RECENT DEVELOPMENT OF DUPLEX STAINLESS STEELS

J.-O. Nilsson, G. Chai, U. Kiviskk

R&D Centre, Sandvik Materials Technology, Sweden

Abstract
Recent development of duplex stainless steels is described. The advent of SAF 2707 HD, a 27Cr-7Ni-5Mo-0.4N duplex steel, shows that it is possible to reach a PRE-value close to 50 without sacrificing the fabricability. Modern methods for simulating the interaction between ferrite and austenite intimates that the steels of tomorrow may be optimized with respect to mechanical as well as corrosion properties. Methods under development presented here are multi-scale modelling of plastic deformation and high resolution electrochemical techniques.

Introduction
Duplex stainless steels (DSS) were first described by Bain and Griffiths in 1927 but it was not until the 1930s that duplex stainless steels (DSS) became commercially available. About 80 years have passed since the first discovery but DSS are still under development. The interest in DSS in recent years derives from the high resistance of newly developed high alloy DSS to chloride induced corrosion. As a matter of fact it is the combination of several properties such as corrosion resistance, mechanical properties, weldability and price that makes the DSS unrivalled in many applications [1, 2]. Despite the attractiveness of DSS they have limitations. 475C-embrittlement sets an upper limit to the temperature range recommended during service. Improper welding or production may cause precipitation of -phase or chromium nitrides resulting in deteriorated mechanical properties and/or corrosion properties. The endeavour to design gradually more corrosion resistant DSS provides a driving force to add more chromium, molybdenum and nitrogen, all of which destabilize the microstructure and promote formation of precipitates. The conflict between microstructural stability on one hand and the incentive to add more alloying elements on the other is a challenge to the designer of the alloys of tomorrow. The characteristic features of DSS, whether it is plastic deformation or corrosion, derive from the interplay between the two constituents ferrite and austenite. With the aid of modern computational tools it has become possible to predict microstructures with great precision and also simulate plastic deformation in a two-phase material such as a DSS. We therefore have powerful tools for simulating this interaction as a means of optimising corrosion and mechanical properties.

Trends in the development of DSS

Two trends in the development of DSS may be identified; one towards lean nickel-poor DSS and one towards highly alloyed so called super duplex stainless steels (SDSS). One advantage of lean DSS is the paucity of nickel, the price of which is high and also shows enormous fluctuations.
F03-9

585

However, if a high resistance to pitting corrosion is required significant amounts of the ferrite stabilizers chromium and molybdenum have to be used which sets a lower limit as regards the nickel concentration. The SDSS UNS S32750, S32760 and S32520 with a PRE-value of about 42 were introduced more than 15 years ago. It was thought that SDSS with a PRE-value well above 42 were remote but very recently a SDSS having a PRE-umber close to 50 has been launched. Existing SDSS have shown some limitations in high temperature sea water. Therefore, there has been a need on the market of a DSS with improved resistance to pitting corrosion. SAF 2707 was developed to meet this need and provided a leap in performance. The nominal composition of this alloy together with SAF 2507 is shown in Table 1 below.
Table 1. Nominal chemical composition of two super duplex stainless steels

Grade SAF 2507 SAF 2707HD

UNS S32750 S32707

Cmax 0.03 0.03

Cr 25 27

Ni 7 7

Mo 4 5

N 0.3 0.4

PRE 42 49

The comparison in Figure 1 shows that the pitting resistance is significantly improved in SAF 2707 compared to SAF 2507. The tests used in this comparative study was a modified version of the ASTM G48 test and a crevice corrosion test in 6% FeCl3 according to the MTI-2 procedure [3]. Critical pitting temperatures (CPT) can also be measured using potentiostatic tests at +600mV. The CPT as a function of the concentration of sodium chloride in the range 3-25% is shown in Figure 2. It is quite apparent that SAF 2707 HD is superior to SAF 2507 in the entire concentration range.
120

100

Temperature C

SAF 2507

80

SAF 2707

60

40

20

CPT G48C

CCT MTI-2

Figure 1. Critical temperature assessed using modified G48A and MTI-2 testing.

Figure 2. CPT obtained in the concentration range 3-25% NaCl.

