPLC-UVD
HPLC-UVD
EPA-624
EPA-624
EPA-601
EPA-601
TEST PROTOCOL
The test protocol is similar to the GSRI and the NSF
test Protocol (6). All testing was done on duplicate
devices. The main test stand (Figure 2) provides
control of the critical parameters. The contaminants
are injected under control of an HPLC pump. This
48
-------
Figure 2. Main test stand.
back-pressure adjustment
recirculation loop
2OO gal.
electrical valve
mechanical valve
jet pump
effluent
sampling
point
motionless mixer Influent
Sampling point
provides very consistent influent concentrations. The
flow is controlled, influent and effluent samples
collected for every sample point, and the temperature
monitored. The test was run continuously for eight
hours per day with an overnight stagnation period.
Influent concentrations were generally chosen to be
approximately 100 to 200 \IQ/\. Chloroform was higher
to approximate the NSF standard requirement.
RESULTS
The results are contained in Tables 6 to 10. The
trihalomethanes group shows only chloroform
breaking through, at a concentration below 5 yg/l.
The halogenated alkanes were subdivided for testing.
The detection limit for 1,2-dibromoethane (EDB) was
reduced to 5 ng/l to accommodate states with limits
of 20 qg/l. The detection of trichlorofluoromethane
and carbon tetrachloride are near detection limits.
The halogenated alkenes include trichloroethylene
and tetrachloroethylene, common pollutants. The
aromatics group includes benzene, which based on
Dobbs isotherm data (8) would be expected to break
through. Our data contradicts that expectation. The
other aromatics are removed. The chlorinated
hydrocarbons, including 1,4-dichlorobenzene are
removed as would be expected based on efficacy for
benzene.
These compounds provide a glimpse of the diversity
of chemical compounds that can be removed by
GAG. The difficulty of completing a project of this
magnitude several times forced a good look at the
trends. The potential use of a surrogate for most of
the compounds tested appeared feasible although not
yet adequately documented or accepted by the
scientific or regulatory community.
GRANULAR ACTIVATED CARBON -
PERFORMANCE
THEORETICAL CONSIDERATIONS
A brief look at the basic principles of GAG adsorption
and some of the molecular properties affecting
adsorbability of chemicals is helpful. Granular
activated carbon has been demonstrated to adsorb a
wide variety of organic chemicals. It has limited
capacity for some classes of organic compounds, and
for water soluble ions and metals. Carbon has been
referred to as "black magic," but this is not really
true. There is a good base of information on how
GAG works (9,10,11). For a chemical to be adsorbed
onto carbon, the attractive forces must be strong
enough to overcome repelling forces. The forces of
attraction are generally agreed to be primarily due to
Van der Waals forces, which are relatively weak. First
49
-------
Tablo 6. Class: Purgeables - Group: Trihalomethanes
Compound
Chloroform
Bromodichloromethano
Oibromochloromethane
Bromform
Tablo 7. Class: Purgeables - Group:
Compound
1,2 Dichloroethane
1,1-DiclitorocUtans
1 ,1 ,2-Tnchtoroetriane
1,2-Dibfomooltiano (EDB)
1,1.1-Tnchtoroethane
1 ,2-DichlofOpropane
Trichlorolluoromelnane
1 ,1 ,2,2-Tetrachtoroelhane
1 ,2,3-Trichloropropane
Carbon TetrachtorWe
1 ,2'Dibromo-3-Chloropropane (DSCP)
Tablo 8. Class: Purgeables - Group:
Compound
1,2-Dicriloroethytene
(ftiris-1 ,2-OichloroGlriene
irans-1 ,3-Dichtoropropylone
Tnchtoroelhytene
1 , 1 ,2,2-Tolrachloroelhylene
Tablo 9. Class: Purgeables - Group:
Compound
Bonzono
Tduono
Chforobonzone
Xytono
EtliylborKeoa
Average Influent Effluent Effluent
(ug/l) @ 500 gal (ug/l) @ 750 gal (ug/l)
414 1.25 1.25
129
115 - -
204
Halogenated Alkanes
Average Influent Effluent Effluent
(ug/l) @ 500 gal (ug/l) @ 750 gal (ug/l)
112
120
164
59 " -
116
118
112 " 0.8
151
146 • -
78 " 0.2
258
Halogenated Alkenes
Average Influent Effluent Effluent
(ug/l) @ 500 gal (ug/l) @ 750 gal (ug/l)
109
96
170 * -
100
92
Aromatics
Average Influent Effluent Effluent'
(ug/l) @ 500 gal (ug/l) @ 750 gal (ug/l)
113
114
107
292
163
Detection Limit
(ug/l)
0.28
0.11
0.13
0.93
Detection Limit
(ug/l)
0.5
0.25
0.5
0.005
0.16
0.34
0.51
0.31,
2.7
0.22
0.09
Detection Limit
(ug/l)
0.17
0.11
0.14
0.28
0.27
Detection Limit
0.82
1.0
0.19
3.9
1.6
Total Loading
(mg)
1,253
366
327
580
Total Loading
(mg)
320
340
467
169
329
334
318
430
413
222
733
Total Loading
(mg)
309
272
482
284
262
Total Loading
(mg)
320
325
304
830
462
* Indicates value below detection limit (99 percent Cl).
50
-------
Table 10. Class: Base/Neutrals - Group: Chlorinated Hydrocarbons
Average Influent Effluent Effluent
Compound (lig/l) @ 500 gal (ug/l) @ 750 gal (ug/l)
1,2 Dichlorobenzene
,3-Dichlorobenzene
1 ,4-Dichlorobenzene
Hexachloroethane
1 ,2,4-Trichlorobenzene
2-Chloronaphthalene
Hexachlorobutadiene
Hexachlorocyclopentadiene
Hexachlorobenzene
150 * -
184
120
111
313
109
154
58
66
Detection Limit
(ng/D
0.75
0.49
1.05
1.06
0.35
0.5
0.24
1.03
0.48
Total Loading
(mg)
429
526
342
318
313
110
444
171
192
* Indicates value below detection limit (99 percent Cl).
the molecules must be brought close to the carbon
surface. Particle size and design are important to
minimize the water thickness around the carbon
particles. Then the chemical must diffuse through the
boundary layer to the carbon surface. It must diffuse
into the pores, until it is tightly retained in the
micropore region of the GAG. The kinetics of the
surface diffusion or pore diffusion are limiting steps.
Crittenden, Snoeyink, and Weber, among others,
have developed theories and applied them to explain
GAG behavior in actual use (12-15). For single
solute models, predictions work quite well. When
multiple solutes are being adsorbed on GAG, there is
a competition for available sites, with the more
strongly attracted molecules displacing the more
weakly retained ones. These competitive effects are
beginning to be understood, such that prediction of
orders of desorption and capacity can be made when
detailed information is available on the GAG being
used. This is of great value to a manufacturer when
designing POU devices and selecting a carbon
source. It is not practical for evaluation of many
different completed units due to the detailed
information that must be developed for the GAG.
However, the principles will be important in
understanding results of dynamic testing.
It is important to note that GAG is not a generic
commodity. The source of the carbon and how it is
prepared for use provides a range of capacities,
selectivities, and kinetics that must be evaluated.
Evaluation of GAG includes numerous tests. Surface
area is important since adsorption is basically a
surface phenomena. The surface area of carbon is
very high, generally in the range of 500 to 1,500 m2/g
(2,445,500 to 7,327,600 sq ft/lb) mostly in internal
pores which are like a maze of interconnecting
channels. Pores below 20 angstroms (7.87 x 10'8 in)
in radius, called micropores, are where most
adsorption takes place with larger transition pores
from 20 to 500 angstroms (7.87 x 10-8 to 1.97 x
10-6 in) providing access and some additional
adsorption. Micropores over 500 angstroms (1.97 x
10'6 in) have little capacity for small molecules. The
distribution of pore sizes and the percentage in the
range of interest determine the actual performance of
carbon. A BET nitrogen surface area measurement
provides a great deal of information on surface area
and pore size distribution. Particle size and particle
size distribution are also important influences in the
kinetics of adsorption.
Characterization of carbon pores includes traditional
tests such as iodine values, generally" agreed to
measure small pores, with a pore radius of less than
20 angstroms (7.87 x 10-8 jn). Carbons may have
iodine values of 600 to 1,200. Molasses number,
based on decolorization of molasses solutions,
represents pore volume in the range of 20 to 500
angstroms (7.84 x 10-8 to 1.97 x 10-6 in). Dye
adsorption, such as methylene blue, provides
additional data on capacity for large molecules.
All of these measurements are useful for preliminary
evaluation of GAG, but must be used with caution
due to the multiplicity of interactions taking place
under dynamic conditions.
APPLICABILITY
To determine whether GAG is applicable to removal
of a chemical, it is necessary to look at some of the
properties of the chemical that affect its adsorption.
Removing a chemical from water is easier if it is not
ve/y soluble (Table 11). The solubility versus GAG
capacity for several VOCs (capacities expressed as
Freundlich adsorption isotherm values), shows a good
correlation between low solubility and good
adsorption. While the capacity in mg/g (Ib/lb) of
carbon may vary with carbon type, the relative order
of elution is predictable.
Other factors, such as the substitution of bromine for
chlorine increase adsorption, as evidenced for the
THMs:
51
-------
CHCIs
CHC^Br
CHCIBr2
CHBrs
Adding chlorine to a molecule
CHCIa
CHCIs
CCI4
Benzene
1 ,4-Dichlorobenzene
Addition of double bonds to
adsorbability:
1,2-Dichloroethane
transl ,2-Dichloroethane
Capacity
X/MCo (5) 100 ug/l
3.05
5.9
10.3
17.0
increases adsorbability:
Capacity
X/MCo @ 100ua/l
0.45
3.05
9.78
13.47
196.4
a molecule increases
Capacity
X/MCo <5> 100 uq/l
2.88
8.32
Capacity
X/MCo @ 100 ug/l
CL CL
\ /
1,1,2-trichloroethylene C = C 25.17
/ \
CL H
Capacity
X/MCo @ 100 ug/l
CL CL
\ /
tetrachloroethylene C = C 122.9
/ \
CL CL
Table 11. Solubility vs. Capacity
Capacity
Compound Solubility® X/MCo @ 100
oompound 20<>c (mg/|) ^ (mg/g)
Methylene Chloride 18,236 0.454
1,2-Dichloroethane 8,690 2.88
Chloroform 8,000 3.05
1,1,1 -Trichloroethane 4,400 7.72
trans 1 ,2-Dichloroethylene 2,190 8.32
Benzene 1,780 13.47
Trichloroethylene 1,100 25.17
1 ,4-Dichlorobenzene 70 196.38
As several of the factors are considered, a significant
increase in capacity results:
CL
Chloroform CL - C -
1
CL
Capacity
X/MCo (5) 100 UQ/I
H 3.05
Capacity
X/MCo @ 100 ug/l
Based on capacity values obtained from Freundlich
adsorption isotherms, it is obvious why chloroform or
1,1,1-trichloroethane would break through first, and
why trichloroethylene and tetrachloroethylene did not
break through in GSRI or field studies.
CAPACITY
Activated carbon has a finite capacity for any one
compound. When multiple compounds occur, they
compete to some extent for the available sites on the
carbon, reducing the capacity for the less strongly
adsorbed compound. Estimating capacity involves a
number of considerations including:
CL H
. - C - C - H 7.7
1,1,1-trichloroethane CL-C-C-H 7.72
CL H
o Which chemicals are present?
• What is the concentration of each?
• What is the maximum capacity of the carbon for
the chemicals?
• What is the maximum effluent concentration
allowed?
52
-------
• What are the kinetics of the compound with the
carbon?
• What are possible competing materials?
• How does the design of the POU device affect
adsorption?
• How is capacity to be expressed?
CRITERIA TO DESIGN AND EVALUATE
GAC-POU DEVICES
There are basically two main criteria that determine
the efficacy of a GAC-based POU device:
• The capacity of the GAC used (isotherm
capacities),
• The design of the final unit to approach maximum
capacity (dynamic testing).
Capacity of the carbon is obviously critical.
Performance cannot be achieved without a GAC that
has a significant capacity for the chemicals of
concern. One driving interest presently is the
category of VOCs as proposed by EPA. This group is
of special interest due to the difficulty of removal from
water. It is a limiting factor in GAC performance. A
GAC can be selected for a POU that provides a
maximum adsorptive capacity for the smaller
molecular weight VOC compounds. Generally, this
comes at the expense of capacity for the higher
molecular weight compounds, but this is not the
limiting factor.
FREUNDLICH ADSORPTION ISOTHERMS
The capacity values used have been isotherm values
that are maximum values at an effluent concentration
of 100 ug/l. To discuss maximum theoretical capacity,
it is necessary to bring in batch equilibrium
experiments, referred to as Freundlich adsorption
isotherms (11,16). Bottle point isotherms are run in a
series of bottles containing different amounts of
carbon, and a water solution of the chemical (e.g. 1
to 2 mg/l of VOC). The bottles are sealed and mixed
for seven days, the carbon filtered out, and the
concentration remaining in the water measured. The
carbon weights and VOC concentrations must be
adjusted to yield final solution concentration of 1 to
200 ug/l. This allows calculation of the quantity
adsorbed on the carbon, expressed as mg/g carbon.
The Freundlich equation is expressed as:
X/M = KC1/N
Plotted as a straight line:
Log(X/M) = Log K + (1/N) Log C
where,
X/M = the amount of component adsorbed (mg/l)
divided by the weight of carbon (g/l)
C = the equilibrium concentration of the
component (mg/l or ug/l)
1/N = slope of the line for the component in
solution
K = Constant for each compound:
(mg/g) x (l/mg)1/N Or (ng/g) x (l/ug)i/N
It must also be stated that to compare isotherm
behavior, the chemicals must be grouped such that
the similar molecules are compared. The adsorption
isotherm curve, when plotted as the logarithm of both
sides of the equation, yields a straight line providing
significant information (Figure 3). One can:
• Screen potential activated carbon samples for use.
• Provide an estimate of the adsorption capacity of a
component by activated carbon.
• Determine if the desired effluent can be obtained
with a given amount of activated carbon.
• Estimate the capacity difference when equilibrium
concentrations are changed.
• Predict the relative breakthrough order of
adsorbates during column studies by comparing
the capacity values obtained from isotherms.
From the isotherm of chloroform, one can find the
capacity of that carbon at an equilibrium
concentration of 100 ug/l, the current MCL for total
trihalomethanes. This carbon could hold 3.05 mg
CHCIa/g carbon (0.003 Ib CHCIa/lb carbon). One can
also readily see the impact of decreasing the
equilibrium concentration to 25 pg/l; capacity is now
1.29 mg/g GAC (0.0013 Ib/lb GAC). Reducing it
further to 5 jig/I, the MCL concentration for most of
the VOCs, the capacity is now down to 0.47 mg/g
GAC (0.0005 Ib/lb GAC). These values may change
with different carbons or different operating
conditions.
Absolute values for adsorption isotherms for the same
chemical can show significant variations between
published data, which may be due to any one or a
combination of factors including the carbon used,
pore size distribution, particle size, the time to
equilibrium, the water temperature, pH, analytical
techniques, and the number of points used to
determine the line, among other factors. One factor is
the presence of hardness ions and TOC. The effects
of deionized water versus a municipal water were
investigated. No significant difference is seen,
53
-------
£
m
ffi
•HOT
tta*l
O
tn
c
o
•MM
•|M|
Q.
O
(0
T3
^J
^^^
.c
•••••
~
•o
HI
CD
U-
£
o
Ci««i
o
^•^
:a
%/S)
/Sjv-V
>-% ^
jg
CO
_ O
IO
o
CO
ll
1- II
CO -^
T- Q.
ii °"
II O
z O
Lf S ii
(5)^. ^ — I CD , f
\ *
c*
i_
m
Ux
•CJ
CO
0 ^
-g ^
CO T3
^^ M
O -o 'E
^ i-§
(§) ?§
^)
B)
E
IO
o
CO
V
;
^
* ^
Ii)
"44 — -
CD
^)
£
CD
CM
1% @
*^)
s *
&?& «»/^~*\
^^ ®
(1
D)
^j
£
CD
1 1 1 1 1 1 1 1 . 1 1 1 1 1 1
O* '" CO*
— o o
.C ""
O (6/6 iri) - UOIJBJIU8OUOO 9SBl|
^ ^--' O
-------
however, the effects of temperature make a dramatic
difference in capacity (Figure 4). As the water gets
colder, the capacity of the carbon increases.
Temperature control, or at least measurement, is an
important element of capacity evaluation.
RELATIVE ISOTHERM VALUES
The real value in using adsorption isotherms for
predicting performance of a POD device is in using
the relative capacities for related compounds. By
plotting the isotherms for related compounds (Figure
5), one can readily see that a unit that removes
chloroform to a concentration of 5 ng/l will also
remove 1,1,1-trichloroethane, trichloroethylene, and
1,4-dichlorobenzene. While the total quantity of
compound absorbed may differ from unit to unit, the
relative capacities, or the order of breakthrough of the
compounds will remain the same. These general
observations were made during the Gulf South
Research Institute study, and also during the
NSF/EPA studies at Silverdale and Rockaway. A
review of the literature on isotherms supports the
relative ranking, although absolute values differ
(8,9,11,17,18).
Therefore, if one compound is chosen as a model, or
surrogate compound, adsorption for the other
compounds is rather certain. By selecting the
concentration of the model compound, e.g.
chloroform, at the highest level anticipated for a
known contaminant, removal is assured for at least
the rated capacity of chloroform, and longer if the
compound is more tightly retained.
DYNAMIC TESTING AND ISOTHERM
CORRELATION
The gap between theory, that is maximum capacity as
determined by isotherm, and practice is a measure of
how well a POD device has been designed. This is
done by dynamic testing. The isotherm capacity is
determined on the GAC under near ideal conditions.
The actual performance of the unit takes all factors
into consideration. The contact time between water
and carbon in POU devices is typically short, in the
order of seconds rather than minutes for large scale
GAC contactors. Kinetics become a factor in
selecting the carbon source. Since the active sites in
GAC are internal and the pores very small, it takes
time for molecules to diffuse into them. Design must
take into account the pore size distribution, particle
size, and particle size distribution. As particle size
gets smaller, adsorption gets faster but backpressure
gets higher and flow rate drops. This may be a
limiting factor. The chromatographic effect, where
strongly adsorbed materials are retained at the early
portion of the bed, reduces competition so only
similar compounds of similar capacity compete for
pores. The mass transfer zone is created where the
first GAC contacted is loaded to its capacity first and
the concentration decreases through the bed until the
compound is completely removed. This wavefront, or
mass transfer zone moves through the carbon bed
until the compound .elutes. Either a long bed, or a
short mass transfer zone is needed for optimum
performance. Again, this is an important criteria when
designing a unit to remove VOCs to below 5 ng/l.
Results of dynamic tests are shown for chloroform
well past the point of breakthrough (Figure 6). It
illustrates that a lot of capacity may be left after
breakthrough reaches 5 ug/l. The typical single solute
chloroform curve with an average influent of 108 pg/l
is shown plus a typical chloroform curve at an
average influent of 442 ug/l. The capacity decreases
significantly as the influent concentration is raised.
The multisolute curve for chloroform at 102 vig/l
shows a reduction in capacity but not as severely as
the 442 ug/l curve. The point of breakthrough is
sooner as the load on the filter increases, but the
difference is less at the low ug/l concentration. This
demonstrates that competition can be accounted for
by increased challenge concentrations, and the
multisolute curve also demonstrates a limited
competition since the total challenge greatly
exceeded the 442 ug/l level.
The effects of multisolute interactions can be seen in
Figure 7. Here, the chloroform curve can be seen in
relation to the other VOCs as part of a 14 component
run. Seven of the VOCs are plotted here. This
illustrates the predictive power of relative isotherms
since the components of the multicomponent mixture
elute in the same relative order as that shown by the
isotherms summarized in Table 12.
In these cases, testing capacity of the POU device for
chloroform would represent a worst case. If
chloroform is removed, so too is every other
compound. If CHCIa testing for capacity is done at a
level equalling the concentration of any single
component it could provide a minimum capacity of
the POU device.
FURTHER EVALUATION TECHNIQUES
The data from dynamic studies provides practical
data, but limited to the specific chemicals tested. The
use of isotherm data to predict elution orders lends
support to the surrogate concept. There are other
theoretical treatments that lend further confirmation
that the practical behavior complies with theory. The
use of Polanyi liquid phase adsorption potential theory
deals with the energy involved in the adsorption
process (10). From limited isotherm data it is possible
to predict capacity for other related compounds.
Further theoretical modelling based on complex
computer programs can provide predicted
performance for single solute or multisolute situations
to compare to actual dynamic data. Successful
prediction of performance for several of the VOCs
confirms that the basic theoretical treatments are
correct. Use of minicolumns to predict the
performance of full scale operations has great
advantages for large scale operations (19,20). It also
55
-------
O
Q.
O
(0
T3
C
3
CD
III
e
O
TTT I I I I T
CO
O
CO
_ o
O)
OJ
. O
o
1
c
CD
O
O
O
Q)
CO
(0
_ o
O
CM
O
O (6/6 in) - uojiBJiuaouoo eseifd PHOS - (l/\l/X)
E
O)
C.
o
Q.
1_
O
13
.C
O
•I—
s
tt)
(1)
ZJ
C3>
56
-------
£
CD
C
O
O
(/)
TJ
CD
LL
in
O
CM
.0
O)
C
o
1
4-1
C
o
o
O
(A
CO
£
Q.
TJ
"5
-T- o
11" i i i • r i t 11 f it i i i i
CO OJ
000
(6/6 id) - uoi.iBJiuaouoo asBMd PHOS - (l/\l/X)
CD
-C
4->
o
C
O
a.
o
CO
T3
T3
3
co
in
57
-------
CO
CD
o
JC
O)
3
O
CO
0>
9U
OQ
O
I
(0
c
Q
£
jo
"o
J5
O
O
o
o
8
o
O
o
in
o
o
o
o
0)
E
m
-------
•b
3
CO
"c
o
E
o
o
o
o
o
o"
00
o
o
o
o
o
CO
o (0
8
tn
+->
c=
(U
c
o
Q.
O
o
O
LO
CO CM T-
'O/O
(U
i_
3
CO
59
-------
PERFORMANCE AND APPLICATION OF RO SYSTEMS
Donald T. Bray
Desalination Systems, Inc.
Escondido, CA 92025
BASIC CONCEPTS
Figure 1 shows some basic concepts related to
reverse osmosis (RO). On the top of the figure there
is a scale in angstroms (A) and pm. These are
referenced to the pore size ranges of the
membrane-filtration business. An angstrom is about
the size of a hydrogen atom (10-8 cm [3.87 x 10-9
in]) and a water molecule is around 2 A (7.9 x 10-9
in). The RO field is at the extreme lower end of the
pore size range, and covers the range 1 to 5 A (3.94
x 10-9 to 2 x 10-8 in). Ultrafiltration (UF) covers the
range from about 10 A to about 1,000 A (3.9 x 10-8
to 3.9 x 10"6 in) micro filtration from 0.1 to 1 pm
(3.9 x 10-6 to 3.9 x 10-5 jn) and general filtration
above about 1 pm (3.9 x 10-5 jn). Other people may
use slightly different divisions. My subsequent
discussions will be concerned with the RO range.
Also shown on Rgure 1 is an important concept: the
basic difference between RO/UF devices and filtration
devices. RO/UF devices are separative devices; they
take a feed stream and separate it into two parts -
a product and a reject. There is no accumulation
within the unit. It can operate continually without
buildup. Filters, on the other hand are accumulators.
Particles are removed from the feed and accumulate
in the unit itself. Hence, the filters have a finite life
and the characteristics of the product are continually
changing. A shut-off provision after a given
throughput is being considered for accumulator type
devices. Such a concept does not apply to separative
devices such as RO, and a different safeguard
approach needs to be taken here.
Rgure 2 shows a highly stylized rendering of the
surface of cellulose acetate (CA). Note the long
roundish particles of CA with various bumps
protruding. Actually the CA molecules are much
longer relative to their diameter and of course are
more twisted and intertwined. One might represent
thin film membranes (TFM) in somewhat the same
fashion but more crooked and with occasional cross
links. The spacing between molecules might vary
from 0 to 5 A (0 to 2 x 10-8 in). The molecules are
in rapid motion, vibrating several thousand times a
second so the spacings are continually changing.
Hence, we have no pore size per se in RO. Only
when we get to UF do we start having definable pore
sizes. Also shown is a water molecule. It is moving
very fast, making thousands of collisions per second,
and under influence of a driving force (in this case
pressure), can enter the spacings between the CA
molecules and diffuse through the network essentially
one at a time. Also shown is a Na+ ion surrounded
by a group of water molecules. Water molecules,
being dipolar, will attach themselves to the + charged
surface of the Na + ion forming a molecule several
times as large. It will move through the lattice in the
same way as water. However, it can't move nearly as
readily as water since it is much larger. Hence, there
is, in effect, a separation of Na + and water molecules
with the water molecules going through the
membrane and the Na+ accumulating on the surface
until back diffusion through the laminar layer to the
reject stream removes it from the unit. One can
imagine that a virus would look like a huge boulder on
the surface. One might also note that the separation
efficiency of the membrane will vary with the shape of
the molecule -- e.g., whether rod-, sheet-, or
ball-shaped. Basically, however, the larger the
molecule, the better the rejection. The above is a
simplistic picture. Surface effects such as the
electron structure of both membrane and diffusing
molecule also come into play, but the physical size
approach is a good first approximation.
TYPES OF SYSTEMS
Figure 3 lists the three types of RO point-of-use
(POU) systems in'use today. I should note that all
use a spiral-wound membrane assembly
configuration. The first commercial unit on the market
was an over-the-counter (OTC) type made and
marketed by Culligan in 1965 to 1966. ft hung on the
wall of the kitchen as shown schematically in Figure
4. The spiral-wound element and pressure vessel
were located inside the product storage tank. Note
locations of feed, reject, overflow, and product. This
unit had several drawbacks and was never very
successful due in part to installation difficulties of
feed, reject, and especially the gravity overflow.
62
-------
Figure 1. Reverse osmosmis - basic considerations.
10
100
1
1,000
I
1
10,000
10
100,000
I
100
1
1,000
mm - milimeter
li - micron
A - angstrom
I
RO
* *
UF
Separative Devices
Feed
Product
Reject
MF
Filtration
Feed
Accumulative Devices
*• Product
Figure 2. CA membrane surface schematic.
Na+ ion
Water Molecule
These early units had no post carbon filters so there
were some taste problems. Also, the gravity flow of
product from the spigot was replaced by air in the top
of the storage tank which introduced dust particles
with each use, so the heterotrophic bacteria species
within the storage tank kept changing. Further, and in
hindsight, the marketing and selling techniques of RO
POL) still needed to be developed.
The next generation of OTC units were counter-top
models that sat on the counter with feed connections
to the sink faucet, and drained directly into the sink.
Figure 3. Types of RO/POU systems.
UTC - Under the Counter
OTC - Over the Counter
OU - Office Unit
This eliminated the installation and overflow problems
of the original, wall mounted unit but resulted in use
of valuable counter space and retained the continued
taste and contamination problem. The third generation
of OTC, as shown in Figure 5 and made by Nimbus
63
-------
Figure 4. OTC concept.
FEED
REJECT
OVER FLOW
DOMESTIC WATER SUPPLY
Water Systems, is of more recent vintage in which
the unit is attached directiy to the faucet with a quick
disconnect. Reject goes to the sink; product to a
collection bottle. The unit is hooked up at night,
removed in the morning and stored in the refrigerator,
as is the product water container. It overcame most
of the difficulties encountered in the other two
approaches but adds the inconvenience of hookup
and removal.
The second type of RO POD system developed was
the office unit (OU), the first of which appeared in
1966. It was really an adaptation of the wall-mounted
unit but utilizing the olla, or water storage
compartment, of the conventional bottle water stands.
Most of the original limitations of the wall-mounted
unit still applied, but were offset to some extent by
increased revenues per unit (to compensate for
installation costs), and use of a product level operated
shut-off valve and post carbon filter. More recently,
there has been a trend toward use of pressurized
product storage tanks, as used in the under-the-
counter (UTC) systems.
The UTC was developed in 1966 and 1967 in an
effort to overcome the limitations of the over-the-
counter and office units. Figure 6 shows the basics of
a UTC unit. All the components except the faucet are
located under-the-sink. Tap water feed under
pressure enters a pressure vessel containing a
spiral-wound membrane element. The reject flows
through a small capillary tube to drain. The pressure
is reduced from line pressure at entrance to
discharge pressure at outlet over the length of the
capillary- The amount of reject is determined by the
inside diameter and length of the capillary tubing. It is
generally set at four to five times product flow rate.
This ratio was chosen based on field experience.
Basically, one can only take out that percentage of
water until saturation of the least soluble component
occurs, which in San Diego, is calcium carbonate at
about 20 percent water removal. The product flows
64
-------
Figure 5. Mint II - hang-on unit.
out of element into a sealed storage tank .at a
continuous but slow rate (10 to 20 ml/min [0.34 to
0.68 oz/min]). The tank has a rubber diaphragm with
a low air pressure on one side (35 to 55 kPa [5 to 8
psi] when tank contains no product water). As the
product flows into the tank, it moves the rubber
diaphragm down, compressing the air. When the
faucet is opened the water is forced out of the tank
by the compressed air to point-of-use.
Some additional accessories need to be added to
make a working sytem, as shown in Figure 7. A post
carbon unit is placed in the product line between the
storage tank and faucet. It serves as a final polisher
to remove any tastes which were picked up from the
tank material or that came through the membrane.
There needs to be either a pressure relief valve on
product side or a shut-off valve on the feed line
operated by tank pressure. If a relief valve is used, it
is generally set at about 1/2 line pressure. When the
tank pressure reaches this point it opens the relief
valve allowing the product to join the reject line. The
unit continues to operate producing good water --
it just goes down the drain. When water is drawn from
the storage tank and the pressure is reduced, the
valve closes and good quality product refills the tank.
Figures. UTC - basics.
/I f
Point of Use
RO
Element
iragm
Pressure
Vessel
Capillary Flow Control
Feed Reject
65
-------
Figure 7. UTC - complete system.
Air Gap
Post Carbon
Polisher
J.
Point of Use
Foed
i—i
Pressure
Relief
Valve
Diaahragm
Product
Air
Sediment Filter
Storage
Tank
Reject
If a shut-off valve is used, it is located in the feed
line, is operated by tank pressure, and is set at 1/2 to
2/3 line pressure. In many locations, a sediment filter
(25 pm) needs to be added. When a TFM is used, a
precarbon filter is also added to the feed line after the
sediment filter to removal chlorine (which damages
the polyamide TFM). The reject is shown flowing to
an air gap above the sink. This is needed to meet
plumbing codes requiring a positive break between
the potable water and the sewer -- i.e., an air gap.
The above UTC concept was developed in 1967 and
is little changed to this day. Essentially, all UTC units
made and in use today, utilize this concept.
USE DATA
Figure 8 shows my estimates of some key use data.
There are about 500,000 units in use today. Total
1987 income to manufacturers was between $15 and
25 million from new units and about $20 million from
replacement parts and elements, for a total of about
$40 million per year. The total income to dealers and
distributors was $50 to 75 million from new
placements and perhaps $20 million from rental or
leased units for a total of about $100 million per year.
The cost to consumers to rent or buy is also given in
the figure. Note that the little Mini II hang-on
competes very well in production and quality, at a
much lower price. It was developed to bring lower
cost, good quality drinking water to the lower income
and elderly as replacement for store or vending
machine-purchased drinking water. To a large
extent, it has succeeded in doing so.
MEMBRANE TYPE
There are two types of membranes on the market
today that account for essentially all the membranes
used in RO POU units. These are the CA types and
the TFM polyamide types. I have grouped all the
cellulosic base membranes into "CA Types." This
includes cellulose acetate, diacetate, blend, triacetate,
and cellulosic esters. There are only minor
differences in their performance or limitations. Most of
the units in the field today use CA, but the
percentage of TFM is increasing and perhaps half of
the new units placed in 1987 were TFM. Figure 9 lists
the characteristics of the two types.
