Ralph W. Buelow
Presented at AWWA Water Quality Technology Conference
Philadelphia, Pennsylvania
December 11, 1979
Drinking Water Research Division
Municipal Environmental Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
/ 1

Laboratory Techniques for Determining Corrosivity of
Water to Asbestos-Cement Pipe
Ralph W. Buelow
The Drinking Water Research Division (DWRD) of the U.S.
Environmental Protection Agency has been researching Asbestos-Cement
(A/C) pipe corrosion and corrosion control for a few years. This
phenomenon is called corrosion because the cementing materials are
being dissolved, rather than a physical wearing of the pipe. A/C
deteriorates when certain aggressive water quality conditions exist
and asbestos fibers are loosened and can be released into the water.
These aggressive waters attack most other piping materials as well.
A/C pipe resistance to chemical attack has been evaluated by
use of an index comprised of the pH, alkalinity and calcium factors
in water.	This index does not work in all cases. Some waters
determined by the index to be non-aggressive may attack A/C pipe
while some determined to be aggressive may not. Iron, manganese,
zinc and silica are examples of substances found in water that can
protect A/C pipe, even when the water is aggressive according to
the calculations. The goal of the DWRD was to develop an experimental
technique for evaluating these factors.
The first experiments were done with large recirculation
systems and full lengths of 4 in (10 cm) and 6 in (15 cm) diameter
A/C pipe assembled in approximately 90 ft (27 m) loops. These
experiments were not successful. Small scale tests using a coupon
of A/C pipe in a recirculation system were then investigated. The
equipment used in the experiments and the reason for certain
techniques will be described.
Figure 1 identifies the parts in the recirculation system used
in the small scale A/C pipe experiments. It also shows the flow
path of the recirculated water. Some of the more important aspects
of this arrangement will be discussed. The water is pumped out of
the bottom of the tank by the recirculating pump (Part No. 10).
The water flows past the A/C pipe coupons placed in PVC pipe nipples
(Part No. 7 and Figure 2) on the way to the pump. The volume of
water pumped through the system Is recorded by a meter (Part No 12)
before returning to the tank.
The most important items in controlling the water quality
during an experiment are the floating cover (Part No. 16 and Figure
3) and the Tygon type plastic tubing circular gasket (Part No. 17).
Fabricating the floating cover to fit the tank closely is better than

Pari No. Description
*1	100 gal. Tank
2	1" Plastic Ball Valve
3	1" Plastic Tee
4	Plastic Boiler Drain
5	1" Plastic Union
6	1 '/2"x1" Reducer Coupling
**7	1'/2"x6" PVC Nipple
8	1V" PVC Coupling
9	y2" PVC 90° Ell
10	Pump-Magnetic Drive
March Model MDXT
11	Vi" PVC 45° Ell
1 2	Water Meter. Plastic
13	V2" PVC Tee
14	W Pipe to Tube Ell
15	V2" I.D. Tygon
***16	Floating Cover
17	Tygon Tube Ring Gasket
18	Tank Legs
19	'/2"PVC Pipe
20	Sample Valve
21	3A" PVC Nipple
22	Rubber Stopper
' 100 gal. lank can be stainless
sieel or plastic.
" Part tt7 serves as the A/C pipe
coupon holder.
"* Floating cover made Irom '/»"
sheet PVC. 1" strip of PVC
cemented around edge of disc to
form floating pan Threaded
fittings in cover require build up
Note: Pipe busings and nipples required to
make connections are not numbered
and described
Asbestos-Cement Pipe Small Scale
Corrosion Experimental Set Up
Figure 1.

purchasing a ready made cover that is poor fitting. The plastic
tubing circular gasket is fitted so that it wipes the sides of the
tank as the water level in the tank drops. The purpose for the
cover and gasket is to prevent exposure of the water to the
If the seal is not substantially complete, the pH of the water will
be difficult to control and require frequent adjustment as carbon diox-
ide from the air causes it to change. Alkalinity control can also
be a problem in systems open to the atmosphere. Water in the systems
tends to establish equilibrium with its environment and if exposed
to the atmosphere will seek equilibrium with it. Experience with
open tanks found frequent pH adjustment required, at least daily.
Further, a closed system is more representative of an actual distri-
bution system.
Figure 2 pictures an A/C pipe coupon being inserted in the PVC
coupon holder. The coupon is small relative to the recirculation water
volume. The intent of this size ratio between the coupon and water
volume was to permit long runs without the need to chaRge water because
of large increases in ion concentrations in the water caused by dissolu-
tion of materials in the coupon. At the same time the coupon is large
enough to contribute measureable changes in the water quality under
aggressive water situations. Large size pipe materials other than
A/C can be cut into coupons and handled in this same manner.
Figure 3 is a view of a floating cover fabricated from a disc
of 1/8" (2.3 mm) PVC sheet with a strip of PVC solvent welded around
the edge to produce the pan. The Tygon ring underneath the cover fills
in the gap between the cover and the sides of the tank. When threaded
fittings are attached to the cover a built—up section is usually
needed because 1/8 inch (3 mm) PVC is too thin to thread. Use of
bulkhead fittings would eliminate the need for this reinforcement.
Additions of chemicals for controlling or adjusting water quality
when the system is in operation are made through the stoppered hole
in the floating cover.
Figure 4 shows the actual arrangement of the parts of a recircu-
lation system. The tanks being used are stainless steel because
these tanks had been used for previous research and were available.
Plastic or fiberglass tanks can also be used and would be preferable
for metal corrosion studies. All of the equipment is essentially
located underneath the tank in order to conserve space and protect
the equipment from being bumped. The pump, mounted on a piece of
plywood that rests on the floor, recirculates water at 2.3 (9 L/min)
to 3 gpm (11 L/min). The pump requires only 25 watts and therefore
does not impart sufficient energy to the system to cause significant
heating of the water in either stainless steel or plastic tanks.
At the top of the tank in Figure 4 the water sampling arrange-
ment is teed off the recirculation line. The size of the sample
line is kept small to minimize flushing needs before sample collec-
tion. The volume of water for samples is also minimized. Conserv-
ing sample water allows the operation of a system for 6 months
without the need for make up water.

