United States
Environmental Protection
Agency
Municipal Environmental Research ^
Laboratory *"
Cincinnati OH 45268
Research and Development
EPA-600/S2-82-026 August 1982
Project Summary
Seattle Distribution System
Corrosion Control Study
Volume 1. Cedar River Water
Pilot Plant Study
Brian P. Hoyt, Carlos E. Herrera, and Gregory J. Kirmeyer
For 6 months, the Seattle Water
Department conducted a corrosion
treatment pilot plant study, obtaining
data on the treatment of Cedar River
water with lime and lime/sodium
bicarbonate. Continuous-flow pipe,
coupon tests were conducted to
determine, corrosion rates, penetra-
tion rates, and corrosion types for
copper, galvanized steel, and black
steel pipes. Metal leaching tests were
conducted using small diameter pipes.
Research showed that using lime and
lime plus sodium bicarbonate will sig-
nificantly reduce corrosion in home
plumbing systems. Copper and gal-
vanized steel showed corrosion rate
reductions of 65% and 69%, respec-
tively, when lime alone was used, and
56% and 58% when lime plus sodium
bicarbonate was used. Both treat-
ments significantly reduced lead and
copper leaching in the metal leaching
tests. The leaching of zinc and cad-
mium from galvanized pipe into stand-
ing water, however, increased with
both treatments. Based on this pilot
study, lime treatment is recommended
for the Cedar River water supply at an
average dosage of 1.7 mg/L CaO.
This dose should achieve an average
distribution system pH of 7.9 and an
alkalinity of 18 mg/L CaC03
This Project Summary was devel-
oped by EPA's Municipal Environ-
mental Research Laboratory, Cincin-
nati, OH, to announce key findings of
the research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
The Seattle Water Department (SWD)
serves an average of 161 MGD of high
quality water to nearly 1 million people
in the greater Seattle area. The water
originates in the Cascades from two
mountain sources — the Cedar and Tolt
Rivers. The watersheds are well pro-
tected and the water requires only disin-
fection with gaseous chlorine to meet
Federal standards. The Cedar River sys-
tem, developed in 1901, serves about
two-thirds of the area; the remaining
third comes from the newer Tolt supply.
These mountain waters, which are pre-
dominantly rainfall and snowmelt
runoff, are very soft in nature and tend
to be highly corrosive to the unlined,
metallic pipes in home plumbing sys-
tems. Corrosion of the plumbing sys-
tems and the associated water quality
degradation has been a major concern
of the Seattle Water Department 'or
many years. Corrosion has caused ti ae
types of problems - aesthetic, economic,
and health. This summary discusses
these problems and the approach to
reducing corrosion in the Cedar distri-
bution area.
Customer complaints of aesthetically
undesirable rusty water, red and blue
stained fixtures, and metallic tastes are
-------
frequently received from within the
Cedar water distribution system. This
problem has been documented by accu-
rate complaint records and by a ques-
tionnaire survey conducted by SWD in
1973. The survey, which was distrib-
uted to 10% of all services within the
direct service area (15,000 customers),
showed that 16.7% of customers in the
Cedar water distribution system expe-
rienced corrosion-related problems.
Piping corrosion in the premise
plumbing systems served by the Cedar
supply also places a significant eco-
nomic burden on the homeowner. The
average estimated life span on hot
water galvanized and copper pipes is
approximately 35 years. The forecasted
average annual cost in 1978 for main-
taining serviceability in these pipes
is estimated to be approximately $4
million.
Metals from corroded pipe leach into
the water. Studies performed from
1972 to 1976 demonstrate that the lev-
els of lead, copper, and iron in overnight
standing Cedar tap water often exceed
the levels defined by the National Inter-
im Primary Drinking Water Regulations
and the National Secondary Drinking
Water Regulations. Cadmium and zinc
were also found to increase after over-
night standing in home plumbing; how-
ever, they rarely exceeded their levels.
These metals originate from copper,
galvanized pipes, and the solders used
in home plumbing systems. Although
the health impact of metal levels from
overnight standing water is notan acute
problem, it is certainly desirable to
reduce exposure where possible.
