United States
Environmental Protection
Agency
Office of Water (4601M)
Office of Ground Water and Drinking Water
Distribution System Issue Paper
Permeation and Leaching
August 15, 2002
-------
PREPARED FOR:
U.S. Environmental Protection Agency
Office of Ground Water and Drinking Water
Standards and Risk Management Division
1200 Pennsylvania Ave., NW
Washington DC 20004
Prepared by:
AWWA
With assistance from
Economic and Engineering Services, Inc
Background and Disclaimer
The USEPA is revising the Total Coliform Rule (TCR) and is considering new possible
distribution system requirements as part of these revisions. As part of this process, the
USEPA is publishing a series of issue papers to present available information on topics
relevant to possible TCR revisions. This paper was developed as part of that effort.
The objectives of the issue papers are to review the available data, information and
research regarding the potential public health risks associated with the distribution
system issues, and where relevant identify areas in which additional research may be
warranted. The issue papers will serve as background material for EPA, expert and
stakeholder discussions. The papers only present available information and do not
represent Agency policy. Some of the papers were prepared by parties outside of EPA;
EPA does not endorse those papers, but is providing them for information and review.
Additional Information
The paper is available at the TCR web site at:
http://www.epa.gov/safewater/disinfection/tcr/requlation revisions.html
Questions or comments regarding this paper may be directed to TCR@epa.gov.
-------
Permeation & Leaching
1.0 General Description of Topics
Distribution system infrastructure and appurtenances, including piping, linings, fixtures, and
solders, can react with the water they supply as well as the external environment. These
interactions can result in degradation of the distributed water. Permeation of plastic pipes and
leaching from linings and metal appurtenances are known pathways for water quality
degradation.
Permeation of piping materials and non-metallic joints can be defined as the passage of
contaminants external to the pipe, through porous, non-metallic materials, into the drinking
water. The problem of permeation is generally limited to plastic, non-metallic materials.
Leaching can be defined as "the dissolution of metals, solids, and chemicals into drinking water"
(Symons et al, 2000). Leaching can result in elevated levels of metals, organic contaminants, or
asbestos in water consumed at the tap. Health effects and mitigation techniques related to
leaching of lead and copper from lead service lines or household plumbing materials are
addressed in the Lead and Copper Rule (USEPA, 1991). Health effects associated leaching of
asbestos fibers from asbestos-cement piping is currently addressed under the Phase II National
Primary Drinking Water Regulations (USEPA, 1991). Thus, this White Paper will focus on
leaching and permeation of organic contaminants and other metals.
1.1 Permeation of Piping and Non-Metallic Joints
Permeation is a physicochemical mass transfer phenomenon involving diffusion of a solute
through a porous medium. The driving force for mass transfer is the presence of an activity (e.g.,
concentration) gradient with respect to the solute. The rate of permeation can be generalized in
simple mathematical terms shown in Equation 1.
(1) N = UA-(Aa)
where: N = Mass Permeation Rate
U = Overall Mass Transfer Coefficient
A = Transfer Area
a = Solute Activity
The overall mass transfer coefficient (U) is a complex function of the following variables:
Solute properties (composition, phase)
Medium properties (composition, pore structure, swollenness)
Solute-medium interaction (equilibrium partitioning, diffusion coefficient)
Pipe flow hydrodynamics (Reynolds number)
Transfer geometry (medium thickness)
Environmental conditions (temperature)
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
1
-------
Permeation of potable water mains and distribution system fittings by external contaminants can
be viewed as a three-step process. First, the solute partitions between the external bulk phase
(e.g., pore water, soil) and the pipe wall exterior. Next, the solute diffuses through the pore
structure of the pipe or fitting. Finally, upon penetration the solute partitions between the
internal bulk phase (e.g., pipe water) and the pipe wall interior.
Permeation can occur either from the vapor or aqueous phase. With respect to permeation of
potable water mains, the contaminants of interest include highly volatile hydrocarbons and
organic solvents. Therefore, both water mains and fittings installed in the vadose and saturated
zones are susceptible to contamination by permeation (DWI0441, 1992).
1.2 Metals and Chemical Leaching
Leaching is a broad category that includes the dissolution of a variety of metals and chemicals
into drinking water. In some instances, it is difficult to differentiate between corrosion and
leaching. Studies have been conducted to determine the rate and extent of leaching from
metallic, plastic, and concrete pipes, as well as various coatings, linings, and sealants. Coatings
and linings are often employed to prevent corrosion of water mains and mitigate red water
problems. Among the more common linings are epoxy resins, cement-mortar, asphalt
(bituminous), and concrete.
2.0 Description of Potential Water Quality Problems
Table 1 provides a summary of potential water quality problems associated with permeation and
leaching.
Table 1
Summary of Potential Water Quality Problems
Permeation
Leaching
Increased VOC content of distributed water2
Increased lead and copper levels1
Vinyl Chloride formation2
Increased asbestos levels1
Aesthetic issues (taste, odor, film formation)
Increased organic contaminants from PE pipe2
Increased metals levels from cement pipe or linings2
Increased organic contaminants from organic linings2
Aesthetic issues (taste, odor, color)
(1) Health effects associated with these parameters have already been addressed by USEPA through existing Safe Drinking
Water Act Regulations.
(2) Potential direct public health impact.
The following discussion focuses on the issues listed in Table 1 that can directly impact public
health (denoted by a number 2) but have not been addressed through existing Safe Drinking
Water Act Regulations within the distribution system.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
2
-------
2.1 Permeation Occurrences and Health Impacts
2.1.1 Dilution Effect
The movement of water through mains acts to dilute contaminants that have permeated the pipe
wall. In a simplified hypothetical model involving clean water flowing unidirectionally through
a pipe section surrounded by contaminated media (of uniform activity), the solute activity in the
pipe water is related to the flow rate (Q) according to equation 2. This is referred to as
convective dilution.
(2) a oc 1 - e(~UA'Q }
The rate and extent of permeation is greatest for small-diameter mains and service lines
(DWI0772, 1997). These water lines contain the highest ratio of mass transfer surface area to
pipe volume, and are often associated with stagnant or low flow conditions (poor convective
dilution). This effect is exacerbated by the greater likelihood of accidental releases of organic
contaminants such as petroleum products on a customer's property and consequently, closer to
the point of withdrawal or consumption.
2.1.2 Case Studies
More than 100 incidents of drinking water contamination resulting from permeation of
subsurface mains and fittings have been reported in the United States (Glaza and Park, 1992).
