Outokumpu Corrosion Management News Acom 3 2014

A CORROSION MANAGEMENT AND APPLICATIONS ENGINEERING MAGAZINE FROM OUTOKUMPU 3/2014

Qualification Testing of Lean Duplex Stainless Steels for the Process Industry

3/2014 | 2

AbstractThe materials specifier needs accepted methods for qualification or assurance testing to verify that supplied materials have acceptable properties. For duplex stainless steels the ASTM A923 standard has found extensive use as a qualification test for the duplex UNS S32205 and superduplex S32750 grades, where the primary cause for concern is the precipitation of intermetallic phases. However, lean duplex grades such as UNS S32101 or UNS S32304 present more of a challenge because these steels are much less sensitive to intermetallic phase precipitation than the higher alloyed duplex grades. The small microstructural changes associated with improper heat treatment are challenging to detect and impact toughness acceptance limits need definition. The ferric chloride immersion test used in ASTM A923 results in sub-ambient temperatures for lean duplex grades and is therefore in many instances impractical to use.

The approach explored in the present work is the use of an inhibited 5% ferric chloride solution containing additions of 1% sodium nitrate. This leads to critical temperatures for the onset of pitting which are around or above ambient. Results are presented showing the influence of the level of nitrate additions on the corrosion performance and also correlated to impact toughness data. Application to various product forms is discussed, together with proposed acceptance criteria. Finally, a limited laboratory intercomparison of the test method is presented.

These results represent steps towards development of a new variant of ASTM A923 for lean duplex stainless steels, which is aimed to remove any remaining uncertainty when specifying such grades.

Key words: stainless, testing, corrosion, sensitization, lean duplex, qualification, acceptance

Qualification Testing of Lean Duplex Stainless Steels for the Process IndustryRachel Pettersson, Carolina Canderyd, Jan Y. Jonsson, Avesta Research Centre, Outokumpu Stainless AB, Avesta, Sweden,Poul-Erik Arnvig, Outokumpu Stainless – North America, Schaumberg, IL, USA.

IntroductionQualification or acceptance testing is often required by the materials purchaser in order to be able to verify that delivered material is in an acceptable condition and will fulfill the property requirements placed on the grade. The ASTM (1) A923 standard is an established test for the standard duplex grade UNS S32205 and the superduplex S32750. Both these grades are sensitive to the precipitation of intermetallic phases, typically in temperature range 800 – 1000°C, Figure 1.

The ASTM A923 test describes three methods. Method A involves electrolytic etching in sodium hydroxide solution to reveal microstructural changes. Intermetallic phases such as sigma phase are usually seen as discrete particles at grain and phase boundaries and can easily be detected using the required magnification of 400 – 500x. According to the standard this type of screening test can be used to pass material which has an acceptable, unaffected structure, but if the microstructure appears affected or possibly affected this must be followed up by testing according methods B or C. Method B involves Charpy-V impact testing at -40°C (-40°F) while Method C involves immersion testing in 6% ferric chloride solution, with the requirement that the weight loss at the specified temperatures should not exceed 10 mdd (mg/dm2/day) measured in a 24 hour test. In practice, these two readily quantifiable tests are often used in preference to the metallographic screening.

(1) ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA, 19428-2959, USA.

Figure 1 TTT diagram shown here as the times to obtain 50% reduction in impact toughness. [2]

Tem

pera

ture

(°C)

Time (h)

11000

1000

900

800

700

600

500

400

3000.01(36 s)

0.1(6 min)

1 10 100

S32750

S32101S32205

S32304

3/2014 | 3

MaterialsThe investigated materials were UNS S32101 (EN 1.4162, Outokumpu LDX 2101®(2)) and UNS S32304 (EN 1.4362, Outokumpu 2304(2)), with the typical compositions given in Table 1. Various product forms were investigated as specified in Table 2. These were investigated in the mill annealed condition, and also after a number of laboratory heat treatments designed to simulate the variation in annealing conditions which are likely to be encountered in different mills. In addition, sensitizing heat treatments were performed at 700°C, which is the temperature of most rapid precipitation of secondary phases according to Figure 1. The sensitizing times ranged from 1 minute to 30 hours and are the holding times at the respective temperature, as detailed in previous work. [3]