BACTERIA - VIRUS
No discussion of RO POU would be complete without
a comment on bacteria/virus. When referring to Figure
2, I noted that viruses are huge compared to the
spacings between molecules. When the membrane is
sound there is no leakage. Membranes today can be
made virtually defect-free. However, current POU
systems cannot be sold as removing all bacteria and
viruses for a few reasons. First, bypass leakage exists
in all elements including the spiral-wound type. This
leakage historically has been on the order of 0.1 to
0.5 percent; i.e., 0.1 to 0.5 percent of the product
does not go through the membrane but through
nonmembrane locations such as the glue lines, leaks,
glue area weepage, O-ring leaks, etc. Secondly,
there is no readily available fast response fail-safe
systems such as a conductivity meter that tells us
66
-------
Figure 8. Miscellaneous data.
Item
Units Placed in 1987
Total Units in Use, end 1987
Manufacturer's Selling Price, $
Cost to Consumer, $ to buy
$/month to rent
Production, gpd
General Ionic, percent CR
General Inorganic Reduction
UTC
60,000-100,000
300,000-400,000
150-250
500-700
16-20
3-10
80-95
Varies
OTC
25,000-50,000
70,000-100,000
30-70
70-150
NA
3-6
85-97
Varies
OU
5,000-10,000
30,000-50,000
350-450
NA
25-40
6-15
80-95
Varies
Figure 9 Membrane types - comparison of merits.
CA Types
TFM Polyamide Types
Advantages
Disadvantages
More experience
Low cost
Good ionic rejection
pH range limited to 3-8
Temperature limited (<100°F)
Subject to bacteria attack
Only fair organic rejection
High flows
High rejections
WidepH range (4-11)
High temperature (120°F)
Bacteria resistant
More expensive
Less experience
Needs carbon prefilter
Tendency to foul
Figure 10. Nimbus N-3A - a 20-year service call record.
Customer Months
per Service Call
25 r-
5 -
Records from Cal Pure
San Diego, CA
(based on local service calls)
68
70
72
74
76
78
80
82
84
86
Year
when a unit fails as regards bacteria/virus removal.
Therefore, for these two reasons, the RO POU
cannot warrant use on a nonpotable water supply. On
the other hand, if I lived on a farm and had no readily
available potable source, I would surely use an RO
POU unit.
As shown in Figure 7, bacteria are always present in
the product part of the system. They are introduced
during installation from dust particles in the air, from
handling, etc. They will build up to a semi steady
state population depending on amount of use,
temperature, species, amount of organics present,
etc. Dr. Lee Rozelle, in his article in this publication,
refer to the July 1987 standard setting a heterotrophic
plate count of 500/ml or less. This number was
related to the ability to accurately test for E. coli and
not the health aspects of heterotrophic bacteria in the
67
-------
drinking water. Several years ago, I tested several
hundred RO POU systems in the San Diego area. I
also tested bottled water cabinet model drinking units.
Essentially, none complied with this criteria. If this
regulation were to be enforced across the board
today, it would destroy the RO POU industry. In fact,
the only unit on the market today that would meet this
criteria is the nonstorage system; i.e., the hang-on
faucet unit. I would suggest a look at developing
different analytical techniques for determination of E.
co// in the presence of other bacteria, if this is the
limitation.
Progress has been made over the years in the
technical service area. I would like to share with you
the record of one dealer - Cal Pure in San Diego.
Rgure 10 shows the service call record for the last 20
years for the Nimbus N-3A model. This data is for
several thousand rental units. I use customer-
months per service call as the criteria of excellence.
This is the average number of months a customer's
unit is in operation between all service calls
regardless of nature of the call. The original design
objective was 12 customer months per service call.
As you can see, it started at around six and gradually
increased to a current level of about 22. If the unit is
owned by the customer rather than rented, the
customer months per service call tend to be several
months longer. This is a remarkable performance
record for a unit that is generally installed by unskilled
labor, operates unattended for long periods of time,
retains a constant reject flow of very small magnitude,
and has check valves that seal under "1" water
pressure over these time periods. I might note that
the dips in the curve were the result of minor system
design changes. One must be very careful of design
changes in these systems. If we plotted the cartridge
life in San Diego instead of customer months against
years, it would add about 10 months to the vertical
part of this curve.
CONCLUSIONS
• RO POU is a relatively old concept ~ going on
20 years. It has been slow in developing but has
picked up speed in the last few years. There are
around 500,000 units in use in the U.S. today.
• The economic value in 1987 was about $40 million
to the system manufacturers and $100 million to
the dealers and distributors.
* The challenge of the immediate future in the
nonhealth-related areas for RO POU is marketing,
sales, and service. We have satisfactory
membranes to cover the needs, and the technical
aspects are sufficiently in hand.
68
-------
PERFORMANCE AND APPLICATION OF ULTRAVIOLET LIGHT SYSTEMS
Clyde Foust
Ideal Horizons Inc.
Rutland, VTOS701
INTRODUCTION
The increasing demand for high quality water,
combined with the reduction in quality of available
water sources, has caused an upsurge in water
treatment technology. Ultraviolet (UV) light as a
disinfectant was first discovered in 1847 and first
commercially applied in 1901. The effectiveness of
UV as a preferential microbiological treatment should
be evident in its choice as the disinfectant in over
2,000 cities in Europe. Ultraviolet light accomplishes
this without using harmful chemicals, some of which
when used can lead to the formation of harmful
secondary chemical products, such as
trihalomethanes. Ultraviolet light does not impart any
taste or odor nor is it possible to overtreat water
using this method. Properly designed units cannot
expose users to any harmful products. The effect of
no measurable residual disinfectant in the water has
been well studied in Europe and found to be of no
concern. Maintenance of these systems is usually
limited to annual servicing.
Most of the commercially available units use low
pressure mercury vapor lamps. Operationally and
mechanically, these lamps are very similar to the
fluorescent lighting with which we are all familiar.
Most units are quite simple in principle because the
mercury vapor bulb in conjunction with a pressure
reaction chamber is all that is required. However, in
application, the units are considerably more
complicated. The commercially available units must
also be fitted with monitors, flow controllers, and
other operational equipment. To treat water with
ultraviolet light, the three parameters necessary to
consider are the organism, dosage, and unit design.
ORGANISMS
The UV spectrum in this article refers to the spectrum
between 200 and 300 nm (7.87 x 10-6 and 1.18 x
10-5 jp). AS a point of reference, the entire visible
spectrum is 380 to 700 nm (1.5 x 10-5 in to 2.76 x
1Q-5 in) with the major spectrals between 450 and
600 nm (1.77 x 10-5 jn and 2.36 x 10-5 in) The
inactivation of organisms occurs through
photophysical damage imparted to the DNA by the
UV light. The low pressure mercury vapor light emits
most of its energy centered about 254 nm (1 x 10"5
in). This allows a high efficiency because the
germicidal action curve for most organisms is
centered at about 260 nm (1.02 x 10-5 in). Since
this high energy wavelength is readily absorbed by
DNA, RNA, protein, and enzymes a single photon
strike affects most organisms.
Most common pathogenic microorganisms have been
tested for their sensitivity to UV light, and the results
have been published in various publications. £. coli
will be reduced to a 0.0001 survival ratio, if treated in
most commercially available units. Giardia lamblia has
a survival ratio of 0.1 with twice the treatment
available in most commercial units.
The sensitivity of organisms is determined by
bioassay methods. Care should be exercised when
using the data, because some results are achieved
using various wavelengths for inactivation. Data using
wavelengths other than 254 nm (1 x 10"5 in) may be
used with corrections allowed for effectiveness.
DOSAGE
Photophysical damage is time dependent with
organisms dying in a constant fraction with increasing
increments of time. This is expressed as the survival
ratio. The survival ratio when plotted on semilog
paper is a straight line. The implications of this effect
are reflected in the sizing of the unit and will be
discussed later. Temperature effect on this process is
negligible. The main factor in sizing is UV dosage.
The. recommended dosage is at least 16,000 yW-
s/cm2 (103,200 nW-s/sq in) based upon an HEW
1966 policy statement. Presently, most manufacturers
treat with a dosage of 30 to 35,000 iiW-s/cm2 (190
to 225,800 pW-s/psi). These dosage rates are
capable of a four-log reduction, which means that
incoming water would be effectively 99.99 percent
treated. Applying these numbers to the effluent of a
well designed and well run sewage treatment plant
would result in the treated water meeting the £. coli
standard for drinking water.
69
-------
UNIT DESIGN
Application of the available technology is quite simple
and easily quantified. Most commercial units are
designed to operate when supplied with water that
meets a known specification. As with all drinking
water supplies, the water supply must meet EPA
nonbiologic standards.
Additionally, the turbidity should be less than 10 NTU.
The turbidity, while not directly measuring
transmission in the wavelengths under examination, is
indicative of transmission. The two most commonly
found UV absorbers are iron and tanin.
Most units presently available are similar in
construction as far as basics. They consist of a
cylindrical reaction chamber with the bulb mounted
along the center axis allowing water to flow parallel to
the bulb. The bulb is separated from the water by a
quartz sleeve. To maintain proper treatment levels,
some type of flow control device is necessary.
Beyond these basics some type of monitoring and/or
shut-off mechanism is required. The minimum
protection would be a visible light metering device to
cause an alarm or effect a shut-down in the failure
mode. Utilization of the visible spectrum light for this
purpose is possible because the low pressure
mercury discharge tubes emit some of their energy in
the visible range. Obviously, this device will not
protect against all modes of failure but is sufficient for
most applications. The next level of protection may be
achieved by using a light sensing device that
measures only light in the UV sterilization range, and
has absolutely no response in any other range. This
more sophisticated unit should be provided with
appropriate time delays and may be slaved to any
alarm or shut-down device.
APPLICATION
Most waters that do not meet the specifications can
be preconditioned to acceptable levels. However, the
minimum preconditioning required is a 5-um
(0.0002-in) prefilter. This is necessary to insure
most particles are removed that could allow the
organism's penumbral or umbral shading.
Start-up and maintenance of these units is quite
simple. After installation and any breach of the
system, the downstream piping should be sterilized.
Normally the unit requires annual servicing, which
consists of cleaning the parts and replacing the bulb.
The NSF presently has no standards concerning UV
water purification. Under their proposed Standard 55,
UV equipment will be accepted for the NSF seal. The
proposed standard is in its first testing stages and
should be established in a relatively short time.
70
-------
PRECOAT CARBON FILTERS AS BARRIERS TO INCIDENTAL MICROBIAL CONTAMINATION
P. Regunathan, W. H. Beauman, and D. J. Jarog
Everpure, Inc.
Westmont, IL 60559
The microbiological characteristics of point-of-use
(POD) filters have been a source of concern in the
minds of many in the regulatory, utility, and academic
fields. These concerns have centered around the
possible growth of bacteria in various POL) devices,
particularly carbon filters, and the probable effect of
such growths on the health of the consuming public.
It is the objective of this article to investigate the role
of a POU precoat carbon filter as a final barrier to
microbial contamination that may otherwise
inadvertently reach the-consumer. A schematic view
of such a possible role in the overall route of water
from source to use is shown in Figure 1, giving the
various acknowledged barriers that are traditionally
depended upon to provide safe water to the
consumer. In this article, a POU fine-filtering precoat
carbon filter has been specifically examined as an
added or final barrier to different types of microbial
contaminants.
POU precoat filters belong to a family of POU fine-
filtering devices that includes, besides precoat filters,
carbon block filters made from powdered carbon,
ceramic filters manufactured with controlled fine
pores, pleated membrane filters with absolute ratings,
and reverse osmosis membrane systems. All these
devices are usually rated as capable of filtering down
to 1 urn (0.00004 in) or less.
Figure 2 shows a schematic view of a precoat carbon
filter where the water enters from the top through an
inlet tube and rubber check valve into the bottom of
the filter, making the dry powdered media there into a
slurry, which is then evenly deposited as a precoat
layer on the surface of a folded septum envelope.
During operation, water intermittently enters the unit,
filters through the precoat cake and the septum
fabrics, travels through the drainage grids that are
inside the septum envelope, and reaches the outlet of
the filter. Precoat filter media in precoat carbon filters
include mostly powdered activated carbon. Table 1
shows the significance of a precoat carbon filter as a
barrier for different types of contaminants. While there
are some specific barrier benefits for organic,
inorganic, and particulate contaminants, this article
focuses on microbial contaminants, which can be
properly divided into bacterial pathogens (coliforms as
indicators), viruses, protozoan cysts or surrogates,
and heterotrophic bacteria. In addition, the use of
silver and, in some tests copper also, has been
studied as bacteriostats in precoat carbon filters.
COLIFORM REMOVAL
A test was set up to compare the abilities of filters
with or without silver or copper in removing and
inhibiting coliforms. Two standard precoat carbon
filters, two containing copper powder added to the
filter media, and two containing silver plated onto a
powder and added to the media, were plumbed into a
test module. The influent water was softened,
dechlorinated, and fine-filtered Westmont tap water
to which a nutrient-deprived suspension of
Enterobacter aerogenes (ATCC 15038) had been
added by a pump just before the module. The
choices of organism, water pretreatment, and other
precautions in culture preparation and transfer were
designed to promote maximum coliform survival.
All six units were operated at a 0.06-l/s (1.0-gpm)
flow rate, with a 30 seconds ON-30 seconds OFF
cycle for eight hours per day, for several days to
simulate the condition of used filters being subjected
to accidental contamination. After 3,785 I (1,000 gal)
had passed through each filter, samples were
regularly collected just before and after overnight and
weekend quiescent periods. All the data collected
along with detailed procedures have been reported
elsewhere (1). Results are shown graphically in
Figures 3 through 6.
Influent coliform levels increased from about
10,000/100 ml (2,960/oz) to more than 100,000/100
ml (29,600/oz) due to improvements in technique.
Effluent levels in standard filters (Figure 3) were
approximately 99 percent lower. Both influent and
effluent levels generally decreased during overnight
and weekend quiescent periods. Filters with copper
(Figure 4) allowed higher running coliform levels in
the effluents than standard filters, probably due to
disruption of the precoat cake by heavy metallic
copper powder, but copper reduced these levels
71
-------
Figure 1. Barriers to contamination.
Infiltration
through
river banks, etc.
Source
Protection
Treatment Plant
Pre-
Dlslnfectlon I
H Coagulation I I
& I— Filtration —I
Settling | I I [
IPost-
Dlslnfectlon
Clear Well
8
Distribution System
Management
C—*"
to User
Figure 2. Prccoat filter design.
Sect A-A
Filter Media
when dry
Exploded View of
Septum Envelope
In Operation
Drainage
' Grids
When powdered
carbon is included in the
filter media, then it is a
precoat carbon filter.
significantly during quiescent periods. Silver-
containing filters (Figure 5) generally reduced the
coliform levels similar to standard filters, but further
exhibited significant inhibitory effect during quiescent
periods. Averaged levels (Figure 6), show the relative
capabilities of these filters with filters containing silver
showing a further 1/2-to-1 log reduction compared
to the other filters.
VIRUS REMOVAL
Standard precoat carbon filters without any
bacteriostats have been recently evaluated for their
ability to remove viruses using a protocol developed
by a U.S. EPA-Army task force (2). Virus reduction
requirements (along with bacterial and cyst
reductions) for a unit to achieve purifier status have
been determined by the task force to be a minimum
of four-log reduction of a mixture of 107 plaque
forming units (PFU) of poliovirus 1 and 107 PFU of
rotavirus Wa or SA-11 per liter (0.26 x 107 PFU of
poliovirus 1 and 0.26 x 107 PFU of rotavirus Wa or
SA-11 per gallon).
This test on the filters was conducted at the
Department of Microbiology and Immunology of the
University of Arizona, Tuscon (3). Three filters were
tested as per protocol (2), plumbed into a test rig
served by large tanks, a pump, bladder tank, solenoid
valves and timer set to operate on a cycle of three
72
-------
Table 1. Significance of Contaminant Reduction
Type of Contaminant
Significance of Precoat Carbon
Filter As a Barrier
Organics
Taste and Odor-Causing
Organics
Volatile Organics
Other Health-Related Trace
Organics
Inorganics
Particulates
Turbidity, etc.
Microbiological
Bacterial Pathogens or
indicators
Viruses
Protozoan Cysts
Heterotrophic Plate Count
Significant, but may not be
health related
Not high enough capacity, low
amount of carbon
May be significant
Generally no removal, unless
heavy metal in precipitated form
Significant as a barrier
May be significant enough barrier
May be significant enough barrier
May be significant enough barrier
minutes ON and 27 minutes OFF for eight hours per
day. Influent (control) and all three effluents were
sampled seven times as per the protocol and were
assayed for PFU. Complete details of procedure and
results have been reported elsewhere (1,3). Table 2
shows the results of this test.
Influent levels shown in the table indicate that the
high challenge levels of 107 PFU/I (0.26 x 10?
PFU/gal) were not achieved in this effort. In spite of
this flaw, the precoat carbon filters appear to have
reduced the virus concentrations of 10^ to 1Q6 PFU/I
(0.26 x 104 to o.26 x 106 PFU/gal) in the influent to a
range of <10 to 104/1 (<2.6 to 0.26 x 104/gal). This
reduction of approximately 99 percent is significant,
even though these filters cannot be said to have
achieved the four-log reduction required for purifier
status. Further efforts are underway to better define
the capabilities in this area.
PROTOZOAN CYST SURROGATE
REMOVAL
Protozoan cysts, being inert and having no capacity
for movement or reproduction, can be treated as
particles that can be removed by filters at the point-
of-use. There is evidence to support the use of
surrogate particles in lieu of live cysts in filtration
tests.
National Sanitation Foundation (NSF) Standard 53 (4)
has detailed procedures for Cyst and Turbidity
Reduction, and requires a filter to reduce particles in
the range of 4 to 6 urn (0.00016 to 0.00024 in) by at
least 99.9 percent throughout the life of the filters.
These procedures were followed to test standard
precoat carbon filters without any bacteriostats for
their ability to remove such surrogates.
Two filters were installed in a test rig with a system of
solenoid valves and a timer set to operate the units
for 1.5 minutes ON and 13.5 minutes OFF for 16
hours per day, feeding Westmont tap water at 60 psig
fortified with fine test dust to 20 to 30 NTU. Samples
were taken at start-up and at morning start-up after
the overnight quiescent periods when the initial flow
rate had been reduced by filter plugging to 75, 50,
and 25 percent. Samples were analyzed with a
particle counter with one channel set to record 4- to
6-um (0.00016- to 0.00024-in) particles.
Results in Table 3 show the filters to be effective in
removing the 4- to 6-um (0.00016- to 0.00024-
in) particles. The filters improve in efficacy as they
become plugged up with fine dust as would be
expected of precoat filters.
The size of 4 to 6 um (0.00016 to 0.00024 in) was
chosen by NSF with amoebic and giardial cysts,
which are two to three times larger, in mind. There
has been some concern about an inadequate factor
of safety in relation to Cryptosporidium cysts, which
are also 4 to 6 um (0.00016 to 0.00024 in) (5). New
data have been gathered recently in the NSF
laboratories using a submicron detector. The
procedure used in this test is the Paniculate
Reduction test described in NSF Standard 42 (6),
which is similar to the earlier test. Differences include
the use of 10 minutes ON and 10 minutes OFF
cycles, collection of samples at the beginning and at
morning start-up after the buildup of pressure drop
to 277 kPa (40 psi) across the filters, and the analysis
with submicron detector of several size ranges.
Results in Table 4 show the abilities of filters to
remove 0.5 um (2 x 10"5 in) particles by more than
99.9 percent, indicating the ability of these filters to
remove all cysts, including Cryptosporidium cysts.
EFFECT ON HETEROTROPHIC
BACTERIA
Scientific studies have addressed concerns regarding
heterotrophic bacterial growth in POU devices,
notably two funded by U.S. EPA. One studied and
quantified the growth in POU products (7,8), and the
other studied the epidemiologic correlation of such
growth with illness (9). Conclusions from these as
well as other studies can be summarized to say that
average increases in bacterial populations in the
effluents of POU units compared to influent levels
were around one order of magnitude, but exposure to
such higher densities was not statistically correlated
with any increase in acute symptoms, either
gastrointestinal or dermatologic, compared to
exposure to unfiltered water.
These studies and results, however, have not abated
the concerns expressed by many (10-12). In this
study, further tests have been conducted to provide
additional information on the growth of heterotrophic
plate count (HPC) organisms in precoat carbon filters
73
-------
11/V 001 J=e( SU1JOJH03
3
tf
•il
•WOOtJid muojuoa iBioi
SUJJOJIKO |B»Ql
74
-------
Table 2. Virus Removal Tests
Plaque-Forming Units/Liter*
Average
Time (days)
1
3
6
7
8
10
10
(stagnant)
Influent
717,000
230,000
203,000
257,000
60,000
53,300
13,000
Unit #1
4,200
2,100
330
670
<10
<10
<10
Unit #2 Unit #3
9,000 1,400
6,100 3,300
330 <10
<10 <10
<10 <10
<10 <10
< 10 <10
(percent)
99.32
98.33
99.89
99.91
> 99.98
> 99.92
> 99.92
* Average of triplicates.
Table 3. NSF Cyst Reduction Test
Flow
Reduction
(percent)
0
25
50
75
No. of 4-6 urn
Particles/mr
Influent
76,492
158,140
348,223
212,783
Unit#1
76
128
35
20
Unit #2
60
92
40
19
Average Reduction
(percent)**
Unitn
99.90
99.92
99.99
99.99
Unit.#2
99.92
99.94
99.99
99.99
* Average of triplicates.
** Minimum required for acceptance by NSF: 99.9 percent.
Table 4. NSF Filtration Efficiency Test
Startup
After
40 psig
Ap
Particle
Size
(nm)
0.5-1
1-5
5-15
15-30
0.5-1
1-5
5-15
15-30
Unit #1
Influent
Counts*
282,165
32,595
395
5
357,930
58,640
625
10
Counts*
117
23
1
0
21
2
0
0
%
Red.
99.96
99.93
NS
NS
99.99
99.99
NS
NS
Unit #2
Counts*
245
26
1
0
261
7
0
0
%
Red.
99.91
99.92
NS
NS
99.92
99.99
NS
NS
* Average of triplicates.
NS - Influent challenge insufficient for significance.
incubation temperature, and time to the researcher,
but requires the conditions to be spelled out along
with the results. The more commonly utilized
conditions appear to be R2A medium, incubation
temperature of 28°C (82°F), and incubation time of
seven days. The number and type of organisms
enumerated under these conditions appear to be
significantly higher and different from those obtained
by SPC procedures. This factor was investigated in
this study also. ; '
!,
In conjunction with the coliform removal studies
discussed earlier, SPC data also were collected in the
same samples that were analyzed for coliforms.
Figure 7 shows all the results of the analyses
indicating no concrete conclusions, except that
generally the softened, carbon filtered influent waters
had higher counts than the filter effluents and that
silver or copper did not have any measurable
inhibitory effect on these bacterial levels.
Figure 7.
HPC organism in various precoat carbon filter
effluents.
S 10.000. ,
<
Effluent* from Filters
itJndjfd
Running •
Ovtrnlgnl 0
Q
1,000 t.ZOO
l.*oo 1.COO
Volume Filtered. Gal
and the effect of bacteriostats, specifically silver, on
such growths.
It is important to point out the differences in the
presently used methodologies used in HPC
procedures. In earlier Standard Methods (13), plate
count organisms were termed Standard Plate Count
(SPG) organisms, and the procedures required the
use of plate count agar with an incubation
temperature of 35 °C (95 °F) and incubation time of
two days. Many of the earlier important studies used
this methodology. The most recent edition of
Standard Methods (14) leaves the choice of media,
A more recent test was conducted using the
Bacteriostatic Test Procedures protocol in NSF
Standard 42 (6). Complete details of test procedures
can be found elsewhere (1,6). Two standard precoat
carbon filters and two similar filters with silvered
powder added to the filter media were simultaneously
tested using three different types of waters. Solenoid
valves and timers were set to operate on a cycle of
three minutes ON and 27 minutes OFF for 16 hours
per day. The test was continued for 6670 I (1,762 gal)
filtered through each unit at an operating pressure of
415 kPa (60 Ib/psi) as required in the test procedures.
The results are shown in Table 5. These SPC data
75
-------
Tablo S.
NSF Bacteriostatic Test Data
Heterotrophic Plate Counts/mr
Walcr Filtered
(gal)
0
0
1
22
73
319
740
980
1.220
1.220
1.460
1,722
1,762
Type oi Water
Quality"
Regular
Regular
Regular
Low
High
Regular
Regular
Regular
Regular
Regular
Regular
High
Low
Duration of Quiescence Just
Priot to Sampling-Hour
0
60
0
8
8
8
8
8
0
60
8
8
8
Filters w/o Silver
Influent
12
35,000
42
9,000
4,400
1,900
3,900
240
11
23,000
135
11,000
3,200
Unit #1
-
-
2
600
530
1,300
97
36
19
3,500
46
69
1,400
Unit #2
-
-
<1
1
<1
6
11
1
<1
3
<1
<1
430
Filters w/Silver"*
Unit #1
-
-
1
1
1
5
1
1
1
15
1
1
330
Unit #2
-
-
<1
<1
<1 •
1
<1
<1
<1
1
<1
<1
4
RoQlrfar - 200-600 ppm TDS, 7.2 + 0.5 pH.
Low - 25-100 ppm TDS, 6.2 ±0.5 pH.
High - > 800 ppm TDS, 9.5 ± 0.5 pH.
Duplicate analyses. 35°C, 48 hr.
Effluent silver; 1-3 ppb.
indicated that both the units without silver, as well as
those with silver were bacteriostatic filters, according
to NSF Standard (6) requirements. Silver, however,
does seem to provide some extra assurance and
capability in providing fairly low counts in the effluents
of both units, while one of the two units without silver
had significantly higher counts than its duplicate unit.
During July 1987 a comprehensive test program was
initiated t& study the growth of SPC and HPC
organisms in standard precoat carbon filters with or
without silver as a bacteriostat, and in granular carbon
bed units. The following were the pertinent factors
controlled or used in the tests:
• Three of each type of unit (standard, w/silver,
granular) with two influent locations.
• 0.03 l/s (0.5 gpm) through each filter and each
influent port.
• Westmont municipal water as supplied.
Total chlorine content in running samples = 1.0 to
1.5 mg/I (1.0 to 1.5 ppm)
Free chlorine content in running samples = trace
levels.
• Three minutes ON, 27 minutes OFF, eight hours
per day only during working days.
* Sampling
- Monday AM at start-up
- Wednesday PM during running
- Thursday AM at start-up
- Friday PM during running
• 1 / (0.26 gal) sample (£ minute run) collected.
Bacteriological sample from this into sterile bottle
with Chambers' neutralizing solution. (Influent
samples always taken before filter effluents.)
• Only R2A medium used in the beginning. Started
with pour plates, switched to spread plates. Started
with 25°C (77°F) for five days, changed to 28°C
(82°F), seven days.
• Tests started 7/17/87.
• Plate count Agar, incubation at 35 °C (95 °F) for
two days routinely used from 9/2/87 for SPC
measurements.
HPC data collected using R2A media for the influent
samples and the three sets of effluent samples are
graphically shown in Figures 8 through 11. Arithmetic
means of all these separate sets of data have been
calculated and are shown in Table 6. Comparison of
the data in these figures and table would indicate that
the use of silver in precoat carbon filters results in
maintaining the organism levels at or below the
influent levels for the different types of samples, i.e.,
running, overnight, and weekend samples. Running
and overnight quiescent samples from standard
precoat carbon filters without silver appear to have 1
to 1-1/2 orders of magnitude higher HPC counts
than influent samples or the effluent samples from
units with silver. Granular carbon filter effluents
yielded generally 1-1/2 orders of magnitude higher
counts than influent waters or precoat units with
silver, when running or overnight samples were
76
-------
£
» | 1
« 5 i
5 3 1
01
-------
compared. Weekend quiescent samples from influent
ports and granular carbon filters were in the same
order of magnitude, while precoat units with or
without silver generally yielded 50 percent lower
numbers.
Tablo 6. Comparison of Bacteriological Data, Arithmetic
Mean HPC*, CFU/ml
Influent
Silver
Standard Granular
Running
Overnight
Weekend
194
2,510
75,700
260
1,870
27,600
7,040
20,900
37.300
8,730
44,700
87,200
" R2A media, 25-28'C, 5-7 days.
Table 7. Comparison of Bacteriological Data, Arithmetic
Mean SPC*. CFU/ml
Influent
Silver
Standard Granular
Running
Overnight
Weekend
33
444
26,700
47
332
7,350 '
23
515
7,240
1,050
4,180
22,600
* SPC: Pour Plates, 35°C. 2 days.
SPC data collected using plate count agar (35 °C
[95°FJ, two days) for similar units are graphically
shown in Figures 12 through 15. Arithmetic mean
values for the separate sets of data have been
calculated and are shown in Table 7. Examination of
these values shows different conclusions from those
reached using HPC data. Unlike earlier tests, these
data show that precoat filter units with and without
silver yield organism counts equal to or lower than
influents when running water, overnight quiescent, or
weekend quiescent samples are compared between
themselves. Only granular carbon bed units yield 1 to
1-1/2 orders of magnitude higher counts in running
and overnight quiescent samples.
Table 8 shows the direct comparison of HPC and
SPC data from all the samples for which both
analyses were performed. This comparison confirms
earlier stated observations and conclusions. Further, it
shows that R2A media at 28 °C (82 °F) for seven days
yields 1/2 to 1-1/2 orders of magnitude higher
counts than those obtained using plate count agar at
35°C (95°F) for two days. The regulatory efforts to
control these organisms at a particular level need to
consider these huge differences in values obtained
from identical samples.
These comparisons indicate that silver when used in
precoat carbon filters has a selective effect on
organisms. Earlier coliform removal tests showed that
silver had a measurable effect on Enterobacter
aerogenes. Silver also seems to have a significant
effect on those organisms that grow in R2A media at
28 °C (82 °F), while it has no measurable effect on
those organisms that grow in plate count agar at
35°C (95°F). Efforts to identify these selective effects
were not successful, because many of the colonies
on the plates could not be identified. The few
identifiable organisms shown in Table 9 did not yield
enough useful information. Further efforts are needed
in this area of activity.
CONCLUSIONS
On the basis of this study, the following conclusions
can be reached:
• Data presented show significant and consistent
reductions by these precoat filters of coliforms (~
99 percent), enteric viruses (~99 percent), and
protozoan cyst/surrogates (>99.9 percent).
• Silver in precoat carbon filters lowers coliform
levels at least 1 log more than the standard
precoat filters. Silver .and copper act slowly to
reduce coliform levels in filters during non-use
periods.
« The decision to use R2A agar at 25 to 28 °C (77 to
82 °F) for seven day incubation procedures instead
of SPC agar pour plates at 35 °C (95 °F) for two
days is not trivial. Not only does the new HPC
procedure yield 1/2 to 1-1/2 logs higher counts, it
also favors a different population and thus can lead
to different overall conclusions.
» Silver appears to effectively control. the filter
effluent HPC levels that are found growing in R2A
media incubated at 25 to 28 °C (77 to 82 °F).
Proper comparisons of all data indicated silver as
bacteriostatic in precoat carbon filters.
• Precoat filters with or without silver appear to act
as barriers to incidental microbial contamination
because they do confer some measurable level of
protection against incidental microbial
contamination of potable water supplies that are
normally safe.
ACKNOWLEDGEMENT
The authors thank Charles Ferrara for producing the
illustrative and graphic figures, Robert Gonzalez for
collecting and assembling data, and Pam Menefee for
producing the manuscript.
78
-------
%
in 3 O
- B 3
« (B C
>- i. 5.