Material modelling of micro mechanical behaviour in DSS

Since the austenitic phase and the ferritic phase have different chemical, physical and mechanical properties, these phases behave differently at the microstructural level. Each phase respond differently to the environments such as corrosion, thermal cycle and loading. For example, the load sharing between the individual phases during loading is different due to the differences in the modulus of elasticity and deformation hardening rate of the individual phases. Strong inter-phase reactions will also result in the formation of micro stresses that maintain their equilibrium among subsets of grains of different orientations [4]. These residual micro stresses can have great effects on SCC, yielding and damage of the material, and consequently affect their strength, deformation and fracture behaviour [5, 6]. Understanding the micromechanical
F03-9

586

reactions is therefore important for the application of the duplex stainless steels and for alloy design and material development. It is difficult to measure the stress-strain behaviour of the individual phases in DSS by the conventional mechanical testing methods due to its fine and heterogeneous microstructure. In-situ diffraction methods using X-ray, synchrotron and neutron are now used to analyze the load sharing, stress interaction between phases and grains and consequently the micro stressstrain behaviour of DSS [4, 7]. Hardness is a measure of the material resistance to plastic deformation. This indicates that the hardness (micro or nano) method can also be used to estimate plastic deformation hardening rate if the size of the austenitic and ferritic phase is sufficient [6]. In recent years, multi-scale material modelling has gained much interest from the researchers in the field of material mechanics. This type of modelling offers the possibility to study the behaviour of single phases, single grains and load sharing between the phases in a multi-phase material. The basic idea in multi-scale material modelling is that the a priori homogenized macro-scale material model is replaced by the homogenized response of a representative volume element (RVE) as shown in Figure 3. Multi-scale material modelling uses micro-scale crystal plasticity and continuum models [6]. Figure 4 shows the results of the multi-scale material modelling for SAF 2507 bar material in as delivered condition; 2507AD during static tensile testing. The ferritic phase is a stronger phase at a total strain less than about 3% and then becomes a softer phase with increasing strain. These observations are similar to the results from the experimental observations as shown in [6].

Figure 3. Multi-scale material modelling of duplex stainless steels

Fatigue is a progressive process. The early stage of fatigue damage is the permanent substructural and microstructural changes (strain localization) and creation of microscopic cracks. Fatigue damage is indicated by the formation of persistent slip bands (PSB) on the mesomicro scales and the subsequent crack initiation. Although much work has been done concerning two-phase or multi-phase metals, it is not clear in which phase or how fatigue damage occurs in these metals. Multi-scale material modelling provides the possibility to study the behaviour of single phases, single grains and load sharing between the phases in a multi-phase material as shown [6]. During cyclic strain loading, DSS materials usually have hardening and then softening processes. The simulations using the multi-scale material modelling show that hardening and softening processes also occur in the austenitic and ferritic phases [6], but behave differently. The ferritic phase has a shorter cyclic hardening period and lower hardening rate compared with the austenitic phase. In this paper, micro material damage is defined as the formation of slip bands in
F03-9

587

the individual phases. Figure 5 shows the micro damage behaviour in 2507AD due to cyclic loading. The accumulated effective plastic slips are mainly in the ferritic phase, but can also be observed in the austenitic phase. This can be explained by the fact that the damage in 2507AD may start in the austenitic phase but finally dominates in the ferritic phase since the weaker phase is the first to become damaged. This indicates that damage and crack initiation in a twophase alloy depend not only on the initial strength of the individual phases, but also their deformation hardening behaviour. The final damage and crack initiation may occur in the weakest phase.

DSS

True stress

2507AD

True strain

Figure 4. Stress versus strain curves for 2507AD by multiscale material modelling.

Figure 5. Representative volume element shows the simulated effective accumulated plastic slip in the 28th cycle for 2507AD. Lighter areas correspond to areas with a high degree of plastic slip.

Corrosion properties
The corrosion properties of duplex stainless steel depend upon the chemical composition as well as the degree of homogeneity of alloying elements. In an entirely austenitic stainless steel the distribution of elements is very homogeneous. However, a complication arises in DSS, in which chromium and molybdenum are partitioned to the ferrite and nitrogen is partitioned to the austenite. As a consequence, the PREN value [8] and the associated resistance to pitting may become notably different in the two phases. This problem in DSS can be circumvented by choosing an annealing temperature at which PREN are equal in austenite and ferrite, whereby equal pitting resistance in ferrite and austenite ensues [9]. As mentioned before -phase can be formed in DSS leading to a decrease in corrosion resistance. However, a finite amount of -phase is required to reduce the pitting corrosion significantly. In a previous investigation [10] the influence of various amounts of -phase on the pitting corrosion behaviour of Sandvik SAF 2507 and Sandvik SAF 2906 was made. It is shown that about 1% of -phase is necessary to significantly deteriorate the pitting corrosion resistance of Sandvik SAF 2507 and Sandvik SAF 2906. Qualitatively similar results have been reported by others for super duplex stainless steels [11]. In seawater Sandvik SAF 2507 with 6% -phase has passed a test at 35C [12]. Similar results have been found for UNS S32760 for which pitting was observed in chlorinated seawater at 35C with 6% -phase while no pitting was observed at 1.5% [13]. Modern electrochemical techniques offer a means of investigating the corrosion properties of DSS. The advantage of these techniques is that the potential (Scanning probe force microscopy
F03-9