The majority of these incidents were associated with gross soil contamination in the area
surrounding the pipe. The occurrence of permeation incidents was equally split between high-
risk locations such as: industrial areas; former sites of fuel stations; and near underground
storage tanks; and low-risk locations such as residential areas. The sources of contamination for
the low-risk areas included disposal and accidental leaking of gasoline, oil, and paint thinner
products (Hoisen et al., 1991a).
Figure 1 illustrates the distribution system materials involved in the reported permeation
incidents. Pipes composed of polymeric materials (i.e., plastics) were involved in 98% of the
incidents. These materials include polybutylene, polyethylene, polyvinyl chloride (PVC), and
acrylonitrile-butadiene-styrene (ABS). No reported incidents of permeation through metal-based
pipe were identified.
Figure 2 illustrates the contaminants involved in reported incidents. The contaminants most
likely to permeate plastic are lipophilic and non-polar in nature. Diesel and petroleum products
(gasoline-range organics) were involved in 89% of the incidents, while volatile chlorinated
solvents accounted for 5% of the incidents. Other contaminants that exhibit high rates of
permeation include (simple) chlorinated aromatics, chlorinated and unchlorinated straight-chain
aliphatic hydrocarbons, and phenolic compounds (Holsen et al., 1991a; Holsen et al., 1991b).
Strongly polar pesticides (e.g. paraquat, malathion, and atrazine) and long-chained (high
molecular weight) hydrocarbons were not permeation threats (DWI0032, 1990; Park et al.,
1991).
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
3
-------
Legend
ABS Acrylonitrile-butadiene-styrene
AC Asbestos Cement
PB Poly butylene
PE Polyethylene
PVC Polyvinyl chloride
AC Gasket
1 % r Materials
Figure 1
Pipe materials involved in U.S. water system permeation incidents (Holsen et al., 1991a)
TCE + FCE
5%
Natural Gas
2%
Other
4%
Gasoline-
Range
Organ ics
89%
Note: Trichloroethene (TCE),
Tetrachloroethylene (PCE),
Figure 2
Contaminants involved in U.S. water system permeation incidents (Holsen et al., 1991)
The occurrence of contamination was generally identified by the customer, and indicated by an
unusual taste and odor in the tap water. For many highly toxic substances incorporated in Figure
2, including benzene, vinyl chloride, and dichloromethane, the taste and odor threshold is well
above the drinking water Maximum Contaminant Level (MCL) (DWI0441, 1992; Glaza and
Park, 1992).
Holsen et al. (1991a) calculated the frequency of permeation incidents for both service
connections (Table 2) and water mains (Table 3). The results further indicate that polymeric
pipelines are most susceptible to permeation.
Table 2
Frequency of Permeation of Service Connections
(incidents per 106 connection-year)
Metal
Polyethylene
Polybutylene
Polyvinyl Chloride
0
3.6
16.5
2.2
Prepared bvAWWA with assistance from Economic and Engineering Services, Inc.
4
-------
Table 3
Frequency of Permeation of Water Mains
(incidents per105 mile-year)
Metal
Concrete
Asbestos Cement
Polyvinyl Chloride
0.1
0
0.3
4.6
Holsen et al. (1991a) conducted an investigation of seven sites where plastic pipe permeation had
been reported. All seven incidents involved polybutylene and polyethylene service connections
and were associated with gross contamination of the soil surrounding the pipe. In one instance
where the cause of contamination was determined to be a 10-gallon gasoline leak onto the road
surface, water quality samples were collected from the service line (after 40 hours of stagnation)
and analyzed for benzene, toluene, ethylbenzene, and xylene (BTEX). The results, which are
presented in Table 4, indicate that the concentrations reached were higher than the numeric value
of two drinking water MCLs (MCL violation would be based on annual average exceeding
numeric value).
Table 4
Analytical Results from a Gasoline Permeation Incident
Contaminant
MCL Qig/L)
Water Sample (jig/L)
Benzene
5
1,300
Toluene
1,000
4,300
Ethylbenzene
700
<500
Xylene
10,000
<500
In March 2000, the Montana DEQ presented water quality results from another incident
involving gasoline-contaminated soil. Benzene permeated a 30-inch black polyethylene service
line to a concentration of 527 |_ig/L, over 100 times the drinking water MCL. The contact time
between the pipe and contaminant was unknown.
Selleck and Marinas (1991) identified another permeation incident involving a polybutylene
service connection. In this case, soil adjacent to the connection had been contaminated with
chlorinated aromatics. The results of soil and service line water samples, which are provided in
Table 5, indicate that contaminants were capable of permeating to concentrations above health
action levels.
Table 5
Analytical Results from a Solvent Permeation Incident
Contaminant
Level of Concern (jig/L)
Water Sample (jig/L)
Wet Soil Sample (g/kg)
1,2-Dichlorobenzene
6001
2,500
<0.010
(1) Numeric value of drinking water MCL; compliance with MCL is based on annual average.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
5
-------
2.1.3 Laboratory Studies
Laboratory studies have been conducted to determine the rate of permeation and time to
penetration for several different combinations of contaminants and pipe materials. The primary
limitation of these studies is the use of stagnant water columns, and the corresponding lack of
convective dilution.
Park et al. (1991) presented the results of pipe-bottle studies using various mixtures of organic
chemicals at varying activities. The findings (Table 6) support an important conclusion about
the synergistic effects of organic mixtures. The addition of a readily permeable organic chemical
to a mixture of relatively non-permeable organic chemicals increases the overall rate of
permeation. This consideration is important because most chemical spills and contamination
events involve mixtures of similar components (Glaza and Park, 1992).
Table 6
Penetration Times for Organic Solutions Through Polybutylene Pipe
Contaminant
Relative
Penetration Time1 (days)
Concentration
Pure
Mixture
Toluene
0.26
38-60
16
1,2-Dichlorobenzene
0.19
> 140
16
Toluene
0.24
40-60
9
Trichloroethylene
0.21
20
9
Trichloroethylene
0.31
9
9
1,2-Dichlorobenzene
0.24
>80
10
(1) Defined as the time required to reach 1 mg/L in the pipe water
Joint gaskets have a high intrinsic permeability, but usually are not the primary pathway for
permeation. The most common gasket materials include styrene-butadiene-rubber (SBR),
chlorinated rubbers, fluorinated rubbers, nitrile rubbers, and ethylene-propylene-diene-
monomers. Gaskets and seals are routinely used to join both iron and plastic pipes. According
to Park et al. (1991), organic chemicals are approximately 5 to 100 times more permeable in
gasket materials compared to polybutylene pipe. However, there are two reasons why most
permeation events do not involve gaskets. First, the mass transfer area associated with gaskets is
considerably smaller than that associated with pipelines (DWI0772, 1997). Second, gaskets are
usually installed in areas where flow is continuous and flow velocities are high, which increases
the dilution effectiveness (Holsen et al., 1991a).