I?1!!
u. .5 «
J3 £
9 Ib
5
if
M
1
a
1
e
IN «
- 3
b. ,S
41
U
0
o>
Oj k M
I « S
H E JS
arc
O
8
79
-------
Table 8. Direct Comparison of SPC vs. R2A, Arithmetic Mean, CFU/ml
Influent
Running 33
Overnight 444
Weekend 86,700
SPC: Pour Plates, 35°C, 2 days
Silver Standard Granular
47 23 1,050
332 515 4,180
7.350 7,240 22,600
Table 9. Documented Organisms 7.
Influent Standard Silver Granular
Psoudomonas
acidovrans
lutoola
paucimobilis
vcsiculans
slutzori
copacia
I'uofcscons
maltoi
Flavobacterium spp.
Agrobacterium
radiobactcr
Acliromobaclcr spp.
X X
X X X X 8.
X X X X
X X X X
XX X
X x
X
X X X X
XXX
x
HPC: R2A Spread Plates, 25-28 °C, 5-7 days
Influent Silver Standard Granular
331 497 4,880 8,050
3,410 3,130 29,200 60,200
108,000 46,700 34,900 126,000
Criteria and Standards Division, Office of Drinking
Water, U.S. EPA. Fact sheet/update, home water
treatment units contract. July 1980.
Bell, F. A., Perry, D. C., Smith, J. K. and Lynch,
S. C. Studies on home treatment systems.
JAWWA 76:2, 126-130, February 1984.
Calderon, R. L. An epidemiological study on the
bacteria colonizing granular activated carbon
point-of-use filters. In Press: Proceedings of
the Water Quality Association Annual
Conventions. Dallas, TX, March 1987.
Geldreich, E. E. et al. Bacterial colonization of
REFERENCES
1. Regunathan, P. and Beauman, W. H.
Microbiological characteristics of point-of-use
precoat carbon filters. JAWWA, 79:10:67, October
1987.
2. Gerba, C. P. and Thurman, R. Towards
developing standard procedures for testing
microbiological water purifiers. Proceedings of the
Third Conference on Progress in Chemical
Disinfection. Binghamton, NY, April 1986.
3. Gerba, C. P. and Kutz, S. M. Evaluation of
Everpure 4C cartridge filters for virus removal.
Unpublished report, University of Arizona,
Tucson, May 1987.
4. National Sanitation Foundation. Standard 53:
drinking water treatment units - health effects.
Ann Arbor, Ml, Rev. June 1982.
5. Culotta, N. J. Personal communication. National
Sanitation Foundation, Ann Arbor, Ml, March 27
1987.
6. National Sanitation Foundation. Standard 42:
drinking water treatment units ~ aesthetic
effects. Ann Arbor, Ml, Rev. June 1982.
point-of-use water treatment devices. JAWWA
77:2, 72-80, February 1985.
11. Reasoner, D. J, et al. Microbiological
characteristics of third-faucet point-of-use
devices. JAWWA, 79:10:60, October 1987.
12. Fed. Reg. 50:219, November 13, 1985.
13. Standard methods for the examination of water
and wastewater. APHA, AWWA, and WPCF.
Washington, D.C. (15th ed., 1980).
14. Standard methods for the examination of water
and wastewater. APHA, AWWA, and WPCF.
Washington, D.C. (16th ed., 1985).
80
-------
MICROBIOLOGICAL STUDIES OF GRANULAR ACTIVATED CARBON POINT-OF-USE
SYSTEMS
Donald J. Reasoner
Drinking Water Research Division
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
INTRODUCTION
Activated carbon, powdered (PAG) or granular (GAG),
has been used in water treatment for many years,
primarily to remove tastes and odors. During the past
15 to 20 years, the availability and sales of small PAC
and GAG filter units, or point-of-use (POD)
treatment units, have increased steadily. Much of the
increased demand for these units has resulted from
consumer concerns over the quality of the water
supplied by local water utilities. These concerns are
stimulated by national and local news reports of
organic and inorganic chemicals in drinking water,
health risks due to long-term ingestion of potentially
carcinogenic compounds in drinking water, taste and
odor problems in the local water supply, and other
problems of treatment and/or distribution. Other
factors contributing to the use of POU devices
Include lack of a centrally treated water supply,
contaminated ground water supplies, and ground
water supplies containing naturally high
concentrations of iron, sulfur, nitrates, or fluorides. An
aspect of GAG POU treatment devices that has been
of concern to the Drinking Water Research Division,
Water Engineering Research Laboratory, U.S. EPA,
Cincinnati is the long-term microbiological quality of
the product water from such devices.
The first phase of studies on the microbiological
characteristics of carbon POU filters was begun in our
laboratories in 1977. This phase examined four
carbon POU filters for variations in bacterial counts,
heterotrophic plate count (HPC), total organic carbon
(TOG), and chloroform (CHCIs) levels. In addition,
one filter was installed on a low-flow drinking water
fountain, and bacteriological quality and THM content
of the product water were monitored weekly. HPC
levels in the effluents from the carbon filters were
always higher than the bacterial levels of the influent
tap water. One of the test filters had consistently
higher HPC levels than the other three test filters, and
morning samples generally had higher HPC levels
than did those in the afternoon. Free-chlorine
removal for the test carbon filters ranged from about
53 to 65 percent in one run, and 77 to 97 percent in a
second run. Total organic carbon removal appeared
to be minimal because influent TOG levels were
always less than 2 mg/l (2 ppm) and effluent levels
were within 0.1 to 0.3 mg/l (0.1 to 0.3 ppm) of the
influent levels one day after filter installation. Two to
four weeks later, effluent levels were essentially the
same. Chloroform removal effectiveness was directly
related to the amount of carbon in the filter. The initial
percentage removal of CHCIa ranged from 100
percent down to about 55 percent, and decreased
with time over a period of 20 weeks. By the end of
the 20-week test period, effluent CHC|3
concentration for three of the test filters exceeded
influent CHCIa concentration, while the fourth filter
unit was still removing about 20 percent of the
influent
The second phase examined four additional GAG
POU filters designed to be installed as stationary
filters under a kitchen sink on the cold water line to a
common mixing faucet. This study examined the HPC
of the product water from the POU filters and the
potential colonization of the test filters when
challenged by pure culture suspensions of
opportunistic and frank bacterial pathogens that might
contaminate a potable water supply. This phase of
the POU study was reported by Geldreich et al. (1).
Variations in HPC levels in the product water between
morning and afternoon samples from the test filters
were similar to those found during the first phase
study. Additionally, it was shown that stagnation
periods from several hours to several weeks resulted
in significantly increased HPC levels in the test filters.
The HPC levels in the product water could be
reduced after a stagnation period by simply flushing
the units thoroughly at full-flow for two to three
minutes before using the water.
The potential for colonization of the test filter units
was, examined by using pure culture suspensions of
Serratia marcescens, Pseudomonas aeruginosa,
81
-------
Enterobacter cloacae, Enterobacter aerogenes,
Escherichia coli, Citrobacter freundii, and Salmonella
typhimurium. £ coli, S. typhimurium, and £ cloacae
did not colonize the filters and were not detected in
the product water after the initial sample. S.
marcescens, P. aeruginosa, £ aerogenes, and C.
freundii persisted and were found in the product water
from some or all of the test filters for periods of time
ranging from five days for C. freundii to 156 days for
S. marcescens and P. aeruginosa.
In addition to following the HPC levels of the POU
filter units and challenging them with the organisms
above, new filter cartridges were installed, disinfected,
and the product water monitored for 12 months for
HPC bacteria. Bacterial cultures were periodically
isolated, purified, and identified*The organisms found
during this 12-month period included C. freundii, E.
aerogenes, £ cloacae, K. pneumoniae, Alcaligenes
spp., Pseudomonas cepacia, P. fluorescens, P.
maltophila, S. marcescens, and S. rubidaea. Not all
of these organisms were isolated from each of the
filter units at the same sampling time, and no
individual unit yielded isolates of all of the organisms.
Only £ cloacae, Alcaligenes spp., and P. maltophila
were found in all the units at some time during the
12-month period, and none of the isolates was
found to be continuously present. Since the POU test
set-up had been disinfected prior to the beginning of
this 12-month study, the organisms isolated must
have been present in the treated distribution water
influent to the test system, and were able to survive
and multiply to some extent in the POU filter
cartidges.
The third phase of the GAG point-of-use study,
currently nearing completion, was designed to
examine GAG POU treatment units intended to be
installed as third-faucet units, thus treating only
water for drinking and cooking, not all of the water
going to the main kitchen faucet. Seven test units
were selected for this phase. The test set-up
configuration from the previous two study phases was
retained, but the plumbing was modified to accept the
additional three test units. Partial results from this
phase were reported earlier as technical conference
presentations and appeared in the October issue of
the Journal of the American Water Works
Association. HPC results from this phase of the POU
study showed several types of variation. POU product
water HPC levels varied depending on the time of
sample collection (morning versus afternoon).
Generally, the afternoon samples contained fewer
HPC bacteria than did the morning samples, reflecting
the effect of flushing on wash-out of bacteria during
the daytime simulated use periods. The magnitude of
the difference beween morning and afternoon HPC
levels varied among the test units. Examination of the
variation in the monthly mean HPG indicated that
there was no single pattern of HPC results for all the
test units other than the changes that occurred with
flushing between the morning and afternoon samples.
Some of the variation in HPC monthly means may be
attributed to the combined influence of seasonal
changes in water temperature and changes in
disinfectant residual. HPC levels for some test units
clearly decreased as water temperature decreased
from September through December. Peaks in HPC
levels usually occurred in July, August, or September,
corresponding to peak water temperatures.
Some HPC variation may reflect the influence of unit
design, volume of GAG in the cartridge, and possibly
the material used in the construction of the cartridge
holder and the cartridge itself. Metal cartridge
housings may contribute to more rapid equilibration of
the water and GAG within those units to ambient
room temperature during nonflow periods. Any
increase in water and GAG cartridge temperature
would result in increased growth of bacteria on the
GAG, resulting in higher HPC levels in the product
water.
The bacterial flora (HPC) of the dechlorinated tap
water influent to the test POU units reflected the
contribution of the GAG dechlorinating filter. This filter
served to remove free chlorine from the distribution
water and seeded the dechlorinated water with
bacteria. Thus, reasonable worst case conditions
were set up for the tests (i.e., no free chlorine
residual and HPC levels greater than the treated
distribution water). The mean HPC of the
dechlorinated tap water was generally lower than that
of the water from the best of the POU test units. POU
test units that included silver as a bacteriostatic agent
were found to have HPC levels as variable as the
nonbacteriostatic POU test units. The bacterial flora
of the bacteriostatic units appeared to be different,
both in colony appearance and variety, from the
bacterial flora of the nonbacteriostatic POU units. The
silver served as a selective agent, allowing silver-
tolerant bacterial strains to grow.
All of the POU test units modified the percentage of
pigmented bacteria found in the product water, as
compared to the percentage of pigmented bacteria
found in the dechlorinated tapwater. Generally, the
percentage of pigmented bacteria present in water
from the bacteriostatic POU test units was lower than
that of the nonbacteriostatic test units. The
dechlorinated tap water usually contained more than
50 percent pigmented bacteria, whereas the
pigmented bacterial content of the water from the
POU units ranged from less than four percent to
about 40 percent depending on whether the samples
were taken in the morning or afternoon. Old filters (on
line for several months) tended to show relatively
stable HPC and pigmented bacterial levels, whereas
newly installed filters had low initial HPC levels that
rapidly increased during the first two to three weeks
of use, and became fairly stable thereafter.
Challenges of test POU filter units with specific
bacterial pathogens (Klebsiella pneumoniae,
82
-------
Aeromonas hydrophila, and V. enterocolitica) during
this third phase study showed that only K.
pneumoniae colonized the test filters for an extended
period of time (2). Aeromonas hydrophila colonized
the POU filters during a warm water (20° C, October
1984) experiment but not during a cooler water
(12°C, February 1986) period. Recently, challenge
experiments with Leg/one/fa pneumophila were
concluded. The results of these experiments indicated
that L. pneumophila apparently did not colonize the
POU test filters. However, recovery methodology for
L. pneumophila is not efficient and lacks sensitivity at
low cell concentrations, and it cannot be stated with
certainty that this organism will not colonize GAG
POU treatment units. Cool or cold water temperatures
and the presence of any disinfectant residual (total
chlorine residual) in the water probably mitigate
against colonization by this organism.
The general implications of the studies conducted in
our laboratory, as well as other published studies, are
that all GAG POU devices become generators of
bacteria due to the large surface area exposed to the
water, and due to adsorption of nutrients from the
water that bacteria are able to use for growth.
However, the potential for adverse human health
effects from ingestion of large numbers of HPC
bacteria in water appears: to be low. To date, there
have been no verified reports of waterborne illness
resulting from consumption of contaminated water
from GAG or other POU treatment devices.
Practical recommendations for users of home POU
treatment devices are as follows:
• Use the POU device only on a microbiologically
safe water supply, unless specifically
recommended by the manufacturer for other
applications as well.
• Prior to using the product water from the POU
device after a prolonged quiescent period (several
hours or overnight), run the water to waste for 30
seconds or longer at full flow. Longer flushing is
desirable after a prolonged nonuse period such as
a vacation.
« Change the filter cartridge(s) at least as frequently
as recommended by the manufacturer, or
preferably more often.
« Adhere to the manufacturer's maintenance
recommendations and specific instructions relative
to changing the filter cartridge(s).
REFERENCES
1. Geldreich, E. E., Taylor, R. H. Blannon, J. C. and
Reasoner, D. J. Bacterial colonization of point-
of-use water treatment devices. JAWWA. 77:72-
80, 1985.
2. Reasoner, D. J., Blannon, J. C. and Geldreich, E.
E. Microbiological findings with point-of-use,
third faucet devices. JAWWA. 79, October 1987.
83
-------
HEALTH STUDIES OF AEROBIC HETEROTROPHIC BACTERIA
COLONIZING GRANULAR ACTIVATED CARBON SYSTEMS
Alfred P. Dufpur
Toxicology and Microbiology Division
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Aerobic heterotrophic bacteria are ubiquitous in the
aquatic environment. Since surface waters frequently
serve as a source for potable water, it is not unusual
to find members of this large heterogeneous group in
drinking water. The types and species of
heterotrophic bacteria found in drinking water have
been described by various authors and, in general,
the organisms are gram negative, nonspore-forming
bacilli. The second column of Table 1 is a composite
list of bacteria isolated from raw and treated drinking
water. This list is for comparison purposes only, and
does not represent all those heterotrophic bacteria
that are indigenous to drinking water distribution
systems. These bacteria occur in drinking water at
densities as high as several hundred thousand per
milliliter in some cases, and they provide a constant
seed for devices used to treat potable water at its
point-of-use. Granular activated carbon (GAG)
filters, in this regard, are of special interest because
of their increased use by the general population and
because they have the capacity to adsorb bacteria
from water. Once adsorbed on the GAG filter, the
bacteria are able to multiply to even higher densities
than observed in the influent water and they, in turn,
slough off into the effluent water. The genera and
types of bacteria found on GAG filters are listed in
column 3 of Table 1, and it is obvious from this list
that the bacteria isolated from filters reflect the
distribution of bacteria observed in drinking water
systems. Gram negative bacteria are predominant,
just as in the water distribution systems. The density
of bacteria observed in GAG filter effluents has been
in the hundreds of thousands, and this frequently
represents an increase of many magnitudes above
the density of heterotrophic bacteria in the influent
water.
The amplification of heterotrophic bacteria by GAG
filters has caused some concern that these bacteria
may pose a health risk to water users. The reason for
this concern is the infrequent observation that some
of the heterotrophic bacteria isolated from drinking
water and GAG filter effluents have been associated
Table 1. Bacterial Isolates Associated with Raw and Treated
Potable Water, Granular Activated Carbon Filters,
and Nosocomial Infections
Source of Bacterial Isolates
Raw/Treated
Potable Water
Bacterial Type (Ref. 1,2)
Acinetobacter
Aeromonas
Alcaligenes
Citrobacter
Enterobacter
Klebsiella
Moraxella
Pseudomonas
Serratia
Flavobacterium
Staphyloccus
Bacillus
Achromobacter
X
X
X
X
X
X
X
X
X
X
X
X
Granular
Activated
Carbon (Ref. 2)
X
X
X
X
X
X
X
X
X
. Nosocomial
Infection
(Ref. 3-8)
X
X
X
X
X
X
with nosocomial or hospital associated infections and
illnesses. The fourth column of Table 1 is a partial list
of the types of bacteria causing infection and illness
under hospital conditions, and which were thought to
be due to contact with drinking water. The water
source, in most cases, was some type of amplifier,
which increased the densities of the organism in the
water that was linked to the patient's illness or
infection. The amplifiers were varied and included
devices such as humidifiers (3,4), dialysis machines,
disinfectant bottles, and water reservoirs for pediatric
isolettes (7). The infections ranged from septicemia
to pneumonia to peritonitis and, in some cases, have
been fatal. In most, if not all, of the cases the normal
body defense mechanisms of the patient had been
compromised in some way. The bacterial isolates in
column 4 of Table 1 were obtained from patients who
suffered from diabetes (5), bronchitis (6), had just
84
-------
undergone open-heart surgery (8), or were on a
regimen of steroid therapy (4). The patients in two
cases, were infants whose immunological systems
had not yet matured (3,7). The linkage of nosocomial
infections to drinking water has been instrumental in
part in promoting the concern about heterotrophic
bacteria in drinking water and GAC filters. The
conclusions drawn from nosocomial infections may,
however, be very misleading in that they do not
address the fact that the patients were usually
compromised in some manner. Infections of this type
are usually caused by bacteria that are avirulent or
have limited virulence, and which seize the
opportunity offered by weakened defense
mechanisms to inflict damage to the host. These
bacteria are called opportunistic pathogens, and they
seldom cause illness in healthy individuals. Although
the available evidence indicating that healthy
individuals are not at risk from these bacteria appears
to be strong, there is no empirical data supporting this
conclusion. Since the Environmental Protection
Agency (EPA) may be placed in the position of
• recommending GAC filters as an alternative form of
water treatment for removing organic chemical
hazards present in drinking water, it is necessary to
know whether or not that hazard is being replaced by
another (i.e., heterotrophic bacteria amplified by the
filter). An epidemiological study conducted by Yale
University was supported by the EPA in order to
determine if adverse health effects are associated
with GAC filter use. The results of the Yale study are
reviewed here to characterize the risks observed in
healthy populations exposed to water treated with
point-of-use GAC filters (9,10).
The study was conducted at a large military
reservation in eastern Connecticut. Families with
children were recruited for the study from the large
residental population of 800 families at this naval
base. A military base population was considered ideal
because the study participants had easy access to
cost-free medical care. This factor was critical since
a laboratory workup of clinical specimens obtained
from participants was a requirement for each illness
or infection where an individual consulted a physician.
The health reporting aspect of the study used two
approaches. The first approach used the calendar
system. Each participating family in this system was
given a health status calendar form on which they
could fill in, on a daily basis, their health status.
Symptomatology, such as vomiting, nausea, diarrhea,
high temperature, skin infections, and rashes, as well
as visits to a doctor were recorded. This was usually
done by the mother of the family or some other
responsible adult. The calendars were usually
collected every two weeks so that any participants
who failed to keep the calendar up-to-date could
be questioned and their previous two-week health
status recorded. The second approach involved
instructing each participating family to go to their
medical facility if they experienced gastrointestinal
illness or skin infections! When a gastrointestinal
illness or skin infection resulted in a clinical bacterial
isolate, the filter associated with the individual from
whom the isolate was obtained would be replaced
with a new filter and the old GAC filter bed would be
analyzed to determine if a bacterial specie could be
isolated that would match the clinical isolate.
Two types of filters were used in the study. One was
a faucet-type filter that attaches to the tap with a
special adapter. The filter was activated by turning a
valve that directed the water through the GAC bed.
The second type of filter was a bypass-type that
tapped directly to the cold water line and delivered
the water through a separate tap attached to the sink.
The source of the water serving the study population
was the Groton, Connecticut city water supply. The
water is obtained from the Great Brook watershed,
and it is sand filtered before being sent into the
distribution system.
Two bacteriological media were used to assay the
water samples during the, course of the study.
Aerobic heterotrophs were enumerated using
Standard Plate Count agar (11) and the R2A medium
of Reasoner and Geldreich (12). A summary of the
results of the bacterial monitoring is shown in Table
2.
It is immediately apparent that the R2A agar detected
much higher densities of heterotrophic bacteria than
the Standard Plate Count agar. This observation has
been noted by others, and it is thought to occur
because the two media detect different parts of the
distribution of heterotrophic bacteria. One other
interesting result of the monitoring is the high initial
densities of heterotrophs observed in the water
samples taken from the faucet filter housing units
without carbon. The cause of these high densities is
unknown. It is possible that the housing units were
contaminated before installation, however, this effect
did not show up in filters with carbon beds. The
greatest exposure to heterotrophic bacteria occurred
with the bypass filters. The exposure to heterotrophic
bacteria was, on the average, about 20 times greater
for bypass-filter users than for control groups
exposed to heterotrophic bacteria in unfiltered tap
water. The faucet-type filter effluents contained
about 12 times more heterotrophs than were found in
the tap water. However, the heterotrophic bacterial
densities in the bypass and faucet-filter effluents
exceeded that of the blank filter housing units by
factors of only six and four, respectively. Thus, the
exposure differences were not as great when the
faucet housing only group was used as a control
population, as was the case in this study. The means
of the heterotroph densities of both the faucet-type
and bypass-type filters were statistically significant
from the mean density of heterotrophs observed in
the blank housing unit effluents.
85
-------
Table 2. Comparison of Bacteria! Densities in Tap Water and GAC Filter Effluents on Standard Plate Count Agar and R2A Agar
Geometric Mean of Heterotrophic Bacteria" (Ref. 10)
Filler Type
SPC Agar
* Numbers In parentheses indicates number of samples analyzed.
R2A Agar
Bypass
Faucet
Faucet Housing (no carbon)
Tap Water
Initial
0.2 (62)
6.0(10)
85.0 (43)
6.0 (215)
Subsequent
1049 (722)
689 (559)
198 (486)
53(1,776)
Initial
0.4
9.0
98.0
11.0
Subsequent
2,042
1,035
289
92
During the course of the study only a few individuals
reported to the medical facility for treatment. No
bacterial specimens were obtained from these
patients and, therefore, no relationship between
clinical isolates and isolates from GAC could be
established.
A summary of the data collected using the calendar
questionnaire system is shown in Table 3. This
system serves a two-fold function. First, it may
detect excess illness in filter users that might not
have been observed clinically and, second, it may
detect a decrease in illness among filter users, if the
filters adsorb potential nonbacterial pathogens that
might occur in the source water. The data in the table
is given in terms of the number of symptomatic
illnesses that occurred per thousand person years of
filter usage. The total person years of usage was 230
for the faucet-type filters and 181 for those that did
not use filters. The data were analyzed statistically,
and no significant differences in the rates of
symptomatic illnesses were observed between the
two user groups and the control group. Thus, even
with the questionnaire data, it could not be shown that
excess illness could be linked to the use of GAC
filters. Conversely, there was no evidence that the
illness rate was lowered in GAC filter users.
TabloS. Comparison of Symptomatic Gl Illness Rates
Observed in Study Groups Using GAC Point-of-
Use Filters and a Control Group
Symptomatic Illness Rate/1,000
Person-Years in Groups (Ref. 10)
Symplon
vomiting
Nausea
Diarrhea
Fovor
Body Aches
Skin Rash
Infoctod Wound
Bypass Fifter
32
45
59
58
46
13
1
Faucet Filter
36
51
76
47
46
12
3
None
33
49
74
62
55
10
3
The conclusion that can be drawn from the results of
the Yale study is that point-of-use granular
activated carbon treated water containing high
densities of heterotrophic bacteria is not a risk factor
for healthy populations.
The Environmental Protection Agency is supporting
further research on the use of GAC filters and health
effects which will be conducted by Yale University.
The research reviewed here addressed exposure to
high densities of heterotrophic bacteria via the
ingestion route. Point-of-entry type filters add a
new dimension to potential exposures since all of the
water entering a home is treated. Amplified
heterotrophic bacterial densities can be disseminated
in aerosols from showerheads and, subsequently,
carried into the body via the respiratory route. The
results of the continuing study may provide some
information on the etiology of respiratory illness in the
United States.
REFERENCES
1. LeChevallier, M. W., Seidler, R. J. and Evans, T.
M. Enumeration and characterization of standard
plate count bacteria in chlorinated and raw water
samples. Appl. Environ. Microbiol. 40:922, 1980.
2. Parsons, F. Microbial flora of granular activated
carbon columns used in water treatment. In:
Wood, P. R., Jackson, D. F., Gervers, J. A.,
Waddell, D. H. and Kaplan, L., Removing
Potential Organic Carcinogens from Drinking
Water, Vol. I. Appendix A, U.S. Environmental
Protection Agency, EPA-600/2-80-130a,
Cincinnati, OH, 1980.
3. Foley, J. F., Gravelle, C. R., Englehard, W. E.
and Chin, T. D.'Y. Achromobacter Septicemia -
fatalities in prematures. Amer. J. Dis. Children.
101:279, 1961.
4. Smith, P. W. and Massanari, R. M. Room
humidifers as a source of Acinetobacter
infections. J. Amer. Med. Assoc. 237:795, 1977.
5. Berkleman, R. L., Godley, J., Weber, J. A.,
Anderson, R. L., Lerner, A. M., Peterson, N. J.
and Allen, J. R. Pseudomonas cepacia peritonitis
associated with contamination of automatic
86
-------
peritoneal dialysis machines. Ann. Int. Med.
96:456, 1982.
6. Mertz, J. J. Scharer, L and McClement, J. H. A
hospital outbreak of Klebsiella pneumonia from
inhalation therapy with contaminated aerosol
solutions. Amer. Rev. Resp. Dis. 95:454, 1967.
7. Scheldt, A., Drusin, L M., Krauss, A. N. and
Machalek, S. G. Nosocomial outbreak of resistant
Serratia in a neonatal intensive care unit. N.Y.
State J. Med. 82:1188, 1982.
8. Herman, L. G. and Fournelle. Flavobacteria: a
water-borne potential pathogen. In: Proceedings
of the Third International Congress on
Chemotherapy, Stuttgart, Germany, July 22-27,
1964.
9. Mood, E. W. and Calderon, R. L. An
epidemiological study on bacteria in point-of-
use activated carbon filters. Draft Report to
Health Effects Research Laboratory, Cincinnati,
OH for Cooperative Agreement CR-811904,
1987.
10. Galderon, R. L. An epidemiological study on the
bacteria colonizing granular activated carbon
point-of-use filters. Point-of-Use. 5:1, 1987.
11. Standard methods for the examination of water
and wastewater. Amer. Public Health Assoc.,
Washington, DC, 15th ed, 1981.
12. Reasoner, D. J. and Geldreich, E. A. A new
medium for the enumeration and subculture of
bacteria from potable water. Appl. Environ.
Microbiol. 49:1, 1985.
87
-------
ACTIVATED ALUMINA FOR POUlPOE REMOVAL OF FLUORIDE AND ARSENIC
Robert L. Lake
Water Treatment Engineers
Scottsdale, AZ 85257
INTRODUCTION
For the past 14 years, we have been treating high
fluoride and/or arsenic water with activated alumina
(AA) in POU/POE applications. Typical systems are:
1) a unit on each drinking fountain at a school; 2) a
unit for each home in a subdivision of over 300
homes; 3) a dual system for a 365-room hotel in
combination with a 185-unit trailer park; 4) a single
water tap for a small trailer park; and 5) all potable
water in a restaurant.
Each system may require a different approach with
respect to installation, monitoring, and servicing. We
are currently the certified operator for over 50 water
systems in Arizona. To monitor fluoride reduction
systems, we use a spectrophotometer and SPADNS
reagent, which is simple, inexpensive, and can be
performed in the field. Arsenic has no simple field
test, but fortunately all of our arsenic-bearing waters
also contain fluoride. We have found that fluoride
breakthrough occurs before arsenic, and by testing
for fluoride we can exchange units before arsenic
breakthrough.
ACTIVATED ALUMINA
Activated alumina is primarily a hydrated aluminum
oxide (AlaOa), which has been heated to a
temperature of 300 to 700°C (570 to 1,290°F). It is
then ground and screened to sizes ranging from
12.7-mm (0.5-in) granules to minus 325-mesh
powders. The optimum size for POU water treatment
applications has been 28 to 48 mesh (44 to 297 um).
AA particles are very irregular and porous with a very
high surface area per unit mass. AA is an ion
exchanger with the capability of exchanging both
anions and cations. Alumina chromatography has
been used successfully for separating organic as well
as inorganic compounds. Acid-treated AA is
primarily an anion exchanger with an anion selectivity
sequence as follows:
OH-, P03-3, F-, Si(OH)30-, As04-3,
rFe(CN)6J-4, AsO3-3, CrO4-2, SO4-2, Cr207-
2, NO2'1, Br-1, CI-, NO3-, MnO4-, CI04-,
CHsCOO-
The anions are listed in their decreasing order of
preference. The more preferred anions will tend to
displace on the AA those anions that are lower in the
sequence. Whereas in most organic anion
exchangers, the fluoride ion is one of the least
preferred, the reverse is true for AA. The OH" ion is
the most preferred and this has been a problem in
POU applications. High alkalinity in water will reduce
the capacity of AA to remove fluoride and arsenic.
This problem is resolved in central plant fluoride
reduction by lowering the pH of the water to be
treated to approximately 5.5. Since such pH
adjustment is not feasible with POU/POE applications,
the lower capacity must be accepted.
Both of the forms of arsenic (As04-3 arsenate and
AsO3"3 arsenite) found in water are in the anion
selective sequence preferred to the SO4"2 sulfate
ion, and activated alumina has been used
successfully as a POU/POE method for the removal
of arsenic. Test facilities in Alaska, Oregon, and New
Hampshire have determined that the activated
alumina systems were very effective and much more
economical then reverse osmosis and others tested.
PREPARATION
Activated alumina such as ALCOA F-1 or Kaiser
Chemical A-2 should be pretreated before
incorporation into a POU system. Pretreatment
consists of a thorough backwash and acid wash.
Aqueous state AA will have a pH of 9 to 10 and
should be acid washed with a pH 2 sulfuric acid
solution to a pH of 5 to 6. Backwashing is essential to
remove the dust and fines in the material. If they are
not removed, the alumina has a tendency to cement
and destroy its adsorption capability. Activated
alumina should always be added to an excess of
water to dispel! the heat generated, which will also
contribute to cementing of the material.
CHEMISTRY
A simplified explanation of the absorption reactions
(A = activated alumina particles):
88
-------
1. Acid pretreatment
A»H2O + H2SO4
(aqueous state)
A»H2SO4 + H20
(acid state)
2. Absorption or ion exchange
A«H2SO4 + 2NaF -»• A«2HF + Na2S04
3. Regeneration
A»2HF + 4NaOH -> A»2NaOH + 2NaF + 2H20
4. Neutralization
A«2NaOH + 2H2SO4 -»• A«H2S04 + Na2S04 + 2H20
(basic state)
REGENERATION
Several methods of regeneration of activated alumina
have been explored over the past 15 years.
Inconsistent results from regenerated media were a
problem in many instances.
A regeneration method that gives uniformly good
results has been recently developed. The process
entails allowing complete replacement of the sorbate
(Fl, AsO4-3, etc.) with OH" and complete
neutralization of the basic state AA with acid. Both of
these reactions are time dependent, and allowing
sufficient time for completion is the key to the
success of this method. It has also been possible to
greatly reduce chemical costs by recirculating the
NaOH regenerant. It appears that the failure to
completely neutralize the AA accounts for most of the
loss of capacity on the media. In larger systems,
where pH control is used, the continuing injection of
acid is neutralizing the media well into the treatment
cycle.
The steps in regeneration are:
1. Backwash - A backwash rate of 19.5 m3/m2/h (8
gpm/sq ft) gives 100 percent expansion of the
alumina bed, and the excess material is contained
in an additional vessel (three to four minutes or
until clear).