Inhibited Ferric Chloride TestsAn inhibited test solution comprising 5 wt% FeCl3 + 1 wt% NaNO3 was used as described previously. [3, 5] Samples were cut to approximate dimensions of 50 x 25 mm. All surfaces were dry ground to a 120 grit finish as used in ASTM A923, followed by an acetone rinse. The exceptions were some tube, rebar and welded samples which were pickled, rather than being ground, in order to retain the original surface which is typical for the products in question. According to ASTM A923 testing of a specimen in the as-fabricated condition, rather than undergoing the prescribed grinding to 120 grit, is permitted if this is relevant to the applica-tion. All samples were weighed to the nearest 0.0001 g and the dimensions measured. The majority of testing was performed at the Avesta Research Centre laboratories (denoted Lab A); while Lab D and Lab N participated in the inter-laboratory comparison.

Testing was performed in glass beakers with a minimum of

600 ml of test solution (150 ml at Lab N). The test temperature was controlled to ± 1 °C. Each test involved immersion for 24 hours, after which the exposed samples were cleaned, rinsed and dried. Samples were then re-weighed and the corrosion rate calculated in units of mg/dm2/day (mdd). Each condition was tested at various temperatures to determine the pitting temperature as a function of the ageing time. A corrosion rate greater than 10 mdd was used as a threshold for unacceptable performance, which is the same acceptance criterion used in the ASTM A923 Method C. The critical temperature was defined as the highest temperature at which the corrosion rate was below this limit.

In order to assess the electrochemical behavior of lean duplex grades in nitrate-inhibited chloride solutions, polarization curves were also obtained at a scan rate of 20 mV/minute in 5.5 wt.% NaCl solutions which were acidified to pH 1.3 and modified by various additions of NaNO3.

The lean duplex grades such as UNS S32101 and UNS S32304 show a very different type of behavior if they are subject to inappropriate heat treatments. Most rapid precipitation of secondary phases occurs at lower temperatures, typically 600 – 1800°C, as seen in Figure 1. This is primarily associated with the precipitation of chromium nitrides and carbides. [2] Intermetallic phases appear first after much longer ageing times, typically in excess of 10 hours at the most sensitive temperature, and are thus very unlikely to be an issue.

The methodology described in ASTM A923 is unfortunately difficult to apply to lean duplex grades for a number of reasons. Firstly, the sodium hydroxide etching in Method A is not suitable for revealing the fine phase boundary precipitates which are involved, even though this can be done by an experienced microscopist working at fairly high magnification. The impact toughness in Method B is not particularly sensitive for lean duplex grades at

-40°C [3] while the ferric chloride testing in Method C results in sub-ambient temperatures which are impractical to work with in many laboratories.

The approach addressed in this work is the use of an inhibited ferric chloride solution containing additions of sodium nitrate. The sodium nitrate acts as an inhibitor, and increases the critical temperatures for the onset of pitting to around or above ambient, thus providing the possibility for a simple and practical acceptance test. This concept has been addressed in a number of recent works on lean duplex grades, [3, 4, 5] and was also discussed many years ago within the ASTM G48 working groups, for lower alloyed austenitic grades.[6] The aim of the present work is to apply the inhibited ferric chloride test to a range of product forms in the two lean duplex grades UNS S32101 and UNS S32304 order to examine its validity and usefulness.

UNS EN C N Cr Ni Mo Mn

S32101 1.4162 0.03 0.22 21.5 1.5 0.3 5

S32304 1.4362 0.02 0.10 23 4.8 0.3 1.5

Table 1 Nominal compositions of the grades investigated.

ProductSteel grade and dimension (mm)

UNS S32101 UNS S32304

Plate 12, 30 20, 30

Sheet 0.5–1.5 1.5–6

Tube Ø27x2 Ø27x1.5

Rebar/bar Ø16–25

Table 2 Tested products from the two grades.

Materials and Experimental Procedures

(2) Trade name.

