Chloride Local Corrosion in Stainless Acom - 3 - 2011
Chloride Local Corrosion in Stainless Acom - 3 - 2011
Chloride Local Corrosion in Stainless Acom - 3 - 2011
3 - 2011
Introduction
The European Drinking Water Directive sets a maximum limit of 250 ppm for chlorides in drinking water but does not contain guidelines for chlorine. The WHO drinking water standard states that 23 ppm chlorine should be added in order to gain a satisfactory disinfection and adequate residual concentration. The residual chlorine has a significant influence on the corrosion behaviour of stainless steels and may have detrimental consequences in the form of localized corrosion if an inappropriate stainless steel grade is used. This article clearly demonstrates that the novel duplex grades LDX 2101 and LDX 2404 provide attractive alternatives for handling potable water and cooling water. They also have a price less affected by nickel price fluctuations and higher strength compared to the standard austenitic grades 4307 and 4404. In 30-day laboratory tests, the lean duplex grade LDX 2101 performed as well as or better than 4307 at both 30C and 50C. It is also shown that the presence of crevices strongly increases the risk for localized corrosion in a chlorinated environment.
www.outokumpu.com
acom 3 - 2011
Summary
In drinking water systems the main stainless steel grades used are the standard austenitic stainless steel grades 4307 (304L) and 4404 (316L), with the grade selection depending on the chloride and chlorine levels in the water. The lean duplex grades LDX 2101 and LDX 2404 provides attractive alternatives, with a more stable price and higher strength level, but there is little available data on their use in drinking water systems. The European Drinking Water Directive sets a maximum limit of 250 ppm (mg/L) for chlorides in drinking water but does not contain guidelines for chlorine. Drinking water is normally treated to give a residual level of 0.2 to 0.5 ppm of chlorine to kill bacteria, but the actual concentrations added are usually higher. The WHO drinking water standard states that 2 3 ppm chlorine should be added to water in order to gain a satisfactory disinfection and adequate residual concentration. For a more effective disinfection the residual amount of free chlorine should exceed 0.5 ppm after at least 30 minutes of contact time at a pH value of 8 or less. The residual chlorine has a significant influence on the corrosion behavior of stainless steels. The remaining of residual chlorine in drinking water is a major factor leading to the ennoblement of the natural potential of stainless steel. This oxidizing effect of chlorine may have detrimental consequences in that stainless steels may suffer from localized corrosion if an inappropriate grade is used. The aim was to understand and determine to what extent residual chlorine levels at various chloride contents will affect the localized corrosion behaviour of the standard austenitic stainless steel grades 4307 and 4404, also the duplex grades LDX 2101, LDX 2404 and 2205. A simulated chlorination system was created in which the specimens were immersed for 30 days at 30C and 50C at chloride levels of 200 ppm and 500 ppm, with residual chlorine levels of 0.2, 0.5 and 1 ppm at pH 6.57.5. The specimens were investigated by visual examination and microscopy. The duplex grades LDX 2404 and 2205 perform very well in all the chlorinated environments tested. The lean duplex grade LDX 2101 performed as well as or better than 304L at both 30C and 50C. The results also indicated that the presence of a crevice increased the risk for localized corrosion in a chlorinated environment. This study demonstrates that duplex stainless steels are good candidates to use in water pipes or water storage tanks.
Keywords: drinking water, chloride, chlorination, total residual chlorine (TRC), localised corrosion, stainless steel.
acom 3 - 2011
1 Introduction
Stainless steel use for drinking water applications is increasing in the world. Stainless steels offer several advantages compared to other materials, such as mild steel, cast iron and copper which have been used for decades. First of all, stainless steels have generally excellent corrosion resistance and require little maintenance. There is no need for any protective coating or any protective system. Correct grade selection and good practice will minimize the risk of any localized corrosion. Therefore there is practically no contamination of water in contact with stainless steel, as has been demonstrated in the investigation [1] shown in Figure 1.
