United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-84-056 Apr. 1984
SERA Project Summary
Biologically Mediated
Corrosion and Water Quality
Deterioration in
Distribution Systems
John T. O'Connor and Shankha K. Banerji
Research was conducted to evaluate
biologically mediated corrosion and
deterioration of water quality in munic-
ipal water distribution systems. The
investigation included (1) a national
questionnaire survey of water utilities,
(2) an in-depth evaluation of five
Missouri water supply systems. (3) a 1 -
year study of nine water distribution
systems across the country, (4) fabrica-
tion of a pipe manifold system to moni-
tor and study water quality changes and
microbiological growths in the labora-
tory, (5) construction of a laboratory-
scale pipe loop study to simulate a
water distribution system and study
water quality changes and corrosion,
(6) studies of biologically mediated
corrosion in a batch reactor system, and
(7) an evaluation of cast iron corrosion
control using polyphosphate formula-
tions.
This Project Summary was developed
by EPA's Municipal Environmental Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
The quality of water leaving water
treatment plants in most municipalities
meets the federal standards setfor drink-
ing waters. Often substantial degradation
occurs in the water quality while water is
being transported through the distribution
system to the consumer. This deteriora-
tion in water quality results in common
nuisances such as iron staining of laundry
and porcelain fixtures, unpleasant tastes
and odors, and the presence of sediments
in drinking water. Many of these quality
changes can be brought about by the
activity of microorganisms. The present
study investigated the biologically medi-
ated water quality changes in actual and
simulated water distribution systems.
Scope
The research program had the following
specific objectives: (1) to determine the
extent and nature of water quality prob-
lems originating in water distribution
systems in the United States, (2) to
conduct laboratory studies to assess the
potential for a treated water to promote
sediment deposition and microbiological
growth in the simulated distribution
systems, (3) to determine the role of
microorganisms in mediating chemical
changes observed in distribution systems,
(4) to determine remedial measures for
controlling sediment deposition, organism
growth, and quality deterioration, and (5)
to prepare a guideline document for use
by water utilities for assessing and con-
trolling water quality problems in distri-
bution systems.
Results and Conclusions
1. A survey conducted to determine
the extent and nature of water
quality problems in distribution
systems in the United States indi-
-------
cated that 60% of the responding
utilities reported taste and odor to
betheir most common water quality
problems. So-called red water
ranked second, with 47.7% of the
utilities reporting this problem.
Complaints of cloudy and black
water were, respectively, the next
two most frequently cited water
quality problems. These responses
indicated that regardless of water
source, utility size, or geographical
location, virtually every utility peri-
odically experiences some water
quality problem that originates in
its distribution system.
2. Evaluations of distribution system
water quality in five Missouri com-
munities indicated that the most
frequently observed water quality
changes included the loss of chlo-
rine residuals, decreases in dis-
solved oxygen concentrations, in-
creases in iron concentration, and
increase in hydrogen sulfide con-
centrations without significant
changes m sulfate ion concentra-
tions. Higher bacterial plate counts
were generally observed at loca-
tions where consumers reported
water quality problems. Higher
standard plate counts (SPC) were
also associated with low chlorine
residuals. With the use of selective
enrichment culture media, a variety
of microorganisms were isolated
from these distribution systems.
They included sulfate-reducing,
sulfate-oxidizing, iron-precipitating,
nitrogen-fixing, nitrifying, denitrify-
ing and stalk-producing microorga-
nisms. These microorganisms
belong to groups that can transform
carbon, nitrogen, sulfur, and iron as
part of their microbiotic cycles.
Water quality changes can thus be
influenced and mediated by orga-
nisms indigenous to these water
distribution systems.
3. An additional 1 -year study of water
quality data from nine water distri-
bution systems throughout the
United States confirmed some of
the observations made from the
Missouri water distribution system
studies. The systems studied were
Tacoma, WA; Chicago, IL; Spring-
field, IL; Champaign, IL; Durham,
NC; Columbia, MO; Tampa, FL;
Minneapolis, MN; and Sioux Falls,
SD.
4. The analysis of hydrant flushing
samples from 12 utilities through-
out the United States indicated that
samples that had low turbidity and
lowtotal and volatile solids had few
organisms as measured by SPC and
on R-2A plates. Samples that were
turbid and brown showed positive
results for most of the microbio-
logical enrichment culture tests
They also exhibited high concentra-
tions of total solids and total iron.