3/2014 | 4

RESULTSElectrochemical tests in nitrate inhibited solutionsPolarization curves in acidified sodium chloride with and without additions of sodium nitrate are shown in Figure 2 and demonstrate the presence of a slight tendency to an active corrosion peak at ~-350mVSCE, which is depressed by nitrate additions. There is also an earlier onset of transpassive corrosion, at ~900 mVSCE in the presence of nitrate. If the breakdown potentials (defined as the potential at which the anodic current density exceeds 100 µA/cm2) are plotted as a function of temperature it is seen that there is a sharp transition between the transpassive behavior at lower temperatures and the pitting behavior, typically at potentials of <300 mVSCE, at higher temperatures. This transition is denoted the critical pitting temperature (CPT) and plotted as a function of the NO3-/Cl- in the third diagram in Figure 2. The CPT shows a virtually linear increase as the level of nitrate is increased in the range 0.5 – 2%. Use of a more dilute solution of 1.1% NaCl + 0.3% NaNO3 (i.e. with concentrations which were employed in [4] gives somewhat higher CPT, as is to be expected.

The open circuit potential measured in the inhibited ferric chloride test (5 wt% FeCl– + 1 wt% NaNO–) is close to 700 mVSCE, and it is seen from Figure 2 that this potential lies in the middle of the transition region and is thus well within the region in which the critical pitting temperatures is largely potential-independent [7].

Immersion testing in inhibited ferric chlorideIn Figure 3 the results of inhibited ferric chloride tests on a range of products of UNS S32101 and UNS S32304 are shown. The solid points represent temperatures at which the weight loss is >10 mdd, thus corresponding to a “fail” if the same acceptance criteria are applied as in ASTM A923. The open points represent a “pass”. The normal procedure used here is to test two specimens, but for clarity only a single point is given at each temperature and a “pass” given only if both specimens have a weight loss of 10 mdd or below.

For UNS S32101 the thicker (30 mm) plate materials all pass the test at 30°C, although there are some cases in which the critical temperature may be higher. The 1.5 mm sheet material has an even more homogenous structure as a result of cold rolling and this give a slightly higher critical temperature of >40°C. After a holding time of 5 minutes at 700°C a drop in the critical tempera-ture is discerned, with slightly more rapid sensitization being seen for the thinner 1.5 mm sheet material.

The trends for UNS S32304 were somewhat less pronounced. The annealed states showed critical temperature ranging from 45°C for the 30 mm plate to 30°C for the 20 mm plate and 1.5 mm sheet. There was little effect of shorter sensitization times, but a clear drop was seen after >1 hour at 700°C. This is in agreement with data reported in the literature that for a significant level of sensitization of the UNS S32304 grade, sensitization times longer than 10 hours are usually needed.[8] In contrast to UNS S32101, the investigated sheet material showed slightly more rapid sensitization than the 30 mm plate, while the 20 mm plate showed very little effect.

The challenge in trying to define an acceptance criterion for a qualification test is that all product forms must pass the test when in a correct mill-finished condition but at the same time fail if the material has been sensitized to a detrimental extent. The annealing temperature and cooling rate affect the phase ratio and also the partitioning of alloying elements between the phases, thus the corrosion resistance. In addition thick plate has a much coarser structure than cold rolled sheet and may show a great

Figure 2 Use of polarization curves in acidified NaCl + NaNO3 to evaluate critical temperatures for the transition between transpassive corrosion and pitting for 3 mm sheet of UNS S32101.

Potential (mVSCE)

Brea

kdow

n po

tent

ial (

mV SC

E)Cu

rren

t den

sity

(µA/

cm2 )

45°C,5.5% NaCl1.5% NaNO3

8°C,5.5 NaCl

1.E -01

1.E +00

1.E +01

1.E +02

1.E +03

1.E +04

-1000 -500 0 500 1000

Temperature (°C)

5.5% NaCl

1.1% NaCl0.3% NaNO3

5.5% NaCl2% NaNO3

5.5% NaCl0.5% NaNO3

0

200

600

400

800

1000

1200

0 10 20 30 40 50 60 70 80

Criti

cal p

ittin

g te

mpe

ratu

re (°

C)

NO3/Cl ratio

0

20

10

40

30

50

60

70

80

0.0 0.2 0.4 0.6

1.1% NaCl

3.3% NaCl

5.5% NaCl

qwertyuiop娒äölkjh

3/2014 | 5

Figure 3 Results of inhibited ferric chloride corrosion testing for different product forms of UNS S32101 (upper three diagrams) and UNS S32304 (lower three). Each set of vertical points to the left hand side of the diagrams represents a different variant of annealing.