Fig. 1 Nickel (Ni) and Chromium (Cr) content of water drawn from
Figure 1 show the leaching values for Cr and Ni were less than 5% of the maximum levels permitted by the European Drinking Water Directive (50 and 20 g/L respectively) [2]. The low leaching levels from the use of stainless steel in the drinking water system are clearly of benefit in this situation. Another point to be considered is the mechanical properties. The good ductility, strength and weldability enable the use of lightweight structures, for example thin walled tubes. Among the stainless steels, the duplex materials exhibit much higher mechanical strength than corresponding austenitic grades as shown in Table 1. Compared to other materials used for applications in the potable water distribution network, duplex grades
Minimum mechanical strengths at 20C of hot rolled plate/cold rolled strip and sheet according to EN 10088-4 and EN 10028-7 when applicable [3, 4, 5].
Table 1
Tensile Strength MPa 520/540 500/520 520/530 520/531 680/700 750/750 700/700 Elongation % 45/45 45/45 45/40 45/41 30/30 25/25 25/20
Outokumpu steel names Austenitic 4301 4307 4401 4404 Duplex LDX 2101 2205
0.2% Yield Strength MPa 210/230 200/220 220/240 220/241 480/530 550/550 460/500
LDX 2404
* LDX 2101 is not yet listed in EN 10028-7. ** LDX 2404 is not yet listed in EN 10088-4 or EN 10028-7. Data for LDX 2404 corresponds to the internal standard AM 641.
acom 3 - 2011
allow a reduction in wall thickness and consequently reduces investment costs. All together stainless steels give a life cycle cost benefit. The two main alloying elements of stainless steels are chromium (Cr) and nickel (Ni). From a general point of view, chromium improves the pitting corrosion resistance whereas nickel additions are made for controlling microstructure. Further alloying elements may be added like molybdenum (Mo) for increasing pitting resistance or nitrogen (N) for improving mechanical properties and resistance to pit initiation. Depending on the stainless steel composition and chloride content of water, these materials may be resistant to aqueous corrosion in a wide range of pH at ambient temperature. Stainless steels ability to resist pitting corrosion may be estimated by calculation of the Pitting Resistance Equivalent Number (PREN). Equation (1) gives the most frequently employed formula for PREN calculation.
PREN= Cr (%) + 3.3 Mo (%) + 16 N (%) Equation (1)
In drinking water systems the main stainless steel grades used are the standard austenitic stainless steel grades 4307 and 4404. The grade selection depends on the chloride levels of the water and also on the severity of the crevices the materials are exposed to, as shown in Table 2 from the Nickel Development Institute. The chloride content of the water is the most important parameter because of its influence on localized corrosion, crevice corrosion in particular. The European Drinking Water Directive sets a maximum limit of 250 ppm (mg/L) for chlorides in drinking water but does not contain guidelines for chlorine [2].
Chloride level guidelines for waters at ambient temperatures [6].
Table 2
Chloride level (ppm, mg/L) < 200 200 1000 1000 3600 > 3600 and sea water
Suitable grades 1.4301 (304), 1.4307 (304L), 1.4404 (316L) 1.4404 (316L), 1.4462 (2205) 1.4462 (2205), 6% Mo Super austenitic, Super duplex 6% Mo Super austenitic, Superduplex
2 Water Chlorination
Chlorination is a one of many methods that can be used to disinfect water and control bacteria. Sodium hypochlorite (NaOCl) is the form of chlorine normally use for chlorination process because it is cheap and easy to dose. When chlorine added to water, it immediately begins to react with compounds found in the water to give hypochlorous acid (HOCl) and hypochlorite (OCl-). The remaining amount is called free residual chlorine. The free residual chlorine is typically measured in drinking water disinfection systems to find if the water contains enough disinfectant. Typical levels of free chlorine in drinking water are 0.2 0.5 ppm [7], but the actual concentrations added are usually higher. The WHO drinking water standard states that 2 3 ppm chlorine should be added to water in order to attain a satisfactory disinfection and maintain residual concentration [8]. The maximum amount of chlorine one can use is 5 ppm. For effective disinfection the residual amount of free chlorine should exceed 0.5 ppm after at least 30 minutes of contact time at a pH value of 8 or less. The residual chlorine has a significant influence on the corrosion behaviour of stainless steels. The remaining residual chlorine in drinking water is thought to be a major factor leading to the ennoblement of the natural potential of stainless steel. This oxidizing effect of chlorine may have detrimental consequences and stainless steels may suffer from localized corrosion if an inappropriate grade is used.