588

(SKPFM)), or current distribution (electrochemical scanning tunnelling microscopy technique (EC-STM)), can be mapped with m resolution. Since an AFM is used for the SKPFM this enables magnetic force microscopy (MFM) to be used to identify in which phase the measurements are performed. Hence, the corrosion properties of each phase in the DSS as well as galvanic interactions can be studied. The general corrosion properties of DSS have been studied in 1 M and 4 M H2SO4 with 1 M NaCl with both EC-STM and SKPFM. The austenite was found to be more noble than the ferrite and consequently more pronounced dissolution of ferrite was observed [14, 15].

Modelling of microstructures
The advent of thermodynamically based computer programs such as Thermo-Calc provided powerful tools for developing new alloys during the 1980s. In fact, SAF 2507 was the first alloy ever to be developed and optimised using computerized techniques that later became known as Thermo-Calc [16]. The main achievement was to define a combination of temperature and composition that led to equal PRE and consequently equal pitting resistance in both phases. The development of SAF 2507 therefore provides a milestone in alloy development, not only within Sandvik but in the steel industry in general. The techniques have since been refined and developed further to include also DICTRA [17], a computer based tool by which diffusion controlled phase transformations can be modelled. Both programs are dependent upon experimental data such as activities, equilibrium tie lines, solubilities, diffusivities and surface energies. It is very often the case that the experimental data are uncertain and therefore limit the accuracy of the calculations. Surface energy is a parameter that is known for being difficult to measure experimentally with accuracy. As a consequence coarsening processes are difficult to model with accuracy. Fortunately, new tools are available for calculating surface energies. Using ab initio calculations based on density functional theory surface and interfacial energies can now be calculated with a precision that is far better than experimental methods can offer. Results from such calculations will provide new and more reliable input data to programs like Thermo-Calc and DICTRA and will therefore contribute to more accurate modelling of materials behaviour in the future.

Future prospects
Although DSS have been produced since the beginning of the 1930s new DSS emerge continually. The trend in alloy development has been to increase the concentrations of chromium, molybdenum and nitrogen so as to improve the resistance to pitting corrosion. Also copper has been added to some DSS to enhance the resistance to general corrosion. As with all remedies there are side-effects; Chromium and molybdenum both promote the formation of intermetallic phases while nitrogen is an ingredient in nitrides of the type Cr2N. As a consequence, production is becoming increasingly difficult leading to intermetallic phase formation if the cooling rate is too slow and Cr2N in the ferrite if it is too rapid. There is also evidence that copper promotes spinodal decomposition of ferrite [18]. It is, therefore, quite obvious that the laws of nature impose fundamental limits in alloy development. However, with more sophisticated production equipment the practical limits are continually pushed forward. As an example, recently developed DSS with a PRE-number close to 50 have been produced, thus confirming that alloys considered visionary a decade ago have now become a reality.

Acknowledgements
This paper is published with permission from Sandvik Materials Technology. The support from Prof. Olle Wijk is gratefully acknowledged.