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60140°F

70168°F

50122°F

40104°F

3086°F

2068°F

1050°F

00.1 1.0 10.0 100.0

S32101, 30 mm

mdd>10mdd<10

mdd>10mdd<10

Annealed

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60

70

50

40

30

20

10

00.1 1.0 10.0 100.0

S32101, 12 mm

Annealed

mdd>10mdd<10

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60

70

50

40

30

20

10

00.1 1.0 10.0 100.0

S32101, 1.5 mm

Annealed

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60140°F

70168°F

50122°F

40104°F

3086°F

2068°F

1050°F

00.1 1.0 10.0 100.0

S32304, 30 mm

mdd>10mdd<10

mdd>10mdd<10

Annealed

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60

70

50

40

30

20

10

00.1 1.0 10.0 100.0

S32304, 12 mm

Annealed

mdd>10mdd<10

Test

tem

pera

ture

(°C)

Time at 700°C/1292°F (hours)

60

70

50

40

30

20

10

00.1 1.0 10.0 100.0

S32304, 1.5 mm

Annealed

degree of segregation, while bar material, and particularly rebar, may have undergone a process annealing step rather than a separate solution annealing stage. Corrosion results from all the product forms tested for UNS S32101 and UNS S32304 are shown in Figure 4 and indicate that for both grades an acceptance temperature of 20 or 25°C fills the specified requirements. For UNS S32101, the data presented in previous work indicated that good welds in 12 – 30 mm plate show a pass temperature in

the range 35 – 45°C and can thus be treated in the same way as the base material. [3] For UNS S32304 even longer sensitization temperatures than 10 hours are really needed to see a clear degradation in corrosion properties, but it can be argued that this is hardly realistically likely to be encountered so this material can be regarded as fairly resistant to structural degradation caused by inappropriate heat treatment.

Figure 4 Compilation of corrosion data for various product forms of UNS S32101 and UNS S32304. For annealed states the critical temperatures are arranged in descending order for clarity. An acceptance limit of 20 – 25°C appears appropriate.

Time at 700°C/1292°F (hours)

Test

tem

oera

ture

(°C)

Annealed

10

00.1

20

30

40

50

60

70

1.0 10.0 100.0

S32304

30 mm plate20 mm plate6 mm plate1.5 mm sheet

Time at 700°C/1292°F (hours)

Test

tem

oera

ture

(°C)

Annealed

1050°F

00.1

2068°F

3086°F

40104°F

50122°F

60140°F

70168°F

1.0 10.0 100.0

S3210130 mm plate12 mm plate1.5 mm sheet27x2 mm tubeBar/rebar

3/2014 | 6

Effect of test variablesAny acceptance test needs to be examined in order to evaluate whether test variables have such significance that they need to be tightly controlled or whether a certain amount of variation can be permitted. Figure 5 shows the application of such sensitivity analysis to various variants of UNS S32101. Two effects appear very clearly: the starting temperature and the presence of weld oxides. If the specimen is placed directly into the preheated test solution, as is specified for the ferric chloride testing of duplex and superduplex grades in ASTM A923, then the critical temperature is lower than if the specimen is placed in a room temperature solution and then heated up to the test temperature. In the case shown, the latter leads to a pass at 60°C, while failures are seen at 35°C for the correct procedure.

