Paper No.

708

EFFECT OF CHLORINE ON COMMON MATERIALS IN FRESH WATER

Arthur H. Tuthill’ Blacksburg, VA Richard E. Avery’ Londondeny , NH

Stephen Lamb’ Huntington, WV Gregory Kobrin’ Beaumont, TX

ABSTRACT

Long term corrosion data for common alloy materials from test spool exposures in chlorinated fresh water are reported. Data from the published literature on the effect of chlorine in fresh water on materials are reviewed. Experience with the high initial dosages used to sterilize potable water systems is reported. As a biocide and an oxidant, chlorine has other and secondary effects on corrosion behavior. Case histories of several of these are reported. Guidelines for alloy usage are developed.

Keywords: Chlorine, stainless steels, carbon steel, cast iron, aluminum alloys, copper alloys, manganese, nickel base alloys, heat tint, fresh water, potable water.

CHLORINE! AS AN OXIDANT

Chlorine is the primary oxidant, other than oxygen (aeration), used in treating cooling water, potable and waste water, and water used in swimming pools. Other oxidants used include potassium permanganate, ozone, chlorine dioxide and bromine. While these other oxidants have important uses, only the effect of chlorine on corrosion behavior is considered in this paper. Chlorine is added to potable water in several forms; as chlorine gas dissolved in “chlorine water,” as liquid sodium hypochlorite, the common household bleaching agent, and sometimes as calcium hypochlorite granules. In whatever form added, chlorine comes to equilibrium at pH 7.5 as hypochlorous acid and hypochlorite.

HOC1 i---1 OCI-

*Consultants to the Nickel Development Institute

(1)

Copyright 01998 by NACE International. Requests for permission to publish this manuscript in any form, in part or in whole must be made in writing to NACE International, Conferences Division. P.O. Box 218340, Houston, Texas 77218-8340. The material presented and the views expressed in this paper are solely those of the author(s) and are not necessarily endorsed by the Association. Printed in the U.S.A.

Below pH 7.5 hypchlorous acid, HOC], predominates; above pH 7.5, hypochloritc, OCT predominates.

Chlorine is a very strong oxidizing agent for all metallic and organic species present. The reaction with iron follows:

HOC] + 2Fe”+ 5H,O --f 2Fe(OH),.l + Cl-+ 5H (2)

Ferric hydroxide precipitates out and is normally deposited on the wall of the piping material. The pH is depressed and the chloride ion concentration is increased by the addition of chlorine.

Chlorine also reacts with manganese and other metallic ions present, although to a lesser degree than with iron. Chlo- rine, in excess of the amount consumed in these reactions with metal ions, reacts with ammonia and organic matter forming chloramine, chloro-organic and other organic compounds. Chlorine, in excess of the demand from metallic and organic com- pounds present, is reported as free available chlorine (FAC), often less precisely as “chlorine residual”.’ Of the common oxidizers, chlorine is the only one that has a residency time long enough to keep potable water disinfected from the treatment plant to point of use. However, chlorine reacts so readily with metallic and organic materials that the residual is consumed within a few days in most waters.

EFFECT OF CHLORINE ON COMMON MATERIALS

Corrosion behavior was studied by the International Nickel Company, INCO, during the post WWII era until 1982 by placing test racks with 2” diameter specimens of different materials in field environments and reporting weight loss, corrosion rate and localized corrosion to those concerned. The data in Figures 1 and 2 were developed in this manner and made available for this report by those receiving the data from INCO. The nominal composition, original alloy designation and current UNS numbers for the materials tested are given in Table I. UNS designations are used in Figures 1 and 2, although the data were originally recorded and reported using the common alloy designations.

Genera1 corrosion rates

Figure 1 shows the weight loss, general corrosion rate for carbon steel (UNS GlOlOO), cast iron (UNS FlOOOl), austenitic cast iron (F41002); 3003 (A93003), 6061 (A96061). and 1100 (A91100) aluminum; C70600, C71500 copper nickel, 85-5-5-5 red brass (C83600) and copper (ClOlOO) in six different fresh water locations with chlorine concentrations form 0 to 20-25 ppm. In some locations two test racks were exposed, one usually for a shorter time than the other. The chlorine concen- trations, as reported by those exposing the test racks., the time of exposure for each test rack and an arbitrary test number are shown on the horizontal axis. Table II identifies the general location of each exposure. Each datum point in Figure 1 is the average of the corrosion rate for the two specimens of each alloy that were exposed on each rack.