F03-9

589

References
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] H.D. SOLOMON and T.M. DEVINE, Proc. Conf. DSS (ed. R.A. Lula), Materials Park, OH, ASM, (1984) , pp. 693-756 J. CHARLES, Proc. Conf DSS 91, Les Ulis, France, Les Editions de Physique, (1991), pp. 3-48. K. GRANSSON, M.-L. NYMAN, M. HOLMQUIST AND E. GOMES, Sandvik SAF 2707 HD, Internal lecture no. S-51-63, 2006. N. JIA, R. Lin PENG, Y.D. WANG, G.C. CHAI, S. JOHANSSON, G. WANG, and P.K. LIAW, Acta Mater., 54, (2006), pp. 3907-3916. G. CHAI and R. LILLBACKA, (2006), Key Engineering Materials 324-325, (2006), p. 1117. R. LILLBACKA, G. CHAI, M. EKH, P. LIU, E. JOHNSON and K. RUNESSON, Acta Mater., 55, 2007, pp. 5359-5368. P.R LIN, J. GIBMEIR, S. EULERT, S. JOHANSSON and G.C. CHAI, Materials Science Forum 524-525, (2006), p. 847. G. HERBSLEB, Werkst. Korros., 33, (1982), pp. 33-34. H. VANNEVIK, J.-O. NILSSON, J. FRODIGH and P. KANGAS, Trans. ISIJ 36, (1996), pp. 807-812. P. KANGAS and J.-O. NILSSON, Proc. Stainless Steel World 05 Conf., Maastricht (2005), KCI Publishing BV, Zutphen, The Netherlands (2005). A. TURNBULL, P.E. FRANCIS, M.P. RYAN, L.P. ORKNEY, A.J. GRIFFITHS AND B. HAWKINS, Corrosion 58 , (2002), p. 12. S. SOLTANIEH, M. KLOCKARS, U. KIVISKK and P. EKLUND, Stainless Steel World 15, (2002), p. 28. R. FRANCIS and G.R. WARBURTON, Proc. Stainless Steel World 99 Conf., Hague (1999), KCI Publishing BV, Zutphen, The Netherlands (1999), p.711 M. FERMENIA, J. PAN, and C. LEYGRAF, Journal of Electrochemistry Society 149, (2002) p. B187 M. FERMENIA, C. CANALISA, J. PAN, and C. LEYGRAF, Journal of Electrochemistry Society 150, (2003) p. B274. B SUNDMAN, B. JANSSON and J-O ANDERSSON, Calphad, 9, (1985), pp 153-190 J-O. ANDERSSON and J. GREN, J. Appl. Phys., 72, (1992), p. 153. H.D. SOLOMON and L.M. LEVINSON, Acta Metal., 26, (1978), pp. 429-442.

F03-9

590

DETECTION OF THE 475C EMBRITTLEMENT IN A LEAN DUPLEX STAINLESS STEEL USING THE ELECTROCHEMICAL POTENTIODYNAMIC REACTIVATION (EPR) TEST

C. Fosca, J. Sakihama

Pontificia Universidad Catlica del Per, Peru

Abstract
The duplex stainless steel UNS S32101 (LDX 2101) is a new leaner DSS that provide a high mechanical resistance with a corrosion behavior, in most cases, better than the traditional austenitic stainless steels. Although DSS are very competitive alloys, they are susceptible to precipitation of secondary phases as the spinodal decomposition of the ferrite, when they are exposed at temperatures between 300-600C. The spinodal decomposition of ferrite can increase the hardness of the DSS but reduce strongly its toughness and corrosion resistance. Microstructure changes for each aging condition were characterized by Electrochemical Potentiodynamic Reactivation EPR test in order to achieve a non destructive method to detect on service detrimental aging conditions in this alloy. The EPR test was carried out using an appropriate electrolyte composition (H2SO4 with addition of KSCN) at 20C. The reactivation potential and scan rate were selected to improve more sensitivity to the microstructural changes.

Introduction
LDX 2101 (EN 1.4162, UNS S32101) is a new low alloyed (lean) Duplex Stainless Steels with low addition of nickel in order to reduce the cost. To assure an adequate phase balance in the microstructure (50% austenite, 50% ferrite), the austenite stability effect of the nickel is replaced with additions of manganese and nitrogen. The mechanical resistance of this alloy is comparable with the DSS 22%Cr-5%Ni (EN 1.4462, UNS S32205) and the corrosion properties are in general better than for austenitic 304 (EN 1.4301) The unique combination of mechanical properties, corrosion resistance and low cost, make this alloy an excellent choice for many applications for which the traditional austenitic stainless steel are usually employed. Its relative low alloy content in comparison with others DSS brings the additional advantage to be less sensitive to the secondary phase precipitation, when these materials are heated in the range of 700- 900C (sigma phase, Chi-, R- phases, carbides) and in the range of 300 - 600C ( - spinodal decomposition of ferrite). All of these phase precipitations produces a strong reduction of the material toughness, known as thermal aging embrittlement of DSS.

F03-10

591

Many researches have investigated and developed different methods to detect and also to quantify the thermal embrittlement in the DSS [1,2,3,4]. Due to the presence of precipitated phases promoted not only the embrittlement but also the localized corrosion of the DSS, electrochemical techniques can be used to detect microstructural changes in these alloys [5]. The Electrochemical Potentiodynamic Reactivation (EPR) test, originally developed to detect the intergranular corrosion of austenitic stainless steel, has been used also to detect the susceptibility to intergranular corrosion of DSS due secondary phases [6,7,8] but there are scarce published results about the use of this electrochemical technique to detect the 475C embrittlement of DSS [9]. The aim of this article is to study the use of the EPR technique on the detection of the aging embrittlement in a new lean duplex stainless steel.