Allowing oxides, from annealing or welding, to remain on the specimen surface also had a critical influence. The presence of weld oxides caused failures due to the weight loss exceeding 10mdd for 6 mm UNS S32101, at temperatures as low as 20°C. However, there were no pits seen on these specimens, so it is apparent that the measured weight loss is related to the dissolu-tion of the weld oxides in the ferric chloride solution rather than to actual corrosion attack. Not until 45°C was pitting attack

observed. If the specimens were instead thoroughly cleaned, using sand blasting followed by laboratory pickling, failure was seen only at 55°C or higher.

Tube material was tested both in the as-fabricated condition, as is permitted in ASTM A923, and after dry surface grinding to 120 mesh finish, which is required for mill products. The effect was minor to negligible, with failure in both cases occurring at 55°C. Likewise use of a newly-prepared test solution, or one which had been made up from a stock test solution prepared a week before, had only minor impact when applied to 12 mm UNS S32101. Cleaning specimens with acetone or a magnesium oxide paste had no significant effect on the weight loss when tested above the critical temperature

Finally, two different test solutions were compared: the 5% FeCl3 + 1% NaNO3 which forms the basis of this work, and the 1% FeCl3 + 0.3% NaNO3 which has been investigated in other work.[4] It can be recalled from Figure 2 that a more dilute NaCl+NaNO3 solution gave rise to a electrochemically evaluated critical pitting temperature which was some 10°C higher. However, the immersion testing actually yielded the same critical temperatures. This may reflect the lower degree of precision typically associated with immersion testing, but merits more extensive investigation.

Figure 5 Effect of test variables on results from the inhibited ferric chloride test. Each marked point represents a single sample.

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Start at

test temp.0 Start

at RT

S32101, 30 mm

mdd>10 failmdd<10 pass

mdd>10 failmdd<10 pass

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0As

welded0 Sand blast

+ pickled

S32101, 6 mm

mdd>10 failPits

mdd<10 pass

mdd 6.2–12.0

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0As rec.0 120#

S32101, 27x2 mm tube

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Start at

test temp.0 Start

at RT

S32101, 12 mm

mdd>10 failmdd<10 pass

mdd>10 failmdd<10 pass

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Acetone0 MgO

S32101, 12 mmmdd

10682

3438

8986

3768

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

05% FeCl3+1% NaNO3

0 1% FeCl3+0.3% NaNO3

S32101, 12 mm

3/2014 | 7

Inter-laboratory comparisonA second aspect necessary to establish the robustness of a test method is to verify that the same, or sufficiently similar, results can be achieved at different laboratories. Three UNS S32101 materials were selected for a small Round Robin: 12 mm plate which had been laboratory annealed, a variant which had been sensitized for 5 minutes at 700°C and 27x2 mm tube material with both an as-received and 120 mesh ground surface. The results in Figure 6 show excellent agreement in the two latter cases even though there were some differences in test procedure. For example Lab A used the prescribed solution volume of 600 mL while Lab N used a smaller amount of 150 mL.

The only discrepancy observed was for the laboratory annealed

material and this was the subject of intense investigation, as witnessed by the large number of test points in the diagram. At Lab A the material passed at 40°C but showed some failures at 45°C, while at Lab N there were passes up to 60°C with only a single failure among sixteen tests conducted at 50°C. Lab D showed results intermediate between these two extremes. An exhaustive search of reasons behind this difference yielded only one tentative explanation: that it could be related to the presence of some remaining surface oxide or underlying depletion. As seen in Figure 5 oxides can have a large effect on the evaluated critical temperature, and it was found that Lab N removed much more material by grinding than did Lab A. There is thus a risk that there may be some surface effects remaining in the Lab A tests.

Comparison with other corrosion testing methodsAs mentioned in the Introduction the use of ASTM A923 Method C for lean duplex grades results in impractically low testing tempera-tures, or at least the requirement of cooling baths which are not standard in many laboratories. The same applies to immersion testing in ferric chloride solution according to ASTM G48. [9] However, a limited comparison with the latter is included in Table 3 in order to put the present results into perspective. It is seen that the critical temperature according to ASTM G48E is 5 – 10°C, which is actually on the low side compared to normal values for

the grade. [11] After sensitization for 5 minutes at 700°C pitting occurs at 0°C. A comparison with results from electrochemical testing in 1M NaCl according to ASTM G150 [10] is also included in Table 3 and shows the same trend of a clear drop as the result of sensitization. The data for UNS S32304 indicates that the electrochemical testing even seems to be more sensitive to inappropriate heat treatment than immersion testing. However, this type of testing is impractical for qualification and acceptance purposes since the necessary equipment and electrochemical know-how are not so extensively available.