2.1.4 Desktop Studies
The large pool of variables and boundary conditions complicate the task of quantifying pipeline
permeation. Selleck and Marinas (1991) developed and presented analytical solutions for the
permeation of hydrophobic contaminants through plastic water mains. The following two
boundary conditions were considered:
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
6
-------
Case 1. New pipe installed in soil subject to gross contamination
Case 2. Previously contaminated pipe flushed with clean water
The analytical solutions were used to calculate penetration times for 3/4-inch polybutylene water
mains exposed to a variety of organic contaminants. The results are presented in Table 7. The
penetration times for the previously contaminated pipe were small, on the order of minutes.
Conversely, the penetration times for the new pipe were on the order of weeks. This
phenomenon has been laboratory-verified for PVC and polyethylene pipes as well. Organic
chemicals and solvents promote swelling of polymeric materials, which in turn increases the rate
of diffusion by several orders of magnitude (Selleck and Marinas, 1991).
Table 7
Penetration Times1 for Polybutylene Service Lines
Contaminant
Case l2
Case 23
Toluene
36 days
54 minutes
1,2-Dichlorobenzene
53 days
20 minutes
1,3 -Dichlorobenzene
65 days
24 minutes
Tetrachloroethylene5
13 days
5.5 minutes
(1) Time to: toluene > 100 |ig/L, 1,2-dichlorobenzene > 10 (xg/L, 1,3-dichlorobenzene > 10 |_ig/L, tetrachloroethylene > 5 |_ig/L
(2) New pipe
(3) Previously contaminated pipe flushed with clean water
These results indicate that a contamination incident cannot be corrected by simply flushing the
line with clean water for a protracted period of time. Although contaminants may be removed
from the internal surface, the pipe will retain its status as a swollen, highly permeable medium
(Selleck and Marinas, 1991).
New PVC pipes exhibit lower permeation rates than new polyethylene or polybutylene pipes,
primarily due to differences in the material matrices (DWI0772, 1997). PVC is an amorphous
glassy polymer, while polyethylene and polybutylene are semicrystalline rubber. At low solute
activities, PVC is virtually impermeable (penetration time of 105 years). However, when
exposed to high activity (e.g., saturated) organic conditions, such as those that would occur
during gross chemical spillage, PVC pipe is softened to the point of failure. As a result,
permeation rates increase dramatically (Selleck and Marinas, 1991; DWI0772, 1997).
Thermodynamic theory indicates that hydrostatic pressure within the pipeline provides negligible
resistance to permeation at the pressure range commonly found in the distribution system
(Selleck and Marinas, 1991).
2.1.5 Health Impacts
Many of the contaminants involved in permeation events are regulated by federal and state
drinking water standards because of known health risks. These include the Phase I, II, lib, and V
Rules of the Safe Drinking Water Act. Maximum Contaminant Levels have been established
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
7
-------
and apply at the point of entry to the distribution system. These water quality standards are
established based on risks associated with the ingestion (consumption) pathway and typically
reflect health consequences resulting from long-term exposures. However, it is important to note
that serious acute and chronic health effects can also occur through other pathways, including
inhalation and dermal sorption (direct contact). Examples of non-consumptive water use where
inhalation and dermal absorption risks may be incurred including showering and hand washing
(Argonne National Laboratory Environmental Assessment Division, 2002). Potential health
risks associated with these routes of exposure include skin rashes and respiratory distress.
The Chemical Health Effects Tables (U.S. Environmental Protection Agency, 2002a) provides a
summary of potential adverse health effects from high/long-term exposure to hazardous
chemicals in drinking water.
2.2 Leaching Occurrences and Health Impacts
ANSI/NSF Standard 61: Drinking Water System Components - Health Effects (NSF
International, 2001) establishes minimum health effects requirements for the chemical
contaminants and impurities that are indirectly imparted (via leaching) to drinking water from
products, components, and materials used in drinking water systems. This standard does not
establish performance, taste and odor, or microbial growth support requirements for drinking
water system products, components, or materials.
The products and materials covered under ANSI/NSF 61 relevant to this White Paper include
protective materials (coatings, linings, solvent additives, etc.), joining and sealing materials
(solvent cements, welding materials, gaskets, etc.), and pipes and related products (pipes, tanks,
fittings, etc.). The NSF Drinking Water Additives Program started with cooperative agreement
from the USEPA in 1985. Thus, there are many distribution system materials and components
that were installed prior to the adoption of NSF standards. Despite the availability of NSF-
approved materials and standards, it is not possible to ensure continuous adherence to existing
standards under all circumstances.
A limited survey of 7 states (OR, CA, CO, FL, WA, TX, and NY) suggests that NSF Standard 61
certification is commonly required by State Health Departments for the components of drinking
water distribution systems. All seven of the primacy agencies interviewed mandate the use of
materials approved under NSF 61.
The overall and general method for certifying a particular product is described herein. First,
product-specific information is reviewed "to determine the appropriate analytical testing and to
ensure that the potential health effects of products or materials are accurately and adequately
identified". This information includes but is not limited to complete formulation information for
each water-contact material (e.g. percent or parts by weight for each chemical in the
formulation), the total volume of water that the product can hold when filled to capacity, the
maximum temperature to which the product, component, or material is exposed during its
intended end use, and a list of published and unpublished toxicological studies relevant to the
chemicals and impurities present in the product, component, or material.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
8
-------
Appropriate analytical testing of the exposure water includes a "minimum test battery"
established for a material of common formulation and/or formulation-dependent analysis for
uncommon or non-standard formulations.
The extraction procedures used to generate exposure water varies by product category (e.g. pipes
and related products, barrier materials, joining and sealing materials, mechanical devices and
others), by application (hot or cold etc.) and by configuration, size and material composition of
the individual product as applicable. Some of the variables to be selected and/or adjusted include
surface area-to-volume ratio, conditioning (e.g. cleaning and pre-soaking), exposure time, and
single vs. multiple time point sampling protocols, filling, emersion or other contact arrangement
and initial test-water quality characteristics.
Four extraction waters are specified and available for exposure:
(1) pH = 5, with 2 mg/L available chlorine and 100 mg/L hardness;
(2) pH = 6.5, with 2 mg/L available chlorine and 100 mg/L hardness;
(3) pH = 8 (organic analysis), with 0 mg/L available chlorine and 100 mg/L hardness;
(4) pH = 10, 2 mg/L available chlorine.