acom 3 - 2011
This work was conducted to understand and determine to what extent total residual chlorine levels at various chloride contents will affect the pitting and crevice corrosion behaviour of the standard austenitic stainless steel grades 4307 and 4404 , also the duplex grades LDX 2101, LDX 2404 and 2205. The recently introduced duplex grades LDX 2101 and LDX 2404 provide an attractive alternative, with a more stable price and higher strength level, but there is little available data on their use in drinking water systems.
Steel grades, surface finish, thickness, PREN values and the chemical composition for materials used in long term chlorination.
Table 3 Thickness (mm) 3 3 3 3 3 Typical composition, weight-% Cr Ni Mo N Others 18.1 17.2 21.5 24.0 22.0 8.1 10.1 1.5 3.6 5.7 2.1 0.3 1.6 3.1 0.22 0.27 0.17 5Mn 3Mn
EN Product Conditions 2B 2B 2E 2E 2E
LDX 2404
2B: Cold rolled, heat treated, pickled, skin passed 2D: Cold rolled, heat treated, pickled 2E: Cold rolled, heat treated, mech. desc, pickled
Coupons of duplicate plain (sheet), welded and crevice specimens with size 60x30x3 mm were used with an as-received surface as show in Figure 2A. All cut edges were wet ground to 320 mesh. The crevice samples had a 12 mm hole placed in the centre of the sample. Samples were bolted together with INCO crevice formers on both sides of specimen (Figure 2B). All crevice formers were tightened with a torque of 1.58 Nm. It was verified that there was no electrical contact between the samples and the screw. Plain (sheet) and welded specimens were suspended in the solution on platinum wires to minimize crevice effects when investigating pitting corrosion.
Fig. 2 Coupons of plain (sheet), welded and crevice specimens used for long term testing. Fig. 2A Fig. 2B
Sheet
Weld
Crevice
acom 3 - 2011
6
Table 4
Welding wire TIG (EN ISO designation) Avesta 308L-Si/MVR-Si (W 19 9 L Si) Avesta 316L-Si/SKR-Si (W 19 12 3 L Si) Avesta LDX 2101 (W 23 7 L) Avesta 2205 (W 22 9 3 N L) Avesta 2205 (W 22 9 3 N L)
Nominal composition, weight-% Cr Mo Ni N Si 20.0 18.5 23.0 23.0 23.0 2.6 <0.5 3.1 3.1 10.5 12.0 7.0 8.5 8.5 0.14 0.17 0.17 0.85 0.85 0.40 0.50 0.50
LDX 2404
The welded samples were obtained by tungsten inert gas welding (TIG). The welding was done with filler material and welding conditions as specified in Table 4 and Table 5 below. This welding process is often used for water applications. All samples have the same thickness of 3 mm. Weld samples were pickled in mixed acid (3M HNO3 and 3M HF). Chloride (Cl-) containing electrolytes with various total residual chlorine (TRC) levels, at pH 6.5-7.5, were prepared from distilled water. Chloride ions were added to the level of 200 ppm and 500 ppm as sodium chloride (NaCl). The solutions were dosed with a stock solution containing 1000 ppm of sodium hypochlorite to obtain various predetermined total residual chlorine concentrations. Total residual chlorine (TRC) is defined as the sum of hypochlorous acid (HClO) and hypochlorite ion (ClO-) concentrations.The amount of residual chlorine was measured with a colorimeter using the diethyl-p-phenylene diamine (DPD) method [10]. Three total residual chlorine concentrations were investigated that correspond to the residual concentration typically used for water disinfection treatments: 0.2, 0.5 and 1 ppm. The open circuit potential (OCP) was monitored for 30 days in the test solutions with the different residual chlorine levels and a temperature of 30C or 50C. The chlorine was dosed once every 57 days to maintain the residual chlorine level. After testing the specimens were examined and the depth of maximum attack was measured with a light optical microscope. A depth exceeding 0.025 mm was defined as localised corrosion.