The corrosion rate for cast iron is slightly greater than that for carbon steel, but follows the same pattern. Both cast iron and carbon steel exhibit increasing corrosion rates as chlorine increases. The maximum rates are above 5 mpy even at low concentrations. Materials that corrode at rates > 5 mpy generally require good protective coatings or inhibitors and substantial maintenance as compared to materials which corrode at l-2 mpy and are generally used bare without coatings.

The corrosion rate for the aluminum alloys and copper alloys is slightly depressed by chlorine up to about 2 mg/l chlorine. The rate increases substantially for aluminum (UNS A93003) at chlorine concentrations, 3-5 mg/l. Copper alloy specimens were not exposed at concentrations above 2 mg/l.

Austenitic nickel cast iron, aluminum and copper alloys all have low general corrosion rates allowing them to be used uncoated in most chlorinated waters.

Localized corrosion

Stainless steels and the nickel base alloys have insignificant genera1 corrosion rates of < 0.1 mpy in all chlorinated fresh waters in these exposures and were omitted from Figure 1 since only zero would have been reported. Corrosion of these materials, when it occurs, is localized and is reported as depth of attack. If the localized corrosion is in a creviced area, it is reported as crevice corrosion. If the localized corrosion occurs on the boldly exposed, uncreviced area, it is reported as pitting.

708l2

Localized corrosion behavior of stainless steels and nickel base alloys, as well as aluminum alloys, which also suffer localized corrosion, is shown is Figure 2. The maximum depth of attack on the two specimens of each alloy exposed on the rack is reported in Figure 2, not the average as in Figure 1.

UNS A93003 aluminum suffers base plate pitting ofjust under IO mils depth in un-chlorinated water and up to 26 mils depth in chlorinated waters. A96061 was resistant in un-chlorinated water but suffered up to 18 mils depth of base pitting in chlorinated waters. Both aluminum alloys exhibit 5-7 mils depth ofcrevice corrosion in un-chlorinated water and up to I5 mils depth in chlorinated waters. While the data show these aluminum alloys will suffer some base plate pitting and crevice corro- sion in chlorinated fresh waters, the low general corrosion rate and the moderate depth of localized corrosion, are within ranges that permit the use of these aluminum alloys in applications, such as slide gates, where some surface corrosion and some maintenance can be tolerated.

Stainless steels S304Oll and S31600 are resistant to localized corrosion in un-chlorinated and chlorinated fresh waters up to 2 ppm chlorine. At 3-5 mg/l chlorine, there was incipient pitting of S30400 base plate and 16 mils depth of attack in creviced areas. S31600 base plate was resistant up to 5 mg/l chlorine, but suffered 5 mils depth of attack in creviced areas, There was no general or localized corrosion of S304OOand S316OO specimens in the 3 1-32 day exposure at 20-25 mgil chlorine. The test rack in this exposure was not fully immersed, but above the rolls of a fruit washing machine and subjected to water spray. These data indicate S30400 and S3 1600, and S30403 and S31603, the low carbon grades used for welded fabrications, should resist long term exposure in most chlorinated fresh waters, which is in agreement with general experience. These data also indicate that for long term, continuous exposure towards the high end of the 3-5 mg/l range of chlorine, S31603 would be a somewhat more conservative choice than S30403.

Types S31700, l7-4PH (Sl7480), JS 700 (N08700), and Carpenter 2OCb3 (N08020). alloy 825 (N08825) and the nickel-chromium-molybdenum alloys C, B, G and 625 (N10002), (NlOOOl), (N06007), (N06625) respectively were resistant to chlorinated waters with up to 2 mg/l chlorine. N08700, N10002, and NO6625 were resistant in waters with 3-5 mg/l chlorine. Nickel-copper alloy 400 (N04400) suffered severe crevice corrosion in chlorinated fresh waters with 2 mg/l or less chlorine. NO8020 suffered 5 mil depth of crevice corrosion and NO8825, 1 mil depth in waters with 3-5 mg/l chlorine. The other alloys were not included in the 3-5 mg/l exposure.