Experimental Procedure
As mentioned before, the material studied was the LDX 2101 duplex stainless steel. The alloy was supplied in the form of a 6 mm plate in as received condition, with a chemical composition described in table 1. The samples were aged at 475 5C during different times (4, 8, 16, 24, 48 and 72hrs) and cooled with water at room temperature.
Table 1. Chemical composition of the LDX 2101 duplex stainless steel.

C 0.035 W 0.015

Si 0.73 Ti 0.01

S 0.002 Sn 0.005

P 0.002 Co 0.043

Mn 4.87 Al 0.023

Ni 1.52 B 0.002

Cr 21.83 Nb 0.008

Mo 0.28 N 0.22

V 0.05 Fe rest

Cu 0.32

Mechanical testing Hardness measurements and impact testing were carried out in the aged samples in order to study the effect of the aging time at 475C on the mechanical properties. Vickers Hardness measurements were made on 10mm x 10mm area specimen with a 20Kg load. Charpy Impact test were conducted at 0C using three specimens for each aging time condition. Because the thickness of the plate, 6x10x55 mm3 reduced charpy - V notch samples were used as shown in figure 1.

Figure 1. Reduced Charpy V-notched samples used in the impact testing

Microstructural analysis The determination of ferrite content was done on a Fisher Ferritescope MP30E working according to DIN 32514-1. For the metallographic analysis the samples were mounted in resin, grinded, polished and etched using the Bloech and Wedl color etching agent.

F03-10

592

Electrochemical Potentiodynamic Reactivation (EPR) testing A modified Double loop Electrochemical Potentiodynamic Reactivation (DL-EPR) test was used to detect possible different reactivation grades as consequence of the aging level at 475C. The samples were mounted in resin and polished. They were submerged with an exposed area of 24 mm2 in a 50 ml glass beaker together with a Pt counter electrode, saturated calomel reference electrode was submerged in a 3.5KCl solution and was connected with a salt bridge to the electrolyte solution (figure 2). The electrolyte composition was optimized to obtain the higher sensitivity of the test for this alloy in the studied conditions. The electrolyte solution was 0.5M H2SO4 with 0.01M KSCN, The solution temperature was maintained at 20C. Before scanning a preconditioning of the surface was carried out at 700 mV (SCE) during 60 seconds with a stabilization time at corrosion potential for another 60 seconds. The potential was scanned from the corrosion potential (open circuit potential) to +200 mV (SCE) and immediately the scan is reversed to reach again the corrosion potential. The scan rate for both scan directions was 2 mV/s.
Counter electrode Calomel Reference electrode

Working electrode 3.5M KCl solution 0.5M H2SO4 + 0.01M KSCN solution Salt bridge

Figure 2. Corrosion cell used for EPR tests.

The sensitization grade is determined by the ratio of the reactivation current (Ir) and the activation current (Ia), using the Ir/Ia ratio as a measure of the degree of sensitization (figure 3).
V (mV) 0.3 0.2 0.1 0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 0 2 4 I (mA) 6 8 10

Reactivation current (Ir)

Activation current (Ia)

Figure 3. Typical DL-EPR curve.

F03-10

593

Results and Discussion

Mechanical properties The aging at 475C of the LDX 2101 duplex stainless steel produced only a slight increase in hardness, observed in the first 4 hours of aging. However the impact test results confirmed the embrittlement effect of the aging treatments. The toughness decreased in about 40% after 72 hours at 475C. A first conclusion from these results is the poor correlation between the hardness values and impact test results, hardness measures should not be used to detect the embrittlement effect of the aging condition.

Figure 4. Hardness measurements and Impact energy (at 0C) vs aging time at 475C for the LDX 2101 duplex stainless steel.

Microstructural analysis The ferrite volume fraction suffers a slight decrease after the aging, as shown in figure 4, from 52.2% to 42.6% after 72 hours, observing the strongest variation during the first 4 hours aging treatment (figure 4). After this time the aged alloy shows stabilization on the % ferrite content for longer aging periods.

F03-10

594

55.0 53.0 51.0

% -Ferrite

49.0 47.0 45.0 43.0 41.0 39.0 37.0 35.0 0 20 40

aging time (hours)

60

80

Figure 5. -Ferrite percentage measured by magnetic induction measurement method vs. aging time.