Methods of preparing these waters are specified and the water or waters to be used for testing are
either specified by material (e.g. the pH 6.5 water and the pH 10 water for copper and copper
alloy pipe and tubing), or by formulation-specific selection criteria. Several entries within the
standard direct the use of a conservative approach, for example:
"Products or materials whose intended uses fall under more than one section of this
Standard shall be evaluated under the section having the most rigorous evaluation
conditions. NOTE - Rigorous conditions are typically associated with shorter
conditioning periods, longer exposure periods, higher surface-to-volume ratios and
higher exposure temperatures."
The results of the laboratory analysis on the exposed extraction water are normalized to
determine the level of contaminants projected "at the tap". The normalized contaminant
concentrations are then compared to various criteria. The water quality criteria used to evaluate
the normalized concentrations come from several sources. There are normative criteria based on
the contaminants regulated by the USEPA and established by Health Canada or addressed by
USEPA guidance. The single product allowable concentration (SPAC) utilized by NSF Standard
61 is one tenth of the regulatory (or guidance) MCL (or MAC), except for radionuclides for
which the values are the same. There are six additional substances for which a SPAC has been
established by NSF International and peer-reviewed. There are over 300 additional substances
listed in NSF Standard 61 as threshold-of-evaluation chemicals, for which too little data exists to
determine a SPAC and for which "a comprehensive literature search for the particular substance
and consideration of structure-activity relationships" must be undertaken if certain threshold
concentrations are exceeded.
NSF 61 also applies "informational drinking water criteria" which includes over 60 additional
compounds with SPACs that have not been peer reviewed.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
9
-------
2.2.1 Plastic Water Mains
Polyvinyl chloride (PVC) mains manufactured prior to 1977 contain elevated levels of vinyl
chloride monomer, which are prone to leaching (Flournoy, 1999). Vinyl chloride is a toxic
chemical with known carcinogenic effects; the drinking water MCL is 2 |_ig/L, enforceable at the
point of entry to the distribution system.
Water quality samples collected from a rural water system in Kansas, which had installed over
100 miles of pre-1977 PVC, contained as much as 14 |j,g/L of VC. Of the 53 samples collected
(from 4 unique sites) over a 6-year period, 55% exceeded the VC MCL. This finding motivated
other states with pre-1977 PVC to monitor for VC within the distribution system. Other VC
MCL exceedances were identified in Missouri, Iowa, Arkansas, and Texas. The study identified
the following factors which promoted VC leaching:
¦ Small-diameter pipes (i.e., < 2 inches)
¦ High temperature (i.e., > 50 °F)
¦ High contact times with pre-1977 PVC pipe (i.e., > 1 day)
No instances of MCL violations associated with leaching from PVC mains manufactured post-
1977 were cited in the literature. Additional research is required to estimate the quantity of pre-
1977 PVC mains still in use.
According to "The Handbook of PVC Pipe" (Uni-Bell, 3rd Edition) PVC pipe and fittings are
immune to nearly all types of corrosion that could result in the leaching of corrosion products.
The Handbook provides detailed data regarding the resistance of PVC pipe to numerous
chemicals and states that PVC pipe and fittings are resistant to chemicals generally found in
water and sewer systems.
Advantages and drawbacks of using plastic water mains are summarized in Table 8.
Table 8
Features of Plastic Pipe
Advantages
Drawbacks
Resistant to chemical corrosion and tuberculation
Do not require linings
Low relative cost
Ease of handling and installation
Vulnerable to permeation by organics
May leach organic chemicals
More frequent mechanical failure
Reduced strength and flexibility (kinking)
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
10
-------
2.2.2 Cement Materials
Cement-based materials include reinforced or pre-stressed concrete pipes, cement-mortar linings,
and asbestos-cement pipe. Two general components of cement-based materials include the
aggregates and the binder. The binder consists of calcium silicates and calcium aluminates in
various proportions depending on the type of the cement (Leroy, Schock, Wagner, and
Holtschulte, 1996).
Several types of degradation of cement materials can occur in the presence of acid waters or
waters aggressive to calcium carbonate (Leroy, Schock, Wagner, and Holtschulte, 1996). Waters
that have a very low ion content are aggressive to calcium hydroxide contained in hydrated
cements (ACIPCO, 2002). These waters will also attack calcium silicate hydrates, which form
the larger portion of cement hydrates. Although calcium silicate hydrates are nearly insoluble,
soft waters can progressively hydrolyze them into silica gels, resulting in a soft surface with
reduced mechanical strength. Calcium hydroxide will also leach from cement-mortar linings
exposed to soft waters. The extent of leaching increases with the aggressiveness of the water and
its residual time in the pipe and is inversely proportional to the diameter of the pipe.
Douglas and Merrill (1991) studied the deterioration of new cement-mortar linings under various
water quality conditions. Field-testing demonstrated that aggressive water is capable of leaching
cement compounds from these linings, causing significant increases in bulk solution pH,
alkalinity, and calcium. The pH of stagnant water held in contact with cement-mortar lined pipes
for one week increased from an initial value of 7 upwards as high as 12. While convective
dilution reduced the magnitude of these water quality changes, it also resulted in a greater overall
extent of leaching, as indicated by the depth of the linings calcium loss. Asphaltic seal coats
significantly reduced cement-mortar deterioration.
Cement materials contain a variety of regulated inorganic chemicals, many of which are prone to
leaching. Guo et al. (1998) conducted laboratory tests to determine the extent of leaching from
ductile iron pipes lined in situ with portland cement (type I) mortar. The pipes were lined and
cured in accordance with ANSI/AWWA Standard C602-89, and subsequently disinfected
according to ANSI/AWWA C651-92. The test water was standard faucet water from a New
Jersey utility. Under static conditions, arsenic, barium, cadmium, and chromium leached from
the lining to maximum values roughly 10-20% of their respective drinking water MCLs.
Further, the acid-soluble contents of arsenic, barium, cadmium, and chromium in the cement
coating applied were only 3, 1.9, 13, and 6.6 percent of the cement industry maximum,
respectively. Therefore, the extent of leaching could have been higher if an alternate cement
lining had been applied. Additional research is necessary to understand the fate of heavy metals
associated with cement leaching, and the degree of accumulation within distribution systems.
According to Berend and Trouwborst (1999), the application of cement-mortar lining can also
lead to aluminum leaching. The aluminum content in cement-mortar linings varies, as shown in
Table 9. One of the predominant crystalline phases in cement is tricalcium aluminate, which is
believed to dissolve according to the following reaction:
(3) Ca3Al206 + 6H20 3Ca2+ + 2A13+ + 120H"
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
11
-------
Table 9
Aluminum Composition of Common Cement Linings (% as Al203)
Blast Furnace
Cement
Pozzolanic
Metallurgical
Cement
High-Alumina
Cement
Portland Cement
Berend and Trouwborst
Study Cement
13%
13%
36%
5%
18.7%
Aluminum poses serious health risks to hemodialysis patients. The European Union defines a
maximum aluminum concentration of 30 |j,g/L in water used for hemodialysis. The USEPA has
established a Secondary MCL range for aluminum of 50-200 ng/L.