2. Upflow regeneration - One percent NaOH (by
weight) is injected while the AA bed is still
expanded. Regenerant is recycled for 30 minutes
at 7.3 m3/m2/h (3 gpm/sq ft).
3. Soak - Up to three hours.
4. Neutralization - Rinse with (pH = 2) dilute sulfuric
acid until pH of alumina bed drops to 4.5-5.
5. Soak - 24 hours or until pH of alumina returns to
10 + due to pore migration within the alumina
particles.
Repeat steps 4 and 5 as needed to complete
neutralizing of alumina bed.
6. Final backwash and refill.
CONCLUSION
Although the EPA has determined that POU
technology is unacceptable for water treatment, we
have successfully protected thousands of people from
the damage caused by fluoride and arsenic in their
drinking water. We have shown that POU treatment is
a cost effective method of treatment in situations
where the cost of central treatment would be
prohibitive.
BIBLIOGRAPHY
1. Bellen, G.E., Anderson, M. and Gottler R.A.
Defluoridation of Drinking Water in Small
Communities. EPA/600/2-85/112, Cincinnati, OH,
January 1986.
2. Bellack, E. 1971. Arsenic Removal from Potable
Water. JAWWA 63:7:454, July 1971.
3. Clifford, D., Matson J., and Kennedy, R. Activated
Alumina: Rediscovered "Adsorbent" for Fluoride,
Humic Acids and Silica. Industrial Water
Engineering, December 1978.
4. Harman, J.A. and Kalichman, S.G. Defluoridation of
Drinking Water in Southern California. JAWWA
57:2:245, February 1965.
5. Kubli, H. On the Separation of Anions by
Adsorption on Alumina. Helvetia Chimica Acta.
Switzerland 3:453, 1947.
6. Maier, F.J. Defluoridation of Municipal Water
Supplies. JAWWA 45:8:879, August 1953.
7. Savinelli, E.A. and Black, A.P. Defluoridation of
Water with Activated Alumina. JAWWA 50:: 1: 33,
January 1958.
8. Singh, G. and Clifford, D.A. The Equilibrium
Fluoride Capacity of Activated Alumina. EPA-
600/S2-81-082, Cincinnati, OH, July 1981.
89
-------
MODELLING POINT-OF-ENTRY RADON REMOVAL BYGAC
Jerry D. Lowry
University of Maine
Orono, ME 04469
Sylvia B. Lowry
Lowry Engineering, Inc.
Thorndike, ME 04986
The health implications of airborne 222Rn in
households are well documented (1-8), as is the
importance of elevated 222Rn in the water supply and
how it contributes to the airborne 222Rn (1,2,9-19).
The feasibility of removing 222Rn from household and
small system water supplies with granular activated
carbon (GAG) or aeration devices has been reported
by various researchers (1,20-23) and a detailed
report for public water supplies has recently been
prepared for the U.S. Environmental Protection
Agency (EPA) (24).
In response to the growing concern about airborne
and waterborne 222Rn, research was initiated in 1980
at the University of Maine to identify technologies to
remove 222Rn from ground water (17,22). Aeration
and GAG were found to be potentially cost effective
treatment processes for point-of-entry applications.
Laboratory testing of full-scale point-of-entry GAG
and diffused bubble aeration units showed that both
methods were effective, but GAG appeared to have
the most promise for household applications. That
study and a previous one documented that an
adsorption/decay steady state is established with a
GAG bed, allowing it to be virtually maintenance free
for an indefinite but long period of time. These studies
led to laboratory research to model the GAG process
for municipal application, field research to develop a
design model for point-of-entry application, and
installation and monitoring of the GAG technology in
more than 100 households throughout the U.S. This
article reports on the findings of the second and third
aspects of the preceding research.
DEVELOPMENT OF GAC MODEL
Previous research indicated that the adsorption/decay
steady state performance could be modelled by first-
order kinetics, allowing the use of the following
equation to describe and predict removal:
where:
Ct = the 222Rn cone, at time t, pCi/l
GO = the initial 222Rn cone., pCi/l
Kss = the steady state adsorption/decay constant,
hr1, and
t = the empty bed detention time (EBDT), hr
This is logical since at steady state, the adsorptive
removal equals the decay of adsorbed 222Rn at all
points in the bed, and decay kinetics are first order.
The actual achievement of the adsorption/decay
steady state is quite complex. It involves nonsteady
state adsorption kinetics, which dominate the early
period, followed by an increasingly significant decay
phenomenon set up by 222Rn and its short-lived
progeny. The result of these processes is the
establishment of the steady state distribution of
222Rn and short-lived progeny on the GAC.
Although the establishment of the steady state is
complex, the performance after steady state is
reached can be accurately described by the simple
steady state model.
The use of the average empty bed detention time (t)
is appropriate since the GAC bed at steady state acts
as a decay storage device. This means that for all
practical purposes, the normal intermittent diurnal flow
experienced in a typical household is not important, in
terms of its effect on the steady state performance.
An analogy to a chromatographic column can be
made, in which the 222Rn travels along the bed at a
relatively slow rate compared to the water, aiad
decays down to the effluent value. The bed is simply
a concentrating device that stores 222Rn and is
equivalent to a plug flow storage tank having a much
greater liquid detention time. For example, a GAC bed
giving a 99 percent reduction is equivalent to an ideal
plug flow decay storage tank having a detention time
90
-------
of 25.3 days. Thus, t should be calculated for the
water volume used over two to three weeks because
this is the period to which the GAG bed is
responding. In a typical household the three week
flow average does not vary significantly, except
perhaps seasonally in a gradual manner.
To accurately document this model and to test the
relative effectiveness of several different GAG
products, a field study was designed to measure the
steady state adsorption/decay constant (Kss). A
Maine household that had an extremely high 222Rn
concentration in its ground water supply was selected
to demonstrate the general applicability of the model
across a wide range of 222Rn levels. The
performance of three GAG products (Table 1) was
examined using a modified commercial treatment unit
(Figure 1). Slotted [(0.30 mm (0.01 inch)] laterals
were installed at depths corresponding to 0.007,
0.014, 0.03, 0.04, 0.06 m3 (0.25, 0.5, 1, 1.5, and 2
cu ft) bed volumes to obtain depth samples. The total
bed volume was 0.07 m3 (2.5 cu ft).
Table 1.
Carbon
A
B
C
D
Summary of Adsorption/Decay Design Constants
Manufacturer Type Kss (hr1)
American Norit
1CI Americas*
Calgon
Barneby Cheney
Peat (8X20)
HD4000 (12X40)
F-400 (12X40)
299 or 1002
1.35
2.09
1.53
3.02
Now manufactured by American Norit.
Water samples for 222Rn analysis were taken directly
from sampling valves with a 10 ml (0.34 oz) syringe,
which was subsequently discharged directly into a
previously prepared liquid scintillation vial containing 5
ml (0.17 oz) of fluor. The vial was immediately
capped, and mailed to the laboratory for counting.
The basic counting procedure used was one
described by Prichard and Geseli (25), except that a
mineral oil- rather than toluene-based fluor was
utilized because of postal regulations. Precision of the
222Rn analysis is a function of the level of 222Rn
present, the counting time, and the time elapsed
between sampling and counting. Typical levels of
uncertainty (2-sigma) for this study are given in
Table 2.
The water use at the household was monitored by a
standard 16 mm (5/8 in) meter and totalizer readings
were taken daily. Water temperature ranged between
6 and 10°C (43 and 50 °F) throughout the study.
The various GAC products were tested sequentially,
each by the same method. The virgin carbon was
placed in the pressure vessel over a gravel support
and commissioned after a backwashing period of 15
to 30 minutes to remove fines. The GAC bed
Figure 1. Experimental GAC vessel for K.sa determination.
Raw Water *V/ ^ **S 2.50 cu ft scnple
Raw Voter
Freeboard
Riser Tube
GflC Bed
Basket Distributor
Support Gravel
lor
>
— 1
=^
i~
b
S^ Control Valvi
1 O.25 cu ft sample
| 0.50 cu ft sample
| 1.00 cu ft sample
] 1.50 cu ft sanpla
] 2.00 cu ft sample
Table 2. Typical Levels of Uncertainty (Counting) for This
Study
222Rn Concentration (pCi/1)
800,000
300,000
40,000
2,000
1,000
500
100
60
Uncertainty (percent)
0.5
0.5
1.0
4.5
6.0
15.0
25.0
45.0
remained in service to allow a steady state to be
achieved (approximately three weeks) and was
monitored for an additional three to four week period.
Fourteen sets of samples (all ports) were taken to
determine Kss.
Typical examples of the results of the field testing are
illustrated by Figures 2 through 4 for GAC B. Figures
2 and 3 show results of the depth removal of 222Rn
and the establishment of the adsorption/decay steady
state. The exact reason for the elevated point for
each depth on day 42 was not known, but suspected
to be caused by desorption brought about by possible
extreme raw water 222Rn variation that was not
documented by sampling the previous week. This
particular well is subject to such variations, and
previous monitoring has documented that the 222Rn
91
-------
Figure 2. Performance for the top portion of the GAG bed for GAC B.
800
O Raw Water
A 0.25 cu ft
a 0.5 cu ft
* I.0 cu ft
10
— ?"*"' 7"" ^st
15 20 25 30
TIME, days
Figure 3. Performance for the bottom portion of the GAC bed for GAC B.
15
14
13
12
II
10
9
8
7
6
5
A
3
2
1
O 1.0 cu ft
A 1.50 cu ft
a 2.00 cu ft
o CIDED—ago-—
10 15 20 25 30
TIME, dcys
35 40
45
50
92
-------
Figure 4. First-order steady-state adsorption-decay relation for GAC B, with 95 percent confidence limits indicated.
14
13
12
11
10
9
8
7
6
Kss = - 2.09/hr
30
90 120 150
EBDT. minutes
180 210 240
Figure 5. Contrast of the steady-state adsorption-decay relation for test carbons A, B, and C with GAC D.
100*
50.
20.
10.-
Hi
2.
1
\
o Carbon R
A Carbon B
a Carbon C
Carbon D
0 X 60 90 120 150 180 210 240
EBDT. minutes
93
-------
variation over a period as short as several days can
be from 150,000 to over 2,000,000 pCi/l. For bed
volumes of greater than 42 I (1.5 cu ft), these
variations in depth are not significant in relation to the
raw water concentrations because there is enough
GAG to provide adequate dampening. They are
apparent in this case due to the extremely high
average raw water 222Rn.
A semi-ln plot of bulk solution 222Rn (In vs. EBDT)
yields a linear relationship with a slope equal to Kss.
This relationship is illustrated in Figure 5 for one of
the carbons tested.
The analysis of variance (ANOVA) for the data is
summarized in Table 3. Although the deviations from
regression were significant statistically, they were
extremely small compared to the variation explained
by regression. This fact is reflected by the relatively
narrow confidence limits (95 percent) around the least
squares regression line in Figure 4.
Tablo 3. Summary of Analysis of Variance for GAC B
Sourco ol Variation df SS MS F
Among Groups (EBOT)
Linear Regression
Deviations from Regression
Wiiliin Groups (ports)
5
1
4
78
534.7
530.6
4.1
10.1
106.9
530.6
1.0
0.13
829*
515*
8.0"
Significant at 0.001 level.
The performance of carbons A, B, and C is
contrasted to that of carbon D, which has been
determined by other research (23,26) to have the
highest Kss value tested to date. A summary of the
KSS values for these carbons is given in Table 1. The
ranking of these carbons bears little relation to how
they performed previously by isotherm testing, and
shows that isotherms are not indicators of how a
carbon will perform at steady state (17). This group of
four carbons contains the best and worst carbons
with respect to 222Rn removal; it is clear that the
type of GAC selected has significant bearing upon the
performance achieved. For example at 99 percent
removal, the required carbon volume for GAC D is 50
percent of that needed using carbon A. For equal bed
volumes, carbon D achieves a 99 percent reduction
compared to an 88 percent reduction with carbon A. It
is interesting to note that the bulk densities for
carbons A and D are approximately 288 and 513
kg/m3 (18 and 32 Ib/cu ft), respectively, and that on a
mass, rather than volume, basis all GAC types tested
are much closer in performance. Because the number
and size of vessels required is determined by the
volumetric performance, this has little practical
significance. Of more importance is the probable
positive influence of decreasing particle size;
however, no studies have documented the magnitude
of this factor for the 222Rn steady state.
Because of the small flow treated and the relatively
narrow range of bed sizes required to cover the entire
range of 222Rn encountered in point-of-entry
applications, the long EBDT is easily satisfied by
commercially available pressure vessels. A range of
bed volumes from 28 to 85 I (1 to 3 cu ft) will remove
in excess of 99 percent of the 222Rn from any
household ground water supply. In contrast, the
relatively long EBDT is of importance with larger
design flows in municipal applications. Compared to
an EBDT of approximately 15 minutes for organics
removal, the required EBDT for 222Rn removal is
quite long. Although the GAC will last indefinitely in
the 222Rn application, the initial high capital cost for
the GAC makes aeration an attractive alternative for
larger water systems.
FIELD EXPERIENCE WITH GAC
TREATMENT
Since 1981, GAC units have been installed in a
significant number of households to remove 222Rn
from the water supply and thereby lower the airborne
222Rn levels in homes. The current number of units
that exist is estimated to be in excess of 500.
Approximately 100 units have been installed and
monitored as a part of a data base for future research
on aspects other than simple removal, such as the
resulting gamma exposure rate from 214Pb and 214Bi
and the long term buildup of 210Pb. In each of these
installations the GAC type, the GAC quantity, and the
installation date are known. In some of the
installations the water use is known and the radium
and uranium content of the raw water has been
documented. This data base is unique in that the
units are installed over a widespread area (Canada
and 12 states in the U.S) and contain elements that
represent the longest operating GAC units for 222Rn.
In addition, they are installed on water supplies that
cover the entire documented range of 222Rn in the
world -- from less than 1,000 pCi/l to in excess of
1,000,000 pCi/l.
Although four GAC types have been used in these
units, over 85 percent contain carbon D and 10
percent contain carbon C. A summary of the steady
state performance of all installations that are routinely
monitored is presented in Figure 6. With the
exception of three units, the performance level in field
installations is very high. Eighty percent of all units
are in the 0.05 m3 (1.7 cu ft) category, with remaining
units ranging from 0.03 to 0.08 rr>3 (1 to 3 cu ft). The
average removal of 222Rn for all units is 96.2
percent.
Elimination of three known prematurely fouled or
malfunctioning units and the units containing carbon
C, yields the histogram summarized in Figure 7. For
these units, the average removal of 222Rn is 98.9
percent, and it is uncommon to monitor a unit and
find less than 99 percent removal. Although the real
94
-------
Figure 6. Histogram of the steady-state performance of 66 GAG treatment units.
22
20
18
£ 16
I—I
5 14
& 12
10
8
6
4
2
o:
10 20 30 40 50 60 70 80
REMOVRL, percent
90 100
Figure 7. Histogram of the steady-state performance of properly operating GAG units conatining GAG 0.
£4
22
20
18
{2 16
§ 14
& 12
a:
$ 10
z 8
4
2
n
-
• •
•
-
•
•
•
•
m
t it til t 11
91 92 93 94 95 96 97 98 99
REMOVflL. percent
100
95
-------
value of theso units lies in future studies involving
222Rn progeny buildup, these data demonstrate the
high degree of removal possible with properly
designed and installed systems.
The reason for the three poorly performing units has
not been determined. Preliminary investigation
appears to indicate water quality as a cause, rather
than the GAG. Although only a few percent of the
existing GAC units exhibit this phenomenon, the
reason for this possible premature fouling should be
determined.
GAC vs. AERATION
It has been documented that aeration is a feasible
method for 222Rn removal in point-of-entry
applications (22,27). But a number of factors have
kept it from becoming as popular a method as GAC
treatment:
• Aeration is performed at atmospheric pressure and,
therefore requires re-pressurization of the water
supply.
• The initial cost of aeration systems designed for
222Rn removal is relatively high, partly due to the
re-pressurization requirement. The installed cost
ranges from $1,700 to over $4,000, compared to
approximately $650 to $1,200 for GAC.
* Several of the currently available aeration methods
have limited removal capabilities. A novel but costly
spray aeration system was developed by the Maine
Department of Human Services, Division of Health
Engineering, and six units are operating in the field.
These units achieve 90 to 95 percent removal but
may be limited to wells containing only 10,000 to
20,000 pCi/l if, for example, the future U.S. EPA
maximum contaminant level (MCL) for 222Rn is set
at 1,000 pCi/l. Although the MCL would apply only
to public water supplies, the real 222Rn issue is in
private household supplies - in many cases
lending institutions are already requiring removal.
These institutions tend to use the MCLs as
guidance for supplies as well.
Packed-tower aeration systems are confined by
the ceiling height in the cellar or living area and are
therefore limited to about 85 to 90 percent
removal. Two such devices are currently available
for household use. These units cost approximately
$3,000.
A multi-staged diffused bubble aeration system
developed for organics removal (28) has been
tested on a supply that contains 250,000 pCi/I and
removed 222Rn to below detection, for virtually
100 percent removal. A less expensive version
designed specifically for 222Rn removal has been
developed that will achieve 99 percent removal. It
will cost approximately $1,700.
• Aeration methods require significant O&M
compared to GAC, which increases the cost
differential over the long term.
Aeration methods have two advantages to be
considered for point-of-entry applications, i.e., they
avoid the elevated gamma exposure rate and long
term buildup of 210Pb that is and may be,
respectively, associated with GAC beds. These topics
are currently the subjects of on-going research on
GAC treatment and are beyond the scope of this
paper; however, economical water jacket shielding
and proper location can minimize increased gamma
exposure. The buildup of 210Pb in these applications
is not well documented, but it could be a concern
from a regulatory point of view. Future documentation
and research on this subject will determine the extent
of and solutions to these problems.
CONCLUSIONS
• A first order model accurately describes the
adsorption/decay steady state removal of 222Rn by
GAC.
• A single design constant, Kss, can be used to rank
a given GAC type for 222Rn removal. The ranking
of a carbon for steady state performance does not
appear to be related to its ranking according to an
adsorption isotherm.
• There is a significant range of the design constant,
Kss, f°r the carbons tested to date, making the
selection of the correct GAC important. This is
especially true for small public water supply
application, where economics are morje sensitive to
vessel size.
• More than 99 percent reduction of 222Rn is
possible with an effective GAC.
• The progeny of 222Rn make it important to
consider the location of and protective shielding for
a GAC bed, to minimize increased gamma
exposure over background levels.
• Limited data indicate a possible premature fouling
of approximately three to five percent of existing
GAC units. The reason for this decreased removal
in these installations is unknown and should be
investigated.
ACKNOWLEDGEMENT
This paper is largely reproduced from a previously
published paper in the Journal AWWA (October
1987).
96
-------
Primary funding for the field research to determine
the steady state design model was provided by the
Office of Research and Development, U.S.
Environmental Protection Agency (EPA), under Grant
No. R8108290. The field monitoring data was
supplied by Lowry Engineering, Inc.
REFERENCES
1. Castren, O. The contribution of bored wells to
respiratory radon daughter exposure in Finland.
Technical Report, Institute of Radiation
Protection, Helsinki, Finland, 1977.
2. Hess, C. T. et al. Radon in potable water supplies
in Maine: the geology, hydrology, physics, and
health effects. Completion Report, Land and
Water Resources Center, University of Maine and
the Office of Water Research and Technology,
U.S. Department of the Interior, Washington, DC,
1979.
3. Evans, R. D. et al. Estimate of risk from
environmental exposure to radon-222 and its
decay products. Nature. Vol. 290, 1981, pp. 98-
100.
4. Harley, N. H. Editorial-radon and lung cancer
in mines and homes. New England Journal of
Medicine. Vol. 310, No. 23, June 7, 1984, pp.
1525-1526.
5. National Academy of Sciences Committee on
Biological Effects of Ionizing Radiation (BEIR).
The effects on populations of exposure to low
levels of ionizing radiation. National Academy
Press, Washington, DC, 1980.
6. U.S. Environmental Protection Agency. A citizens
guide to radon: what it is and what to do about it.
August, 1986.
7. U.S. Environmental Protection Agency. Radon
reduction methods: a homeowners guide. U.S.
EPA OPA-86-005, August, 1986.
8. Cothern, R. C. Estimating the health risks of
Radon in drinking water. Journal of the American
Water Works Asso. Vol. 79, No. 4, April, 1987.
9. Castren, O. et al. High natural radioactivity of
bored wells as a radiation hygienic problem in
Finland. Presented at the 1977 International
Radiation Protection Association Fourth
International Congress, Paris, France.
10. Duncan, D. L, Gesell, T. F. and Johnson, R. H.
Radon-222 in potable water. Proceedings of the
10th Midyear Topical Symposium: Natural
Radioactivity in Man's Environment, Health
Physics Society, Saratoga Springs, NY, October,
1976.
11. Hess, C. T. et al. Investigation of Rn-222, Ra-
226, and U in air and groundwater of Maine.
Completion Report B-017-ME, Land and Water
Resources Center, University of Maine and the
Office of Water Research and Technology, U.S.
Department of the Interior, Washington, DC,
1981.
12. Partridge, J. E., Horton, T. R. and Sensintaffer, E.
L. A study of radon-222 released during typical
household activities. ORP/EERF-79-1, U.S.
Environmental Protection Agency, Office of
Radiation Programs, Eastern Environmental
Radiation Facility, Montgomery, AL, March, 1979.
13. Weiffenbach, C. V. Radon in air and water: health
risks and control measures. Technology Transfer
Report, Land and Water Resources Center,
University of Maine, June, 1986.
14. Horton, T. R. Methods and results of EPA's study
of radon in drinking water. EPA 520/5-83-027,
U.S. Environmental Protection Agency, Office of
Radiation Protection Programs-Eastern
Environmental Radiation Facility, Montgomery,
AL, December, 1983.
15. Aldrich, L. K., Sasser, M. K. and Conners, D. A.
Evaluations of radon concentrations in North
Carolina ground water supplies. Department of
Human Resources Division of Facility Services-
Radiation Protection Branch, Raleigh, NC,
January, 1975.
16. Lowry, J. D. and Moreau, E. Removal of extreme
radon and uranium from a water supply.
Proceedings of the 1986 National Conference on
Environmental Engineering, Environmental
Engineering Division, American Society of Civil
Engineers, Cincinnati, OH, July, 1986.
17. Lowry, J. D. and Brandow, J. E. Removal of
radon from water supplies. Journal of
Environmental Engineering, Environmental
Engineering Division, American Society of Civil
Engineers. Vol. 111, No. 4, August, 1985.
18. Lowry, J. D. Radon at home. Civil Engineering.
Vol. 57, No. 2, February, 1987.
19. Prichard, H. M. The transfer of radon from
domestic water to indoor air. Journal of the
American Water Works Asso. Vol. 79, No. 4,
April, 1987.
20. Hoather, R. C. and Rackham, R. F. Some
observations on radon in waters and its removal
by aeration. Proceedings of the Institution of Civil
Engineers, Great George Street, London, S.W.1.,
December 7, 1962, pp. 13-22.
97
-------
21. Smith, B. M. et al. Natural radioactivity in ground
water supplies in Maine and New Hampshire.
Journal of the American Water Works
Association. Vol. 53, No. 1, January, 1961, pp.
75-88.
22. Lowry, J. D. et al. Point-of-entry removal of
radon from drinking water. Journal of the
American Water Works Asso. Vol. 79, No. 4,
April, 1987.
23. Lowry, J. D. Extreme levels of Rn-222 and U in
a private water supply. Proceedings of the
Conference on Radon, Radium, and Other
Radioactivity in Ground Water: Hydrologic Impact
and Application to Indoor Airborne Contamination,
National Water Well Asso., Somerset, NJ, April
7-9, 1987.
24. Malcolm Pirnie, Inc. Technologies and costs for
the removal of radon from potable water supplies.
Draft of report to EPA under Contract No. 68-
01-6989, January 6, 1987.
25. Pritchard, H. M. and Gesell, T. F. Rapid
measurements of radon-222 concentrations in
water with a commercial liquid scintillation
counter. Health Physics. Vol. 33, 1977, pp. 577-
581.
26. Pinnette, J. M.S. thesis in Civil Engineering,
University of Maine, Orono, ME, 1985.
27. Hinkley, W. W. Experimental water treatment for a
drilled well with the world's highest known
radon-222 levels. Maine Oept. of Human
Services, Div. of Health Engineering, State
House, Augusta, ME, 1982.
28. Lowry, J. D. and Lowry, S. B. Restoration of
gasoline-contaminated household water supplies
to drinking water quality. Proceedings of the
Eastern Regional Ground Water Conference,
National Water Well Asso., Portland, ME, July,
1985.
98
-------
POINT-OF-ENTRY ACTIVATED CARBON TREATMENT LAKE CARMEL - PUTNAM COUNTY
George A. Stasko
Bureau of Public Water Supply Protection
NY State Department of Health
Albany, NY 12237
Lake Carmel is a small lake located approximately 80
km (50 mi) north of New York City. The hills
surrounding the lake were extensively developed in
the 1930s with seasonal residences. Since then most
of these residences have been converted to year
round housing. Lot sizes range from 370 to 1,110 m2
(4,000 to 12,000 sq ft) and each lot has a well and
septic system (see Figure 1).
In the 1970s, some of the residents complained about
petroleum odors in their water. When individual action
did not bring the desired results they formed a
Citizens' Advisory Committee. The committee
enlisted the aid of their legislators with the result that
investigations of the area were made by the
Department of Transportation, the State Health
Department, and the Putnam County Health
Department.
INVESTIGATION RESULTS
The residents believed that the ground water
contamination came from a petroleum spill or from a
waste site. However, as a result of the investigations,
it was concluded that the residents had contaminated
their own water supplies by localized petroleum leaks
and spills, and by the chemicals flushed into their
septic systems. In addition, it was discovered that
some wells had elevated nitrate levels and high
coliform counts.
BACTERIOLOGICAL QUALITY
Bacteriological test results indicated widespread
bacterial contamination. Approximately 40 percent of
the samples were above the standard of 1 coliform
organism per 100 ml (0.3 organism per oz). Counts
as high as 245 coliform organisms per 100 ml (72 per
oz) were found (see Table 1).
Coliform levels varied considerably from well to well
and in the same well over a period of time. In the
Fitzsimmons well} 245 coliform organisms per 100 ml
(72 per oz) were present on February 4, 1982. Two
weeks later the count was 3 per 100 ml (0.9 per oz)
and four weeks after the second sample the count
was 134 per 100 ml (39 per oz). In the Saver well on
February 18, 1982 the count was 189 coliform
organisms per 100 ml (56 per oz), and on March 23,
1982 the count was less than 1 coliform organism per
100 ml (0.3 per oz). Because of this variability, the
health department concluded that all wells would be
subject to bacterial contamination at some point in
time and therefore the treatment system must include
disinfection.
ORGANIC CHEMICAL QUALITY
Volatile Organic Chemicals
Detection of volatile organic chemicals varied from
well to well, and with the laboratory, performing the
analyses. All the highest results were obtained from
one laboratory. Other laboratories detected the same
chemicals but at lower levels. Sampling for the three
studies was not coordinated, leaving gaps in the data.
Benzene, toluene, and xylene were detected at high
levels confirming residents' complaints of petroleum
tastes and odors. In addition, several solvents were
found including carbon tetrachloride,
tetrachloroethylene, trichloroethylene, and 1,1,1-
trichloroethane. Table 2 summarizes the results of
tests for volatile organic chemicals.
Base/Neutral Chemicals
Of the base/neutral organic chemicals tested, only bis
(2-ethylhexyl) phthalate was found in detectable
quantities (see Table 3).
Pesticides/Herbicides
No pesticides or herbicides were detected. For the
ch'emicals tested see Table 4.
INORGANIC CHEMICAL QUALITY
The average inorganic water quality of the wells
tested was soft and corrosive, but no chemicals
exceeded standards. Both sodium and nitrates were
significantly higher than background levels in area
ground water, indicating that leachate from the septic
systems was reaching the wells. Only in one instance
did nitrate exceed the standard of 10 mg/l (10 ppm). It
99
-------
Figure 1. Location map.
LOCATION
MAP
N.Y.S.D.PW BUREAU OF PROGRAMMING
U.S.G.S. LAKE CARMEL QUADRANGLE - 7'/2 MIN. SERIES
SCALE' I" = 2000'
100
-------
Table 1. Bacteriological Quality of the Ground Water in the
Lake Camel Project Area*
Table 2. Organic Chemical Quality of the Ground Water in
the Lake Carmel Area (Volatile Compounds)
Name Collection
Backer
Baisley
Bao
Behnken
Cuomo
Fitzsimmons
Fitzsimmons
Fitzsimmons
Greco
Laconte
Laconte
Lawton
MacNeil
Madden
Mahoney
Micciche
Nappi
Placek
Porrino
Prisco
Sauer
Sauer
Sheridan
Sutor
Zasso
Zasso
3124182
4/15/82
4/27/82
2/18/82
4/21/82
2/4/82
2/18/82
3/16/82
2/4/82
2/4/82
3/16/82
3/24/82
3/23/82
4/15/82
2/4/82
4/21/82
4/21/82
1/5/82
3/23/82
4/15/82
2/18/82
3/23/82
2/4/82
2/4/82
3/18/82
4/21/82
Total Coliform Above Health
per 100 ml Limits"
57
<1
<1
6
<1
245
3
134
<1
91
17
92
<1
<1
<1
<1
<1
38
<1
<1
189
<1
<1
<1
<1
<1
Yes
No
No
Yes
No
Yes
Yes
Yes
No
Yes
Yes
Yes
No
No
No
No
No
Yes
No
No
Yes
No
No
No
Yes
No
* All samples were collected by the Putnam County Department'
of Health and analyzed by Sanitary Science & Laboratories,
Inc., Newburgh, New York.
" The NYS Dept. of Health limits total coliforms to 1/100 ml.
Volatile Compound
Acrdein
Acrylonitrile
Benzene
Bromodichlorometriane
Bromoform
Bromomethane
carbon Tetrachloride
Chlorobenzene
Chlorodibromomethane
Chloroethane
2-Chloroethyl vinyl ether
Chloroform
Chloromethane
Dichlorodifluoromethane
1,1-Dichloroethane
1 ,2-Oichloroethane
1,1-Dichloroethylene
Trans-1 ,2-Dichloroethylene
1 ,2-Dichloropropane
1,3-Dichloropropane
Ethylbenzene
Methylene chloride
1 ,1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,1 ,1 -Trichloroethane
1 ,1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Vinyl chloride
Xylene
High Value
(WM
ND
ND
820
4
ND
ND
5.9
ND
ND
ND
ND
14
ND
ND
ND
ND
ND
ND
ND
ND
110
ND
ND
560
320
14.2
ND
13
2
2
490
Mean Value*
±Std. Dev. (jig/l)
ND
ND
67.7 ±176.3 (21)
2.90 + 1.15(3)
ND
ND
4.50 + 1.98 (5)
ND
ND
ND
ND
5.03 ±4.81 (8)
ND
ND
ND
ND
ND
ND
ND
ND
38.3 ±62.1 (3)
ND
ND
100.2 + 215.3(11)
8.5 + 88.0(13)
6.91 ±5.10 (9)
ND
5.48 + 3.86(10)
1.25 ±0.50 (4)
2±0(2)
108.4 ±178.1 (7)
was decided that no treatment was needed to remove
inorganic chemicals. See Table 5 for a summary of
inorganic chemical test results.
Alternate Solutions
As a result of the investigations, the health
department recommended that a public water system
be provided for the affected area. An engineering
consultant was hired to conduct a feasibility study.
The study found that a public water system would
cost in excess of $1,200 per year per homeowner.
Expensive rock cut for the distribution system was the
main reason why the public water system would be
so costly. Because the public water system was
impractical, it was decided to study the feasibility of a
point-of-entry solution.