Observing the figures 4 and 5 it finds a good correlation between the hardness variation and the ferrite content in the microstructure for the aged samples. However, neither the hardness measurement nor the ferrite content determination can be used as indirect method to detect 475C-embrittlement in this alloy because the measured values fall within the typical ranges observed in normal microstructural conditions, where there is no embrittlement problem. The metallographic analysis by optical microscopy using color etching showed clear differences in the microstructure for the aged samples. Figure 6a shows the microstructure of the DSS basis material, where the dark phase is ferrite and the light phase is austenite. Figures 6b and 6c show indications in ferrite phase after 24 and 72 hours aging heat treatment The observation of microstructural changes by optical microscopy for low aging temperatures in DSS is very difficult. Commonly the evidence of a possible spinodal decomposition of ferrite can be obtained observing the microstructure by TEM [3] because the size of Cr-rich precipitates is about a few nanometers. The color etching method, used in this study, can detect difference between the micrographies for the 475C aging conditions by optical metallography. The optical micrographies do not revealed directly the precipitates but yes its effect on the interference film, formed during the color etching on the surface of the metallographic sample. These micrographic differences were evident from 24 hours aging treatment (figure 6.b).

F03-10

595

EPR testing Figure 7 shows the results of the double loop EPR tests, in function of the coefficient Ir/Ia versus aging time at 475C. Here we can observe an increase of the sensitization, due presumably to the precipitation of Cr-rich phase into the ferrite and the resultant depletion of Cr around it.
0.25 0.2 0.15
Ir/Ia

0.1 0.05 0 0 10 20 30 40 50 60 70 80 aging time (hours)

Figure 7. Sensitization by DL-EPR test (Ir/Ia) vs aging time at 475C

There is a slightly similar tendency between the impact toughness results (figure 4) and the Ir/Ia ratio obtained by EPR test (figure 7) in the studied aging conditions. The decrease in impact energy of the aged alloys is consequence of the diminution of the dislocation mobility due to -ferrite precipitation. On the other hand, the sensitization index is related to the degree of the Cr depletion around this precipitates. In the studied conditions, both mechanisms probably have been rising with the aging time.

F03-10

596

Conclusions
DSS LDX 2101, suffers 475C - aging embrittlement, reducing its toughness after 72 hours about 40% of its as received condition. Bloech and Wedl Color etching allowed to identify micrographies associated to aging conditions in the DSS. The Ir/Ia ratio of DL-EPR test was found to be a good measure of the degree of 475C embrittlement in the studied aging conditions due to its good correlation with the impact test results.

Acknowledgments
The authors are grateful to Outokumpu S.A. (Barcelona) for supplying the duplex stainless steel.

References
[1] EVANSON, S.; OTAKA, M.; HASEGAWA, K. Use of Squid Magnetic Sensor to Detect Aging Effects in Duplex Stainless Steel British Journal of Non-Destructive Testing, Vol. 32, No. 5, 1990, pp. 238-240 Y ISHIKAWA, M OHTAKA, S TSUCHIYA, T YOSHIMURA, Atom-probe study of the aging embrittlement of cast duplex stainless steel. SME International Journal Series A, Vol. 38, p. 384,1995. S.S.M. TAVARESA, R.F. DE NORONHA, M.R. DA SILVA, J.M. NETO, S. PAIRIS 475C Embrittlement in a Duplex Stainless Steel UNS S31803 Materials Research, Vol. 4, No. 4, 237-240, 2001. H. OGI, M. HIRAO Ultrasonic Noise Relaxation for Evaluating Thermal Aging Embrittlement of Duplex Stainless Steels. Research in Nondestructive Evaluation, Vol. 9, No. 3, November 1997. CHAN-JIN PARK, HYUK-SANG KWON. Electrochemical noise analysis of localized corrosion of duplex stainless steel aged at 475C. Materials Chemistry and Physics 91 (2005) 355360. C. Fosca. Influencia de la fase sigma sobre la resistencia a la corrosin por picaduras de aceros inoxidables duplex y su caracterizacin electroqumica mediante la tcnica EPR" tesis doctoral, Universidad Complutense de Madrid, Espaa. 1995. T. AMADOU, C. BRAHAM, and H. SIDHOM. Double Loop Electrochemical Potentiokinetic Reactivation Test Optimization in Checking of Duplex Stainless Steel Intergranular Corrosion Susceptibility. Metallurgical and Materials Transactions A Volume 35A, November 2004. 3499-3513. C.J. PARK, V. SHANKAR RAO, AND H.S. KWON. Effects of Sigma Phase on the Initiation and Propagation of Pitting Corrosion of Duplex Stainless Steel Corrosion, Vol. 61, No. 1, 2005, 76-83. CHAN-JIN PARK, HYUK-SANG KWON. Effects of aging at 475C on corrosion properties of tungsten-containing duplex stainless steels Corrosion Science, 44 (2002) 28172830.