In the study by Berend and Trouwborst (1999), the installation of 7,200 feet of cement-mortar
lined ductile iron pipe caused aluminum levels in a water supply to increase from 5 |j,g/L to 690
Hg/L over the course of 2 months. More than two years later, aluminum continued to leach from
the lining and produce water with over 100 ig/L of aluminum. This was attributed to several
illnesses and a 32% mortality rate at a receiving dialysis center. The water in contact with the
pipe was seawater that had been desalinated and subsequently treated with coal-filtration,
fluoridation, and UV disinfection. The water was aggressive (maximum Langelier Index
between -0.5 and -1.5), soft (hardness 15-20 mg/L as CaCCb), of low alkalinity (no data) and
high pH (8.5 to 9.5). The pipe had been lined with cement mortar at the factory by a rotary
centrifugal process.
Aggressive, soft, and poorly buffered (i.e., low alkalinity) waters promote aluminum leaching
from cementitious materials. These are the same water quality conditions that are conducive to
leaching of lead and copper. The Lead and Copper Rule (USEPA, 1991) specifies a minimum
pH within the distribution system of 7.0, but does not specify minimum alkalinity levels.
Utilities are required to maintain optimal water quality parameters at the point of entry to the
distribution system and at several locations within the distribution system to minimize lead and
copper leaching at the tap.
The extent of leaching is also strongly related to the contact time between the water and the
cement-mortar lining. In the study by Berend and Trouwborst (1999), the average contact time
was 2.3 days.
Polyphosphate corrosion inhibitors attack and soften cement linings, thereby accelerating
cementitious leaching. These corrosion inhibitors can also chelate and complex with soluble
calcium and aluminum.
2.2.3 Bituminous Coatings and Linings
Asphalts and coal tars are bituminous materials that occur naturally or are derived from non-
destructive separation of petroleum fractions. Bituminous coatings contain a complex mixture of
many chemical compounds, the exact composition of which depends upon the individual source.
Historically, the use of bituminous coatings has resulted in problems when exposed to soft,
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
12
-------
acidic waters. Among these problems are red water discoloration, extensive tuberculation, and
loss of carrying capacity.
The material properties of the coating systems are enhanced by the use of additives, the most
obvious of which are the solvents used as a vehicle for application. However, these additives can
diffuse through the coating and into the pipe water. Elevated levels of alkyl benzenes and
polycyclic aromatic hydrocarbons (PAHs) have been reported in reservoirs with new bituminous
coatings and linings (Yoo et al, 1984; Krasner and Means, 1985; Alben, 1980). While few
additives from coating materials are included in prioritized lists of SOCs to be monitored in
drinking water, proper application, curing, and testing prior to returning coated surfaces to
service will minimize subsequent exposure to solvents used in application.
Alben et al (1989) studied leaching of organic contaminants from flat steel panels lined with
various coatings, including vinyl, chlorinated rubber, epoxy, asphalt, and coal tar. Emphasis was
given to the rate of leachate production and leachate composition. The test water was GAC
processed tap water with a pH of 8 to 9. Leaching rates (mg/m -day or |j,g/L-day) were assessed
over a period of 30 days. Organic contaminants were found at parts-per-billion levels in water
compared to parts-per-thousand levels in the coating.
Concentrations of total organic carbon (TOC) in leachate from asphalt were the lowest of the
five coatings studied. Alkyl benzenes constituted 72% of the leachate from asphalt coating.
Based on the results of 30-day laboratory experiments with asphalt-coated test panels (coating
thickness 536 |j,m), the rates of leaching were predicted for full-scale storage tanks with volume
3 2 3 2
to surface area ratios ranging from 8.2 to 37.7 ft /ft (2.5 to 11.5 m /m ). Initial rates of leaching
for the semi-volatile base-neutrals were found to range from 0.4 to 1.7 |j,g/L-day; after 30 days
the rates dropped to 0.2 to 1.0 |j,g/L-day.
Concentrations of TOC in leachate from coal tar were the second lowest of the five coatings
studied (2 bituminous and 3 polymeric coatings). PAHs accounted for essentially all of the TOC
in leachate from coal tar coating. Based on the results of 30-day laboratory experiments with
coal tar-coated test panels (coating thickness 2,746 |j,m), the rates of leaching were predicted for
3 2
full-scale storage tanks with volume to surface area ratios ranging from 8.2 to 37.7 ft /ft (2.5 to
3 2
11.5 m /m ). Initial rates of leaching for the PAHs were found to range from 2.6 to 12 |j,g/L-day;
after 30 days the rates were similar at 2.3 to 11 ng/L-day.
There have been incidents of PAH leaching from coal tar linings that have ultimately led to taste
and odor complaints. In one particular case, a butterfly valve failed while in the closed position,
thus creating an artificial dead-end along a pipe stretch coated with coal tar. Upon repairing the
valve and reopening it, the utility received several odor complaints. Water quality sampling
revealed that anthracene, naphthalene, and other PAHs were present at concentrations in excess
of 5,000 |j,g/L (Satchwill, 2002).
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
13
-------
2.2.4 Epoxy Coatings and Linings
Epoxy coating starts out as a dry powder which is produced by combining organic epoxy resins
with appropriate curing agents, fillers, and pigments. When heated, the powder melts and its
constituents react to form complex cross-linked polymers.
Alben et al (1989) studied leaching of organic contaminants from epoxy-coated flat steel panels,
with emphasis on the rate of leachate production and leachate composition. The test water was
GAC processed tap water with a pH of 8 to 9. Leaching rates (mg/m -day or |j,g/L-day) were
assessed over a period of 30 days and found to decay as a simple exponential.
Concentrations of TOC in leachate from epoxy resin were the highest of the five coatings studied
(2 bituminous and 3 polymeric coatings). The solvents methyl isobutyl ketone (MIBK), ortho-,
meta-, and para-xylene represented a major portion (51%) of leachate from the epoxy coating.
Based on the results of 30-day laboratory experiments with epoxy-coated test panels (coating
thickness 393 |j,m), the rates of leaching were predicted for full-scale storage tanks with volume
3 2 3 2
to surface area ratios ranging from 8.2 to 37.7 ft /ft (2.5 to 11.5 m /m ). Initial rates of leaching
for the solvents were found to range from 40 to 187 |j,g/L-day; after 30 days the rates dropped to
8 to 37 |j,g/L-day.