The Citizens' Advisory Committee was able to secure
an imminent threat grant from the U.S. Department of
Housing and Urban Development. They provided
$165,000 to design, purchase, and install point-of-
entry treatment systems. Because the Citizens'
Advisory Committee could not receive the money, it
ND - Not detectable.
* Numbers in parentheses indicate sample size.
was given to the town of Kent, which in turn hired an
engineering firm to design the treatment systems.
Water Treatment Systems
The health department worked closely with the
consultants to develop a water treatment system that
would adequately treat the water and would satisfy all
regulatory concerns. The design that resulted evolved
from design criteria developed by the health
department's Ad Hoc Committee on Removal of
Synthetic Organic Chemicals from Drinking Water,
and published in an interim report, entitled Point-of-
Use Activated Carbon Treatment Systems.
A schematic of the treatment system is shown in
Figure 2. The treatment system consists of:
• A raw water tap located immediately after the
homeowner's pressure tank used to collect
untreated water samples.
101
-------
Tablo 3. Organic Chemical Quality of the Ground Water in
the Lake Carmel Area (Base/Neutral Compounds)
High Value Mean Value*
Baso/Noutral Compound
Aconaphthene
Aconaphthylone
Anthracene
Bonzo (a) anthracene
Bonzo (b) fluoroanthene
Bonzo (k) fluoroanthene
Bonzo (a) pyreno
Bonzo (Q.h.O perylone
Bonzidino
Bis (2-chtoroethyl) ether
Bis (2-chtoroelhoxy) methane
Bis (2-elhylhexyl) phthalate
Bis (2-chtoroisopropyl) ether
4-BromophenyI phenyl ether
Butylbonzylphthalate
2-Chtoronaphthatene
4-Chtorophonylphenylether
Chrysone
Dibonzo (a,h) anthracene
Di-N-Butylphthalate
1 ,2-Dtchkx obonzeno
1 ,3-Dtchlorobenzene
1 ,4-Diehtorobenzeno
3.3'-Dichtorobenzidine
Diothylphthalate
DimothylphthafatQ
2.4-Dinitrotoluono
2,6-Dinitrotoluone
Di-octyl-phthalate
1 ,2-Diphonylhydrazine
Fluoroantheno
Fluorono
Hcxachlorobenzcne
Hoxchlorobutadiene
Hoxchtofoe thane
Hoxachtorocyclopentadiene
IncJono (i ,2,3-cd) pyrene
Isophorone
Naphthalono
Nitrobonzone
N-Nitrosodimothylamine
N-Nit/osodi-N-propylamine
N-Nitrosodiphenylamine
Phonan throne
Pyrono
1 ,2.4-Trichlorobenzene
2,3.7,8-Tetrachlorodibenzo-
p-dtoxm
NO • Not detectable.
(ug/l) ±Std.
ND
ND
ND
ND
ND
ND
NO
ND
ND
ND
ND
Dev. (ug/l)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
35 23 + 17.0(2)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
* Numbers in parentheses indicate sample size.
• A water meter to measure the amount
processed.
of water
Table 4. Organic Chemical Quality of the Ground Water in
the Lake Carmel Area
High Value Mean Value
Pesticide/Herbicide (ug/l) ± Std. Dev. (ug/l)
Aldrin ND ND
-BHC
-BHC
-BHC
-BHC
Chlordane ND ND
Dieldrin ND ND
-Endosulfan
-Endosulfan
Endosulfan sulfate ND ND
Endrin ND ND
Endrin aldehyde ND ND
Heptachlor ND ND
Heptachlor Epoxide ND ND
4,4'-DDT ND ND
4,4'-DDE ND ND
4,4'-DDD ND ND
PCB 1016 ND ND
PCB 1221 ND ND
PCB 1232 ND ND
PCB 1242 ND ND
PCB 1248 ND ND
PCB 1254 ND ND
PCB 1260 ND ND
Toxaphene ND ND
ND - Not detectable.
• Two 5-pm (0.0002-in) cartridge-type prefilters
in parallel to prevent the activated carbon filters
from clogging due to particulate matter.
Backwashing of the activated carbon filters is not
recommended because of the operational problems
this would cause, due to the difficulty of disposing
of the backwash water and the difficulty in
obtaining enough treated water at adequate
pressure to provide an adequate backwash.
• Two activated carbon filters in series. Each filter
consists of 25.4-cm (10-in) diameter fiberglass
tank containing 18 kg (40 Ib) of virgin activated
carbon. Bed depth is 91 cm (36 in) and each
cylinder has a empty bed contact time of
approximately 2.5 minutes at a flow rate of 0.32 l/s
(5 gpm). The theoretical lifetime of this treatment
system is 36 months based on an influent Benzene
r*nnr*^ntrsitinn r\f *}&&. nn/1 RcnTT'ona at thio lowal
\j\J\ IL*d fllaLHJI I Ul <£*T*T JJ-lj' <• DtSI IZ.t'I It* dl 11 Ho ItJVfc?!
was chosen as the critical design factor because it
will give a 99 percent assurance that the treatment
system would meet any influent organic chemical
challenge encountered during the testing program.
The filters are operated in series with the lead
cylinder changed yearly. This is done, even though
the theoretical lifetime is 18 months, to provide a
factor of safety. When the lead cylinder is removed
102
-------
Table 5.
Inorganic Chemical Quality of the Ground Water in
the Lake Carmel Area
Element
Aluminum (Al)
Arsenic (As)
Barium (Ba)
Beryllium (Be)
Boron (B)
Cadmium (Cd)
Calcium (Ca)
Chromium (Cr)
Cobalt (Co)
Copper (Cu)
Iron (Fe)
Lead (Pb)
Magnesium (Mg)
Manganese (Mn)
Molybdenum (Mo)
Nickel (Ni)
Phosphorous (P)
Potassium (K)
Selenium (Se)
Silicon (Si)
Silver (Ag)
Sodium (Na)
Vanadium (V)
Zinc (Zn)
Nitrate
High Value
(van)
0.326
0.060
0.290
0
0.849
0
56.81
0.005
0
0.078
0.268
0.030
15.68
0.372
0.014
0.009
0.095
4.29
0
8.12
0
169.7
0
0.068
13.67
Mean Value*
±Std. Dev. (us/1)
0.095 + 0.155(4)
0.015 + 0.030(4)
0.115 + 0.122(4)
0 + 0(4)
0.212 + 0.425(4)
0 + 0(4)
30.27 ±16.04 (12)
0.001 ±0.002 (4)
0±0(4)
0.033+0.039(4)
0.100 + 0.097(12)
0.008 + 0.015(4)
10.47 + 5.45(12)
0.081 ±0.126 (12)
0.005 ±0.007 (4)
0.002 + 0.005(4)
0.063 ±0.043 (4)
2.18 ±2.42 (4)
0±0(4)
4.69 ±2.75 (4)
0 + 0(4)
85.28 + 73.49(4)
0 + 0(4)
0.033 + 0.038 (4)
5.65 ±3.47 (23)
* Numbers in parentheses indicate sample size.
the lag cylinder is moved to the lead position and
the new cylinder is placed in the lag position.
• A valving arrangement is provided to allow for
water use during the cylinder changing procedure.
• Pressure gauges are provided before and after the
treatment system to determine head loss across
the system.
• Ultraviolet light disinfection is provided after the
activated carbon units to destroy bacteria that
break through the filter system. A light sensor with
a visual alarm is provided on the ultraviolet light
unit to inform the homeowner of proper
disinfection.
WATER TREATMENT SYSTEM COST
To receive as many bids as possible, the engineer
arranged for the system to be bid in six separate
contracts. Low bid results are listed in Table 6.
WATER SYSTEM MANAGEMENT
The original concept for management of the water
treatment systems'was for the Town of Kent to own
the systems and to be responsible for their operation
and maintenance. This responsibility for maintenance
could be carried out by their own personnel or under
contract by a qualified agent. The County and State
Health Departments would provide technical
assistance and some monitoring and regulatory
oversight.
However, the town turned the responsibility for
operation and maintenance over to the homeowners.
They formed a not-for-profit corporation, the Lake
Carmel Water Quality Improvement District
(LCWQID). The corporation consists of all the home
owners who have treatment systems. The
homeowners elect a President, Vice-President,
Secretary, Treasurer, and a seven member Board of
Directors. The Board of Directors consists of the
Officers of the Corporation, who were also the active
members of the Citizens' Advisory Committee, in
addition to three maintenance men.
Participation in the district is voluntary. Sixty-seven
of the 110 eligible homes received treatment
systems.
OPERATION AND MAINTENANCE
Operation and maintenance consists of changing one
of the activated carbon cylinders each year and
changing the bulb on the ultraviolet unit every nine
months. At the beginning of each year, the home
owner is given enough refill cartridges for the
prefilters and is expected to change them when
necessary.
Recharge of the activated carbon cylinders is
accomplished by taking the spent cylinder to a town-
provided workshed. The used activated carbon is
emptied and the fiberglass cylinder is refilled with a
bed of sand and 18 kg (40 Ib) of virgin activated
carbon. The spent carbon is disposed of at a landfill.
Maintenance men make house calls to repair leaks
and to clean the quartz tube on the ultraviolet unit.
They are paid on a flat rate per item basis.
ANNUAL COSTS
For the first four years of operation the annual cost
for operation, maintenance, and monitoring has been
$250 per treatment system. The annual charge has
recently been raised to $320 per year paid on a
quarterly basis. This amounts to a total annual budget
of $20,480 for the district.
WATER TREATMENT SYSTEM
PERFORMANCE
Although there is no legal requirement for the district
to monitor and report the performance of the water
treatment systems, they have tried to follow the
103
-------
Figure 2. Water treatment system.
-S-c-fl
PRE FILTER .
•f**-1!1—-ty—ir;
>—ULTRAVIOLET
" DISINFECTION
UNIT
\
3/4"X 5/8
WATER METER
ACTIVATED
CARBON UNITS
REINSTALL EXISTING
WATER SOFTENER UNITS
(IF ANY) THIS LOCATION
Table 6.
Item
Water Treatment System Cost
Cost ($)
1 Water Motor
8 Gale Valves
i Check Valve
3 Sampling Taps
2 Pressure Gaufles
2 Cartridoe Filter Units
2 Fiberglass Cylinders
80 Pounds Activated Carbon @ $0.90/lb
1 Ultraviolet Disinfection Unit
Installation
Total System Cost
150.60
67.84
140.60
72.00
392.00
494.00
1,317.04
guidance given them to sample at least 10 percent of
the systems each year.
BACTERIOLOGICAL PERFORMANCE
During the years of 1984, 1985, and 1986, 21 paired
samples were collected and analyzed for coliform
organisms. Of these, three untreated samples had
high coliform counts and one treated sample had a
count of 2 coliform organisms per 100 ml (0.6 per
02). This is a great improvement over the original
sampling where 40 percent of the drinking water
samples had high coliform counts.
ORGANIC PERFORMANCE
The district has not had enough money to do
adequate organic chemical monitoring. In 1984, 10
paired samples were collected and analyzed for
benzene, toluene, and xylene. The detection limit was
too high to show system performance, but the sample
results did indicate that these chemicals were not
present in either the untreated or treated water above
the guideline levels of 5 jag/l for benzene, and 50 pg/l
for toluene and xylene. In 1985, four paired samples
of these same chemicals were analyzed with all
results below 1 pg/l.
In 1986 and 1987, the New York State (NYS)
Department of Transportation collected a series of
volatile organic samples at one residence because of
a nearby chemical spill. On December 23, 1986,
samples were collected before and after the treatment
system. The results are shown in Table 7.
Table 7.
NYS DOT Volatile Organic Samples: December,
1986
Chemical Before Treatmen AfterTreatmen
(pg/i) (ps/i)
Toluene
Ethylbenzene
p-Xylene
m-Xylene
o-Xylene
n-Propylbenzene
1,3,5 Trimethylbenzene
1,2,4 Trimethylbenzene
Cyclopropylbenzene
Total
5 <1
3 <1
2 <1
6 <1
4 <1
2 <1
4 <1
4 <1
5
35 <1
On February 5, 1987, samples were collected before
and after the treatment system. The results are
shown in Table 8.
On March 7, 1987, no contaminants were detected in
a treated sample, and on May 8, 1987 and August 12,
1987, no contaminants were detected in samples
collected before and after treatment.
104
-------
Table 8. NYS DOT Volatile Organic Samples: February, 1987
Chemical Before Treatmen AfterTreatmen
(ng/i) (ua/i)
Toluene
m-Xylene
o-Xylene
o-Chlorotoluene
p-Chlorotoluene
1 ,3,5 Trimethylbenzene
Cyclopropylbenzene
o Dichlorobenzene
Hexachlorobutadiene
Total
5
2
1
2
1
2
3
5
6
27
5
2
1
2
<1
<1
2
4
<5
16
The results of all organic samples except one show
removal of the tested organic chemicals to below
detectable levels. In the sample collected on February
5, 1987, contaminants were detected in the treated
water sample at 5 jig/l (5 ppb) or less. The limited
test results are not comprehensive enough to make a
definitive statement on the removal of organic
contaminants, but they do give an indication that the
treatment system is performing satisfactorily.
105
-------
COMMUNITY DEMONSTRATION OF POU SYSTEMS REMOVAL OF ARSENIC AND
FLUORIDE: SAN YSIDRO, NEW MEXICO
Karen Rogers
Leedshill-Herkenhoff, Inc.
Albuquerque, NM 87103
San Ysidro is a small, rural village of approximately
200 people located in the north central part of the
State of New Mexico, approximately 50 minutes north
of Albuquerque. The village is at least 200 years old.
U was settled by Spanish colonists on a land grant
from Spain in 1786. San Ysidro lies between the
lands of the Zia and Jemez Pueblo Indians along the
Jemez River. Life in the community is simple and
relaxed and most residents live there for exactly that
reason. The mean annual income for families in San
Ysidro is $13,500. Fifty-eight percent of families
earn less than $10,000 a year.
The village water supply is collected in an infiltration
gallery into which ground water is drawn. This local
ground water contains leachate from geothermal
activity in the area's abundant mineral deposits and is
therefore high in mineral content. The ground water
exceeds the standards and/or maximum contaminant
levels (MCLs) for arsenic, fluoride, iron, manganese,
chloride, and total dissolved solids. The contaminants
of concern in the village water supply are arsenic V
and III and fluoride, which exceed the MCLs by three
to four times (Table 1).
Tablo 1. Average Water Quality in San Ysidro, New Mexico
Cone. Max. Contaminant Level Avg. After
(mfl/I)
or Rec. Std. (mg/1)
RO (mg/l)
Iron
Manganese
Cntondo
FkwkJo
Arsonlc V & 111
IDS
2.0
0.2
325.0
5.2
0.22
1,000.0
0.3
0.05
250.0
1.8"
0.05
500.0
0.015
<0.01
12.50
0.40
<0.01
< 180.0
* RoconlJy revised (o 4.0.
Prior to discussing the point-of-use (POU)
treatment that is installed in San Ysidro, a better
understanding of the San Ysidro water system is
necessary, including some general problems which
have a direct bearing on the future success or failure
of the point-of-use devices.
As mentioned before, the village water supply source
is an infiltration gallery that produces an average of
27,000 gpd in winter and 36,000 gpd in summer from
the ground water. The infiltration gallery has a storage
capacity of 17,000 gallons. The village currently uses
an average of 30,000 gpd. This equates to about 150
gpd per person. This consumption rate pushes the
production/storage capacity limits of the gallery.
There is a 20,000-gal elevated storage tank
connected into the piping system that should be
providing the additional capacity the village needs, but
it has seldom been a functioning unit for several
reasons. First, the pumps that are currently located in
the infiltration gallery do not have adequate controls
to allow them to operate appropriately to maintain an
adequate supply of water in the system. There is no
remote readout on the status of the pumps or system.
The only way to know there is a problem is when a
faucet is opened and no water comes out. Secondly,
the village does not have one person who knows the
system and who has the responsibility to keep it
operating. A village employee, the major or one of the
village council usually goes down to turn the pumps
on when someone calls to complain about the low
pressure or to report that they have no water. The
pumps should be running all night, but because of
other problems with the controls and overheating
pump motors, someone would need to monitor them
all night to insure safe operation; however, this is not
an acceptable solution. So, the village has a long
history of water supply problems including low water
pressure, no water at all, and quality problems
including taste, color, clarity, and odor in addition to
the contaminant levels discussed earlier.
A number of alternatives have recently been
investigated by engineers at Leedshill-Herkenhoff to
aid the village in obtaining a higher quantity and
quality of water. When the village first was found to
be in violation of the Safe Drinking Water Act (SDWA)
106
-------
for levels of arsenic and fluoride in the water supply,
four deep test wells were drilled to determine if there
was a better source of water available. The best of
these wells only had water equal in quality to the
water in the infiltration gallery. It was at this point that
treatment options began to be studied. The
recommendation for added system capacity at this
point was to increase the size of the infiltration
gallery. The village obtained bids to perform this work
but all of them exceeded the available funds because
of the high costs of dewatering the site during
construction. The village then decided to add the best
test well that had been drilled previously to the
system. This new well was recently developed at
43,200 gpd and is being pumped into the infiltration
gallery temporarily to supplement the system. The
new well will be permanently tied into the system
under a project which is currently being reviewed by
the state for approval. This project will also provide an
automatic control system to regulate pumping from
the well or the gallery to maintain a beneficial water
level in the village storage tank. This control system
will have remote readouts in the village office to
indicate if the system is functioning normally or to
show there is a problem.
A variance from the SDWA for arsenic and fluoride
was granted to the' village while research was
performed by Dennis Clifford of the University of
Houston to determine an economical and effective
solution to the contaminant problem. The treatment
systems studied were activated alumina and reverse
osmosis (RO). Central and point-of-use treatment
were considered. Central treatment of the entire water
supply was not considered feasible for many reasons.
First, there is a disposal problem with both the
arsenic-contaminated sludge from activated alumina
column regeneration and the reject brine from the
reverse osmosis unit. Secondly, the costs of central
treatment were considered to be higher than point-
of-use treatment. And lastly, central treatment was
considered too complicated to be efficiently operated
in a community the size of San Ysidro. The results of
the study indicated the best solution to be point-of-
use treatment with reverse osmosis units. A pilot unit,
a Culligan H-82 with a spiral-wound polyamide
membrane, was installed in the community center to
assure the effectiveness of the membrane and the
acceptability of the unit to the community. This unit
makes about 5 gpd of water with a reject rate of
about 10 to 20 gpd, which is discharged into the
user's septic tank. That test unit has now been in
service for about three years with little maintenance
required.
Since arsenic and fluoride are only considered
harmful in water used for human consumption, a
point-of-use unit for treatment was needed for only
water used for drinking and cooking. A single large
RO unit for only the drinking and cooking water
supply for the village was considered, but there were
still concerns about disposal of larger quantities of
reject water and there was also doubt that the people
would be as willing to use the treated water if they
had to travel somewhere to get it. The EPA was also
very interested in trying point-of-use in a small
town. All of these factors made the decision to try the
units in the individual homes a fairly easy one. It was
decided that the best place to install the treatment
units would be in the home's kitchen, preferably
under the kitchen sink with a separate faucet on the
sink for dispensing the treated water and a small tank
under the sink for storing the treated water.
Once it was determined that point-of-use reverse
osmosis might be a good solution for the village, a
proposal was made to the EPA to obtain a grant to
purchase, install, service, and monitor the units and
to study the overall feasibility of point-of-use
treatment in a small community. A Request for
Proposal for engineering services was generated and
Leedshill-Herkenhoff was retained by the village to
oversee the project. A Request for Proposal then was
written to obtain the units in addition to a
maintenance contract for a period of 14 months.
Culligan was awarded the job and unit installations
began in June 1986. A public hearing was held in
which the proposal was brought before the villagers to
explain the problem with water quality and to discuss
the procedures needed to get the units installed,
maintained and tested during the study period. An
ordinance was passed by the village which made the
use of village water contingent upon installation of the
RO unit in the home. Each water customer also had
to sign a permission form to allow the village to install
the unit in their home and to allow access to the unit
for testing and maintenance. A few reluctant villagers
did not want the units installed in their homes. The
primary reason given was that they did not think they
needed them. After all, people in the town had been
drinking the water for years and it did not seem to
hurt them. Another reason was the permission they
had to give the village to be able to enter their homes
to install, test, and maintain the units. The reluctant
few were inevitably persuaded, however, when they
were informed their water was going to be shut off if
they did not comply. There are still a few people in
the community who do not drink the water from the
RO units. They say they do not like the taste of the
treated water and are either getting water elsewhere
or drinking the untreated water.
Currently, 70 units are available for testing on this
project. There are a few units in unoccupied homes
which the village has been reluctant to take the
initiative to remove. The 70 units are tested every
other month for arsenic and fluoride and
approximately every three to four months for chloride,
iron, and manganese. A smaller sample group of
about 30 units is' being sampled for bacteria. The
testing portion of this project has been difficult at
times because of the inability to obtain samples when
107
-------
homeowners are away. Some residents of San Ysidro
are home only on weekends, and many work during
the day. The sampling has been done by a village
employee. He has reported that he has trouble from
close to half the residents when getting samples.
Most of the complaints are about the inconvenience
of having to let him in. One resident draws her own
samples because she refuses to let the village
employee in her home despite the permission to enter
that she signed. It usually takes 2-1/2 days to collect
samples from 25 to 30 homes.
Because of the time frame involved, we have only
been able to obtain 10 bacteria samples during each
sampling. These are usually taken on a weekday
morning. All of the samples are picked up and taken
to an independent laboratory in Albuquerque the
afternoon of the day the bacteria samples are drawn.
There have been occasional problems with coliform
counts in the RO units. Out of 96 tests that have
been performed over the last 10 months, there have
been six positive tests ranging from one to TNTC (too
numerous to count). The tests have also been
differentiating between conforms and noncoliforms.
We have been getting positive noncoliform readings
also, almost always in conjunction with a high coliform
count. Our procedure has been to have Culligan
replace the filters and the RO module and disinfect
the tubing and tank on the units that have shown
coliforms in the treated water. The EPA lab in
Cincinnati will be attempting to identify the species of
coliform we are seeing as soon as we get another
positive test. (This was done recently and the
coliforms were identified as E. coli.) Culligan has
recently recommended an alternative disinfection
technique utilizing hydrogen peroxide that will be less
expensive. The replacement costs of the filters and
membrane run about $200 (see author's note).
There are a few possibilities for the source of the
coliforms that we have been trying to pinpoint. It is
possible that the low system pressure may be
inducing back siphonage from some cross-
connections in the individual homes. It was recently
discerned that many of the units installed in San
Ysidro had not been installed with an air gap on the
discharge line from the RO module. These could very
well be the cross-connections that have been
causing us to see coliforms in testing. Discussions
with the installer of the units revealed some
disinformation regarding the air gaps. He felt the air
gaps were not really necessary and that it was just
one more place for the units to develop leaks. In
further discussion it was explained that without the air
gaps, especially in San Ysidro, where we have
frequently seen low or no pressure on the water
system, the likelihood of back siphonage is much
greater. The installer is currently rectifying this
problem.
It is also possible that the source could be
somewhere in the system. The village water supply is
chlorinated by a hypochlorination system at the
infiltration gallery, The village has had problems in the
past with the chlorination system. San Ysidro was on
a boil order a few years ago for a coliform infraction
prior to the installation of the RO units. During the
test period we have not had a positive coliform test in
the system, but the monthly system sample is
obtained very close to the chlorinator. The piping
system in the village is arranged in a three-spoke
system with the water supply at the hub and the
pipes dead-ended at the edges of town. This
arrangement could encourage bacterial growth in the
stagnant ends of the pipe but the positive test
locations do not seem to support this theory. The
locations of homes with positive results are not in any
particular location on the system. The State of New
Mexico Environmental Improvement Division (EID)
has encouraged the village to monitor the chlorine
residual and even provided them with a monitoring
device, but they have been uninterested in using it. If
an adequate chlorine residual was maintained in the
system, it might reduce the coliform problems in the
units by removing a potential bacteria source. We will
be recommending to the village that they start
monitoring their system chlorine residual especially at
homes that show a positive bacteria count in the RO
unit. The Village should also retest the treated water
and test the untreated water at that home after a
positive test result. In the one home in which we were
able to do this, the test had shown six coliforms and
three noncoliforms. The retest showed 0 coliforms
and 0 noncoliforms in the RO treated water and 137
noncoliforms in the regular sink water. These retests
could prevent some unnecessary maintenance costs.
We will also be recommending to the village that we
continue testing each unit for bacteria at least every
three months until all of the air gaps are installed
properly and they have the new pumping system
maintaining a minimum system pressure of at least 20
psi, and until they have not had a positive test result
for at least six months. After those conditions are
met, we feel they should be able to decrease testing
to every six months for each unit.
Initial evaluations of the test results obtained from the
RO units indicate removal rates for arsenic, fluoride,
chloride, iron, and manganese to be consistent with
the manufacturer's data for the units despite the low
system pressure experienced frequently in San
Ysidro. The arsenic and fluoride testing seem to
indicate that some membranes may need to be
replaced as often as once a year. We had five units
that had fluoride and arsenic tests at or close to the
MCLs in July and August. It appears there will be two
to three months' warning on these breakthroughs.
The test indicator on the unit was also showing a red
light on four of these units, so this will help the
customer to determine if his unit needs servicing.
Since the testing has shown that fluoride tends to
break through just prior to the arsenic, we
recommend monitoring fluoride levels in the units
108
-------
every three months to determine when the unit needs
servicing.
The most prevalent maintenance problems are leaks.
The most likely place for the leak to occur is at the
drain clamp. This problem is predominant because
the clamp is located where it is easily bumped and it
is just tightened around the pipe rather than being
attached to it. The second most common problem is
breakage of the faucet handle. San Ysidro anticipates
having their village maintenance man trained by
Culligan so that they can take care of routine
problems themselves and thus reduce their future
maintenance costs.
Expenses for the study period have been broken
down as follows:
Initial units $290.00 per unit
Installation 35.50 per unit
Maintenance 8.60 per unit per month
Testing 25.00 per unit per month
Replacement or New Units 350.00 per unit
The future costs should run approximately the same
except for the testing costs. It will be possible to
virtually eliminate the laboratory testing for arsenic
and fluoride by monitoring the conductivity of the
effluent water. This is the theory behind the RO test
module currently installed in each unit in San Ysidro.
If the mathematical correlation between conductivity
and fluoride and arsenic concentration is established,
each unit can be tested on-the-spot by the village
employee during a periodic inspection. Laboratory
testing could be reduced to random sampling every
six months to insure continued correlation between
concentration and conductivity. This procedure was
discussed with and tentatively accepted by our state
EID representatives. The costs that we do not have
information on as yet are the insurance costs for
liability and replacement. However, the reduced costs
of using a village employee for routine unit
maintenance should at least offset the insurance
costs. This would increase the user's bill by
approximately $12 per month (the current bill
averages $10 per month). Preliminary studies by
Dennis Clifford indicated costs of point-of-use RO
to be $10 to 15 per customer per month and the
costs of central RO treatment to be $30 to 40 per
customer per month based on the current
consumption rate (see author's note).
We are currently working with San Ysidro's attorney
and the State of New Mexico Environmental
Improvement Division to develop an ordinance to
govern the policy on the RO after the grant term is up
and the village is under the state's jurisdiction again.
We anticipate that the village will continue to own,
maintain, and test the units in the future. The village
will also have to obtain a special liability policy for the
units for any water damage claims that may be made.
Another insurance issue that came up during the
study period was damage to the unit. We had one
unit destroyed in a house fire and two units needed
various major parts replaced because of freeze
damage. In the future we feel the way to handle these
issue will be to have the homeowner be responsible
for these costs.
Special provisions will be needed for commercial
establishments. We feel the best way to handle this
issue in the future will be to have the village lease a
properly sized unit to the business, arrange for a
maintenance contract with the manufacturer, and then
add these charges to the business' water bill. This
should be easier than having the businesses or the
village purchasing the units outright and should also
give the village flexibility with new businesses that
may require smaller or larger units.
On the whole, community reception to the units has
been positive. Most villagers like the taste of the
treated water, especially for coffee and ice. There are
still some residents, primarily those who have lived in
San Ysidro all their lives, who do not like and do not
drink the treated water. They still drink the untreated
water. There are also a couple of residents who still
bring in their drinking water from elsewhere. There
are a few villagers who are on the water system but
who do not have indoor plumbing, a sink, or other
convenient place to install the unit. They have
expressed a desire to obtain the units but have had to
be turned down until they have a place for the unit to
be installed. There are also some residents who are
dissatisfied with their individual wells who are
considering getting on the village water system and
are interested in the RO units. Eighty units were
purchased initially for the Village and currently 79 are
installed. This number has been sufficient for the
study period, but a few extra will be purchased so
there will be some spares available in the near future.
In conclusion, I would like to summarize the pros and
cons that I've seen with this project. First the negative
issues: sampling costs can be much higher for
multiple point testing of point-of-use systems than
single point testing of central treatment. With point-
of-use, control of the treatment process is dispersed
from a central point to multiple points. To maintain the
same level of control, more regulating and monitoring
must take place. Obtaining samples can be difficult
and time consuming, contributing to higher costs and
decreased control over testing. With a point-of-use
treatment system, another factor that must be
considered is initial and continuing education of the
consumer. New members of the community must be
indoctrinated to the system, and existing users must
be reminded periodically of their responsibilities.
Point-of-use also inherently generates more
bookkeeping for the village clerk. The responsibility
for tracking, testing, and maintenance for each unit
will be a part of the job. Another problem that is an
inherent part of point-of-use treatment is do-it-
109
-------
yourself plumbers. It is an irresistible urge for some
people to tamper with or try to fix the units
themselves since the unit is located in an accessible
part of the home.
The positive side of point-of-use in San Ysidro is
much more encouraging. The RO units have very
much improved the aesthetic as well as health quality
of the village's drinking water. The units are simple to
install and maintain. The units are undoubtedly the
least expensive solution to treatment for the village
and its residents. The vast majority (90 to 95 percent)
are happy with the system and the water it produces.
This project has been challenging for the community
of San Ysidro. If the units are to continue to function
as part of the village's treatment system, the
community will need encouragement and technical
and regulatory support from the State of New
Mexico's Environmental Improvement Division. The
village will need to begin recordkeeping on each unit
and diligently maintain those records to insure the
maintenance and testing continues as required. We
wilt also have to see how the community will respond
to the additional costs on their water bills. The future
success of this project will depend heavily on the
abilities of the Village of San Ysidro to cope with the
recordkeeping for testing and maintenance of the
units, but this would surely be a problem for central
treatment as well. Point-of-use treatment is the
best choice for San Ysidro's water system at this time
and will continue to be the best solution until
population growth in the village will support the costs
of a central treatment plant.
Author's Note: This report has been revised to reflect
pertinent additional data and information obtained
since it was presented.
110
-------
FLORIDA'S FUNDING FOR CONTAMINATION CORRECTION
Glenn Dykes
State of Florida
Department of Environmental Regulation
Tallahassee, FL 32399
For years Florida has had considerable concern for
drinking water quality because of the expanding use
of its ground water resources. The potential for
contaminating these resources is great in view of the
state's rapid growth and the vulnerability of this
valuable source of potable water. Water quality
problems have been confirmed by extensive analytical
work, and now these contamination issues must be
addressed.
Consequently, full evaluation and correction of all
possible contamination would take considerable
money, time, and effort. An important aspect of the
use of EDB was the fact that the state, under contract
to the citrus grove owners, applied the chemical for
nematode control. Since the state and Federal
agriculture departments approved of the chemical
usage, it found broad acceptance in the agricultural
community.
DEFINING THE PROBLEM
Over 90 percent of the state's drinking water comes
from underground resources. Included in these
supplies are approximately 10,000 wells serving
public and semi-public facilities along with several
hundred thousand private wells, serving individual
homes. There is a wide disparity in the quality and
the vulnerability of the ground water resources
meeting the demands of these water supply wells.
Many of the public systems utilize the deeper
limestone strata while the private wells primarily tap
the shallower aquifers. These two sources have
different problems and concerns, but both are
susceptible to contamination that has been verified by
analytical work on these resources.