Grade Condition5% FeCl3 + 1% NaNO3

Max pass temp. (mdd < 10)

6% FeCl3+1% HClASTM G48 E

1M NaClASTM G150

UNS S32101 (12 mm)

Normal values – 15°C 15–20°C

Lab annealed 40°C 5°C (visible pit)10°C (mdd>10)

15.4°C

5 min 700°C 25°C 0°C 9.6°C

UNS S32304(20 mm)

Normal values 11 – – 20 – 25°C

Annealed 30°C – 27°C

1 h, 700°C 25°C – 6°C

10 h, 700°C 25°C – 2.8°C

Table 3 Comparison between different corrosion test methods.

Figure 6 Results of inter-laboratory comparisons show some difference for solution annealed material (left) but good agreement for sensitized material and tube.

mdd>10 failmdd<10 pass

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Lab A Lab D Lab N

S32101, 12 mm. Lab annealed

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Lab A

As-received 120# ground

Lab N Lab NLab A

S32101, 27x2 mm tube

Test

tem

pera

ture

(°C)

60

70

50

40

30

20

10

0Lab A Lab N

S32101, 12 mm. Sense 700°C/5 min

3/2014 | 8

Comparison to metallographic evaluation and impact toughnessAn extensive evaluation of the metallography of sensitization in the lean duplex grades UNS S32101 and UNS S32304 has been presented recently.[3] The 40% sodium hydroxide (NaOH) etchant specified in ASTM A923 was used, with an applied voltage of 2 V for 10 seconds (the specification in the standard is 1 – 3V for 5 – 60 seconds) and it was found that this did give some indication of detrimental phase boundary precipitates, but that such identification was challenging, as seen in Figure 7. An alternative method of electrolytic etching in 10% oxalic acid (C2H2O4) at room temperature and 6 – 7 V for 10 seconds was found to be much more revealing of the fine carbides in the phase boundaries and nitrides within the ferrite phase, and to give a good first indication of sensitization. This etchant is therefore recommended as a preferable equivalent to ASTM A923 Method A for lean duplex grades. It should, however, be pointed out that this metallographic evaluation can be more of a challenge for thinner gauge materials and that very low but non-detrimental levels of precipitates are also often present in mill annealed material.

The use of impact toughness as a qualification or acceptance test for lean duplex grades was also investigated in and it was concluded that a good sensitivity could be achieved by using a test temperature of 20°C for UNS S32101 and 20°C or perhaps even better -10°C for UNS S32304. [3] Two examples are shown in

Figure 7 Microstructure of UNS S32101, 12 mm plate, sensitized for 5 minutes at 700°C showing the advantages of electrolytic etching in oxalic acid over sodium hydroxide

Figure 8 and indicate that a clear distinction between annealed and sensitized material can be detected. The impact toughness for 12 mm plate material of UNS S32101 has decreased from ~180J in the solution annealed condition to 50 – 60J after 5 minutes at 700°C. This degradation is also seen in the corrosion properties and in the oxalic acid etched microstructure. The corresponding curve for 30 mm plate of UNS S32304 indicates that the impact toughness drops after 0.5 to 1h to a level around 150J at room temperature or 100J at -10°C. However, the real drop in toughness for this grade is not really seen until sigma phase begins to precipitate at <10 hours, as indicated in Figure 1. Testing at -10°C appears to be slightly more sensitive to microstructural impairment for UNS S32304 but this slight advantage has to be weighed against the disadvantages of multiple testing temperatures and need for cooling specimens.

Both corrosion and impact toughness testing seem to give similar results, in terms of the time at 700°C before a drop in properties is seen, although impact toughness may be marginally more sensitive. Both tests should nevertheless be interchangeable to use for approving or rejecting a lean duplex batch of material. A requirement should be placed on the number of specimens to be tested, and it would be appropriate to require a single pass, or a pass of both retested specimens in case of a first failure, as is specified in ASTM A923 today for standard and superduplex stainless steels.