Many of the samples showed the
presence of sulfate-reducing bac-
teria. In most samples analyzed, the
diluted SPC media developed more
colony growth compared with R-2A
media.
5. The results of studies using a pipe
manifold system to simulate a water
distribution system showed that the
microbial growth on pipes are re-
lated to TOC levels in the water.
TOC levels of 5.0 mg/L and less
limit the amount of microbial growth
in pipe manifold systems. The addi-
tion of chlorine reduced the accu-
mulation of microbial growth. Free
chlorine residuals of 1.0 mg/L and
higher suppressed microbial growth
significantly. Chloramine doses in
the range of 1 to 2 mg/L were not
as effective.
The presence of methane in
source water stimulates microbial
growth. Methane concentrations of
about 9 to 13 mg/L in the influent
water increased the amount of
microbial growth in the system
markedly. Studies of the effect of
pipe materials on sediment accumu-
lation and biological growth on the
pipe surfaces showed that steel
pipes developed the largest amount
of attached sediment compared
with PVC or copper pipes. The major
portion of the accumulation on steel
pipes was nonfilterable, inert resi-
due. Despite less accumulation on
PVC and copper pipes, bacterial
plate count data indicated that a
comparable number of microoga-
nisms existed in the sediments of
all pipe materials. Thus the potential
for microbiologically mediated re-
actions may be equal in all three
pipe materials. Bacterial plate
counts of materials accumulated on
pipe surfaces indicated that neither
chlorine nor chloramine residuals
prevented microorganisms from
accumulating within the sediments
attached to the pipe surface. This
result may be especially important
to understanding how microbial
processes can proceed in water
mains, even where low bacterial
counts and sufficient disinfectant
residuals are observed in samples
of the bulk water.
6. Laboratory-scale PVC and steel pipe
loop systems were used to simulate
piping in water distribution systems.
The oxidation of iron metal in the
pipe loops was primarily responsible
for contributing iron to the water.
The oxidation of the iron in turn
contributed to the depletion in dis-
solved oxygen.
The TOC concentration in water
was important in estimating the
potential capacity for organism
growth in the pipe loop system.
Since the influent test water con-
tained exceptionally small amounts
of TOC (0.5 mg/L), relatively little
microbial activity was observed.
Bacterial plate counts of water were
generally low when they were
enumerated according to standard
methods. Higher plate counts were
obtained when the organisms were
incubated at a lower temperature
(25°C)orforanextended incubation
period (7 days). As with earlier field
survey experiments, a wide variety
of physiologically different popula-
tions of microorganisms were
observed in the pipe loop systems.
An evaluation of the effect of micro-
bial growth on the corrosion rate of
cast iron and steel pipe in the loop
system was inconclusive.
7. Batch culture corrosion studies of
cast iron test specimens exposed to
pipe loop water indicated that
dissolved oxygen was the most
important controlling parameter in
iron oxidation. The relative effect of
oxygen on the corrosion rate of cast
iron exposed to tap water was much
greater than any other influence,
including that of microbial activity.
The corrosion intensity of cast iron
after 26 days of exposure to an
aqueous system with more than 7
mg/L dissolved oxygen was found
to be about 2000 mg per square
decimeter (dm2), whereas in a sys-
tem with less than 2 mg/Ldissolved
oxygen, the cast iron corrosion
intensity was only 200 mg/dm2. In
the presence of microorganisms,
the corrosion intensity of cast i
test specimens increased to about
-------
500 mg/dm2 for specimens exposed
to pipe loop water for 26 days.
Visual evidence of localized corro-
sion on cast iron specimens in the
presence of microorganisms was
clearly evident in this study. A major
influence of microorganisms in
water distribution systems appears
to be that of promoting localized pits
on pipe surfaces. Decreases in
alkalinity were observed in all cast
iron corrosion tests, possibly as a
result of the formation of ferrous
carbonate from oxidized iron and
bicarbonate ions present in the
system.