These findings were further supported by field studies. Solvents (MIBK, xylenes) from an epoxy
coating were also detected in effluent from 2 of 3 storage tanks that were monitored one month
after application of the coating (Alben et al, 1989).
The extent of leaching of organic contaminants from epoxy resin linings was found to be
strongly dependent on the duration of the curing process (DWI0032). Longer curing periods
produced more stable linings.
The City of Calgary performed leaching studies with pipe coupons coated with two-component
epoxy resins (Satchwill, 2002). Five different NSF and AWWA approved epoxy resins were
independently investigated. Application and curing of the epoxy resin involved the following
steps: (1) applying the initial coating followed by verification that application meets wet film
thickness specification, (2) 24 hours of air drying followed by verification that application meets
dry film thickness specification, (3) applying a second epoxy coating (followed by verification),
and (4) 24 hours of air drying (followed by verification.) The coupons were then filled with
water and held for 72 hours prior to analysis for TOC and VOCs. The results, which are
provided in Table 10, illustrate the extreme extent of leaching from new epoxy resins into
stagnant water columns.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
14
-------
Table 10
72-Hour Epoxy Resin Leaching Results
Coating
BTEX (mg/L)
TOC (mg/L)
Epoxy 1
Epoxy 2
Epoxy 3
Epoxy 4
Epoxy 5
13.2
4.9
48.0
0.2
25.6
166
34
279
345
143
Benzyl alcohol and epoxy resin oligomers have been identified in leachate from epoxy coated
water mains (Crathorne et al, 1984; Watts et al, 1983).
3.0 Prevention and Mitigation Methods
3.1 Indicators
The majority of the contaminants of concern with respect to permeation and leaching events are
organic chemicals, including volatile, semi-volatile, and non-volatile substances. While many of
these contaminants are regulated under national and/or state drinking water standards, and are
thus monitored at the source, others fall into the class of unregulated organics. Some of these
unregulated contaminants are potential health concerns and are listed on the Contaminant
Candidate List (CCL) under the Unregulated Contaminant Monitoring Rule (UCMR).
Source water quality monitoring will not aid in detecting contamination events that occur within
the distribution system, including permeation and leaching. However, there are other methods
and indicators that can be used to prevent and identify contamination events. The following
potential indicators are addressed:
Taste and odor of tap water
Installation inspection/Site use review
Environmental monitoring
Release to service monitoring
3.1.1 Taste and Odor of Water
Taste and odor (T&O) complaints have previously led to the identification of gross subsurface
contamination. The key question in this regard is whether or not T&O detection is useful for
identifying a contamination event before significant health risks are incurred (through
consumption). Table 11 provides a comparison of risk thresholds versus T&O thresholds for
several organic substances commonly involved in permeation and leaching events. The effect of
mixing chemicals on the T&O thresholds is unknown.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc. 15
-------
Table 11
Organoleptic versus Risk Thresholds for Pure Compounds
Contaminant
MCL
Taste Threshold
Odor Threshold
Detectable
Before Risk?
(3)
Benzene
5 (J-g/L
500-4,500 (jg/L
2,000 (jg/L
No
Ethylbenzene
700 (xg/L
29 (jg/L
29 (jg/L
Yes
Xylenes
10 mg/L
No Data
2.2 mg/L
Yes
Naphthalene
300 ixg/L1
No Data
300 (jg/L
No
1,2-Dichloroethane
5 (J-g/L
29,000 (jg/L
No Data
No
Tetrachloroethylene (PCE)
5 (J-g/L
300 (jg/L
No Data
No
Methyl Isobutyl Ketone (MIBK)
500 (jg/L2
No Data
20,000 (jg/L
No
Vinyl Chloride
2 (J-g/L
None
None
No
Styrene
100 (jg/L
730 (jg/L
No Data
No
(1) State of New Jersey MCL
(2) State of Wisconsin MCL
(3) Health risk level based on long-term exposure.
In most instances, the risk threshold is substantially lower than either the taste or odor thresholds.
However, while there are contaminants for which the organoleptic thresholds are lower than the
levels where long-term exposure is a concern, the most prevalent sources of contamination are
events like petroleum product releases where a combination of compounds with varying
organoleptic and risk thresholds will be present.
3.1.2 Installation Inspection/ Site Use Review
There are currently no certification requirements in published literature for installers or
inspectors with regard to identifying potential subsurface contamination or installing pipes or
fittings in regions of potential contamination. Standards ANSI/AWWA C900 through C950
cover plastic pipe and recommend consultation with the pipe manufacturer prior to selecting a
material if a water main must pass through an area of gross contamination. The UPC specifies
that clean soil be used as backfill around buried pipes.
Activities and procedures that are typical of construction within areas believed to be at
significant risk of contamination include reviewing site use(s) and performing environmental
monitoring during installation.
Historical and current site use will allow the construction team to assess the probability of
contamination as well as the specific contaminants for which to look. The contamination of soil
and/or groundwater by highly permeable substances can often be identified during construction
activities through the use of air quality monitoring equipment and/or passive soil gas surveys.
Gas chromatographic instruments, including photo-ionization detectors (PIDs) and flame-
ionization detectors (FIDs), are tools that can be used to detect hydrocarbons in the vapor phase
(or headspace) at the parts-per-million level. Passive soil gas sampling involves the use of
specially designed sorbents to capture volatile and semi-volatile organic compounds in soil and
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
16
-------
water. Passive soil gas sampling provides quantitative, time-integrated data of organic
contamination.
3.1.3 Monitoring
Monitoring is critical to ensure that a new or rehabilitated main can be safely released to service.
Some utilities have developed water quality "release to service" standards for new and repaired
mains. Typical monitoring parameters include total coliform bacteria, HPC, pH, disinfectant
residual, turbidity, and odor. However, monitoring for additional parameters may be warranted
based on knowledge of materials used and/or site-specific water quality or environmental
conditions. For example, monitoring for BTEX and MIBK following application of an epoxy
coating can be used to evaluate the extent of leaching and potentially identify problems with the
lining material or manufacturers defects.
3.2 Maintenance
Stagnation of water can exacerbate permeation and leaching incidents. Poor convective dilution
of permeated or leached solvents is associated with stagnant areas of the distribution system.
The extent of leaching of cementitious and organic material exhibits some relationship to the
contact time between the pipe and internal water. Unidirectional flushing and reservoir turnover
can be used to encourage fluid movement, minimize residence time, and replace stagnant water.