These data strongly support the widespread use of, and preference for the common, Types 304/304L and 316/316L, UNS S30403 and S31603, stainless steels for chlorinated fresh water service as encountered in potable and waste water treat- ment plants, cooling water systems, swimming pools and other similar applications. These data also support use of austenitic nickel cast iron, aluminum and copper alloys in chlorinated waters up to 2 mg/l chlorine and possibly higher, in applications where some corrosion and occasional maintenance is acceptable. The data further indicate that well applied coatings or inhibi- tors are needed to keep the corrosion rate of carbon steel and cast iron below 5 mpy even in mildly chlorinated waters,

Data from the literature

There are numerous papers published on the effect of chlorine on materials in saline waters, but few on the effect of chlorine on materials in fresh water. Boffardi found that additions of up to 0.5 mg/l chlorine had little significant effect on the normal corrosion rate of carbon steel in potable water in 3-4 day tests. Above 0.5 mg/l chlorine the corrosion rate increased rapidly reaching 0.63 mm/y (25 mpy) at 1.0 mg/l chlorine, Figure 3.2 The much lower corrosion rates from Figure I as com- pared to Boffardi’s data in Figure 3, are believed due to the reduction in corrosion rate that occurs in longer exposures with scale formation. Boffardi’s data is for 3-4 day exposures of bare steel before any significant film or scale could form. The data in Figure I are for 30 to 365 day exposures, time enough for films/scale to form and reduce the initial unfilmed corrosion rates that Boffardi measured.

Disinfection

It is common practice to disinfect potable water systems with 25 mg/l minimum chlorine for 24 hours before placing the system in service; after major overhauls; or long outages. AWWA Standard C653 “Disinfection of Water Treatment Plants” requires injection of sufficient chlorine to produce 25 mg/l minimum chlorine and tested at the end of 12 hrs to insure that the concentration has not dropped below I5 mg/l. If below I5 mg/l the disinfection must be repeated. Contrary to what might be expected from the long term data just reviewed, short term exposure to 25-50 mg/l chlorine dosages appears to be beneficial, not detrimental, to performance of stainless steel. Lewus, et. al., in studies of metal pickup in potable water reported, “A significant feature of the static exposure tests, (at 50 ppm Cl2 for 24 hrs) was that the release into water was reduced for the second exposure of the sample to the solution”. Refer to Table III. For continuous exposure at 1.5-3.0 ppm Cl2 metal release rates decreased

703l3

markedly with time of exposure as shown below.3 In other work the author has been engaged in, the beneficial effect of short term exposures to 2.5 ppm chlorine in reducing metal pickup in ultra high purity water systems has been observed. It seems likely that these short term exposures allow chlorine to oxidize some of the unoxidized material in the film, enhancing its corrosion resistance. Long term exposures to more modest levels of chlorine are likely to be detrimental, however, as the data in Figures 1 and 2 show.

OTHER EFFECTS OF CHLORINE

The principal reason chlorine is added to potable water is for disinfection, i.e. control of bacterial harmful to humans. Chlorine also controls bacteria that are harmful to metals and in this respect is quite beneficial to the performance of stainless steels and other materials used for handling fresh water. When bacterial harmful to metals are uncontrolled, or in special situations where chlorine is not fully effective, microbiologically influenced corrosion (MIC) of steel, cast iron, stainless steel and other materials may occur.

Precipitation of iron and manganese

A second reason chlorine is added in potable water treatment is to precipitate iron and, to a lesser extent, manganese so these materials can be filtered from potable water. Potassium permanganate is often added, in addition to chlorine, in order to more effectively precipitate manganese. The ferric hydroxide and manganic hydroxide precipitates form a black deposit on walls of the pipe. The deposit is not harmful to stainless steel base metal or the weld itself, but has led to corrosion in the heat affected zone (HAZ) of welds covered by heat tint scale in fresh waters of low chloride ion content in which under deposit corrosion would otherwise be unlikely to occur.4 The HAZ of welds from which heat tint had been removed with rotating fiber brushes were as resistant as the base metal although covered with the black deposit, indicating 304L in the absence of the heat tint scale is resistant to under deposit corrosion as would be expected in a low chloride fresh water.

The dark heat tint scale formed alongside welds during welding of stainless steel is a thicker chromium oxide scale with a mixture of iron, nickel and other oxides. In the formation of this heat tint scale, chromium diffuses outward from the base metal in the zone heated to temperatures where chromium diffusion will occur. The diffusion of chromium into the scale leaves a thin chromium reduced layerjust beneath the heat tint scale. This thin layer is lower in chromium, the primary constituent that gives stainless steel its good corrosion resistance. Corrosion, that would not occur elsewhere, can initiate in such HAZs unless the heat tint scale and the thin chromium depleted layer just beneath are removed. Removal by rotating fiber brush, pickling or electropolishing readily restores base metal corrosion resistance.