Early in the organic quality testing work by the U.S.
EPA, several of Florida's public supplies were
analyzed. The EPA efforts and testing by the state's
own laboratories confirmed suspicions of
contamination of the state's ground water supplies.
From these early endeavors, the state developed a
very comprehensive set of rules and regulations
governing ground water protection, as well as the first
regulations in the nation setting maximum
contaminant levels (MCLs) for volatile organic
compounds. The driving force for these actions was
the realization that the agricultural chemicals, aldicarb
and ethylene dibromide (EDB), had also been found
in samples from many of the private wells. From
further assessment of this problem, it was also
learned that these chemicals, particularly EDB, had
widespread usage throughout the state.
LEGISLATIVE EFFORTS
The findings of organic contamination; in both the
public and private potable water supplies brought
about vocal manifestations of the electorate's
concerns. The legislature started looking for ways to
correct these problems. In 1983 they passed the
Water Quality Assurance Act (WQAA), which was
very broad legislation addressing a wide variety of
items related to ground water contamination and
providing funding for their resolution. This act spelled
out numerous ground water protection issues, which
included requiring public water suppliers to investigate
a broad spectrum of contaminants, studying private
wells, and funding related research and emergency
corrective actions. Law makers the following year
provided $3.1 million For solving the EDB
contamination problem caused by the state's own
activities. The 1986 legislature broadened the WQAA
and provided additional funding to correct EDB
contamination and other health related water quality
problems over which the well owners had no control.
This legislature also provided funding to address
contamination from leaking underground petroleum
storage tanks through the establishment of the Inland
Protection Trust Fund (IPTF). Backed by this
authority, the Florida Department of Environmental
Regulation (FDER) could restore or replace
contaminated private wells (WQAA and IPTF) and
public wells (IPTF) without waiting for legal
determinations of responsibility. Funds for correcting-
problems in public supplies were not available under
WQAA though assistance was provided with the
state-affected EDB monies.
111
-------
RESTORATION/REPLACEMENT OF
SUPPLIES
The Ground Water Contamination Task Force, which
was organized to address EDB and other contaminant
problems, had to evaluate the potential concerns and
how to approach each one. Initially, activities were
directed at EDB. The group determined that
correction would have to be provided for the entire
household, mandating point-of-entry treatment as
the only approach. Through research funding, data
were developed to indicate that granular activated
carbon (GAG) would effectively adsorb EDB. The
state collaborated with EPA to fund a project to
provide definitive numbers for both packed tower
aeration and GAG that would assist in treatment unit
designs. Research funds were also used to evaluate
the viability of replacing contaminated supplies with
properly constructed wells into the deeper aquifers. In
the final analysis, this approach did not prove feasible
for widespread application.
The use of GAG filters was determined to be the best
alternative to correct the numerous private wells
contaminated with EDB. Since the volume of water,
usage habits, and other parameters that influence the
effectiveness of GAG were all undefined, a very
conservative filter design was devised. A 0.06-m3
(2-cu ft) GAG filter was selected with a 5-pm
(0.0002-in) pre-filter, a water meter, and an
ultraviolet (UV) light included in the standard unit
(Type I). These units were installed to handle
contamination up to 10 ug/l. Higher contamination
warranted additional GAG filter units (Type II). If the
expected water consumption was more than 10 gpm,
larger GAG units were provided to handle increased
flow. Formal bid proposals were solicited to obtain a
qualified supplier, since it was realized that a large
number of units would be needed. The proposals also
included operation and maintenance items such as
the planned replacement every six months of GAG
material and UV lights.
Of the 12,400 wells analyzed for EDB through
October 1987, 1,530 were found to exceed the MCL
of 0.02 pg/l. Most of these (~ 1,400) served private
residences. Where possible, a permanent solution to
correct the problem, such as connections to existing
community supplies, was utilized. At present we have
over 550 Type I and 60 Type II units installed on
state-affected wells. There are also 230 Type I and
2 Type II GAG filters on non-state:affected wells.
Because of different statutory requirements, records
on each program are separately maintained. The
overall EDB corrective effort has also required the
installation of larger units on some of the public and
semi-public systems. There are seven larger GAG
units between 50 and 200 gpm capacity and three
municipal systems with capacities up to 3,000 gpm.
Under the state's current contract, the Type I
installation cost is $1,000, and the cost of Type II
units is $1,050. The annual carbon and UV light
replacement cost is $890 for the residential type unit.
The overall program is currently being re-evaluated
to determine if the longevity of the GAG filters can be
extended past the current six-month replacement
cycle. It is estimated that each month's extension
would save approximately $40,000. If the foregoing
GAG units and maintenance costs are evaluated, one
can easily see that annual outlay is quite large. The
annual maintenance cost for the residential units and
the aforementioned larger systems will exceed $1
million when the remaining planned units are installed.
In view of the potential future maintenance cost, the
state has attempted to evaluate the economic
feasibility of extending existing water lines to replace
the residential supply wells. To determine the current
cost of the filter installations for their lifetime, the
annual replacement cost was projected for io years
and then returned to present worth using five percent
inflation and eight percent interest. This value was
then added to the installation cost of the filter to
determine the most economical approach; i.e.,
individual GAG filters, extension of water lines, or a
new central system. Under the current contract, the
calculated cost for use in this feasibility analysis is
approximately $8,600. Since additional organic
sampling is required to insure that public health
concerns are not compromised, an additional annual
cost of $400 should be added. This would add
approximately $3,400 to the aforementioned present
cost consideration. With the logical addition of
sampling cost, $12,000 should be used to determine
the economic feasibility to provide a permanent
alternative to the point-of-entry solution.
SUMMARY AND CONCLUSIONS
Garbon filtration has been found to be a satisfactory
method of removing EDB from contaminated supplies.
The FDER is also utilizing GAG filters in correcting
contamination problems created by leaking
underground petroleum storage tanks under the IPTF
program. Current testing indicates that GAG will
remove benzene and other hydrocarbons and thereby
solve some of the problems related to these
contaminants.
The use of point-of-entry solutions must consider
the long term cost in considering the economic
viability. The cost as shown in the foregoing
discussion can be large. The annual maintenance and
sampling cost must be given a thorough evaluation
before determining that point-of-entry devices are
to be placed on all residences in a community. It is
always gratifying when a permanent solution can be
found and the utility does not have to worry about the
maintenance problems that always seem to plague
these small installations. The present projected cost
of $12,000 per house with a Type I connection would
go a long way to provide a central system for the
whole community.
112
-------
MONITORING AND MAINTENANCE PROGRAMS FOR POUlPOE
Gordon E. Bellen
Thomas G. Stevens
National Sanitation Foundation
Ann Arbor, Ml 481 OS
INTRODUCTION
Small communities with organic or inorganic
contaminants in their drinking water supplies often
lack the financial resources to solve their problems.
Economies of scale prohibit construction of a central
treatment system for contaminant removal in many
cases. Construction of an alternate well or connection
to a neighboring water supply may not be feasible.
One alternative solution, which has been -receiving
more attention' in recent years, is treatment of
contaminated w^ter at the point-of-use (POD) or
point-of-entry (POE).
All POD devices are designed to treat only water
intended for consumption. Approaches to POU
treatment include batch process treatment, faucet-
mounted devices, in-line devices, and line-bypass
devices. A batch process device treats one batch of
water at a time, is not connected to the water supply,
and may rest on the kitchen countertop. Faucet-
mounted devices are attached directly to the faucet.
In-line devices are installed between the cold water
supply and the kitchen faucet, and generally treat the
entire kitchen cold water supply. With the line-
bypass approach, the cold water line is tapped to
provide influent to a treatment device, which may be
installed under the kitchen sink, and a separate tap is
provided at the sink for treated water.
POE water treatment treats all water entering the
home and has been proposed for contaminant
removal where potential health risks associated with
skin contact and inhalation exist (1). Because they
treat all water entering the home, POE devices must
be much larger (in terms of volume treated) than
POU devices. The length of time in service between
media replacements however, is typically 25 percent
of that of POU devices.
The U.S. EPA has specified when, and under what
conditions, POE and POU water treatment can be
used (2). Although not considered Best Available
Technology (BAT), POE is an acceptable method for
a community water supply to come into compliance
With the Drinking Water Regulations. POU may be
used as an additional control measure during the
period of a variance or exemption, as a condition of
the variance or exemption. If either approach is used,
the EPA has specified conditions that must be met:
• Central Control - Regardless of ownership of the
treatment device, the public water authority will be
responsible for operating and maintaining all parts
of the treatment system.
• Effective Monitoring - A monitoring plan must be
approved by the state before a POU or POE
system is installed. The plan must assure that
devices provide health protection equivalent to
central water treatment. Physical condition of
equipment and total volume of water treated must
be monitored as well.
• Application of Effective Technology - AH devices
must have certified performance (or rigorous
design review) and must pass field testing.
» Maintenance of Microbiological Safety - Control
techniques such as backwashing, disinfection, and
monitoring, are suggested by the EPA to maintain
microbiological safety.
• Protection of All Consumers - Every building must
have equipment that is adequately installed,
monitored, and maintained. Responsibilities for this
equipment may transfer with ownership of the
property.
This paper discusses each of these points and
suggests ways in which a community might comply
with EPA requirements.
CENTRAL CONTROL
It is important that all aspects of POU/POE treatment
come under central control to assure adequate
protection of public health. If a public water system is
already in place, the existing organization can assume
administration of the POU/POE district. If a public
113
-------
water system is not in place, a water quality district
should be formed in a progression of steps similar to
those in Table 1 (3). In a previous study (4), total
administrative costs for operating a water quality
district were estimated (1985 dollars) to be $1.23 per
customer per month. These costs included quarterly
monitoring costs, administration, and distribution
system maintenance (POU maintenance not included)
for reduction of fluoride in drinking water with POU
treatment. Monitoring costs may be higher for some
contaminants, but labor costs can be lower if
community volunteers are used.
Tfibto 1. Chronological Steps for Formation of a Water
Quality District
Process Step
Development phase
Identify problem
Consult regulatory agencies
Water testing
Make preliminary plans and maps
pproyai phase
Estimate costs
Hold public hearing
Property owner petition
District formed by resolution of county/state supervisors
Directors appointed
Agreement with town board and property owners for cost
recovery
Operation Phase
Obtain funding
Pdot demonstration
Select equipment
Equipment installation
Authorize payments
Monitoring and maintenance
Feedback and education
APPLICATION OF APPROPRIATE
TECHNOLOGY
To achieve compliance, the EPA stipulates application
of appropriate technology. Table 2 lists currently
available technologies and the contaminants they are
effective in removing. Selection of appropriate
technology from this list is not straightforward, since
variable water qualities may make one technology
better than another. For example, activated alumina is
effective in fluoride reduction, but in the presence of
high alkalinity and/or arsenic, its capacity may be
reduced (4,5).
Communities lacking expertise should seek
knowledgeable sources of information concerning
treatment techniques appropriate for their water. An
initial consultation with the local or state health
department is a good first step. Other organizations
that can provide information are listed in Table 3. In
addition to those organizations, consulting
engineering firms can be hired. Regardless of the
Table 2. POU/POE Treatment Technologies*
Treatment Type
Reverse Osmosis**
Cation Exchange
Anion Exchange
NIPDWR
Contaminants
Arsenic*", Barium,,
Cadmium, Chromium,
Lead, Mercury, Silver,
Fluoride, Nitrate,
Selenium, Radium,
some organics,
herbicides, and
pesticides
Barium, Cadmium,
Chromium III, Lead,
Mercury, Radium
Nitrate, Selenium VI,
Arsenic III, Arsenic V,
Chromium VI
Other
Contaminants
Total dissolved
solids, Copper,
Chloride,
Sulfate.foaming
agents, corrosion
Copper, Zinc,
Ironf, Manganese
Chloride,
corrosion, Sulfate
Activated Alumina
Direct (Mechanical)
Filtration
Activated Carbon
Distillation
Fluoride,
Arsenic.Selenium IV
Turbidity
Organics, Organic
Mercury
Metals, high molecular
weight organics
Cysts ,
Color, foaming
agents, taste, and
odor
Total dissolved
solids, Chloride,
Sulfate
Taken from the Statement of the Water Qualify Association
to the EPA, EPA, December 13, 1983.
Results of reverse osmosis treatment may vary between
pressurized and nonpressurized units, membrane type, and
configuration.
Arsenic (+3) is poorly removed with reverse osmosis.
Low levels.
Table 3.
Organizations Providing Water Treatment
Information or Services
Organization
Service
American Water Works Association
666 West Quincy Avenue
Denver, CO 80235
303/794-7711
National Demonstration Water Project
1725 DeSales Street, NW - Suite 402
Washington, DC 20036
202/659-0661
National Sanitation Foundation
3475 Plymouth Road
Ann Arbor, Ml 48105
313/769-8010
National Water Well Assocation
500 West Wilson Bridge Road
Worthington, OH 43085
614/846-9355
Water Quality Association
4151 Naperville Road
Lisle, IL 60532
312/369-1600
Water l3ata Base
Educational Materials
Technical Information
Educational Information
Relating to Rural
Communities
Product Testing/Listing
Performance Standards
Technical Information
Technical Assistance
Ground Water
Information
Lists of Manufacturers
and Distributors
Technical Information
source used, professional guidance in the selection of
equipment is important.
114
-------
The EPA requires certification of performance and
field testing of POU and POE devices. Certification
can be accomplished by the state or a third-party
acceptable to the state (2). The National Sanitation
Foundation (NSF) has several performance standards
for POU/POE devices that address performance for
products making contaminant reduction claims for
primary and secondary regulated chemicals, reverse
osmosis equipment, cation exchange water softeners,
and ultraviolet disinfection equipment. A standard for
distillation equipment is currently being written. These
standards are in the public domain and can be used
by anyone as a basis for product certification. NSF
also conducts a product certification program. NSF
listings of products are available on hard copy or
through computer access (see Table 3 for address
and phone number of NSF).
Field testing is important. Product certification testing
with standardized test waters may not accurately
indicate treatment capability or capacity for the water
a community needs to treat. For treatment
technologies with finite usable capacities (e.g., carbon
and activated alumina), field tests should be run to
exhaustion to establish the useful life of the device
(useful life defined as volume of water treated to
breakthrough). Accelerated field tests using surface
adsorbents like carbon and activated alumina will
provide conservative estimates of useful life. Tests of
devices with media which can be regenerated will
provide estimates of annual regeneration costs. Other
devices like reverse osmosis and distillation, that do
not have a readily definable useful life, can be
evaluated quickly for percent removal of undesirable
contaminants.
EFFECTIVE MONITORING
The EPA requires monitoring to assure protection of
the public health comparable to central treatment. To
achieve that goal, a monitoring program should be
established which provides reasonable assurance that
all water provided at the tap is in compliance with the
National Primary Drinking Water Regulation. The
results of field testing should help in that regard. In
addition, the total volume of water treated and
physical condition of each unit must be monitored (2).
The system performance monitoring program will be
influenced by field test results, community
experience, and whether treatment is intended to
achieve compliance. A rigorous and conservative
monitoring program, assuming one technology or type
of device is being used, is outlined in Table 4. This
program addresses POU/POE treatment on a
distribution system. Treatment of a system of
individual wells would require more frequent sampling.
This monitoring program is intended to assure
adequate operation of the POU/POE treatment
system. Additional source water monitoring is
required by the EPA. A description follows.
Table 4. Suggested Minimum Monitoring Program*
Task Frequency
Contaminant Monitoring
First Year Minimum of seven devices per
quarter; if useful life is less than
one year, test at each quarter of
estimated life.
Minimum of seven devices per
quarter; select some known high
volume users.
Minimum of three per quarter. Test
minimum of seven units at
replacement to reconfirm useful life
estimtes.
Minimum of seven per quarter or
number required by EPA population
based monitoring plans, whichever
is higher.
If positive coliform results
obtained.
(May be provided by
homeowners)
25 percent of devices/quarter.
10 percent of devices/quarter.
Second Year
After Useful Life
Established
Microbiological Sampling
Routine Monitoring
Heterotrophic Plate
Counts and Conforms
Fecal Conforms
Treated Volume
Recordings"
First Year
After First Year
In addition to other system monitoring requirements.
Volume recorder for treatment unit, not whole house meter.
Field test results should be used to estimate useful
life of component parts (e.g., media, cartridges,
prefilters, etc.), although manufacturers and/or
consultants can also provide guidance. In addition, a
sampling plan that confirms total community
compliance is necessary. The plan should provide the
minimum number of samples to accurately and
statistically represent the number of installations. In
this example, seven installations were assumed to be
the sample size. If the useful life is one year or
greater, sample quarterly. For less than one year,
sample at intervals of 25 percent of useful life. At
least seven of the devices should have effluent
samples checked immediately after installation.
Continue quarterly sampling of seven of the devices
so that devices have been tested for treatment
performance throughout the first year of operation.
Quarterly sampling of a minimum of seven devices
should be continued until the useful life of all
components under normal operating conditions can
be more precisely defined. Once this more precise
useful life has been established, replacement can be
tracked based on the volume of water treated. Actual
performance sampling can then be based on
recommended sampling frequencies for central water
supplies.
If an effluent sample from a device is positive prior to
the estimated useful life, resample to confirm that
breakthrough has occurred. Replace the treatment
component of a device if breakthrough is verified.
Test an additional seven devices to determine if early
breakthrough is occurring throughout the system.
115
-------
The sampling may be reduced as community
experience increases. In most cases, it will be less
expensive to replace components prematurely than to
sample frequently enough to optimize component life.
Water meters are available with automatic shut-offs,
alarms, and even telemetry. A comprehensive treated
water volume monitoring system can be easy and
economical to establish and operate. User histories
should be established to guide meter reading
schedules.
If useful life is monitored based on gallons of water
treated, one of the more important effluent samples
for the device becomes the sample taken immediately
after installation. This sample serves two purposes: 1)
it assures that the device is operating, and 2) it can
identify media contamination. Media contamination is
rare, but has been noted on occasion (6,7).
The physical condition of the device should be
verified upon installation and an operational check of
the device should be part of the installation
procedure. Most problems with faulty devices or
installations will occur within the first few weeks of
operation (4,6,7). After this initial period, spot checks
of installations coincident with monitoring and/or
meter readings should suffice. However, homeowners
should have access to 24-hour repair service.
MAINTENANCE OF MICROBIOLOGICAL
SAFETY
The EPA requires that communities using POU/POE
devices for compliance treatment assure that the
treated water is microbiologically safe. However,
microbiological safety has not been clearly defined.
Testing for coliform organisms does not necessarily
indicate the presence or absence of other pathogens.
Heterotropic plate counts are even more ambiguous.
Heterotrophic bacteria will colonize on carbon and
other surfaces, but efforts to colonize pathogens on
carbon in the presence of competing bacteria have
not been successful (8,9). The infrequent contact of
pathogens in a water supply with a POU/POE device
should not result in a colonization of pathogens
(4,6,7). Therefore, POU/POE devices may pose no
greater risk of increased pathogens than if they were
not installed. POU/POE devices should not be used
with water of unknown microbiological quality (10,11).
A monitoring program should include heterotrophic
plate counts and coliform counts. The EPA has
determined that it is important to keep heterotrophic
plate counts below 500 per ml to reduce interference
with coliform counts (12). It has been demonstrated
that flushing (running water through) devices will
reduce heterotrophic counts (4,7,10). Consequently, it
is important to use standard water sampling methods
for microbiological analyses. These methods include
disinfecting the sample tap and running water for two
minutes prior to sampling (13), which should provide
adequate flushing.
If devices show greater than 500 organisms per ml,
using this procedure, the device or component should
be replaced. In place disinfection of a device or
component is not recommended since bacteria
colonized on carbon are less susceptible to
disinfection (9).
During the first year of operation, monitoring should
include sampling a minimum of seven devices per
quarter for microbial analyses. More sampling will be
necessary for larger communities. Standard EPA
community sampling frequencies for conforms based
on population served should be followed (12).
In addition to monitoring, preventive measures can
also be taken. Silver-impregnated devices may
provide some protection against coliform organisms,
but they will not typically reduce hererotrophic plate
counts (7,10,14). The addition of a POU/POE
disinfection device is also an alternative. Table 5 lists
disinfection technology that can be applied with
POU/POE devices (15).
Table 5. Currently Available Water Disinfection Technology
Applicable to POE/POU Treatment
Chlorination
Liquid Chemical Feeders
Other Halogens
Resin Based Brominators
Resin Based lodinators
Ozonators
Electrolytic Generation
Ultraviolet Light
Flow Through Irradiation
SUMMARY
Providing water to consumers that meets the U.S.
EPA National Primary Drinking Water Regulations
requires system organization, maintenance, and
monitoring. This is true whether central, POU, or POE
treatment is used. The goals for protection of health
are the same regardless of the method used to attain
the goals.
While POU and POE technologies are not recognized
as best available technologies in the regulations, they
are considered to be acceptable for use if specified
conditions of system control, monitoring,
effectiveness, and public health protection are
assured. Effective and reasonable monitoring and
maintenance programs can be developed to meet the
requirements of the regulations, whether within an
existing water system organization or by means of an
organization established specifically for operation and
maintenance of POU/POE systems.
116
-------
For water treatment professionals with experience
only in central treatment, it may, at first, seem
extremely difficult to achieve comparable health
protection with POU or POE devices. However, as
experience is gained with new approaches to
community treatment, POU or POE water treatment
districts may offer attractive benefits for some
communities.
REFERENCES
1. Andelman, J. Non-ingestion exposures to
chemicals in potable water. Center for
Environmental Epidemiology, University of
Pittsburgh, Pittsburgh, PA, 1984.
2. National primary drinking water regulations;
synthetic organic chemicals; monitoring and
unregulated contaminants. Federal Register Vol.
52, No. 130, July 8, 1987.
3. Bellen, G.E., Anderson, M.A and Gottler, R.
Management of point-of-use, drinking water
treatment systems. Final Report U.S. EPA
Contract R809248010. National Sanitation
Foundation, Ann Arbor, Ml, 1985.
4. Bellen, G.E., Anderson, M.A and Gottler, R.
Defluoridation of Drinking Water in Small
Communities. U.S. EPA Contract No.
R809248010. National Sanitation Foundation, Ann
Arbor, Ml, 1985.
5. Singh, G. and Clifford, O.A. The equilibrium
fluoride capacity of activated alumina. Project
summary, EPA-600/52-81-082. July 1981.
6. DeFilippi, J.A. and Baier, J.H. Point-of-use and
point-of-entry treatment on Long Island.
Journal AWWA, Vol. 79, No. 10. October 1987.
7. Bellen, G.E., Anderson, M.A. and Gottler, R.
Point-of-use reduction of volatile halogenated
organics. U.S. EPA Contract R809248010.
National Sanitation Foundation, Ann Arbor, Ml,
July 1985.
8. Geldrich, E.E. et al. Bacterial Colonizing of
point-of-use water treatment devices. Journal
AWWA, Vol. 77, No. 2. February 1985.
9. McFeters, G.A. et al. Bacteria attached to
granular activated carbon in drinking water. U.S.
EPA 1600/M-87/003. Cincinnati, OH.
10. Reasoner, D.J. et al. Microbiological
characteristics of third faucet point-of-use
devices. Journal AWWA, Vol. 79, No. 10. October
1987.
11. National Sanitation Foundation. Standard 53:
drinking water treatment units - health effects.
Ann Arbor, Ml (revised June 1982).
12. National primary drinking water regulations;
filtration and disinfection: turbidity, Giardia
lamblia, viruses, Legionella, and heterotrophic
bacteria; proposed rule. Federal Register Vol. 52,
No. 212. November 3, 1987.
13. Standard methods for the examination of water
and wastewater. 16th edition, APHA. Washington,
DC, 1985.
14. Regunathan, P. and Bauman, W.H.
Microbiological characteristics of point-of-use
precoat carbon filters. Journal AWWA, Vol. 79,
No. 10. October 1987.
15. Bellen, G.E., Gottler, R.A. and Dormand-
Herrera, R. Survey and evaluation of currently
available disinfection technology suitable for
passenger cruise vessel use. Centers for disease
control. Contract No. 200-80-0535. National
Sanitation Foundation, Ann Arbor, Ml. September
1981.
117
-------
PO1NT-OF-USEAND POINT-OF-ENTRY TREATMENT DEVICES USEDATSUPERFUND SITES
TO REMEDIATE CONTAMINATED DRINKING WATER
Sheri L Bianchin
U.S. EPA-Region V
Chicago, IL 60604
Hazardous waste is one of this nation's greatest
concerns. In response to that concern, a law was
enacted to deal with the hazardous waste problem.
This law is the Comprehensive Environmental
Response, Compensation, and Liability Act
(CERCLA), and is referred to as the Superfund Law.
This law provided broad Federal authority and
resources to investigate and to respond directly to
releases (or threatened releases) of hazardous
substances that may endanger human health or the
environment. Costs for the first five years of the
Superfund program were covered by a $1.6 billion
Hazardous Substance Response Trust Fund
established to pay for cleanup of abandoned or
uncontrolled hazardous waste sites. The law also
authorized enforcement action and cost recovery
from those responsible for the release.
CERCLA was revised in 1986 as the Superfund
Amendments and Reauthorization Act (SARA). The
purpose of the revision was to renew and strengthen
the Superfund Program. SARA reauthorizes the
program for five years and increases the size of the
fund to $8.5 billion.
SARA gives the United States Environmental
Protection Agency (U.S. EPA) the authority and
responsibility to control the actual or potential release
of hazardous substances that pose a threat to human
health or welfare or the environment. Other Federal
agencies will provide assistance as necessary during
response. A comprehensive regulation known as the
National Contingency Plan (NCP) describes the
guidelines and procedures for implementing this law.
The law, SARA, requires that hazardous waste
cleanups do the following:
* Protect human health and the environment;
• Provide for a cost-effective solution with an
emphasis on treatment and permanent destruction
over off-site disposal; and
* Compliance with all Applicable or Relevant and
Appropriate Requirements (ARAR).,
ARARs are those standards or criteria promulgated
under state or Federal law to specifically address the
abatement of contamination by a hazardous
substance, cleanup standards, or advisories.
Federal ARARs may be derived from the following:
« Safe Drinking Water Act (SDWA);
• Resource Conservation and Recovery Act (RCRA);
• Clean Water Act;
« Clean Air Act;
« Toxic Substances Control Act;
• Federal Insecticide, Fungicide, and Rodenticide
Act; and
« Great Lakes Water Quality Act.
Over two-thirds of the Superfund , actions to date,
deal with a contaminated drinking water supply.
Where SDWA Standards are applicable to the
Superfund cleanup, maximum contaminant levels
(MCLs) are usually used. A MCL is an enforceable
standard for each contaminant, which the act directs
U.S. EPA to set as close to the maximum
contaminant level goal (MCLG) as feasible. Decision
on the level of a MCL that is "feasible" includes
consideration of the best technology treatment
techniques and laboratory analyses that are available,
taking cost into consideration.
On the other hand, a MCLG is a nonenforceable
health goal. It is a numerical limit set for each
contaminant at the level at which no adverse health
effects on persons can be expected, with an
adequate margin of safety.
MCLGs may be used as cleanup criteria on a site-
specific determination. One factor in this
determination is whether multiple contaminants or
multiple pathways of exposure exist on the site.
Also important in determining the ARARs from the
SDWA is the use or potential use of the water that is
or is likely to may become contaminated.
The NCP lays out three types of responses for
incidents involving hazardous waste. These
118
-------
responses are immediate removal, planned removal,
and remedial.
A removal action is designed to be a short-term
action to stabilize or clean up a hazardous site that
poses an immediate threat to human, health or the
environment. Typical removal actions include
removing tanks or drums of hazardous substances on
the surface, installing fencing or other security
measures, and providing a temporary alternate source
of drinking water. Removals may be divided into two
categories: immediate and planned removal. U.S.
EPA's policy has been that in order for U.S. EPA to
initiate a removal action for contaminated drinking
water, the level or concentration of the contaminant
typically should exceed the 10-day health advisory.
This policy is expected to become more stringent in
the near future.
An immediate removal or a time-critical removal is a
prompt response taken to prevent immediate and
significant harm to human life or the environment. By
statute, the action must be completed within one year
and the cost of the action shall not exceed $2 million.
Immediate removals are taken to bring a release of
hazardous substances under control; they are not
intended to eliminate completely every long-term
problem. Immediate implementability is the major
consideration in choosing a remedy.
The other type of removal is termed a planned or non
time-critical removal. This type of removal is an
expedited, but not necessarily immediate, response.
A planned removal action is also limited by time and
monies involved in the cleanup.
Typically removal actions are completed by U.S. EPA,
whereby U.S. EPA will subsequently attempt cost
recovery from any identified responsible parties.
A remedial response entails a long and complicated
process aimed at identifying and completing a
permanent remedy to remediate and abate the
hazards at a site. A remedial action is designed such
that a thorough study is completed prior to the design
and construction of a selected remedy. Technical
measures can be selected only after evaluation of all
feasible alternatives on the basis of economic,
engineering, and environmental factors. Specifically
addressed in a study are the ability to protect public
health; technical feasibility; environmental
effectiveness; ability to meet ARARs; compatibility
with other Federal, state, and local laws;
constructibility; reliability; cost; and community
acceptance. The intent is to derive the maximum
benefit from Superfund as a whole. EPA can only
conduct remedial responses to those sites on the
National Priority List (NPL). The NPL is a list of the
nation's most serious hazardous waste sites.
Typically sites are identified for listing by the state. A
preliminary assessment is performed on each site.
The sites are scored by the Hazard Ranking System
(HRS). The HRS looks at potential pollutant pathways
that may reach a receptor, like the ground and
surface water pathways that may affect drinking
water. Scores greater than 28.5 are listed on the
NPL.
After a site is included on the NPL, a remedial action
is planned in a series of defined steps. These steps
are as follows:
• Remedial Investigation/Feasibility Study (RI/FS);
• Remedy Selection; and
• Remedial Design/Remedial Action (RD/RA).
A RI/FS is utilized to examine the type and extent of
the contamination, and identifies and screens the
possible remedies. Remedies selected must strive to
be of a permanent nature. When the final decision on
a remedial action or an operable unit of a remedy is
reached in the remedial process, they are
documented in a Record of Decision (ROD). The last
phase in the remedial process, the RD/RA, is the
design and construction of the selected remedy.
A Superfund action is typically funded by two
mechanisms. The first type is a Superfund-funded
remediation where no responsible parties have been
identified, or no legal agreement for the responsible
parties to conduct the work can be reached. The
second type is a Responsible Party-funded
remediation, where the responsible party pays for the
cleanup, and EPA serves to oversee the action.
The first step in any Superfund action, whether it is a
removal or remedial action, is to identify and confirm
the extent and types of contamination that exist. Next
the levels identified are compared with the standards
and health effects information. Last, the method of
correction and time required to finish the project are
determined. The remedy selected will depend upon
whether the action is a removal or a remedial action.
As a short-term response to alleviate the immediate
danger of contaminated drinking water, bottled water
is routinely utilized. This type of response is most
often used in a removal action.
In the interim, the following are commonly considered
alternatives where drinking water is a concern:
• Continue providing bottled water;
• Provide alternative water;
• Installation of Point-of-Use (POU) Treatment
Devices; and
• Installation of Point-of-Entry (POE) Treatment
Devices.
These types of responses may be used in either a
removal or remedial action.
A POU treatment device is one used at the tap to
purge the water of contamination prior to drinking. A
119
-------
POE treatment device is one used as water enters a
house to purge the water of contamination. A POE or
"whole house" treatment unit treats the entire
household water supply. POU/POE devices typically
involve aeration (air stripping), adsorption (granular
activated carbon), or reverse osmosis.
As a longer term response, or as a permanent
remedy, the following are commonly considered
remedial alternatives for a contaminated drinking
water supply:
• Connect to a community water system;
• Provide a new water source (well or surface); and
• Maintain individual treatment unit previously
installed.