Fig. 2. Coupon Holder & Coupon

Fig. 3. Floating Cover

Fig. 4. Recirculation System

Figure 5 is a photo of a section of lead pipe installed in the
recirculation line following the pump. Pipe materials of small
diameter can be placed in the system in whole sections.
After choosing the materials to be tested and the water quality
conditions to be used in the test, the material to be tested is
mounted in the system. The recirculation tank is then filled with
water and adjusted to the desired quality. The first step in water
quality attainment can be accomplished with a blend of tap water
and deionized or distilled water. This blend then can be further
adjusted to provide the desired alkalinity, pH, and so forth. If
one desires to test a natural water for its corrosive tendencies,
this water can be used "as is" and weekly chlorination and pH adjust-
ments (described as follows) are all that is needed to maintain this
water quality close to original conditions.
After the tank is filled, the floating cover is put in place,
the recirculation tubing connected, the valves opened and the pump
started. Analyses of the water quality will determine what chemical
additions are necessary. Chemical additions are made easily while
the system is in operation by removing the rubber stopper in the cover
and pouring the necessary amount through the opening.
Following the initial water quality adjustment, analysis of the
water quality need only be done once a week. The analysis should in-
clude the parameters related to the aggressive nature of the recir-
culation water, the residual chlorine concentration and chemicals
that would be contributed to the water by the pipe material being
tested. When testing A/C pipe the parameters measured are tempera-
ture, pH, and concentrations of calcium, alkalinity, residual
chlorine, total dissolved solids, and any corrosion control sub-
stance being used.
The need for testing only once a week results from sealing the
system. Without the floating cover and tube ring seal, the pH would
require adjustment more than once a day when using an unstable water.
Usually the pH varies only hundredths of a unit during a weeks time
under sealed conditions. The weekly addition of chlorine, needed to
prevent growth of organisms, influences the pH more than anything else.
The pH is usually raised by adding sodium hydroxide and lowered by
adding hydrochloric acid.
During an experiment, changes in the concentrations of substances
in the water that are contributed by the material being tested indicate
whether or not corrosion is taking place. When testing A/C pipe if
the calcium concentration in the water increases only 1 to 4 mg/L as
CaC03 over a 6 month period then the A/C pipe has not been attacked
and softened. If the calcium concentration in the water increases
continually during a 6 month run and increases 8 to 10 mg/L as CaC03


total during this time, then the A/C pipe has been softened on the
surface and is being attacked by the water. Figure 6 is an example
of both situations.
At the end of an experimental run a physical examination of the
the coupon reveals whether the A/C has been softened or not.
Photographs of the pipe surface by a scanning electron microscope
show whether the surface is protected by a coating or is being
attacked. Information regarding some of the early experiments
performed with this equipment and other A/C pipe performance infor-
mation is available in the Journ. AWWA.. (2)
An important side development that resulted from use of sealed
recirculation systems was the devising of a reliable pH measuring
procedure. Unless the sample is sealed from contact with the air
when pH is being measured in a sample water that is not stable, the
pH will change considerably during measurement. Figure 7 pictures
the arrangement for measuring pH under sealed conditions. Essentially
this requires the boring of a stopper with a hole that will accept
the pH electrode and fit the sample container. This procedure has
been explained in detail in a paper that has been submitted to Journ.
AWWA for publication.^) Pre-publication information about the method
can be obtained from U.S. EPA, Drinking Water Research Division, 26
W. St. Clair St., Cincinnati, Ohio 45268.
Many studies over the past 18 months with both metal and A/C pipe
have shown that the technique described herein permit reliable small
scale corrosion and corrosion control experiments. Utilities inter-
ested in the corrosivity of their own water should consider using
this method for evaluation.
1.	AWWA Standard for Asbestos-Cement Pressure Pipe, 4 in. Through 24
in., For Water and Other Liquids. AWWA C400-77, Revision of
C400-75, AWWA, Denver, Colorado 1977.
2.	Buelow, R. W., Millette, J. R., McFarren, E. F., and Symons, J. M.,
The Behavior of Asbestos-Cement Pipe Under Various Water Quality
Conditions: A Progress Report., Journ. AWWA. 72, (Feb. 1980).
3. Schock, M. R., Mueller, W. M. and Buelow, R. W. Laboratory Tech-
nique For Measurement of pH For Corrosion Control Studies and
Water Not In Equillibrium With The Atmosphere.

25. t
Fig. 6, Calcium Concentration increase During Experiment

Fig. 7. Sealed PH Measuring Unit
t u s. GOVERNMENT PBINTING OfFICE: 1980-657-146/5590


TA 447 .B83
Buelow, Ralph W.
Laboratory techniques for
determining corrosivity of

GJt*M Ml Idmtef
Atlanta, Cja,