Laboratory investigations, conducted by the University of Tennessee, using polarization techniques showed the base metal when clean and free of heat tint scale to be passive and resistant as is normal behavior for stainless steel, Figure 4.5 The vertical section of the dotted curve shows the normal passive range typical of stainless steel. The solid line for the HAZ as welded (with the heat tint scale intact and a chromium reduced layer just beneath) has no significant vertical section, i.e. no significant passive range.

Gallionella bacteria - manganese - chlorine

In an earlier instance, severe localized corrosion of stainless steel occurred in a chlorinated, high manganese fresh water where gallionella bacterial were present in substantial quantities.6

It is clear that in addition to the direct effect of chlorine on materials as reported in Figures 1 and 2, chlorine can enhance performance by controlling bacteria, or alter the environment by precipitating deposits which may degrade perfor- mance, or even enter into synergistic effects when significant gallionella bacterial are present.

Intermediate and continuous chlorine injection in cooling water

Chlorine is also used to treat cooling water for process units and air conditioning, although its use is being increasingly curtailed, sometimes with adverse effects as the following case shows. In this case chlorine had been injected into a cooling water system using 304 heat exchanger tubing for about 8 years with excellent tubing performance. Dosage was continuous but low, OS-I.0 mg/l chlorine. Restrictions on chlorine discharge forced a change to intermittent injection, I2 mg/l for 6 minutes, declining to 0.5-1.0 mg/l at the end of the hour. After 8 more years, the 304 tubing was uniformly pitted in some cases to perforation due to exposure to the higher chlorine residual for extended periods of time.

70al4

Introduction of, and control of, the chlorine addition

Chlorine is available as liquid chlorine in steel containers, as sodium hypochlorite solution or as calcium hypochlorite granules and sometimes as chlorine dioxide. Liquid chlorine is drawn from the bottles through gas flow meters and normally introduced as “chlorine water,” water in which 50-100 ppm of gaseous chlorine is dissolved. Liquid sodium hypochlorite may also be metered through small metering pumps into water to be treated. It is important that the chlorine water or sodium hypochlorite be introduced into the center of the pipe to mix thoroughly with the full volume of the water as intended, and not at the pipe wall where high concentrations can run down the side of the pipe causing localized corrosion, It is also important to carefully monitor and control the addition and avoid overdosing which has led to severe corrosion of stainless steel product water piping in several Mideast desalination plants. When calcium hypochlorite granules are used, they should not be broadcast in a manner that permits settling on stainless steel piping. The micro-environment around calcium hypochlorite granules resting on wet stainless steel piping can lead to serious pitting.

Venting of chlorine vapors

Chorine vapors above chlorinated water has caused general pitting and staining of stainless steel piping only half full; in ladders just above the water line in clear wells; and on the outside of piping in poorly ventilated chambers. One utility that has experienced the staining of ladders and structurals in clear wells has not yet, after 30 plus years of service, found it necessary to replace, treating the corrosion as a cosmetic rather than a structural problem. In other instances where chlorine vapors have not been adequately vented from piping, replacement has been required in a few years. Watanabe’s data, Figure 6, shows the rapid drop in pH with time that occurs in water condensed from the gaseous phase of unvented receiving tanks. In 24 month tests in the gaseous phase of chlorinated water storage tanks, Watanabe found duplex alloy Type 329, UN S32900 to be resistant to the rusting that occurs on 304L and 316L.7 These investigations show that adequate venting is essential to prevent rusting of stainless steel in the gaseous phase above chlorinated water. These investigations also show that Type 329, UN S32900 would be an excellent candidate for the upper section of stainless steel tanks containing chlorinated water, an area which cannot easily be vented.

DISTRIBUTION SYSTEMS

New York City, Tokyo, Seoul, Sweden, Switzerland, and Finland have begun using stainless steel in critical sections of their distribution systems. Germany has used stainless steel as linings in underground reservoirs. While no specific dam on the chlorine concentrations in these distribution systems are available, 2-3 mg/l of chlorine may be reasonable assumed. Enough chlorine must be added to the finished water at the treatment plant to control bacteria to the point of use.