Causes of Corrosion
The corrosiveness of Cedar water
results from several related factors,
including:
• Acidity as indicated by low pH. The
raw Cedar water pH is approximately
7.6; after chlorination and fluoridation,
pH is reduced to a range of 6.8 to 7.2.
• Dissolved oxygen concentration at
saturated conditions.
• Insufficient calcium and bicarbon-
ate alkalinity in the water to form pro-
tective calcium carbonate films on pipe
surfaces.
• A relatively high [halogen + sul-
fate]/alkalinity ratio ([halogen + SO**]/
alk) of 0.5 to 0.8 that results in condi-
tions favorable to pitting corrosion.
In 1970, three factors combined to
intensify the corrosiveness of this water
supply. (1) To decrease the occurrence
of positive bacteriological samples
within the distribution system, the chlo-
rine dosage at the open distribution
reservoir outlets was increased. (2) At
the request of the U.S. Public Health
Service, ammoniation of the water
supply was stopped to enable .a free
chlorine residual to be maintained
throughout the distribution system. This
change from combined chlorination to
free chlorination was implemented to
provide quicker, more effective disinfec-
tion of the unfiltered water supply. (3)
Based on a vote of the Seattle citizens in
1968, fluoridation with hydrofluosilicic
acid began in 1970.
Internal Corrosion Study
In December 1975, the City of Seattle
retained a consulting engineering firm
to perform a detailed analysis of the cor-
rosion problem and to recommend pos-
sible solutions. The Internal Corrosion
Study, which included a 9-month pilot
plant investigation, confirmed the cor-
rosiveness of Seattle water, the causes
of corrosion, and the impactsassociated
with the corrosive water and evaluated
alternative measures to reduce the cor-
rosiveness of the water supply. Alterna-
tive methods to reduce corrosion
included changing the methods of disin-
fection and fluoridation, blending the
water supply with groundwater sup-
plies, and adding corrosion inhibitor
chemicals.
Based on the findings of this study, an
Internal Corrosion Control Manage-
ment Plan was developed. Because the
very low level of mineral solids, pH, and
alkalinity constitutes the major causes
of the waters' corrosiveness, this plan
was designed to correct the natura I defi-
ciency of minerals in Seattle's water
through chemical addition.
The consultant recommended water
quality goals using various chemical
combinations that included the addition
of lime and sodium bicarbonate. The
actual selection of chemical combina-
tions and optimum dosages became the
task of the Seattle Water Department.
Purpose and Scope of Work
This research effort was done to
determine which treatment better con-
trols plumbing corrosion caused by
Cedar water — lime plus sodium bicar-
bonate or lime alone. The scope of work
included:
1. Determining the corrosion rates,
penetration rates and corrosion
types for copper, galvanized steel,
and iron pipe exposed to control.
lime treated, and lime/sodiurr
bicarbonate treated water.
2. Predicting increases or reductions
of residential pipe life spans.
3. Determining metal leaching levels
from galvanized pipe, copper pipe,
and lead/tin solder associated
with each treatment.
4. Establishing the required optimal
full-scale chemical dosages for
both treatments.
From June 1979 to December 1979,
three continuous flow corrosion test
apparatus (control, lime, and lime/bi-
carbonate treatment) were operated at
Seattle's Beacon Hill Gate House. This
test site was chosen because it was
close to the laboratory and had the same
water quality that is delivered to the
SWD customer. The targeted water
quality characteristics for tests are pre-
sented in Table 1.
Corrosion coupon tests were used to
document average corrosion rates
based on weight loss, penetration rates
based on pit depths, and corrosion types
based on visual observations.
To evaluate the effects of treatment
on overnight standing, SWD developed
a new test method: small diameter pipe
sections attached to the main test appa-
ratus were used in determining copper,
lead, cadmium, and zinc leaching levels
from galvanized pipe, copper pipe, and
lead/tin solder.
Operation, Results, and
Evaluation
The pilot test apparatus consisted of a
continuous flow test unit and a metal
pick-up test unit (Figure 1).