Flushing does not suppress the process of leaching, but the movement of fluid helps to prevent
the accumulation of contaminants in a localized area. Studies have demonstrated the
ineffectiveness of flushing to mitigate a permeation incident. While flushing may replace the
contaminated water, the pipeline retains its swollen, highly permeable state. In these instances,
pipe replacement is an effective practice.
With respect to permeation, most incidents occur along service connections due to their small
diameter and frequent stagnation, however, service lines are not owned by the utility (USEPA,
1999). Thus, in the event of a permeation event involving a service connection the consumer
would need to flush the tap long enough to draw water from a point beyond the service line to
avoid exposure to potentially contaminated water.
3.2 Design and Installation
Prevention of permeation and leaching requires proper materials selection and installation
practices. The phenomena of permeation and leaching require consideration of the environments
surrounding and within the pipe, respectively. From a design perspective, permeation can be
precluded by identifying the area of installation and recognizing the potential for past or future
gross spillage of organic chemicals and petroleum products.
ANSI/AWWA Standards C900 through C950 state that if a water main must pass through an
area of gross contamination, the manufacturer should be consulted regarding the permeation of
pipe walls and joint fittings prior to selecting the material. The Standards also note that research
has documented that pipe materials such as polyethylene, polybutylene, polyvinyl chloride, and
asbestos cement; and elastomers, may be subject to permeation by lower molecular weight
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
17
-------
organic solvents or petroleum products. Ten State Standards (Great Lakes... 1997) states that
non-permeable materials shall be used for all portions of the system including water mains,
service connections, and hydrant leads in areas of distribution systems where the groundwater is
contaminated by organic compounds. The Uniform Plumbing Code does not discuss permeation
directly. It does state that clean soil will be used as backfill around buried pipes and seems to
recognize some potential for contamination from the pipe-embedment environment The UPC
does not, however, specifically address installation through contaminated soils or permeation as
a pathway of exposure.
To prevent leaching from cement-based pipes and linings, knowledge of water quality is
necessary when determining the appropriateness of certain materials. ANSI/AWWA Standard
C401-98 provides guidance on the application of asbestos cement pipes based on internal water
quality characteristics, as summarized in Table 12.
Table 12
Aggressiveness of Water towards Asbestos-Cement Pipe
Water Type
pH + log(AH)
Langelier Index
Recommended Pipe Type
Highly Aggressive
< 10.0
<-2.0
*
Moderately Aggressive
10.0-11.9
-2.0 to-0.1
II
Nonaggressive
> 12.0
>0
I and II
Source: ANSI/AWWA Standard C401-98
*The serviceability of pipe for such applications should be established by the purchaser in conjunction with the manufacturer
Type I - no limit on uncombined calcium hydroxide
Type II - 1.0% or less uncombined calcium hydroxide
A = Alkalinity as mg/L CaC03; H = Calcium hardness as mg/L CaC03
The rate of leaching of organic additives was found to decrease exponentially with time.
Therefore, it is recommended that newly lined pipes be pre-soaked prior to release to service.
Normal hydrostatic testing and disinfection activities will help remediate leaching. Extending
the curing process will also help improve the stability of epoxy linings.
Additional Standards applicable to design, installation, and application of pipelines, linings, and
coatings are summarized below:
AWWA Manual M9 provides information related to design, handling, delivery,
laying, field testing, and disinfection of concrete pressure pipe.
AWWA Standard C600-606 covers pipe installation procedures
ANSI/AWWA C104 through C153 covers ductile iron pipe, fittings, linings,
coatings, gaskets, and joints.
ANSI/AWWA C200 through 222 covers steel pipe, linings, coatings, and joints.
ANSI/AWWA C300 through 304 covers concrete pipe.
ANSI/AWWA C400 through C403 covers AC Pipe.
ANSI/AWWA C800-89 covers service lines.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
18
-------
3.3 Source Water Treatment
Water quality plays in important role in the process of materials leaching. Key parameters in this
regard include pH, alkalinity, and dissolved inorganic ions. Water quality parameter
requirements for minimizing soluble lead and copper levels are provided in the Lead and Copper
Rule (USEPA, 1991). Suitable water quality conditions for the installation of asbestos-cement
pipe were summarized previously.
Based on a study by Brend and Trouwborst (1999), aluminum leaching from portland cement
and blast furnace slag cement only occurred in low alkalinity water. Based on this study, the
following recommendations were established:
In-situ linings of ordinary portland cement should not be used if the alkalinity is less than
55 mg/L as CaCC>3.
Factory-applied cement linings should not be used if the alkalinity is below 25 mg/L as
CaC03.
Vik and Hedberg (1991) provided the following recommended water quality conditions to avoid
leaching of cement-based materials, including AC and concrete pipes and cement mortar lining.
pH > 7
Alkalinity >15 mg/L as CaCC>3
Calcium >10 mg/L
Sulfate < 200 mg/L
Aggressive CO2 < 5 mg/L
4.0 Summary
Distribution system infrastructure and appurtenances including piping, linings, fixtures, and
solders can react with the water they supply as well as the external environment. Permeation and
leaching are two mechanisms that can result in the degradation of the distributed water.
Leaching from cement linings can occur in soft, aggressive, poorly buffered waters. Under static
conditions, metals such as aluminum, arsenic, barium, chromium, and cadmium can leach from
cement linings, even when NSF approved materials are used and linings are applied according to
AWWA standards. Current provisions of the Safe Drinking Water Act do not require
monitoring for heavy metals beyond the point of entry to the distribution system, and additional
research would be required to assess the degree of metals accumulation within the distribution
system. Vinyl chloride can leach from pre-1977 PVC pipe. No instances of MCL violations
were cited in association with post 1977 PVC pipe. Volatile organics present in the ground
water or vadose zone can permeate plastic piping and gaskets. Permeation is typically most
severe for small diameter, low-flow pipes. The literature cites instances where MCL violations
have occurred at the point of consumption, although current provisions of the Safe Drinking
Water Act do not require monitoring for volatile organic compounds beyond the point of entry to
the distribution system. In most instances, the risk threshold of chemical contaminants is
substantially lower than either the taste or odor thresholds, suggesting that utilities cannot rely
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
19
-------
confidently on customers' perception of taste and odor for identifying contamination events.
Unidirectional flushing can be used to rid the distribution system of stagnant, contaminated
water, but additional research is needed to determine the fraction of heavy metals and organics
that can be removed through flushing. Permeated plastic piping must be replaced since the
piping retains its swollen porous state after permeation. NSF Standard 61 and numerous
AWWA Standards have been developed to prevent the degradation of drinking water due to
contact with piping materials. Materials selection, design, and installation considerations based
on water quality and environmental conditions are addressed in these Standards.