These types of responses may be used in either a
removal or remedial response. Next discussed will be
case studies at two Superfund sites. The first case
that will be discussed is the Byron Johnson Salvage
yard in Byron, IL. See Figures 1 and 2 for maps of
the Superfund site investigation area.
Byron Johnson is a 20-acre salvage yard located in
a rural area of northern Illinois. The site is owned by
three individuals, and it has been determined that
domestic waste and metallic debris were deposited at
the site. It is also suspected that open dumping has
occurred. A description of the site history follows.
In the 1960s, the salvage yard was operated as a
Junk yard, where miscellaneous wastes and debris
were brought for disposal. In 1970 to 1972, the Illinois
Environmental Protection Agency (IEPA) conducted
periodic inspections to identify any operating
deficiencies. In 1972, IEPA ordered closure of the
salvage yard. The salvage yard ceased operation in
1974. In December 1982, the site was placed onto
the NPL In May 1983, under agreement with U.S.
EPA, the IEPA performed a state-lead RI/FS. This
study specifically focused on contamination directly
on or below the site; ground water contamination
potentially emanating from the site was not
addressed.
Through 1984 and 1985, the U.S. EPA, IEPA, and the
Illinois Department of Public Health (IDPH) continued
to monitor the contamination levels in residential wells
located nearby and down-gradient from the site.
Through periodic sampling, off-site ground water
contamination by volatile organic compounds (VOCs)
was documented. It was found that private wells
contained trichloroethylene (TCE) in concentrations
up to 710 ug/l. In June 1984, the IEPA completed a
RI/FS and signed a ROD to remove drums of waste
and contaminated soil from the site. In July 1984, the
U.S. EPA placed the residents whose water exceeded
200 ug/I in the Dirk Farm area (i.e., those residents
along Acorn and Razorville Roads) on bottled water
as a temporary measure. Late in 1984, the U.S. EPA
contracted to have a RI/FS performed at the site. In
July 1985, a U.S. EPA action was started to augment
the data collected from the IEPA RI/FS.
In October 1985, U.S. EPA conducted a phased FS
to expand the scope of the study to the Rock River
Terrace subdivision which is about one and one-half
miles down gradient from the site. The objective of
the study was to investigate the potential health threat
due to the exposure to the contaminated water supply
and evaluation of alternative water supply and
treatment options that would ensure a safe water
supply to Rock River Terrace residents. Sampling
results indicated that the ground water was
contaminated with levels of TCE up to 48 ug/l TCE.
Although the level detected was below the 10-day
health advisory, it is above the drinking water
standard of 5 ug/l. Also, since the residents are
located in the direction of the contaminant plume, it
was determined that a planned removal action was
warranted.
In May 1986, U.S. EPA, by a removal action, installed
carbon adsorption POU treatment devices for those
residents on bottled water in the Dirk Farm area as an
interim measure to remove TCE from the water.
In June 1986, U.S. EPA completed a study that
focused on the potential ground water problem for the
Rock River Terrace. It was determined that both of
the major aquifers in the area were contaminated to
some extent by VOCs. This contamination extends to
outlying locations 0;8 km (1 mi) northwest and 0.8 km
(1 mi) north of the site. In addition, slight
contamination by cyanide and some inorganic
compounds exist in the groundwater beneath the
salvage yard. In addition, the study identified and
evaluated alternatives for replacing or treating
contaminated water from private wells. Based upon
the RI/FS, the alternatives for treating or replacing
water from Rock River Terrace wells were narrowed
down to the three alternatives listed below, and a
detailed analysis was conducted on each one.
Alternative 1
Connection to the Byron Municipal Facility. This
alternative was estimated to take one to two years to
complete, and was estimated to cost approximately
$900,000.
Alternative 2
Supply bottled water to homes with contaminated
wells. This alternative would not provide water for
bathing and washing. The annual cost of this
alternative was estimated at $91,150.
Alternative 3
Treatment of water from affected wells to remove
contaminants through carbon adsorption. It was
estimated that it would cost $26,000 to install POU
treatment devices and $115,000 to install "whole
120
-------
Figure 1. Location of Byron Johnson salvage yard, Byron, IL. investigation area.
r
T
ToF
PROJEC
LOCATION-
I—South Branch //if
Woodland Cre«k ff]
II / // i ^..J
BYRON SALVAGE YARD ^l
POWER PLANT
BYRON NUCLEAR
IRlt'S FARM
SCALE: 1"-4500'
north
121
-------
Figure 2. Supcrfund site investigation area, Byron, IL.
8UPERFUND STUDY AREA
house" POE treatment devices. Upon
commencement of construction, it would take
approximately two to three months to install the units.
The results of the study indicated Alternative 3 to be
the most economically feasible while providing a safe
and reliable drinking water supply for affected
residents. The whole house units would be installed
at Rock River Terrace homes that are occupied on a
year-round basis. Periodic monitoring would be
conducted to ensure that contaminants are being
effectively removed. The carbon for these units would
be replaced when necessary. Since the carbon
replacement rate is dependent on many factors
including the level of contamination, water
temperature, pH, chemical makeup of the water, and
water usage, monitoring of the carbon bed is
necessary.
In July 1986, U.S. EPA initiated a monthly sampling
program for the residents in the Dirk Farm area to
monitor the efficiency of the POU treatment devices.
The units will be replaced when necessary.
In July 1986, IEPA signed a ROD for design and
construction of a municipal water line to distribute
potable water from the City of Byron municipal water
supply to the residents at Rock River Terrace and
Dirk Farm areas (Acorn and Razorville Road
residents). This action along with a monitoring plan
122
-------
constitutes the permanent remedy to the
contaminated water problem.
In September 1986, U.S. EPA issued a ROD to install
carbon adsorption POE "whole house" treatment
devices in the Rock River Terrace. The POE systems
were either placed in an outdoor insulated shed or in
the basement. The units were engineered and
installed by North American Aqua, Vandalia, Ml. Each
system consists of a 0.5 urn (2 x 10-5 in) prefilter
followed by two granular activated carbon (GAG) units
in series. Each GAC contains 50 kg (110 Ib) of
Calgon carbon. Each unit is 137 cm (54 in) tall and is
designed for a flow of 0.47 l/s (7.5 gpm). The system
is monitored on a monthly basis before and after each
carbon tank to assure the efficiency is maintained.
Upon breakthrough the carbon will be replaced. To
date, sampling results have shown no breakthrough
of the carbon.
The next Superfund site that will be discussed is the
Main Street Well Field and associated actions in
Elkhart, IN. See Figures 3, 4, and 5 for maps of the
Superfund site investigation areas.
Elkhart is located in North Central Indiana at the
confluence of the St. Joseph's and Elkhart Rivers.
The population of Elkhart is approximately 65,000.
The city is diversified in the manufacturing operations
that the city is known for; especially Pharmaceuticals,
band instruments, recreational vehicles, and
injection-molded plastics.
The surface geology of Elkhart consists of a typical
glacial deposit created from various types of sand and
gravel, forming an extensive outwash aquifer
permeating up to 53 m (175 ft). An intermediate
nonpermeable clay bed confines a deeper aquifer.
To date at least five separate Superfund actions have
been taken in and around Elkhart, IN. One of these
actions was a remedial-type response, and four of
the actions were removal-type responses. Although
all are referred to as the Main Street Well Field, each
removal is separate and distinct from the actions at
the actual North Main Street Well Field Site listed on
the NPL.
The remedial action is referred to as the North Main
Street Well Field. Through routine monitoring, ground
water in 9 of the 17 wells at the Municipal Facility
were found to be contaminated with approximately 95
ug/l TCE. The site was added to the NPL on
December 1982. Through the Superfund process,
U.S. EPA and Indiana Department of Environmental
Management (IDEM) decided to install packed air
stripping towers at the municipal city water utility to
meet the drinking water standard.
In the fall of 1987, Calgon constructed the three 17-
m (55-ft) air stripping towers, while the U.S. Army
Corps of Engineers supervised the construction. Each
concurrent flow tower is 3 m (10 ft) in diameter and
contains 9 m (30 ft) of polypropylene packing media.
An estimated 19 to 23 million I (5 to 6 million gal) of
water are treated per day. The total cost for
construction is $2.5 million. The annual O&M cost is
estimated to be between $81,000 and $106,000.
In addition to the remedial action, at least four
separate Superfund-related removal actions have
been taken. Two of those actions have been referred
to as the East Jackson area and County 1 area.
The contamination in the East Jackson area was first
recognized in the fall of 1984, when a citizen sampled
his well water. Results of the sample analysis
exhibited levels of TCE above 200 ug/l. These levels
exceeded the 10-day health advisory of 200 ug/l.
When the County Health Department was contacted
with the information, it initiated an extensive sampling
program. When the County Health Department
confirmed widespread contamination, which was more
extensive than anticipated, it contacted the U.S. EPA.
In May, 1985, U.S. EPA conducted extensive
sampling of the area whereby over 500 samples were
collected. The results of the sampling program
showed that heavy contamination existed, in the East
Jackson area where over 80 wells were found to have
ground water contaminated TCE in excess of 200
ug/l, and 15 of these wells contained levels of TCE
above 1,500 pg/l TCE.
Representatives from the Center for Disease Control
(CDC) advised U.S. EPA that contamination greater
than 1,500 ug/l is unfit for bathing and other
household uses because of the inhalation and
absorption dangers. As an immediate short-term
remedy to the contaminated water, U.S. EPA placed
approximately 800 residents on bottled water for
drinking purposes within 36 hours.
U.S. EPA initially decided to extend the water main to
the 15 homes with the highest levels of
contamination. However, the U.S.EPA decided to
blanket the area with an alternative source of water,
due to the severity of contamination, its extent, and
direction of ground water flow. In October 1985, the
U.S. EPA on scene coordinators (OSCs), Jack
Barnette and Ken Theisen, were charged with the
responsibility of coordinating the installation of
Additional footage of water main. In total,
approximately 4,420 m (14,500 ft) of water mains
were provided to the town, and installed at some 300
homes and businesses. Construction of the first
group of mains, amounting to 884 m (2,900 ft), was
completed on December 1985. However, since
various delays were experienced, the project was not
completed until September 1986.
In addition, at 11 homes where the water exhibited
minor contamination, POU devices were installed
123
-------
Figure 5. Investigation area, Elkhart, IN.
•a
Q.
id
126
-------
because these homes were not adjacent to the water
main.
Since the area of contamination extends over 3.2 km
(2 mi), it is believed that the contamination originated
from more than one source. It is suspected that one
of the potentially responsible parties (PRP) is the
former Accra-Pac site where 13 underground
storage tanks were removed by U.S. EPA.
The other removal action near the western end of
Elkhart that will be discussed is the County Road 1
area in Osceola.
In June 1986, a resident analyzed the water and
found contamination of TCE at approximately 800
ug/l, exceeding the 10-day health advisory of 200
ug/l, and contamination of 480 ug/l of carbon
tetrachloride, exceeding the 10-day health advisory
of 20 ug/l.
During July and August through extensive sampling,
U.S. EPA tracked the contaminant plume from the
County 1 area to its discharge into the St. Joseph
River. The sampling also indicated levels of TCE as
high as 5,000 iig/l, and levels of carbon tetrachloride
as high as 7,500 ug/l. As an immediate response to
the contamination in this area, U.S EPA provided
bottled water to all affected residents, and advised
them of the risks. In addition, the worst homes were
advised not to use their water for any reason.
U.S. EPA decided that it would take too long to hook
these residents to the municipal water supply and
therefore decided to equip the affected residents with
POE treatment devices. In the County 1 Area, U.S.
EPA installed whole house activated carbon treatment
devices onto 54 homes, and installed POU treatment
devices for 22 homes located outside the
contaminant plume. The last filter was installed on
April 3, 1987. All units were engineered and installed
by North American Aqua. U.S. EPA also initiated an
extensive monitoring program, whereby the POU
treatment devices are periodically sampled and
analysis of the water is performed to check the
efficiency of the treatment devices. IDEM has
pledged to sample the affected homes and advise the
homeowners.
It is suspected that the Conrail Yard south of the
County Road 1 area is a potentially responsible party
for the source of contamination. The site is being
scored for the NPL.
Following are some of the future developments in the
area of POU/POE treatment devices used at
Superfund sites.
U.S. EPA has funded a pilot project in the East
Jackson area of Elkhart, IN. As a prototype, two of
the homes with contaminated water were equipped
with a packed air stripper, along with two GAG units
in series. The air stripper was placed in the basement
with the GAC units and it is vented outside. The air
stripper is manufactured by Tykk. The unit has a 40:1
air to water ratio, and operates at a rate of 0.32 l/s (5
gpm). The air stripper is packed with 2.5 cm (1 in)
diameter polypropylene cylinders. The cost of the
POE GAC unit with air stripper is approximately
$4,000. North American Aqua recommends flushing
the system anytime when water has had to stand for
more than a day without use. U.S. EPA is considering
putting on an ultraviolet light for potential bacteria
problems because it is suspected that a potential
exists for buildup on the media. Since U.S. EPA is
still gathering .data on the system, no formal results
are yet available.
U.S. EPA Hazardous Waste Environmental Research
Laboratory (HWERL) has funded a 12-month study
for the purpose of producing a Guidance Manual for
OSCs to use POE treatment devices. The study is
expected to consist of first collecting POU/POE data
from various Superfund sites, and following up with a
pilot study to fill in the missing information. Frank Bell
and James Goodrich are the U.S. EPA technical
advisors.
Another project in the works is the National Register
, for Drinking Water Treatment Technology (The
Register). Working on this project are James
Goodrich of U.S. EPA's Drinking Water Research
Division, along with Harry VonHuben and Sheri
Bianchin of U.S. EPA Region V, Drinking Water
Section. The purpose of the project is to create a
National Register or data base of nontraditional,
innovative water treatment systems which are being
utilized to treat contaminated drinking water supplies.
The majority of these technologies are being used for
water supplies that have been adversely affected by
Superfund sites. Additionally, Region V will serve as
the test region for the Register's development.
The objective of the project is to systematically
collect, organize and disseminate information on
those treatment technologies which have been
already implemented on either a pilot or full-scale
basis at the affected supplies. The data base
information will include: manufacturers and designers
of the units; the design specifications; a comparison
of the design performance versus the actual
performance, capital and operation and maintenance
costs; and a follow-up on operation problems and
benefits, among other items. Presently data collection
questionaires are being developed. The project then
entails soliciting data through the U.S. EPA Regional
offices. The data will be stored on a personal
computer (PC) at the U.S. EPA Research Division,
Cincinnati. Additionally, each region will be provided
with a diskette for its own use and; the National
Technical Information Service (NTIS) will have a
paper copy of information. We anticipate that the
127
-------
summary report and the PC diskette will be available
by the close of 1988 for Region V.
In conclusion, the Superfund program is a significant
part of our national response to one of the major
environmental challenges of the decade. The program
is a coordinated effort of Federal, state and local
governments, private industry and citizens. However,
since the problems are widespread and each is
unique, new and existing technology is needed to
remediate the hazards. In a field where no clear cut
answers exist, the use of POU/POE, an available
technology, has gained more acceptance in the
usage and remediation of hazards where drinking
water is a problem.
128
-------
NEW DEVELOPMENTS IN POINT-OF-USElPOINT-OF-ENTRYDRINKING WATER TREATMENT
Gary L. Hatch
Ametekjnc.
Sheboygan.WI 53081
INTRODUCTION
Over the last several years the U.S. EPA has
promulgated National Primary Drinking Water
Regulations for a number of specific drinking water
contaminants, the latest of which is a group of volatile
organic contaminants (VOCs) (1). These regulations,
as established by the Safe Drinking Water Act, have
provided impetus for the development and application
of point-of-use (POU) and point-of-entry (POE)
drinking water treatment technology to help solve
small community water systems' and individual home
owner's water contamination problems. The above
referenced action by the EPA actually allows the use
of POE technologies (with conditions) as acceptable
means of compliance with the VOC regulations. POU
technologies may also be used, but under more
restricted conditions.
BACKGROUND
New developments in POU/POE technology primarily
have been for contaminant-specific applications. For
example, radon has come to national attention
recently as being a serious health threat to private
well owners. This problem is now being solved by
new POE systems that employ old or well-known
technologies, such as the use of activated carbon and
aeration. Lowry (2) has conducted many studies to
demonstrate the effectiveness of granular activated
carbon systems and aeration systems for removing
high levels of radon from drinking water.
Other new developments have been in the area of
water disinfection. In the last few years, the well-
known technologies of ^ultraviolet light and ozonation
have been designed into packaged systems hardware
for POU/POE application (3-6). Of these, the new
POE ozonation systems that are becoming
commercially available may be the most promising
from the standpoint of overall treatment capabilities.
POE ozonation systems combine the technologies of
filtration and adsorption to provide for effective
removal of some inorganic and organic contaminants
while at the same time providing for disinfection
against bacteria, viruses, and protozoan cysts. Rice
(4-6) has given a detailed description of the use of
ozone and ozonation systems for POU/POE
applications.
During the last 5 to 10 years, the technology of home
reverse osmosis drinking water treatment systems
has advanced to become a viable treatment method
for reducing certain health-threatening inorganic
contaminants. These systems are most applicable for
reducing nitrate, fluoride, and arsenic. POU reverse
osmosis systems have been tested successfully for
the reduction of fluoride and arsenic in a small New
Mexico community (7). Proper system and -membrane
application is very important because various
membranes (e.g., cellulose acetate versus thin-film
composite) have different performance characteristics
(8), especially where nitrate reduction is concerned.
In the last 10 to 20 years, a relatively new approach
to using halogens for water disinfection has been
developed. This technology is based on the
combining of the halogen (in the form of a polyhalide)
with an ion exchange resin. The remainder of this
paper will provide a detailed look at the history,
development, and uses of the halogenated resins for
POU/POE water disinfection.
HALOGENATED RESINS FOR WATER
DISINFECTION
H/STORY OF DEVELOPMENT
Initial work (9) in the field of halogenated resins began
in 1957 when simple ion association experiments
were conducted by adding bromine and iodine to
anion exchange resins. This work revealed that these
resins have an unusually high affinity for halogens,
especially for iodine. Approximately 10 years later, the
first halogenated resins (10) were developed to
control microorganisms in water. Since then a
number of improvements and modifications have
been made in the formulations (11-17) of
brominated and iodinated resins to enhance their
chemical characteristics and their anti-
microbiological performance.
The initial and primary application of the original resin
systems was to use the resin as a way of metering
the halogen (e.g., bromine or iodine) into a stream or
129
-------
body of water. The halogen, then, was still used in
the conventional way (as a residual) for controlling
microorganisms.
A major discovery indicating the potential for using
halogenated resins in point-of-use water
disinfection was made by Mills (11) in 1969. This
discovery was that when microbiologically
contaminated water is passed through a bed of the
halogenated resin, an instantaneous and complete kill
can be achieved. Therefore, a reservoir for holding
the water and the required contact time needed for
conventional residual disinfection action is not
necessary when using a properly designed
halogenated resin water disinfection system. The
superfluous halogen residual -released by the resin
then can be removed or deactivated if desired,
immediately upon emerging from the bed.
Later discoveries fay Lambert and Fina (18) have not
only helped to explain the unique mechanism that
provides for the instantaneous disinfection, but have
shown also that halogen release by the resin is not
necessary to achieve the instantaneous
microorganism kill. Their studies with an insoluble
triiodide resin have shown that kill is achieved through
physical contact of the organism with the resin beads,
and that the disinfecting quantities of halogen are
produced virtually on demand.
HALOGENS
The halogens are a group of chemical elements
comprised of fluorine, chlorine, bromine, iodine, and
astatine. Of these, only chlorine, bromine, and iodine
are used in water disinfection, and only bromine and
iodine are capable of forming the polyhalides, which
bind to the anion exchange resins. Table 1 illustrates
(he polyhalides formed from bromine and iodine and
how an aqueous triiodide solution is made. The higher
polyiodides ((5" and (7") are made simply by adding
the correct stoichiometric amount of iodine to the
sodium iodine along with a very critical and minimal
amount of water).
Tablo 1. Halogens Most Capable of Forming Polyhalides
• Bcomino
• lodino
Exampto;
ra, Brs". Bf>"
', is'.'/
I2 •*• Nal
lodino Sodium Iodide
H2O
Na+ + I3'
Water Solution of Sodium Triiodide
The corresponding polyhalide resins are made by
addition of the aqueous polyhalide solution to the
anion exchange resin (usually in chloride form) under
very carefully controlled conditions.
RESINS
In general, ion exchange resins consist of two main
types - cation exchange resins (those that exchange
positively charged ions, such as calcium [Ca + 2] for
sodium [Na + ] in the water softening process, as
shown in Figure 1); and anion exchange resins (those
that exchange negatively charged ions, such as
triiodide [la"], for chloride [Cl"], as shown in Figure
2). These resins are usually made from the
polystyrene polymer backbone and differ only by their
specific functional groups.
The cation exchange resin contains the negatively
charged sulfonic acid functional group: R-SOs",
where R is the polystyrene backbone. These
negatively charged functional groups attract and hold
on to the positively charged cations. Depending on
their relative concentrations and relative affinities for
the sulfonic acid functional site, different cations can
exchange with others, as depicted in Figure 1.
The anion exchange resins used for making the
halogenated resins also usually have the polystyrene
backbone, but have the positively charged
quarternary ammonium functional group: R-
CH2N + (013)3, where R is, again, the polystyrene
backbone. Here, the positively charged functional site
holds the negatively charged anion. Anion exchange
occurs when the relative affinity for one anion wins
out over another, such as in Figure 2 where the
resin's affinity for triiodide is much greater than for
chloride.
When this triiodide resin is made properly, virtually
nondetectable levels of iodine are found in the post-
column water effluent. Furthermore, when this resin is
challenged with water high in salt content, such as
with chloride or sulfate, no triiodide exchange occurs
and only low levels of iodide ion (I") are found in the
effluent (13). Table 2 shows how effective this resin is
against five different kinds of bacteria and the
polydma virus (13,19). Other halogenated resins also
demonstrate highly effective anti-microbial action as
depicted in Table 3 (11,14,17).
Another type, of resin, polyvinylpyridine, has been
used to make halogenated resins (12,17). In this
resin, the functional group attached to the vinyl
polymer backbone is the pyridine molecule (see
Figure 3). The unique feature of this resin is that the
neutral (no ionic charge) functional group has a high
affinity for the free halogens, specifically iodine and
bromine. Therefore, when making halogenated resins
with polyvinylpyridine, the free halogen need only be
used. Use of the halide salt (Br~ or I*) to make the
negatively charged polyhalide is not necessary. The
halogenated polyvinylpyridine resins exhibit anti-
microbial action similar to the halogenated anion
exchange resins (see Table 3).
LIMITATIONS
Many of the variables that adversely affect the
conventional water disinfection methods also affect
130
-------
Figure 1. Softening process with sodium-form cation exhange resin.
-fCH-CH2-)-
so:
Na+
Ca
+2
Figure 2. Anion exchange resins used in making halogenated resins.
CH2
(CH3) N
(CH3) Nt
n
+2Na
.+
-fCH-CH24-
n
+CI
Cl
U.S. Patent 3,817,860
Lambert & Fina
131
-------
Tabto 2. Anti-Bacterial Efficiency of the Trioidide Resin
Organisms per ml
Organism
Salmonella typhtmlurium"
Escherichla coif
Pseudomonas aeniginosa"
Staphylococcus aureusf
Streptococcus faecalisr
Polyoma virus"*
Feed
1 X105
3x105
1.3x105
1.8x104
1.1 X104
1.2x106
Effluent
0
0
0
0
0
0
* Data from reference 13. 6.5 ml of resin at 20 ml/min flow rate.
" Data from reference 19.30 ml of resin at 30 ml/min flow rate.
TabloS. Anti-Bacterial Efficiency of Various Halogenated
Resins
£ co// per ml
Rosin
Feed
Eff.
Halogen
Residual in
Effluent
Pofybromida, 5% Btz as Bra""
Mixed-form, potyiodide""
Mixed-form, lodine/Bromine-
polyvinylpyrfcline"~
1 x 10s 0 2 mg/l as Br2
3 x 105 0 1 mg/l as I2
9x105 0 10 mg/l as l2
* Data from reference 11.50 ml of resin at 57 ml/min.
"* Data from reference 14.50 ml of resin at 185 ml/min.
•"* Dala from reference 17.50 ml of resin at 140 ml/min.
the disinfection process of the halogenated resins.
Table 4 lists these major limitations as well as some
other concerns of halogenated resins that may limit
their uses and applications.
In a conventional disinfection process, such as a
municipal treatment plant where chlorination is used,
the water usually is subjected to a series of
pretreatment and in -some cases, post-treatment
processes. Many water treatment plants pretreat the
raw water by using methods such as flocculation,
sedimentation, and/or filtration. Where necessary, pH
adjustment can also be done. These pretreatment
steps are necessary to insure that the optimum
disinfection conditions are met. If low temperature or
halogen demand dictates, higher levels of disinfectant
can be added.
For halogenated resins, the water entering the bed
also must be of a reasonable quality such that the
disinfection process is not jeopardized. Therefore, a
preapplication water analysis must always be
conducted so that adequate pretreatment needs can
be determined.
pW
If high pH is encountered (above 9), a preacidification
step must be incorporated into the treatment system
to lower the pH to betwen 7 and 8. Figure 4 shows
the effects of high pH on the triiodide resin (20). At
above pH 9, iodinated resins begin to release high
levels of iodine that can diminish the kill efficiency
and life of the resin bed, as well as stress and
shorten the life of an iodine scavenger system.
Lowering the pH to less than 5 is not recommended
because of evidence that at low pH virucidal activity
of the halogenated resins is diminished (19).
HIGH TDS AND HALOGEN DEMAND
The potential problem of extremely high TDS (1,000
to 15,000 mg/l or greater) could be addressed by
pretreatment with demineralizing resins or reverse
osmosis. These high levels of TDS could promote
additional ion exchange of halide ions and halogen
which, again, would result in lowering the kill
efficiency and life of the resin. Similar pretreatment
measures can be taken against fnorganic
contaminants that create halogen demand (e.g.,
sulfide or sulfite).
TEMPERATURE
Low operating temperature can reduce the
antimicrobial efficiency of the halogenated resins (21).
However, this is presumed not to be as critical as in
the conventional disinfection process where the
"concentration X time" constant must be maintained
to assure adequate disinfection (22). The residence
time within the resin bed can be increased by
enlarging the bed size or lowering the flow rate to
assure disinfection. Ideally, any halogenated resin
disinfection system should be designed to operate at
the anticipated minimum temperature. Unusually high
operating temperatures such as 90 to 100°F (32 to
38°C), which may be encountered in the tropics,
would most likely enhance the disinfecting action, but
cause a slight increased release of halogen.
RESIN FOULING
Since the kill mechanism relies on physical contact
(or very near contact) of the microorganism with the
resin beads or particles, adequate protection against
resin fouling or coating of the beads is a must. The
effects of resin fouling are evident even in a simple
softening or demineralizing process. Resin fouling or
coating by iron floe or organic material, will prevent
the dissolved ions from contacting or entering the
resin beads, and thereby preclude the softening or
demineralizing action. This same phenomenon would
likewise preclude the disinfecting action of the
halogenated resins. Obviously, the potential for resin
fouling is an extremely critical factor that must not be
overlooked when considering halogenated resins for
water disinfection.
PROTOZOAN CYSTS
Any water disinfection process must be totally and
reliably effective against all disease-causing
organisms. A major limitation of some halogenated
resins is that they are not effective against certain
types of protozoan cysts, specifically Giardia lambda.
132
-------
Figure 3. Most recently developed halogenated resins use neutral (non-charged) resin.
-fCH-CH24-
n
O) + Br2 + I.
'NT
Polyvinylpyridine
-fCH-CH_)-
n
\/Brz
U.S. Patent 4,594,392
Hatch
Table 4. Limitations of Halogenated Resins
pH 9 (max.)
High TDS (approx. 1,500 mg/1)
Temperature
Resin fouling - iron, TOG, turbidity
Halogen demand
Resistance of Giardia or
amoebic cysts
Physiological concerns of
iodine and bromine
Monitoring
Tests on the triiodide (13") and penta-iodide (Is")
resins have shown that only the pentaiodide resin is
effective against Giardia (23). Fortunately, since these
organisms are relatively large (typically 7 to 15 nm
[0.0003 to 0.0006 in]), they can be physically
removed by adequate pre or post-filtration thereby
precluding total reliance on the halogenated resin for
a cyst kill.
PHYS/Oi-OG/CAL CONCERNS
Since iodine does have an effect on thyroid
metabolism, it is not recommended for long-term
continuous use (24). Therefore, the best application
of halogenated resins containing iodine would seem
to be in portable emergency devices where use is
short-term or intermittent. Several such devices are
currently on the market.
The U.S. EPA does indicate that iodine could be used
for long-term continuous application if an adequate
post-treatment scavenger system were employed to
remove any iodine species from the water (24).
Physiological concerns with bromine are less than
those with iodine, but some limitations on bromine
levels in drinking water are recommended (25).
MONITORING
Any water treatment device or process that has a
health effects claim should have a reliable and
Figure 4. Effect of pH on iodine release from triiodide resin.
ppm Iodine
60 -
40 -
20 -
7 8
PH
10
11
convenient way of letting the user know when
exhaustion occurs or servicing is required. This is an
obvious understatement when dealing with
microbiological purification of water. One way to
accomplish performance indication or lack thereof is
to incorporate an automatic shut-off or remote
sensing metering device into the system. The
National Sanitation Foundation Standard No. 53 (26)
lists other alternatives.
133
-------
SYSTEM APPLICATIONS
The extremely efficient, once-through disinfecting
action of the halogenated resins make them ideal
candidates for use in small-scale point-of-use
water disinfection systems. Other than previously/
mentioned portable emergency water disinfection
devices, numerous single-tap applications are
possible. Systems or units with replaceable
cartridge-like containers of the resin could be used
where drinking water is supplied for transient
populations, such as for water fountains in parks and
other entertainment facilities, or public buildings.
Multi-tap or high volume drinking water needs could
be met by employing the resin in multiple-cartridge
housings or in a single tank-type container such as
a softener resin tank. As previously indicated for
long-term use, adequate post-treatment measures
must be taken to remove or reduce the halogens to
an acceptable level.
Of course, as cautioned before, the critical water
quality parameters must be predetermined to aid in
system design, and most importantly, adequate
monitoring and servicing is a must.
CONCLUSIONS
Use of halogenated resins in point-of-use water
disinfection can offer a unique opportunity to help
solve some of the current problems associated with
conventional small-scale chlorination disinfection
systems. Some advantages are:
• No electrical connections (depending on monitoring
requirements);
• Easy to install and service - no handling or
storage of chemicals;
• Virtually no "down" time;
* Minimal space requirements;
• More efficient use of disinfectant; and
* Potential to integrate well with other point-of-use
technology.
However, even though these resins have remarkable
disinfecting capabilities, proper design, thorough
testing, and prudent application are absolutely
necessary to assure maximum reliability of systems in
which they are used. Currently available point-of-
use water treatment technology can be properly
combined with halogenated resins to offer another
alternative for providing microbiologically safe drinking
water.
All areas of POU/POE water treatment technology are
advancing rapidly. Much of this technology is based
on newly engineered systems, which by redesigning
existing or well known technologies, can solve almost
any water contamination problem. As revealed in the
newly passed Safe Drinking Water Act, the U.S. EPA
now recognizes that fact. The accompanying
technologies of monitoring and serviceability are also
advancing rapidly. As advances continue, the
POU/POE industry can meet the challenge of
combining reasonable cost, reliability, monitoring, and
serviceability to provide safe, reliable, and economical
methods of drinking water treatment for the consumer
who does not have the option of central treatment.
REFERENCES
1. U.S. Environmental Protection Agency. National
primary drinking water regulations - synthetic
organic chemicals; monitoring for unregulated
contaminants; final rule. Federal Register 50:130,
25690, July8, 1987.
2. Lowry, J. D. et al. Point-of-entry removal of
radon from drinking water. JAWWA. 79:4, 162,
1987.