The continuous flow test unit con-
tained four 10.2-cm-long pipe segment
coupons (2.54-cm diameter) of copper,
galvanized iron, and black steel pipe. A
velocity of 0.30 m/s through the pipe
coupons was held stable by the use of a
constant head reservoir. Periodically
throughout the study, the coupons were
removed from the test loop, cleaned,
and weighed. Knowing the original
weight of each coupon, weight loss was
calculated and plotted versus time for
each metal and treatment (Figure 2).
Average corrosion rates were then cal-
culated from these curves.
As shown in Table 2, both treatments
showed substantially lower corrosion
rates than the control. For each pipe
analyzed, lime treatment produced
lower corrosion than the lime/bicar-
bonate treatments.
Penetration rates were determined
for the copper and black steel coupons.
-------
Table 1. Target Water Quality Characteristics
treatment
pH Total Alkalinity
\lone 7.2 16
ime 7.90 19
jme + Bicarbonate 7.95 22
Constant head
reservoir
(
I
I I
4«
/ r ^
Chemical V^_^/ '
Dump
Static 1 1 J
mixftr I i |
nalvanhed steal pipe coupons
J L. _|
1 1 1
Black steel pipe coupons
i i ,,r
i i i
Copper pipe coupons ^_^
i , £ b-
Hardness
19
20
20
orrosion
nhibitor
hemicafs
u
Water meter
-L
J—L
Metal leaching loop: Galvanized steel pipe, iron pipe,
copper pipe plus 50/50 load/tin solder,
and copper pipe.
Figure 1. Pilot plant flow schematic.
8.0 Y-
s
7.0
6.0
5.0
4.0
•S
^ 3.0
^ 2.0
1.0
Control (Loop #1)
Lime treatment (Loop H2J
Lime/sodium bicarbonate
treatment (Loop #3)
50
150
100
Time (days)
Figure 2. Cedar pilot plant test results - copper coupons weight /oss.
2OO
The coupons were split into quarters
and the pit depths were measured using
either a pointed tip micrometer (iron
specimens) or a binocular microscope
with a graduated fine focus adjustment
(iron and copper specimens). Penetra-
tion rates were then calculated in mils
per year (mpy) based on the average of
the 10 deepest pits on each specimen.
The penetration rate results were not
as promising as the corrosion rate
results. Average iron penetration rates
for the control, lime treatment, and
lime/bicarbonate treatment were 26.7,
25.2, and 23.5 mpy, respectively; these
results demonstrated only small rate
reductions with treatment.
The metal leaching test unit consisted
of 25-cm and 51 -cm lengths of 0.635-
cm diameter pipes attached to the con-
tinuous flow test unit. The leaching pipe
sections were made of copper, copper
plus 50/50 lead-tin (Pb-Sn) solder, gal-
vanized steel (new), and black iron pipe.
These sections were analyzed for
copper, lead, zinc, cadium, and iron
leaching. Two velocities were used
(0.061 m/s and 0.15 m/s) in establish-
ing corrosion films in the metal leaching
sections. The pipe sections were then
periodically removed and dissolved
metals were measured in the laboratory
after approximately a 24-hour contact
with test water.
The treatments resulted in substan-
tial reductions of lead and copper leach-
ing from copper pipe plus 50/50
lead-tin solder, and lead leaching from
galvanized pipe. The treatment, how-
ever, increased the zinc and cadmium
leaching from galvanized pipe. This
occurred from the localized breakdown
of the zinc passivation film under high
pH and standing water conditions.
The metal leaching tests were con-
ducted at an initial velocity of 0.061
m/s; to determine the possible effect of
scouring at higher velocities, the veloc-
ity was later increased to 0.15 m/s. In
most cases, as exemplified in Figure 3,
the velocity increase caused a very
noticeable increase in the metals leached
from the control loops. These increases,
however, did not occur in the loops
that were exposed to chemically treated
water.
Conclusions
The SWD pilot plant test program
demonstrated that lime treatment is the
more appropriate chemical treatment
for the Cedar water supply (Table 3).
Based on weight loss analysis, it proved
superior to lime/bicarbonate treatment
-------
Table 2. Corrosion Rates
Treatment
Copper Corrosion
Rate fmpyj
Galvanized Pipe
Iron Corrosion Corrosion
Rate fmpy) Rate (mpy)
None
Lime
Lime + Bicarbonate
0.66
0.23
0.29
6.66
6.28
2.32
0.74
0.94
' Rate not calculated because of weight loss determination errors.