Joint design Butt joint Butt joint Butt joint Bead on plate Bead on plate
LDX 2404
acom 3 - 2011
The stainless steel samples were immersed in the test solutions with 200 ppm and 500 ppm chloride at 30C and 50C for 24 hours before the start of chlorination. The open circuit potential (OCP) usually stabilised after ~4 hours and was typically found to lie in the range 190220 mV for the sheet specimens after 24 hours. The values were somewhat higher for the weld and crevice specimens. The addition of sodium hypochlorite gave a strong increase in the open circuit potential. After a certain time, typically 1024 hours the potential stabilised and the OCPMax could be measured as shown in Figure 3. The result shows that OCPMax increases with TRC level because the oxidising power of the solution increases, Table 6.
Fig. 3 Evaluation of maximum open circuit potential in chlorinated water.
800 700
OCPMax 720 mVSCE
10
15 Time (days)
20
25
30
Average OCPMax of five different steel grades in water containing chloride and total residual chlorine at 30C and 50C.
0.2 ppm TRC, 30C 425 460 493 NT 480 345 375 370 NT 397
0.5 ppm TRC, 30C 544 556 619 NT 589 594 588 549 NT 625
1 ppm TRC, 30C 722 794 819 770 747 652 679 722 771 673
0.2 ppm TRC, 0.5 ppm TRC, 50C 50C 429 393 363 NT 362 349 356 427 NT 379 454 445 444 NT 535 347 399 367 NT 486
1 ppm TRC, 50C 723 683 679 682 671 736 730 712 720 683
acom 3 - 2011
The OCPMax after chlorination compared to the situation before chlorination is shown in Figure 4. The increase in OCP was about 200 mVSCE for 0.2 ppm TRC, about 300 mVSCE for 0.5 ppm TRC and about 500 mVSCE for 1 ppm TRC. This indicates that even at low TRC concentrations the open circuit potential increases.
Fig. 4 The potential increase (OCPMax-OCP) versus total residual chlorine (TRC)
0.2
0.4
0.6
0.8
1.0
1.2
The occurrence of localised corrosion is frequently seen as a drop in the open circuit potential, as illustrated in Figure 5. After 30 days, visual and microscopy examination showed that pitting had occurred for the welded 4307 and LDX 2101 (Figure 7A). These both showed a rapid drop in OCP during testing. No corrosion was seen for the welded 2205 which maintained a high OCP throughout the test.
Fig. 5 Corrosion potential change of TIG welded specimens of 4307 , LDX 2101
and 2205 in 500 ppm chloride and 1 ppm TRC at 50C showing the potential drop associated with the onset of pitting corrosion.
900 800 700 600 Potential (mVSCE) 500 400 300 200 100 0 -100 0 100
4307-weld 2205-weld LDX 2101-weld Pitting corrosion 50C, 500 PPM CI-, 1 PPM TRC No Pitting corrosion
Pitting corrosion
200
300
400
500
600
700
Time (hours)
acom 3 - 2011
for 4307 and 2205 in 200 ppm chloride and 1 ppm TRC at 30C.
900 800 700 600 500 400 300 200 100 0 0 100 200
No crevice corrosion
4307-weld 4307-crevice
Potential (mVSCE)
Pitting corrosion
Crevice corrosion
300
400
500
600
700
Time (hours)
Crevice corrosion
Figure 6 shows the OCP change of TIG welded and creviced samples of 4307 and 2205 in 200 ppm chloride and 1 ppm TRC at 30C. Visual examination showed that localized corrosion had occurred for 4307 (Figure 7B) but not for 2205.