3. Foust, C. Performance and application of UV
systems. Proceedings. Conference on Point-
of-Use Treatment of Drinking Water. U.S.
EPA/AWWA, Cincinnati, OH. October 6-8, 1987.
4. Rice, R.G. Ozone for point-of-use/point-of-
entry application, Part I. Water Technology. 10:3,
22, May 1987.
5. Rice, R.G. Ozone for point-of-use/point-of-
entry application, Part II. Water Technology. 10:4,
28, June 1987.
6. Rice, R.G. Ozone for point-of-use/point-of-
entry application, Part III. Water Technology. 10:5,
27, August 1987.
7. Rogers, K. Community demonstration of POU
systems. Proceedings, Conference on Point-
of-Use Treatment of Drinking Water. U.S.
EPA/AWWA, Cincinnati, OH. October 6-8, 1987.
8. Slovak, J. and Slovak, R. Developments in
membrane technology. Water Technology, 10:5,
15, August 1987.
9. Aveston, J. and Everest, D.A. Chem. Ind.
(London). 1238, 1957.
10. Mills, J. F. U.S. Patent 3,316,173. April 25, 1967.
11. Mills, J.F. U.S. Patent 3,462,363. August 19,
1969.
134
-------
12. Katchalski, E. et al. U.S. Patent 3,565,872.
February 23, 1971.
13. Lambert, J. L. and Fina, L. R. U.S. Patent
3,817,860. June 18, 1974.
14. Hatch, G. L. U.S. Patent 4,187,183. February 5,
1983.
15. Gartner, W. J. U.S. Patent 4,420,590. December
13, 1983.
16. Beauman, W. H. et al. U.S. Patent 4,594,361.
June 10, 1986.
17. Hatch, G. L. U.S. Patent 4,594,392. June 10,
1986.
18. Lambert, J. L. and Fina, L. R. Proceedings,
Second World Conference, International Water
Resources Association, Vol. II, pp. 53-59, New
Delhi, 1975.
19. Hassouna, N. Doctoral dissertation. Kansas State
University, 1973.
20. Hatch, G. L. et al. Ind. Eng. Chem. Prod. Res.
Dev. 19, 259, 1980.
21. Kao, I. C. et al. J. Ferment. Techno). 51, 159,
1973.
22. Regunathan, P. and Beauman, W. H. Fourth
Domestic Water Quality Symposium. Water
Quality Assoc. and Amer. Soc. of Agricultural
Engineers, Technical Papers, p. 54, Chicago,
1985.
23. Marchin, G. L. et al., Appl. Envtl. Microbiol. 46:5,
965-9, 1983. r
24. Cotruvo, J. &.' Policy on iodine disinfection.
Memorandum to G.A. Jones. March 3, 1982.
25. Drinking Water and Health, Vol. 3, pp. 181-187.
National Academy Press, Washington, DC, 1980.
26. National Sanitation Foundation Standard No. 53.
Drinking water treatment units - health effects,
N.S.F., Ann Arbor, Ml, June 1982.
135
-------
POUlPOE POINT OF VIEW
Frank A. Bell, Jr.
Criteria and Standards Division
Office of Drinking Water
U.S. Environmental Protection Agency
Cincinnati, OH 45268
INTRODUCTION
I would like to preface my remarks by saying that I
will not be addressing official Environmental
Protection Agency (EPA) policy, since that has
already been covered by Steve Clark. My remarks will
also not cover pesticide regulatory programs in EPA,
since that has also been covered by another speaker.
Instead I will address two basic positive options for
handling the broad field of POU/POE water treatment
in terms of claims control and consumer service.
OPTION #1: A DIRECT REGULATORY
PROGRAM
From my understanding, such a program would
involve a verification of claims being made by the
various industry segments. To be universally
applicable, such a program would have to identify and
circumscribe acceptable claims and then have units
tested against the claims. This would be an incredibly
complex operation. It also might be considered a
restraint of trade. Some questions that it would
present are:
• How would claims be limited?
• Would the direct regulatory program deal only with
physically or chemically verifiable claims? If so,
would noncovered units be exempted, such as
magnetic treatment units and good-health units
(ones that treat the water to improve its good-
health qualities without any change in chemistry)?
If such units are exempted, then a substantial area
of consumer concern will remain unattended.
• Would nonhealth or esthetic treatment units be
covered? If yes, then the magnitude and
complexity of the program would become serious
problems, since from a review of water
conditioning treatment (1) some 26 possible
esthetic effects/treatments emerge.
If, on the other hand, esthetic treatment units are
excluded, what will be done about the incidental
health effects claims of the esthetic units? For
example, a taste and odor activated carbon filter may
also remove 10 to 20 percent of the trihalomethanes
(significant, but not enough to be considered
adequate from a health standpoint). If the 10 to 20
percent reduction claim is allowed, how can it be
described so as to inform the consumer properly?
• The currency question will provide further
enormous problems. Manufacturers are constantly
changing their products. How will these changes
be monitored? How will retests and quality control
be verified? Who will certify treatment capabilities
for the manufacturers' new products? For a
government agency to attempt to regulate, monitor,
and control the entire water quality industry would
represent an enormous and continuing
bureaucratic effort.
CONCLUSION
While a direct regulatory program has some appeal
from a simplistic viewpoint, it can introduce a host of
hard-to-answer questions, and in the end, provide
more problems than solutions.
OPTION #2: COOPERATIVE INDUSTRY/
GOVERNMENT/THIRD PARTY EFFORT
Historically, EPA has vigorously supported this
second option. I have personally participated in
several productive efforts that have involved many
cooperative parties, including:
• The Gulf South Research Institute's landmark
study of over 30 commercially available activated
carbon and other units for organic chemical
reduction capabilities (2). Before this study, no
authentic, independent information on unit
capabilities existed. Industry cooperated in protocol
review and development, and in the conduct and
review of the study and its results. Industry
136
-------
technical representatives helped us to avoid doing
unwise things and in having a valid, widely-
accepted study result.
• Establishment of third-party standards and listing
program under the National Sanitation Foundation
(NSF). NSF has provided leadership Jo bring
industry, government and other interests-together
for establishment of a number of drinking water
treatment unit standards, with three currently in
effect: Standard 42 (Esthetics), Standard 53
(Health), and Standard 58 (Reverse Osmosis), and
with two other standards nearing .adoption.
Strengths of an NSF program include the
standards consensus development process (giving
all interests a chance to be heard) and its testing,
monitoring, and control capabilities. NSF has a
continuing presence and mechanism to keep up
, with new product developments and control of
i product quality over time.
i
• Water Quality Association (WQA) industry survey
for use of solvents in the manufacture of -water
treatment units. In response to EPA's concern
regarding possible solvent contamination in
drinking water from, some home treatment units,
WQA conducted an industry survey and developed
guidelines for the use of solvents (3). This survey
and guideline development has raised the
consciousness of the water quality industry and, I
believe, minimized the inadvertent contamination of
water from home treatment units.
I could enumerate other areas of industry/government
cooperation to include the microbiological purifier
guide standards project or the concern with
contamination from ion exchange-resins. However, I
believe the cooperative atmosphere is well developed.
CONCLUSION
While cooperative efforts may suffer slightly by not
having regulatory backing, they can sometimes be the
best solution, particularly if the interests of the parties
are sufficiently served by such efforts. In this
situation, I believe Option #2 for cooperation is the
optimum choice.
RECOMMENDATIONS
Additional steps need to be taken to strengthen
consumer protection in areasof POU/POE.
• The water quality industry needs to continue and
expand education and training efforts aimed at
raising the knowledge and professionalism of water
quality contractors, salesmen, and technicians.
Eventually, the contributions of the water quality-
industry will be limited or expanded depending on
the professionalism and credibility of its personnel.-.
• More- attention needs to be given to the expansion
and utilization of the- third-party standards and
certification program. Specifie areas for attention.
are information and education .programs for local
and state government personnel and for
consumers regarding -the- 1MSF- --programs; and
greater utilization ,«f 4h» •NSFlisting service by
water- quality- product manufacturers. The third-
party program will not be truly-effective until it is
more widely recognized and utilized, particularly by
local and state-regulatory, officials,
• Finally' all Federal, state, and local officials and
industry and other interested parties need to
maintain; open channels of -communication to
examine areas where problems arise. Continuing
the positive patterns of the past will enable us to
have a productive result in the future/
REFERENCES
1. McGown, Wes. Sensitivity: a key water
conditioning skill. Water Technology.
September/October 1982, pp. 2-5.
2. Bell, Frank A. etal. Studies on home water
treatment systems. JAWWA. April 1984, pp.. 126-
130,
3. Water Quality Association. Voluntary guidelines for
the use of solwentSi July 1987.
137
-------
AWWA VIEWPOINT ON HOME TREA TMENT UNITS
JonDeBoer
American Water Works Association
Denver, CO 80235
INTRODUCTION
In order to present the viewpoint of the American
Water Works Association (AWWA) regarding point-
of-use (POU) or point-of-entry (POE) treatment
units, it is necessary to provide a brief overview of
AWWA.
Many are familiar with the Journal AWWA, perhaps
Mainstream, OpFlow, or Water World News. In many
cases these publications are the only perceptions
people have of AWWA. But AWWA is not just the
efforts of its staff, it's the efforts of the people
Involved in the industry — the engineers who
design systems, the people in the field who install
systems, and those who treat and analyze water.
AWWA is involved not only in drinking water, but also
water for all uses. This includes not only commercial,
industrial, and other city uses, but also many uses
within the home such as drinking water, water used
for cooking and cleaning, and water used for
consumption at taps other than the kitchen sink.
AWWA is a service organization, and when individuals
have questions, we are available to provide answers.
The organization is made up of members divided into
sections (regions of the country). Each section elects
a member to the Board of Directors, the controlling
body of the organization. The Board of Directors
internally selects an Executive Committee, which
oversees the entire operation. AWWA is divided into
four councils, the Water Utility Council, which is our
newest council, the Technical and Professional
Council, the Standards Council, and the General
Policy Council.
THE POINT-OF-USE/ POINT-OF-
ENTRY ISSUE
Each of these councils is involved with POU/POE
issues. The Water Utility Council is involved since this
issue impacts utilities, particularly from a legislative
and regulatory viewpoint. They are primarily
responsible for providing input to the regulatory and
legislative positions developed on both national and
state levels. The General Policy Council is involved
because AWWA is in the process of establishing a
policy or position on POU/POE. The General Policy
Council is set up only to do this: to oversee the
development and planning of Association policy. The
Technical and Professional Council is involved
because there are technical issues. The Technical
and Professional Council's scope is to deal with the
technical issues in the water industry. They are not
involved in the development of standards; this is
reserved for the Standards Council.
This paper discusses AWWA's position and
viewpoint. It is necessary to understand that AWWA
has two types of statements. One is a policy
statement, the other a position statement. The
difference between these two is the way they are
developed and used. Policy statements have been in
existence for a long time; they are long-standing
positions, and have been thoroughly researched.
They are selected by a rigorous process, and they
require Board of Directors' reaffirmation at least every
five years.
Position statements, on the other hand,,are a position
on a specific issue. They are often developed in
response to regulation or legislation. They are
somewhat more fluid, and rather than Board of
Directors approval, they are considered official policy
once approved by the Executive Committee.
However, the Board of Directors reaffirms a position
statement at its next meeting, and it must be
reviewed annually to ensure that it is still a current
and valid position of the Association.
LEGAL REQUIREMENTS
The primary legal issue, from the AWWA viewpoint, is
the Safe Drinking Water Act Amendments and the
requirements the Act contains for improving water
quality. The utilities and the bulk of our membership
believe that all water should be treated to acceptable
quality for all users and for all uses. Frequently, the
home treatment industry believes that it is more
appropriate to provide marginally treated water for
general use and provide independent treatment of
specific water streams that have higher use
138
-------
requirements. We have seen comments published
where municipally supplied water is described as raw
or untreated water. A POE device might be supplied
to provide treated water for other specific uses such
as cleaning and laundry. Finally, further treatment by
an under-the-sink device might be used to
increase the quality of drinking water. AWWA believes
that all water supplied to a home should be
considered drinkable.
By law, in the development of the National Primary
Drinking Water Regulations, the purveyor is
responsible to the tap. That doesn't mean through the
meter to the connection. It means to the tap, and it
means to every tap within the home. The regulations
are to provide protection against possible, as well as
proven, harmful agents, which expand the number of
contaminants that may or probably will be regulated in
the future.
In addition, EPA is required to define acceptable
treatment techniques. Best available technology does
not include point-of-use devices for VOCs; that
question is still open .for regulation in the future. In
other words, each best available technology will be
defined based on the contaminants that are
considered.
In addition, affordability can be taken into account by
the agency in defining best available technology; this
is one criteria that has been tentatively defined. An
increase in the. annual water bill for removing a
contaminant shall be limited to no more than one
percent of the household income, with a total annual
water bill of no more than two percent of income.
That doesn't leave a great deal of money for either
point-of-use or central treatment. The goal of the
Safe Drinking Water Act and the objective of the
AWWA is to consistently meet the needs of the
public. We feel that these needs must be met for all
people, not just selected populations who can afford
to install a treatment system.. .. .
AWWA'S POSITION
The AWWA position on POU/POE treatment devices
is: aesthetic treatment of potable water is something
that AWWA has no objection to and never has had.
We commend the Water Quality Association for its
establishment of Voluntary Guidelines and the use of
the Industry Review Panel. We encourage, not only
the manufacturers, but also the utilities, to use this
system to ensure that promotion of products is fair
and accurate.. These products must be properly
advertised and must not' condemn municipal water
supplies. That kind of advertising does neither our
industry nor the home treatment industry any good.
Advertising is only one form of promotion, the other is
the actual demonstration or application of a unit in the
home. Frequently we hear cases where the term
"snake oil salesman" might apply. These individuals
do not do our industry any good, they do not do the
POU/POE industry any good, and we don't think
they're appropriate. •
Treatment for health effects to meet current
regulations is another matter. This is where AWWA
has developed a draft position statement. The
position statement development procedure includes
review by numerous councils, divisions, and
committees, based on their expertise of a specific
topic. Following these reviews, the General Policy
Council reviews the entire position statement and
develops a final draft based on the comments that
have been received by all the review bodies within
the Association. In the final approval process there
may be some modifications to the wording and
language, but not to the specific intent of the
statement. The following will describe the general
content of the statement without using the specific
words:
AWWA believes that POU/POE devices are not
appropriate alternatives or replacements for central
treatment of drinking water. Central treatment of
drinking water is the alternative of choice. There are
occasional situations where POU/POE treatment
might be appropriate. But the condition that we
believe is necessary is central control, although
ownership may not be necessary. Rental/lease
situations may be entirely appropriate for POU/POE
treatment, but the control still has to be under the
auspices of the water utility. .
In addition, there has to be an effective monitoring
program established. POU/POE units have to be
properly applied. That means there has to be an
engineering and health review of the units which
includes how they are going to be installed in the
home. Microbiological safety must be maintained. We
have to protect all consumers at all points where they
are going to consume water, both intentionally and
inadvertently. We can't assume that people aren't
going to come into internal contact with water simply
because we didn't design it to be a point of drinking
water. We believe that there should be no increase in
the risk over a centrally treated supply.
A POU/POE device is installed specifically for
modification of water quality in a beneficial way.
However, there are risks. There are possibilities that,
in addition to the beneficial modification, there is a
potential for adverse, modification of the water. We
believe that consumers need to be educated to
understand the potential for adverse modification with
POU/POE devices.
This is the basic content of the AWWA draft position
statement. It has been developed as a consensus of
the entire industry, not restricted to utility input. There
are numerous people such as academics,
consultants, utilities, and manufacturers on the
139
-------
committees and councils, who have reviewed this
position statement. It is not a frivolous matter. As
stated earlier, it is in the final stages of development
at this point. Once approved, it will be published in
Mainstream. We hope that this discussion has
brought a better understanding of the AWWA position
on point-of-use devices.
140
-------
POUlPOE - POINT OF VIEW - ASSOCIATION OF STATE DRINKING WATER
ADMINISTRATORS (ASDWA)
Barker G. Ham///
Bureau of Safe Drinking Water
New Jersey Department of Environmental Protection
Trenton, NJ 08625
Before I make my disclaimer about who I actually
represent, as my function as a Chief of the Bureau of
Safe Drinking Water in New Jersey, I would like to
comment on the previously mentioned New Jersey
study. We are still trying to deal with the ramifications
of actually having two incidents within one county in
New Jersey where there were installations of
treatment units that did not prove satisfactory. We
have not been able to get the county health officials
to allow any installations of home treatment units for
health-related matters since then; the consequences
of what happens when things go wrong can take a
long time to overcome at the local level. Now that I've
said that, as an official from New Jersey, I would also
like to say that to represent all 50 states is obviously
an impossible task. The views that 1 have are mostly
my own, but I have gone over them with some of the
people within the association. There will be
differences in how the individual states react to
point-of-use treatment, so I'm talking about things
in a very general sense in terms of representing the
association.
I've always been impressed with EPA's research
efforts, their speakers, and the type of research that
they do. As a state official, I was very impressed with
the technical presentation by the point-of-use
treatment industry at this meeting. I have not seen
too many presentations by pointof-use treatment
people before, and I was very surprised by the
technical expertise of the point-of-use treatment
industry represented at this meeting. I think this is
something that the state officials need to take back to
their, local people: there is good technical work being
done within the point-of-use treatment industry.
The good aspect of having a large number of
manufacturers and installers is that it allows for a lot
of attention to local needs. A lot of the needs within
the point-of-use treatment industry are of a very
localized nature. There are different needs for
example, in Arizona, New Jersey, or in Florida, and
it's important to have different manufacturers and
installers serve different needs. It was good to see
how many actual field experiences have been a
success, and to have everyone learn from these field
experiences. 1 also think the beginning of concensus
industry standards is a good thing, whether it's for
aesthetic effects and performance characteristics
done by the Water Quality Association (WQA) or
whether it's the health related work of the National
Sanitation Foundation (NSF). I think its a good
beginning.
There is however a need for what I consider the
transfer of technology. We need to transfer
information about the industry from the national level
to the local and individual level. This technology
transfer will determine how much regulation gets
involved in this entire process. We are going to need
to get technical information to the local health officials
because if good things don't happen at the local level,
the response is legislation. To accomplish that, I think
we need to continue to build on some sort of design
and performance guidelines, both within the WQA and
the NSF or any third-party certification program that
emerges. I think it is a very good idea that we
consider things like the use of a surrogate parameter
like chloroform to see how design standards are
made useful, and to look at the application of
performance standards. We will need to apply
performance standards and to realize that there still
will be various state standards, not only in what will
be applied, but also variations in what the maximum
contaminant levels are. I think one of the first
problems that the industry is going to have to deal
with is the various states' standards. For example,
fcom my own experience, for a chemical like 1,1,1-
trichloroethane we are going to wind up a year from
now with three standards in use. There's going to be
the national standard, 200 ppb, which EPA has
regulated, a standard in New Jersey that's going to
be 26 ppb; New York is well on its way to having a
standard of 5 ppb. The different local and state health
officials in those areas will need to have enough
performance information about GAG units to be able
to make good decisions on what size unit to put in for
what size household. It's going to be important for the
141
-------
manufacturers in the industry to develop information
that can be easily understood and transferred to the
local level for these types of differences in regulatory
efforts.
( want to discuss the down side of having a large
number of manufacturers -- it makes it harder to
get all this done on a consensus basis. It is tougher
to transfer the information, and it also means that
problems associated with the introduction of new
technologies are blown out of proportion. Something
that I see in the future in terms of New Jersey, is
home treatment or point-of-use units for volatile
organic chemicals, both within private wells and what
I consider the new EPA classification of water
systems that are non-transient. It's an area that
hasn't been talked about a lot during this conference,
but I think that these types of water systems coming
under new federal regulation have more monitoring,
are going to have the ability to meet these new
standards, and are going to cause both the states and
industry to address those problems. I think radon will
also continue to be an area where home treatment
units are going to be an integral part of any state
strategy. I think there are some nitrate problems in
New Jersey, and I assume there are other states that
are going to have those problems as well. I also think
lead is an issue that was not addressed here.
Depending upon where EPA goes with its regulations,
there coufd be a need to look very closely at lead
standards and home treatment units, both in terms of
private individual wells and in terms of public water
supplies for those spots that can't be treated
otherwise.
When I look down the road a little bit, I see four
different options. We could have a huge federal
program or a partial federal program which has
registration, certification, design, and performance
standards. This will go a long way to do certain things
in terms of assuring public health, but it also stifles
new innovative designs and puts a financial burden
on the whole process.
We can have individual state programs with individual
sets of state regulations and state registration and
certification programs in some states; in other states
there won't be many programs. I'm not sure that
people will like that. I know that the person from
California didn't say that their program started from
the regulatory effort of the state officials. It was
started by the legislature when they had problems in
that state. I don't know of many state regulatory
programs that would initiate a program themselves.
We don't necessarily look for more work to do unless
we feel that there's a huge public health gain. We're
not sure in these instances whether there is a gain or
whether there isn't.
Third is the direction we seem to be going, which is
consensus guidelines, third-party certification, and
local and state application of these guidelines and
certifications. This has a lot of benefits to it, both from
the industry standpoint and the state standpoint. But
it's got to do the job, and it's got to be perceived by
the public, the federal agencies, and the legislatures
of doing the job of protecting public health. And if it
doesn't do that job, then we're going to wind up with
a bigger regulatory effort since one of the overriding
principles is that the level of regulatory effort
corresponds to the level of public health concern. If
you have more concern that things aren't working
right, the chances of legislation and regulation are
going to increase.
The fourth option would sort of go backwards -
not to do anything at the federal and state level and
let all the local people take care of the problems. It's
a local individual choice. My general belief is that we
have the opportunity to make a good start, and we've
had a good start at consensus guidelines and third-
party certifications. We are going to have to continue
that effort and make it work. The less that we make it
work, the more likely we will get federal regulations or
individual certification programs and legislation for
each state.
142
-------
POUlPOE:AN INDUSTRY PERSPECTIVE
Donna Cirolia
Water Quality Association
Washington. DC 20005
In order for the audience to better understand where
the point-of-use/point-of-entry (POU/POE) water
quality improvement industry is today, and where it's
going, it is important to gain a historieal perspective
on where the industry has been. Back in 1976, the
Water Quality Association (WQA) had only 877
members. Today we have over 2,000 members. At
our convention, which we hold every year in March,
we used to have nearly 100 exhibitors. In the last five
years, the number of exhibitors has increased to over
200. Coupled with this expansion, the direction of the
industry has changed from products dealing solely
with aesthetics, to products reducing contaminants.
However, this transition into the health arena has led
to a number of new issues and concerns.
WQA and others have developed programs to
address these areas, including product validation and
personnel certification, which is a WQA program
designed to educate our dealers who serve the
consumer.
I certainly agree with some of the other panelists that
our industry needs to do more in terms of credibility
and accountability for both products and personnel.
However, we have made great strides. Today, I may
be preaching to the choir, but it's important that the
industry members in the audience encourage your
distributors and dealers to become better educated.
We have some fabulous companies that provide
quality products and quality service. I find it
discouraging to see that some people outside of our
industry unfairly judge the entire industry by the
minority that use misleading sales tactics. WQA
recognized the need to address the issue of
misleading advertising that occurs in our industry, as
in others, and developed the Water Quality
Improvement Industry Voluntary Product Promotion
Guidelines. This program is a positive step in the right
direction.
The industry's advertising, products, and services are
constantly improving. Our efforts have been met with
greater recognition by EPA and the states, and this
seminar proves that.
Where our industry needs to do more is in our
relationship with the water utilities. However, they also
need to do a bit more. I was disappointed that very
few water utility people attended this conference,
particularly since AWWA was a co-sponsor. It would
be far better for both industries not to dwell on the
negative aspects, but rather change with the times
and adjust their attitudes so that we can both be
partners in providing quality water. We have products
that can solve health problems, and also improve the
aesthetic quality of centrally treated water, if the
consumer so chooses. This is .based on our system
of free enterprise, where individuals have the right to
choose products of their liking. Utilities should not feel
that the mere existence of the POU/POE industry
undermines the quality of water which they serve to
their customers. If a consumer does not like the taste,
odor, or color of their water, they have every right to
install a POU device.
Presently, a few states are considering their own
mandated product validation standard to provide
consumer protection. If WQA really felt that this was
the answer on how to provide consumer protection,
we would be out there promoting this type of
legislation. However, we really don't feel that product
validation is the solution. In every state there are
consumer fraud laws; the problem is in the lack of
enforcement. We encourage attorneys general and
county consumer affairs bureaus to enforce their
consumer fraud laws against those companies that
use misleading sales tactics. Trying to regulate the
industry in terms of consumer protection through
product validation is not the answer. This will just end
up increasing the cost of the products and services to
the consumer. The voluntary standards that exist
today are in a constant state of flux, being periodically
revised to meet the demands placed on the products
in the marketplace. Also, it takes a long time to both
develop and revise consensus standards.
Realistically, this type of program would be very
costly for the states to implement and enforce. I
would rather see the marketplace driving our
members to get their products validated, than the
regulator saying, "Oh, I think we need to protect the
consumer via product validation." WQA urges the
143
-------
states to use their existing consumer fraud laws, and
if they're not strong enough, then enhance them. This
is a more realistic and productive way to provide
consumer protection.
Finally, the industry is at the crossroads of trying to
meet the needs of many small systems' and
nontransient, noncommunity systems that are facing
the same heavy regulatory burden as other public
water systems. We're not saying that point-of-use
and point-of-entry are the only answers for these
systems, but they should be considered as options.
Regionalization may be cost prohibitive if the next
system is many miles away. These systems are going
to be forced to look at new, innovative solutions
including point-of-use and point-of-entry. The
industry has made great technological strides
concerning the monitoring and maintenance of
point-of-use/point-of-entry units. We do not have
all the answers but we would certainly look forward to
working with any state or local government agency
that is interested in the POU/POE option.
In conclusion, WQA realizes that our industry needs
to do more in terms of credibility and accountability.
However, I think this also has to be met by a greater
willingness, particularly by the water utilities, to realize
that central treatment is not the only answer. We're
really -partners in providing quality water to the
consumer. That's who we're trying to please. I just
hope that we'll have the opportunity to have more
conferences such as this one so that all the
interested parties including Federal, state, and local
regulators; water utilities; and POU/POE
manufacturers and dealers can maintain a continuous
dialogue on these important and critical issues. We
are all partners in, providing and assuring safe and
aesthetically pleasing water to the consuming public.
144
-------
POINT-OF-USE TREATMENT OF DRINKING WATER: COMMENTS
SueLofgren
Tempe.AZ 85282
In three years on the National Drinking Water
Advisory Council, as the EPA struggled to implement
the Safe Drinking Water Act, I could see that,
particularly as we dealt with public notification, people
were confused and afraid. They would rush out to do
one of two things: buy bottled water or buy a home
treatment device. It's a natural reaction. They didn't
wait for the public utility to come up with a new
treatment for the contaminant they thought was there.
As a result, I kept saying to EPA, "This law is going
to give those two industries a license to steal." At that
time there was nothing to regulate those industries.
Therefore, EPA had a responsibility to do something
to help guide the public on as to the degree of
effectiveness of these devices.
There were two items I bugged EPA about: point-
of-use and ground water. At that time, nobody
thought ground water was a problem. In the 10 years
since that time, I've seen real progress made in both
areas. The industry has made an immense drive
forward in terms of trying to police its own, to make
sure that the devices on the market are effective, that
the advertising is true, and that manufacturers aren't
resorting to scare tactics.
However, we still have a long way to go. What we
have been talking about today, and what I think the
New Jersey study points out, is that what is still
needed is something for the consumer. I'm speaking
from the perspective of the public — the average
person. Where does an individual go to get something
to take care of his concern? If it isn't odor or taste,
smell, or sight - if it's health concerns, how does
he know what to get? First of all, he doesn't really
know what his own health concern is. What is the
contaminant in the water? What device does he need
to remove that particular contaminant? People call me
in Arizona and think that I am the person that knows
something about water since I seem to sit on
everybody's committee. But, the state also gets
called, and they don't know how to respond. So there
is a real need for something -- somewhere that the
public can go to get these kinds of questions
answered.
Perhaps the validation of every manufactured device
is the answer in some form. You can't rely on the
consumer report - that doesn't give you the kind
of information you need, because there's a diversity
of contaminants out there. So, I really think EPA, the
states, and industry have got to come to grips with
this. I would suggest that states like New Jersey,
California, and some of the others who have done
some exploring along the lines of validating or testing,
get together and provide a central repository. EPA or
the Association of State Drinking Water
Administrators might be the ones who deal with this.
Once an individual knows what to buy, he doesn't
know how to maintain it. There are usually
instructions, but those instructions get lost. How
many people keep all the instructions that come with
anything they buy that is mechanical? There are
filters and cartridges that have to be changed. When
do you change them? Maybe you remember the first
time, but then maybe you don't. Assuming you even
remember, where do you go to buy that filter, that
cartridge?
I'm speaking from personal knowledge. My husband
went out and bought a faucet treatment device while I
was on the council, and I said "What are you getting
this for?" He said, "Well, in Arizona they dry up the
canals pnce a year, and when they do,, instead of
surface water, our utility provides us with ground
water which tastes bad. " So he wanted something
for taste. I said, "Well, in that case, that's not so bad,
but are you sure it's not one of those that leaches
silver into the system, into our water?" "I don't
know." Well, neither did I to be truthful. That's the
sort of thing I'm talking about.
More recently my son called me from Prescott, and I
live in Tempe, that's about 130 miles away. He said,
"I've got a friend who needs to get a cartridge for his
under-the-sink system and he can't find a dealer
up here. Where does he go?" I said, "Look it up in
the Yellow Pages. If you can't find it there, he'll have
to write the manufacturer." These are the sort of
things that people shouldn't have to deal with.
The water quality industry also deals with bottled
water. One of the things that we found out on the
Drinking Water Council was that the bottled water
145
-------
industry is regulated by FDA. EPA does not have
responsibility or any enforcement ability. All they can
do is set the standards for FDA to enforce. Of course,
FDA had many other concerns and drinking water
was at the bottom of the totem pole. When we finally
got to talk to the director of FDA who was responsible
for this area, we found out that, not only was it on the
bottom of the totem pole, it basically never got any
attention. If you import mineral water, and it says
"mineral water" — nobody looks at it. It was rather
amazing to us to find out how little attention is paid to
bottled water.
Let me speak from where I am in terms of a person
who is dealing with public involvement programs and.
the needs that I can see out there. I would say that
you need to involve the public more in whatever
you're trying to accomplish, whether you're an
agency person or industry person. You need to reach
the public with the information.
The future is going to call for increased use of the
point-of-use and point-of-entry devices in areas
where people are scattered. In Arizona for instance,
our Indian tribes, which are really scattered, will
definitely be looking at this as a potential way to
resolve some of their problems. I urge you as an
industry to continue to look at alternative forms of
treatment because very few people can handle the
expense of many of the conventional treatment
processes for some of these contaminants. However,
there have to be ways of assuring that those kinds of
treatment devices are monitored and maintained. You
cannot rely on the local person to be the person to
maintain the devices. You may be able to train
someone, but sooner or later they're gone and the
next person who takes over won't be trained, and
then you may really have a very bad situation. That is
going to have to be one of the key points, how to
keep something that is effective, in use and properly
maintained.
Finally, I'll just go back to making a pitch for efforts to
increase the educational level as we deal with
drinking water and as we discover new contaminants.
And, as we try to explain to the public that they aren't
going to die tomorrow, that these contaminants, these
MCLs, are based on drinking two liters of water over
a lifetime of 70 years, that some of these things are
naturally occurring, and that the danger is not
necessarily immediate, at the same time explain that
one needs to do something about it and to do it
effectively.
146
*U.S. Government Printing Office 1988: 548-158/87002
-------
fi> O
c a =; a $-~o
>mc
-
§
2>;
=0$
S I
s 3
(0
••»
D
31
TJ
O
D
------- |