Figure 8 Corrosion resistance and impact toughness for tested 12 mm plate of UNS S32101 and 30 mm plate of UNS S32304 showing how the impact toughness drop occurs marginally before the effect on corrosion resistance.

Test

tem

oera

ture

(°C)

Char

py im

pact

toug

hnes

s (J

)

Time at 700°C/1292°F (hours)Annealed

10

00.1

20

30

40

50

60

70

200

150

100

50

01.0 10.0 100.0

S32101, 12 mm

S32205

CV (J)mdd<10

mdd<10CV (J)

Test

tem

oera

ture

(°C)

Char

py im

pact

toug

hnes

s (J

)

Time at 700°C/1292°F (hours)Annealed

10

00.1

20

30

40

50

60

70

200

250

300

350

150

100

50

01.0 10.0 100.0

S32304, 30 mm

CV (RT)mdd<10CV (-10C)

(a) 40% sodium hydroxide. 2V, 10 seconds

(b) 10% oxalic acid. 6 – 7V, 10 seconds

3/2014 | 9

Future prospectsThe ASTM subcommittee A01-14 has recently established a working group to look into the question of standardization of acceptance testing for lean duplex grades. [12] This group will establish a proposal for a formal standardization of the test methods as well as guidelines for acceptance criteria for different alloys and products. The indications from the present work are that the lean duplex UNS S32101and UNS S32304 in the dimensions and products investigated here could be expected to pass an inhibited ferric chloride corrosion test at 20 – 25°C if correctly heat treated. The impact toughness values attainable seem to surpass the minimum values in the transverse direction specified in the materials standard EN 10088-2/4 of 40J for UNS S32101 and

60J for UNS S32304. Thus it may be appropriate to specify somewhat higher limits based on the present data. Suggestions for limits and a comparison with the current acceptance criteria for duplex and superduplex grades according to ASTM A923 are given in Table 4. From the data presented here it seems that UNS S32101 reasonably attains 60 – 80J at room temperature, while the corresponding value for UNS S32304 may be in excess of 100J However, there are indications that -10°C might be a more sensitive impact test temperature for this steel. [3] Whatever criteria are set, the guiding principle must be that the proposed limits should be sensitive to different degrees of sensitization and be capable of clearly distinguishing materials with detrimental levels of carbides and nitrides from properly annealed states.

Grade/condition A: Etching

B: Requirements impact

C: Requirements corrosion

ASTM A923

UNS S32205 base material

40% NaOH electrolytic

54J -40°C (40ft-lb -40°F)

6% FeCl3,<10 mdd 25°C (77°F)

UNS S32205 weld metal

40% NaOH electrolytic

34J -40°C (25 ft-lb, -40°F)

6% FeCl3<10 mdd 22°C (72°F)

UNS S32750base material

40% NaOH electrolytic

To be agreed 6% FeCl3<10 mdd 40°C (104°F)

Grade/condition Proposal – etching Proposal – impact Proposal – corrosion

Lean duplex (this work and [3])

UNS S32101 base material

10% C2H2O4 60–80J, RT5% FeCl3 + 1% NaNO3

< 10 mdd, 20–25°C

UNS S32304 base material

10% C2H2O4

100J, RT or 60–80J, -10°C

5% FeCl3 + 1% NaNO3

< 10 mdd, 20–25°C

Table 4 Acceptance criteria for UNS S32205 and UNS S32750 from ASTM A923 and comparison with proposed criteria for lean duplex grades UNS S32101 and UNS S32304 from this work.

3/2014 | 10

Reproduced with permission from NACE International, Houston, TX. All rights reserved. Paper No. C2012-0001527 presented at CORROSION/2012, Salt Lake City, UT. © NACE International 2014.

ConclusionsThere are good possibilities to be able to define a qualification or acceptance test for the lean duplex grades UNS S32101 and UNS S32304 which parallel the methods specified in ASTM A923 for the higher alloyed duplex steels UNS S32205 and S32507.