4.3 Visual examination after 30 days.
Samples were examined after exposure in the 200 ppm and 500 ppm chloride solutions with different total residual chlorine levels at 30C and 50C for 30 days. A summary of the results from this investigation is shown in Table 7. Where corrosion occurred, the cells are filled dark blue and where no corrosion occurred the cells are light blue. Table 7 show that the lean duplex LDX 2101 was found to be at least as resistant as 4307. In all experimental conditions tested, the duplex grades LDX 2404 and 2205 perform very well with no significant localised attack. Both these grades have a high PREN (>30), whereas for the grades with PREN<30 some localised attack was observed. The results show that the alloying elements influence the localised corrosion resistance of stainless steel. For the austenitic steels, the corrosion resistance for molybdenum (Mo) containing grade (4404) is higher than for the molybdenum (Mo) free grade (4307). A higher chromium (Cr) level in combination with nitrogen (N) addition has the same positive influence for duplex grades.
acom 3 - 2011
10
Table 7
Test condition Temp. (C) 30 Chloride (ppm) 200 TRC* (ppm) 0.2 0.5 1.0 30 500 0.2 0.5 1.0 50 200 0.2 0.5 1.0 50 500 0.2 0.5 1.0 P 4307 W C P 4404 W C
*TRC = Total residual chlorine, P = Plain (sheet) sample, W = Welded sample, C = Creviced sample No corrosion Corrosion Not tested in this study Crevice corrosion not observed; possibly due to loosening of the screw, but expected based on 30C results.
The results also indicated that the presence of a crevice increases the risk for localized corrosion in chlorinated environments. Special attention should be taken, to avoid crevices in construction, since residual chlorine solution can remain in crevice areas and cause corrosion.
4.4 Comparison with engineering diagrams.
Engineering diagrams for a given steel grade as a function of temperature and chloride content are a useful illustration of the risk areas for localized corrosion in drinking water applications [11]. These diagrams are based on a combination of laboratory testing and extensive practical experience and provide a useful reference base for the present investigation. As can be seen in Figure 8 there is excellent agreement between the diagram and the present data
Fig. 8 Engineering diagram indicating the maximum temperatures and chloride concentration allowed in slightly
C 100 90 80 70 60 50 40 30 20
chlorinated (<1 mg/L) drinking water for 4307 and 4404 [11].
C 100 90 80 70 60 50 40 30 10 100 1000 Chloride, ppm 20 10 100 1000 Chloride, ppm
Green: No corrosion
acom 3 - 2011
11
for 4404 tested with 1 ppm TRC: pitting corrosion occurred only at 500 ppm chloride and 1ppm TRC at 50C for the sheet specimen, and this point is above the line. For 4307 all four of condition tested showed pitting with 1ppm TRC, and should thus lie above the boundary line. If, however the comparison is made to the 0.5 ppm data, the curve seems instead slightly too conservative. The overall agreement is thus very good, and underlines the point that the chloride tolerance of different stainless steel grades is very sensitive to the chlorination level. A summary of chlorination limits for different grades from this investigation are shown in Table 8.
Chlorination limits which did not cause corrosion in the 30 days immersion tests for different grades depending on chloride content.
Table 8
Test condition Temp. (C) 30 30 50 50 Chloride (ppm) 200 500 200 500 P 0.5 0.5 0.5 0.5
TRC limits (ppm) for different grades depending on chloride content 4307 W 0.5 0.5 0.2 C NT 0.2 NT P 1.0 1.0 1.0 0.5 4404 W 1.0 1.0 1.0 0.5 C 1.0 0.2 0.5 0.2 LDX 2101 P W C 1.0 0.5 0.5 0.5 1.0 0.5 0.5 0.2 <0.2 <0.2 0.2 <0.2 LDX 2404 P W C 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 P 1.0 1.0 1.0 1.0 2205 W 1.0 1.0 1.0 1.0 C 1.0 1.0 1.0 1.0
0.5 <0.2
P = plain (sheet) sample; W = welded sample; C = creviced sample; TRC = total residual chlorine; NT= not tested
It is important that a material is not exposed to excessive levels of residual chlorine. For effective disinfection the residual chlorine should exceed 0.5 ppm after at least 30 minutes of contact time [8]. During practical operation, the chloride content will most probably be lower than during this test. Thus, there is a good chance that the 4307, LDX 2101 and 4404 can be used successfully for normal service in water piping systems as long as problematic crevices can be avoided. In doubtful cases upgrading to LDX 2404 or 2205 may be advisable.