The only long term report on performance of stainless steel in distribution systems is from the Tokyo Water Depart- ment, Figure 5. In 1980, after a 10 year materials suitability investigation and evaluation program, the Tokyo Water Department began replacing all leaking and new installation connectors with Type 3 I6 stainless steel. The 50 mm diameter connector joins the ductile iron submain in the street to the meter at the dwelling, Figure I. The stainless steel connector is insulated at both ends, Figure 2 shows that by 1987 35% of the dwellings in Tokyo had been fitted with Type 316 stainless steel connectors. In a 1991 interview with Tokyo Water Department officials arranged by the Japanese Stainless Steel Association, it was reported that the 3 I6 connector replacement program had reached 50% and was scheduled to reach 100% by the year 2000. Tokyo Water Department also reported that as of 1991 they had no indication of internal or external soil corrosion of these Type 316 connec- tors.

SUMMARY AND CONCLUSIONS

I. The corrosion rate of carbon steel and cast iron increases significantly with as little as 0.5 mg/l chlorine and continues to increase as the residual increases. The rates are high enough to indicate coatings or inhibitors and maintenance are needed for these materials in chlorinated fresh waters.

2.

3.

Austenitic nickel cast iron corrosion behavior is not significantly affected up to 2 mgfl chlorine in fresh water.

Aluminum alloys suffer measurable general corrosion and localized corrosion in raw and chlorinated raw water, but not enough to preclude their use for applications where some corrosion and maintenance can be tolerated.

4. Copper base alloy corrosion behavior is not significantly affected up to 2 mg/l chlorine in fresh water.

70855

5.

6.

7.

8.

9.

10.

il.

12.

“Water Supply Operations, Water Treatment,” AWWA, (Denver CO: 1995)

B.P. Boffardi., “Corrosion and Deposit Control in Mill Water Supply” TAPPI Engineering Conference (1992). pp 953-

M.O. Lewus., D. Dulieu, W. Tupelhome and SC. Holeson, “ A Study of the Potential for the Migration of Metals from Stainless Steel Systems into Chloride and Hypochlorite Bearing Waters”

4.

5.

R.E. Avery, R.W. Lutey, J. Musick,, K.E. Pinnow, and A.H. Tuthill, Materials Performance 35,9(1996): pp 59-62

Ping Li, R.A. Buchanan, CD. Landin, A.H. Tuthill, R.E. Avery, P. Angell, and D.E. Sachs, “Effects of Weldments in Microbial Fresh Waters,” CORROSlON/95, paper no. 192 (Houston TX: NACE International, 1995).

6. J. Treberg, K. Pinnow, L Remerski, “The Role of Manganese Fixing Bacteria on the Corrosion Resistance of Stainless Steel,” CORROSlON/90, paper no, I51 (Houston TX: NACE International, 1990).

7. Watanabe, S. et al., “Exposure Tests for stainless steel in water facilities”, Stainless, (1995). No. 4, p. 2.

Chlorine has multiple effects on the corrosion behavior of Types 304/304L and 316/316L stainless steels. The first effect is beneficial in limiting bacterial activity that, in the absence of chlorine or other biocides, might lead to micro- biologically influenced corrosion (MIC). Initial short term 25-50 mg/l dosages for 24 hours for disinfection also appear to enhance the resistance of the normal protective film on stainless steel and are rated beneficial.

Long term corrosion test coupon data in chlorinated fresh waters with up to 5 mg/l chlorine support the widespread use of stainless steel in municipal waste water and potable water treatment plants; fresh water cooled condensers and heat exchangers; swimming pools and other similar fresh water applications.

For applications where chlorine concentrations are expected to be in the 4-5 mg/l range for much of the time, Type 3 16L would be a more conservative choice as compared to Type 304L.

Prevent or remove heat tint scale. Chlorine injection into raw water results in the precipitation of insoluble ferric hydroxide, manganic and other metallic hydroxides. Under deposit type crevice corrosion or MIC has occurred in the heat affected zone of welds covered by heat tint scale in low chloride waters where iron manganese deposits covered the weld area.

Other precautionary measures associated with chlorine injection include injection in the center of the pipe, avoiding overdosing, and venting of the moist chlorine vapors from confined areas.

The clean, chlorinated fresh water of distribution systems appears be an ideal environment for long term, minimal maintenance, performance of stainless steel.

More highly alloyed stainless steels and chromium-containing nickel base alloys are not corroded in waters with up to 5 mgiJ chlorine. However, nickel-copper alloy 400 can suffer severe crevice corrosion in mildly chlorinated fresh water.

These data and case histories indicate that in fresh water, chlorine has both beneficial and detrimental effects on the common grades of stainless steel. Conclusions 5-10 above are guidelines for usage and should be followed for best performance until more data on the effect of chlorine and other oxidants on the common grades of stainless steel are developed and published.