The results of batch tests to
determine the effectiveness of
hosphate compound (sodium meta-
phosphate, sodium hexameta phos-
phate, and zinc polyphosphate)
additions to water supplies for
corrosion control showed that these
compounds did not effectively
decrease the corrosion rate of cast
iron in a galvanic test system (both
under unsaturated and super-satu-
rated conditions with respect to
calcium carbonate) until a sufficient
amount of phosphate (greater than
20 mg/L as PzOs) had been added. In
nongalvanic test systems, zinc poly-
phosphate did not decrease cast iron
corrosion rates until the phosphate
dosage was greater than 10 mg/L
as p20s. The corrosion rate of cast
iron at a polyphosphate dosage of 20
mg/L as PzOs can be reduced by
increasing the total alkalinity, pH,
and calcium concentration, as well
as by decreasing the concentration
of dissolved oxygen. In continuous
test flow systems, a polyphosphate
dosage of 20 mg/L decreased the
cast iron corrosion by about 56%,
whereas a dosage of 10 mg/L as
PzOs produced no corrosion inhibi-
tion after an exposure period of 28
days. Conventional corrosion indices
such as Langelier's saturation index
and Ryznar's stability index were
unable to indicate the CaCOadeposi-
tion or dissolution tendencies of
phosphate-enriched waters, since
the equilibrium chemistry of the
phosphate-enriched water is differ-
ent from that of the natural
bicarbonate-buffered water. Addi-
tion of zinc polyphosphate at a
concentration of 10 and 20 mg/L as
PzOs increased biological growths
(as indicated by standard plate
counts) up to 80 times as much as
the untreated control. In both cases,
a slight dissolved oxygen reduction
was observed in the phosphate-
enriched systems. In actual distri-
bution system management, addi-
tional chlorine residuals may be
needed to control microbiological
growth caused by phosphate addi-
tions to water distribution systems.
Reports Based on This Research
The following publications were based
on research conducted under this grant:
Banerji, S. K., Knocke, W. R., Lee, S. H.,
and O'Connor, J. T., "Biologically Medi-
ated Water Quality Changes in Water."
Proceedings, American Water Works
Assoc. Annual Conference, Part 1,
1977.
Lee, S. H., O'Connor, J. T., and Banerji, S.
K., "Biologically Mediated Deterioration
of Water Quality in Distribution Sys-
tems." Proceedings, American Water
Works Assoc. 5th Annual Water Quality
Technology Conference, Kansas City,
MO, December 1977.
Banerji, S. K., "Water Quality Deteriora-
tion in Mains." Journal of the Missouri
Water & Sewage Conference, pp 46-
56, 1978.
Lee, S. H., O'Connor, J. T., and Banerji, S.
K., "Biologically Mediated Corrosion
and Water Quality Distribution Sys-
tems." Proceedings, American Water
Works Assoc. 7th Annual Water Quality
Technology Conference, Philadelphia,
PA, pp 137-166, 1979.
Lee, S. H., O'Connor, J. T., and Banerji, S.
K., "Biologically Mediated Corrosion
and Its Effects on Water Quality in
Distribution Systems." Journal of the
American Water Works Assoc., Vol. 72,
pp 636-644, 1980.
Prahash, T. M., Banerji, S. K., and
O'Connor, J. T., "Development of a Pipe
Manifold System for Monitoring Water
Quality in Water Distribution Systems."
Proceedings, American Water Works
Assoc. Annual Conference, Part 1,
1982.
Banerji, S. K., O'Connor, J. T., and Huang,
D. J. S., "Polyphosphate Additions for
Corrosion Control in Water Distribution
Systems." Proceedings, ASCE National
Conference on Environmental Engi-
neering, Boulder, CD, pp 97-109, 1983.
The full report was submitted in fulfill-
ment of Grant No.R-804444 by the
University of Missouri-Columbia under
the sponsorship of the U.S. Environmental
Protection Agency.
J. T. O'Connor and Shankha K. Banerji are with the University of Missouri,
Columbia, MO 6521 J.
Eugene W. Rice and Raymond H. Taylor are the EPA Project Officers (see below).
The complete report, entitled "Biologically Mediated Corrosion and Water Quality
Deterioration in Distribution Systems," (Order No. PB 84-157 494; Cost:
$32.50, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officers can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Penalty for Private Use $300
4J.S. OFFICIAL MAIL'
Environmental Protection
Agency
Offirial RiifimasR
Information
Cincinnati OH 45268
/^
/£
("
^,
'«X
•>-\nr-Ni(:-Ty
^Bffc* hn.'/ATE
/JSE S3^D
/ » *
nu\O J/i'^ VETER
UnVJ/ 30179*7
^
Pi> U u u u 4
------- |