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

F03-10

597

F03-10

598

MILL EXPOSURE TESTS OF DUPLEX STAINLESS STEEL LDX 2101 IN RECYCLED FIBER APPLICATIONS

T. Laitinen1, L. Wegrelius2, A. Bergquist2

Metso Paper, Finland, 2Outokumpu Stainless AB, Sweden

Abstract
The performance of a newly developed duplex stainless steel LDX 2101 (EN 1.4162) was compared with traditional construction material, in the first place EN 1.4432 (316L). Test coupons were exposed in recycled fiber (RCF) mill environments while erosion corrosion tests were performed in the laboratory. RCF mill tests were performed in Scandinavia, Europe and Asia. Test coupons were installed in the liquid phase inside pulpers and other RCF line equipment for 2 7 months. The RCF mill environment can contain large amounts of abrasive particles like sand, metal, glass and plastic. Erosion corrosion tests were performed in laboratory environments containing chloride- and sulphate ions or in sulphuric acid loaded with 100 g/l silica sand at a temperature of 50oC. The in-plant tests showed that in mild RCF environment, with chloride levels below 150 mg/l, no measurable difference regarding corrosion performance existed between the tested steel grades LDX 2101 and 1.4432. On the other hand, in more aggressive environments, grade 1.4432 was slightly more corrosion resistant. In the erosion corrosion tests, grade LDX 2101 performed best in all tested environment/load combinations. The results were expected considering, that 1.4432 is slightly more corrosion resistant in acidic and near neutral solutions containing chlorides, while LDX 2101 is more wear resistant.

Introduction
Recycled pulp is made from paper and board, which has been used and then recovered, by one of various waste collection schemes. Waste paper contains impurities of various kinds, e.g. plastics, metals, printing ink and netting spines from books, which cause a more abrasive condition than in other kind of pulping processes. The austenitic stainless steel grade 1.4432 has so far been the dominating construction material for equipment in RCF processes. However, duplex stainless steels are becoming more and more frequently used in many applications, including pulp and paper processes. The mechanical strength of duplex stainless steels is approximately twice as high as for conventional austenitic grades, implying not only benefits in terms of reduced gauges and reduced weight, but also a higher resistance towards abrasive conditions. LDX 2101 is the latest contribution in the duplex stainless steel family with the same desirable properties as other duplex stainless steels - good corrosion resistance, welding and engineering properties. LDX 2101 has a low nickel content that implies a low and stable price and together with the high strength it is a cost efficient alternative to the more traditional austenitic stainless
F03-11

599

steel grades. The purpose of this investigation was to study the performance of LDX 2101 in recycled fiber processes.

Experimental
Test materials Table 1 shows the typical composition, the pitting resistance equivalence (PRE = %Cr + 3.3x%Mo + 16x%N) and proof strength of the investigated stainless steel grades.
Table 1. Typical composition in weight %, PRE values and proof strength of investigated stainless steel grades. Grade EN ASTM C Mn Cr Ni LDX 2101 1.4162 S32101 0.03 5 21.5 1.5 4301 1.4301 304 0.04 18.1 8.1 4404 1.4404 316L 0.02 17.2 10.1 4432 1.4432 316L 0.02 16.9 10.7 * Hot rolled plate, min values at 20 oC according to EN 10088. LDX 2101 is a registrated trade name by Outokumpu. Mo 0.3 2.1 2.6 N 0.22 0.05 0.05 0.05 PRE 26 19 25 26 Rp0,2* 450 210 220 220

Mill exposure tests The test coupons, for the mill exposures, were water-cut from hot rolled plate, measuring 60 x 60 x 8 mm. The coupons had mill surface finish (hot rolled, heat treated and pickled) and the cut edges were dry-ground to 320 grit. The coupons were mounted on an insulated bolt with flat polytetrafluoroethylene (PTFE) crevice washers separating the coupons from the test rack. Each test rack contained three coupons in the same steel grade bolted to a plate made of the stainless steel grade 2205. The bolts were assembled to the rack using a torque of 4 Nm. The test rack assembly is illustrated in Figure 1.

Figure 1. Test rack for mill exposure and coupon assembly.