REFERENCES

ACKNOWLEDGMENT

The authors wish to thank the Nickel Development Institute for their support in the preparation of this paper

7OBl6

Table I. Nominal Composition Wt %, Common Name, UNS Number

Common IUNS 1 C 1 Cr INiIMoIMnICuI Other 1304 1 S30400 1 0.08 1 19 1 9 1 - 1 2 1 - I I

I316 1 S31600 1 0.08 I 17 I 12 I 2.5 I 2 I - I I

I 1

Allov G 1 NO6007 1 0.05 1 22 I1 ieml 6 1.5 2 Cb 2, Co 2.5, Fe 20

$0 1 ClJ 2.5 Alloy B NlOOOl 0.12 1 Rem Z

Alloy C N10002 0.08 15 Rem 16 1 , I 1 Co 2.5. W 4. Fe 5.5 ~~ Alloy 625 NO6625 0.1 21 Rem 9 0.5 I 1 Cb 4. Ti 0.4. Fe 5 I

(*IRem = remainder of weight 96

Table II. The Location And Nature Of The Waters In Which Test Racks Were Exposed

Text No. General Location And Water

l&2 Untreated Lake Ontario water at potable water treatment plant.

3,4, S&6 Treated Lake Ontario water at different locations in potable water treatment plant.

7&S Treated Midwestern water containing 790 ppm chlorides at potable water treatment plant.

9& 10 Chlorine treated fresh water at Florida orange juice plant. Test rack was above the rolls and

covered by the spray.

Table III. Static Exposure Results Hypochlorite solutions at 35”C, 21 days3

Material

1.4301

(304)

1.4404

(316L)

1.4462

(2205)

Solution

50 ppm Cl 1”’

50 ppm Cl 2”d

50 mm Cl 1”’

50 mm Cl 2”*

50 pm Cl 1”’

50 ppm Cl 2”d

Fe

(ppb)

22

18

170

180

45

30

Ni Cr

(ppb) (ppb)

760 585

42 80

6.5 45

7.5 25

160 430

13 40

Table IV. Metal Transfer Rates For Qpe 304 at 35°C Into Hypochlorite Treated Recirculated Water3

r Hypochlorite

Addition

1.5-3.0 Cl2 ppm

Continuous

Chloride

Content (ppm)

295

Elapsed

Time

hrs

24

480

2690

Transfer Rates

ug hr’ m*

Ni Cr

16.5 60

1.5 32

0.37 0.73

708/8

+F41002 ---

0

Chorine Co2ncentratLn (mg/;)

Zhlorinelevcb (mg/l) 0 0.2 0.8-1.0 2 3-5 20-z

Days exposal 365 114 366 366 92 92 Et 2

rest number 1 3 2

f 6 9 10

Figure 1 Corrosion Behavior of Various Alloys in Raw and Chlorine-Treated Fresh Water

?a Chlorine 3-5 ma/l T&s7&0

20 L Crcvicz Cormsim Na Jnchdccl in This Test

10

10

22

s A93003 A96061 530400 531600 531700 NO8020 S17400 NO8825 NO6625 NO6007 NlOOOl NlOOO2 NO8700 NO4400

5% :

Clhrine 0.81 .O mg/l Tests4&5

Pm

g w 10 e ‘0 u A93003 A96061 S30400 531600 S31700 NO8020 517400 NO8825 NO6625 NO6007 NlOOOl N10002 N087M) NO4400

E” Chlcflna 0.2 mg/l Test 3

L .- 3

x ‘O

A93003 A96061 S30400 S31600 S31700 NO8020 S17400 NO8825 NO6625 NO6007 NlOOOl NlOOOZ NO8700 NO4400

3o Chlotinr 0 mgA Tests1 82

Figure 2 Maximum Depth of Localized Corrosion of Aluminum and Stainless Steels Versus Chlorine Concentration

0 0.2 0.4 0.6 0.8 1.0 1.2

AW~O. chlorine . mg/L

FIGURE 3 - Effect of Free Chlorine on Corrosion of Mild Steel

1,600 304L-308 Filler Metal I

.OOlti Current Density @A/cm*)

FIGURE 4 - Polarization curves for 304L as welded and pickled

708/l 1

FIGURE 5 - Increasing Use of Stainless Connector Piping and Decreasing Water Leakage Rate - Tokyo

FIGURE 6 -Results of measurement of chlorine iron and pH of the water condensed in the gaseous phase of receiving

tank (in cases where forced exhaust is not performed for the gaseous phase).’

7ow12