acom 3 - 2011
12
5 CONCLUSION
In long-term (30 days) immersion tests, the highest alloyed duplex grades 2205 and LDX 2404 performed very well in the chlorinated environments tested (200 or 500 ppm chloride, 30C or 50C). No pitting, crevice corrosion or weld attack was seen in any of the environments for these grades. The lean duplex grade LDX 2101 performed as well as or better than 4307 (304L) at all conditions tested. In the pitting test it performed as well as 4404 (316L) in 200 ppm chloride at 30C. Chlorine solution with significant residual chlorine concentrations can remain in crevice areas and cause corrosion, and therefore special attention should be taken in construction. The lean duplex steel LDX 2101 is a good candidate for water piping systems and tanks, when the water is mildly chlorinated. In more severe condition the higher alloyed LDX 2404 or 2205 are more suitable. Material selection guidelines depending on chloride content, chlorine dosage and temperature are shown in Table 9 and Table 10 below. In order to ensure good performance deposits and surface contamination should be avoided.
Summarised results of 30 day tests in chlorinated solutions containing 200 ppm chloride at 30C or 50C.
BM 304L LDX 2101 Temperature 50C 304L LDX 2101 304L LDX 2101 304L Temperature 30C LDX 2101 304L LDX 2101 304L LDX 2101
316L LDX 2404 2205 316L LDX 2404 2205 316L LDX 2404 2205 316L LDX 2404 2205 316L (LDX 2404) 2205 316L (LDX 2404 )
304L LDX 2101 304L LDX 2101 304L LDX 2101 304L LDX 2101 304L (LDX 2101) (304L) (LDX 2101 )
2205 316L LDX 2404 2205 (316L) (LDX 2404) (2205) (316L) (LDX 2404 )
2205
(Red)-possibly corrosion, not tested in this study (Green)-possibly no corrosion, not tested in this study
acom 3 - 2011
13
Table 10
Summarised results of 30 day tests in chlorinated solutions containing 500 ppm chloride at 30C or 50C.
BM 304L LDX 2101 Temperature 50C 304L LDX 2101 304L LDX 2101 304L Temperature 30C LDX 2101 304L LDX 2101 304L LDX 2101
Weld 316L LDX 2404 2205 316L LDX 2404 2205 316L LDX 2404 2205 316L LDX 2404 2205 316L (LDX 2404 )
304L LDX 2101 304L LDX 2101 304L LDX 2101 304L LDX 2101 304L LDX 2101 304L
(LDX 2404 )
(LDX 2101 )
(2205) 316L LDX 2404 2205 (316L) (LDX 2404 ) (2205) (316L)
(LDX 2404 )
LDX 2101
(Red)-possibly corrosion, not tested in this study (Green)-possibly no corrosion, not tested in this study
6 REFERENCES
[1] C.A. Powell and W.Strassburg, Stainless Steel for Potable Water Service, 2nd European Stainless Steel Congress, Dsseldorf, 1996. [2] European Drinking Water Council Directive 98/83/EC, Nov, 1998. [3] Outokumpu data sheet, Standard Cr-Ni stainless steel. [4] Outokumpu data sheet, Standard Cr-Ni-Mo stainless steel. [5] Outokumpu data sheet, Duplex stainless steel [6] Peter Cutler, Stainless steel and drinking water around the world, Nickel Development institute (NiDi). [7] The chlorine institute.INC, Chlorine effect on health and the environment, 3th Edition-Nov.1999. [8] Guidelines for Drinking Water Quality, 3rd Edition, 2008. [9] Avesta Welding handbook, 3rd Edition-Dec, 2007. [10] Pradyot Patnaik, (1995), Deans Analytical Chemistry Handbook, McGraw Hill, New York. [11] Outokumpu, Corrosion Handbook, 10th Edition-Nov, 2009.
Comments on acom and its articles or suggestions on future articles are appreciated and should be sent to the editor Andreas Persson at [email protected] This document is for information only and seeks to provide professionals with the best possible information to enable them to make appropriate choices. Although every effort has been made to ensure the accuracy of the information provided in this document, Outokumpu can not accept any responsibility for any loss, damage or other consequence resulting from the use of this publication. The information provided herein may be subject to alterations without notice.
www.outokumpu.com