Immersion testing with weight loss evaluation in a 5% ferric chloride solution inhibited by the addition of 1% sodium nitrate is suitable for differentiation between acceptable solution annealed states and unacceptable detrimental sensitizing heat treatments. The proposed acceptance test temperature is 20 or 25°C. This has the advantage of avoiding the need for cooling baths for the sub-ambient temperatures which would be required in uninhibited ferric chloride.

The test method has shown good inter-laboratory comparability and robustness in terms of surface grinding, cleaning and the way in which the solution is prepared. However, it is sensitive to the presence of residual weld oxides. It is also important that the specimen be placed in the preheated test solution to avoid false passing results which arise if the specimen is heated from ambient in the solution.

There is good correlation with between loss of corrosion resistance and a drop in impact toughness. A screening evaluation based on electrolytic etching in 10% oxalic acid can be obtained to give a good first indication of a substandard material before proceeding to corrosion or impact toughness testing. Definition of acceptance limits requires further work, preferably under the auspices of the working committee established under ASTM A01.14.

AcknowledgementsAnette Wallin, ARC, Outokumpu Stainless AB is gratefully acknowledged for carrying out the immersion tests in FeCl3+NaNO3 and Sukanya Mameng for the electrochemical work. Thanks are also expressed to James D. Fritz, TMR, for valuable discussions and to laboratories N and D for participating in Round Robin testing.

References[1] ASTM A923-08 Standard Test Methods for Detecting

Detrimental Intermetallic Phase in Duplex Austenitic/Ferritic Stainless Steels.

[2] H Liu, P Johansson and M. Liljas: Structural evolution of LDX2101 (EN 1.4162) during isothermal ageing at 600-850°C. Proc. 6th European Stainless Steel Conference, Helsinki (2008), p555-560

[3] J. Y. Jonsson, C. Canderyd, R. Pettersson: Optimisation of a qualification test method for lean duplex stainless steels. Paper 28 presented at 7th European Stainless Steel Science and Market conference, September 2011, Como, Italy

[4] P. Boillot, R. Bergeron, J. Peultier, K. Wiegers and T. Ladwein: Investigations on standard corrosion test for quality control of lean duplex stainless steel. Proc. 8th Duplex Stainless Steels conference, Beaune (2010), Beaune.

[5] J. D. Fritz, P-E. Arnvig, J. Y. Jonsson, R. Pettersson and S. Randström: Evaluation of possible test methods for qualifying lean duplex stainless steel Proc. Stainless Steel World, Houston, Texas (2010).

[6] Arne Bergqvist, personal communication

[7] R. Qvarfort: The Avesta cell – a new tool for studying pitting. ACOM 1988 vol 2–3 p 2–5

[8] M. Liljas, P Johansson, H-P Liu, C-O Olsson: Development of a lean duplex stainless steel. Steel Research International. Vol. 79, no. 6, pp. 466–473. June 2008

[9] ASTM G48-11 Standard Test Methods for Pitting and Crevice Corrosion Resistance of Stainless Steels and Related Alloys by Use of Ferric Chloride Solution

[10] ASTM G150 -99 (2010) Standard Test Method for Electrochemical Critical Pitting Temperature Testing of Stainless Steel

[11] “Outokumpu Corrosion Handbook”, 10th Edition, 2009, Outokumpu Oyj, Espoo, Finland

[12] www.astm.org

[email protected]

We work with our customers and partners to create long lasting solutions for the tools of modern life and the world’s most critical problems: clean energy, clean water and efficient infrastructure. Because we believe in a world that lasts forever.

Working towards forever.

1548.EN

-GB, Art 58, 10, 14.

Information given in this brochure may be subject to alterations without notice. Care has been taken to ensure that the contents of this publication are accurate but Outokumpu and its affiliated companies do not accept responsibility for errors or for information which is found to be misleading. Suggestions for or descriptions of the end use or application of products or methods of working are for information only and Outokumpu and its affiliated companies accept no liability in respect thereof. Before using products supplied or manufactured by the company the customer should satisfy himself of their suitability.