1.2
1.0
0.8
%
u
0.6
0.2
• = Control (Loop #1)
A = Lime treatment (Loop #2)
B = Lime/sodium bicarbonate
treatment (Loop #3)
0 20 40 60 80 WO 120 140 [ 160 180 200
0.061 m/s 0.15 m/s
Time (days)
Figure 3. Cedar pilot plant test results - copper leaching from copper piping
and SO/SO lead/tin solder.
in reducing copper, galvanized steel, and
black steel corrosion rates. As a result,
estimated pipe life spans for these
materials were greatly increased. Com-
parable copper and lead leaching reduc-
tions (from the copper plus lead-tin
solder test sections) and copper leach-
ing reductions (from the copper test sec-
tions) were obtained using both treat-
ment methods.
The lime/bicarbonate treatment pro-
duced distinctly better reduction than
the lime treatment in only two evalua-
tion criteria: lead leaching from galvan-
ized steel pipe and black steel penetra-
tion rates.
The control tests demonstrated the
greatest corrosion rates and the short-
est estimated pipe life spans. Metal
leachate levels were also greater than
under treatment conditions, with two
exceptions — zinc and cadmium leach-
ing from new galvanized steel pipe actu-
ally increased with treatment.
Recommendations
Since lime treatment produced the
better overall protection against corro-
sion and metal leaching, it should be
used for full-scale treatment of the
Cedar supply. Table 4 details the recom-
mended lime dosage, target water qual-
ity , and chemical costs for this
treatment.
The full report was submitted in fulfill
ment of Grant No. R-806686-010 by
the Seattle Water Department under
the sponsorship of the U.S. Environ-
mental Protection Agency.
Table 3. Percent Reductions In Cedar Pilot Test Results
Copper
Corrosion
Copper/
Lead /Tin Solder
Corrosion
Galvanized Steel
Corrosion
Black Steel
Corrosion
Treatment
-Q
*1
13
1.1
al
-J N
8>
N
ad Leach
15 m/s)
a
ad Leach
15 m/s)
u
!
10
8 Corrosion rate by weight loss - The percent reduction of corrosion rates based on 1 weight loss.
* Metal leaching - The average reduction of metal concentrations during the 0.15 m/s flow testing period.
c m/s - meters per second.
d Pitting - The percent reduction of penetration rates based on pitting data.
"Reduction not calculated because of weight loss determination errors.
18 i
13
a
None
Lime
Lime +
Bicarbonate
0
65
56
0
46
47
0
78
76
0
81
84
0
69
58
0
49
72
0
-138
-152
0
-32
-51
0
6
e
0
6
12
-------
Table 4. Recommended Lime Dosage,
Parameter
Treatment Plant COz (mg/L)
Treatment Plant pH
Average Total Alkalinity (mg/L CaCOs)
Atmospheric Equilibrium pH f@ 15°C)
Minimum Distribution pH
Estimated Average Distribution pH
Average Distribution Conductivity (/jmho)
{Halogen + S0*]/Alk
Average Lime Dosage (mg/L CaO)
Lime Cost per Year (1980 $}
Target Water Quality And Chemical Costs
Present With
Conditions Lime Treatment
2.7
7.21
16
7.79
7.0
7.15
54.5
.5:8
b
b
0.0
S.3
19
7.86
7.65"
7.90
58.2
0.4-0.7
1.7
13,000
a By calculation.
b No present treatment.
Brian P. Hoyt, Carlos E. Herrera. and Gregory J. Kirmeyer are with the Seattle
Water Department, Seattle, WA 98144.
Marvin C. Gardels is the EPA Project Officer (see below).
The complete report, entitled "Seattle Distribution System Corrosion Control
Study. Volume I. Cedar River Water Pilot Plant Study," (Order No. PB 82-231
820; Cost: $9.00, 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 Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
*USGPO: 1982 — 559-092/0449
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use S300
RETURN POSTAGE GUARANTEED
Pij 0000329
u s ENVXK PKtntcuuN
REGION 5 LIBRARY
230 b OEARttORN STREET
CHICAGO IL 60604
------- |