Bibliography
ACIPCO. "Linings." June 12, 2002.
Alben, K., Bruchet, A., and Shpirt, E., January 1989. Leachate from Organic Coating Materials
Used in Potable Water Distribution Systems. Denver, CO, AWWARF.
ANSI/AWWA Standard C401-98. The Selection of Asbestos-Cement Pressure Pipe, 4 In.
through 16 In. (100 mm through 400 mm), for Water Distribution Systems. AWWA, Denver,
CO.
Argonne National Laboratory Environmental Assessment Division. "Identify Potential Exposure
Pathways." The Ecological Risk Assessment Process. June 12, 2002.
Benjamin, M., H. Sontheimer, P. Leroy, 1996. Internal Corrosion of Water Distribution
Systems. AWWARF and TZW. Denver, Co.
Berend, Kenrick and Trouwborst, Tom, July 1999, "Cement-mortar pipes as a source of
aluminum, "JournalAWWA, Vol. 91 No. 7, pp. 91-100.
Bureau of Environmental Health Assessment and Massachusetts Department of Public Health,
October 1997, "Health consultation, Duxbury Water Distribution, Duxbury, Plymouth County,
Massachusetts," http://www.atsdr.cdc.gov/HAC/PHA/duxbury/dux_pl.html
Douglas, B.D., Merrill, D.T., December 1991. Control of Water Quality Deterioration Caused
By Corrosion of Cement-Mortar Pipe Linings. Denver, CO: AWWARF.
DWI0032, March 1990, "Effect of Distribution on Organic Contaminants in Potable Water,"
FWR.
DWI0441, May 1992, "Effects of Organic Chemicals in Contaminated Land on Buried
Services," FWR.
DWI0772, December 1997, "Permeation of Benzene, Trichloroethene and Tetrachloroethene
through Plastic Pipe," FWR.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
20
-------
Flournoy, Ralph L., Monroe, David, Chestnut, Naima-Halim, and Kumar, Vinita, June 1999,
"Health Effects from Vinyl Chloride Monomer Leaching from pre-1977 PVC pipe," 1999
AWWA Annual Conference Proceedings.
Glaza, Edward C. and Park, Jae K., July 1992, "Permeation of Organic Contaminants Through
Gasketed Pipe Joints," Journal AWWA, Vol. 84 No. 7, pp. 92-100.
Great Lakes Upper Mississippi River Board of State Public Health & Environmental Managers.
1997. Recommended Standards for Water Works. Albany, NY: Health Education Services.
Guo, Qizhong, Toomuluri, Parvati J., and Eckert, James O., March 1998, "Leachability of
regulated metals from cement-mortar linings," Journal AWWA, Vol. 90 No. 3, pp. 62-73.
Hammer, Bill, March 2000, News Release - Montana Department of Environmental Quality,
http://www.dep.state.mt.us/press/press2000/PR031700.asp.
Holsen, Thomas M., Park, Jae K., Jenkins, David, and Selleck, Robert E., August 1991,
"Contamination of Potable Water by Permeation of Plastic Pipe," Journal AWWA, Vol. 83 No. 8,
pp. 53-56.
Holsen, Thomas M., Park, Jae K., Bontoux, Laurent, Jenkins, David, and Selleck, Robert E.,
November 1991, "The Effect of Soils on the Permeation of Plastic Pipes by Organic Chemicals,"
Journal AWWA, Vol. 83 No. 11, pp. 85-91.
Kirmeyer, G., M. LeChevallier, M. Friedman, J. Funk, K. Martel, M. Karim, and J. Harbour.
2001. Pathogen Intrusion Into the Distribution System. Denver, CO: AWWARF and AWWA.
Leroy, P., M. Schock, I. Wagner, and H. Holtschulte. 1996. Internal Corrosion of Water
Distribution Systems. AWWARF and TZW. Denver, Co.
National Sanitation Foundation. 2001. NSF Standard 61 Drinking Water System Components
Health Effects. ANSI/NSF Standard 61. National Sanitation Foundation International.
Park, Jae K., Bontoux, Laurent, Holsen Thomas M., Jenkins David, and Selleck, Robert E.,
October 1991, "Permeation of Polybutylene Pipe and Gasket Material by Organic Chemicals,"
Journal AWWA, Vol. 83 No. 10, pp. 71-78.
"Plastics Contaminate Tap Water with Hormone Disruptors," ENDS Report, June 1999, Issue
293
Satchwill, Trevor. "The Impact of Pipe Coatings on Drinking Water Quality." June 12, 2002.
Selleck, Robert E. and Marinas, Benito J., July 1991, "Analyzing the Permeation of Organic
Chemicals Through Plastic Pipes," Journal AWWA, Vol. 83 No. 7, pp. 92-97.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
21
-------
Snoeyink, V., and I. Wagner. 1996. Internal Corrosion of Water Distribution Systems.
AWWARF and TZW. Denver, Co.
State of California. June 12, 2002.
Symons, James M., L. C. Bradley, T. C. Cleveland, 2000. The Drinking Water Dictionary.
AWWA, Denver, CO.
Thompson, C. and Jenkins, D., 1987. Review of Water Industry Plastic Pipe Practices. Denver,
CO. AWWARF.
Uni-Bell PVC Pipe Association. Handbook of PVC Pipe, Third Edition. Dallas, Texas.
USEPA, "Technical Factsheet on Benzene." June 12, 2002.
USEPA, "Technical Factsheet on Ethylbenzene." June 12, 2002.
USEPA, "Technical Factsheet on Vinyl Chloride." June 12, 2002.
USEPA, "Technical Factsheet on Styrene " June 12, 2002.
U.S. Environmental Protection Agency. 2002a. Chemical Health Effects Tables.
http://www.epa. gov/safewater/tcrdsr. html. May 7, 2002.
USEPA, July 1998, Volatile Organic Chemicals Rule (Phase 1) (Final), Fed. Reg.
USEPA, 1994. Office of Pollution Prevention, Chemical Fact Sheet. EPA 749-F-94-021
USEPA, July 1991, Lead and Copper Rule (Final), Fed. Reg.
USEPA, 1991, Phase II National Primary Drinking Water Regulations, Fed. Reg.
Vik, E.A., and T. Hedberg. 1991. Corrosion and Corrosion Control in Drinking Water Systems.
In Proc. Of a Corrosion Workshop and Seminar in Oslo, Norway, March 19-21, 1990. Houston,
Texas: National Association of Corrosion Engineers (NACE).
Virginia Department of Health. "Petroleum Products." June 12, 2002.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
22
------- |