Before the exposure the coupons were weighed, the dimensions measured and the surface roughness characterized by Optical Confocal Microscopy. After the exposure each rack was dismantled, the coupons brushed under running water and cleaned in 20% nitric acid solution at room temperature for at least 15 minutes and dried. After cleaning the coupons were weighed, the surface roughness (Ra) was measured, and they were examined in a binocular at 20X magnification. The mill exposure tests were performed in several RCF mills in Scandinavia (4 mills), Europe (2 mills) and Asia (1 mill). The test coupons were installed inside the pulping line equipment and were thus exposed in liquid or in liquid + gas environments. The process equipments are specified in Table 2 and the process environment is given in Table 3. Pulp made of old corrugated cardboard (OCC-pulp) is typically not bleached, as was the case with all the OCC lines in this study. De-inked pulp (DIP) is usually bleached with H2O2, dithionite (hydrosulphite), formamidine sulphinic acid (FAS) or their combination. In Mill A, DIP was bleached with H2O2, in Mill B with H2O2 and FAS and in Mill G with H2O2. Samples of the process environment were taken from the test sites and analyzed by Dionex ion chromatography
F03-11

600

system equipped with an AS50 autosampler, a LC25 chromatography oven, an EG40 eluent generator, and an IC25 ion chromatograph. The required concentration for the mobile phase (potassium hydroxide) was made simultaneously in the eluent generator module by using the Dionex EGC II KOH cartridge. The samples were processed with conservatives due to several days delay to the analysis. [1]
Table 2. Exposure test conditions. Test site Mill A Mill B Mill C Mill D Mill E Mill F Mill G Location Scandinavia Scandinavia Scandinavia Scandinavia Europe Europe Asia RCF pulp DIP DIP OCC OCC OCC OCC DIP Fresh water [m3/t pulp] 30 30 20 Not known 1,3 6 15 - 20 Equipment HC pulper GapWasher Drum pulper Fine screen Support structure Coarse screen LC pulper Drum pulper Bleaching H2O2 FAS and H2O2 None None None None H2O2 Environment Liquid + gas Liquid + gas Liquid + gas Liquid Liquid Liquid + gas Liquid + gas

Pulpers are used for disintegrating the raw material (pulp bale, recycled paper, broke from paper machines) into a form ready for pumping. Pulpers operate either in low consistency (LC) or in high consistency (HC). Drum pulpers also remove large impurity particles and ink from recycled paper. GapWashers are used for washing ash, ink and stickies from pulp and for pulp thickening up to 10%. Screens are used to separate fibers by fiber length.
Table 3 The chemistry of the media at the different test sites. Test site Exposure time [months] SO42- SO32- S2O32- Molar ratio Temp pH Clo [mg/l] [mg/l] [mg/l] [mg/l] [Cl-]/ [SO42-] [ C] 37.8 48.7 48.8 NA NA NA NA 7.2 6.9 7.1 NA NA NA NA 37 114 80 322 787 861 165 523 703 945 100 1243 625 193 NA 4 163 10 0 0 29 1 1 305 2 21 0 2 0,19 0,44 0,23 8,7 1,7 3,7 2,3 Molar ratio ([Cl-]+[SO42-])/ [S2O32-] 730 1200 4,4 570 192 350

Mill A 2, 4, 6.5 Mill B 2, 4, 6.5 Mill C 2, 4, 6.5 Mill D 6? Mill E 7 Mill F 4 Mill G NA not analysed.

Laboratory erosion corrosion tests Cylindrical test samples, in grade 1.4301 and 1.4404, were prepared from 12 mm bar by turning to a diameter of 8 mm. Test samples, in grade 1.4432 and LDX 2101, were prepared by cutting cylindrical 8 mm samples from 10 mm thick plate. The cylindrical samples were ground with 180 grit paper, cleaned with ethanol and dried. The sample area was measured in 1 mm2 accuracy and sample weight in 0,1 mg accuracy. Erosion corrosion tests were executed in a decanting tank, where rotation speed of the horizontal agitator was 425 rpm. Test samples were attached on the rim of the decanting tank and exposed to the test environment for 5 to 24 h. The erosion corrosion test environments contained either chloride together with sulphate or plain sulphuric acid. Silica sand was used as the wearing agent. The test environments are presented in Table 4.

F03-11

601

Table 4. Erosion corrosion test environmens. Cl[mg/l] 200 1000 200 1000 SO42[mg/l] 542 2708 542 2708 Molar ratio [Cl-]/ [SO42-] 1 1 1 1 H2SO4 [g/l] 49 Silica sand [mg/l] 100 100 100 100 100 pH 5 5 5 5 ~0 Temp [oC] 50 50 50 50 50 Immersion [h] 0 0 168 168 0 Wearing [h] 24 24 5 5 5

Test1Env1 Test1Env2 Test2Env1 Test2Env2 Test3Env3

Results
Mill exposure tests The process environments of the mill sites are presented in Table 3. Mills A, B, C and D were in Scandinavia, Mills E and F in Europe and Mill G in Asia. H2O2 bleaching residues of the process samples were not analyzed because H2O2 residues do not stand the preservation process. The European mills E and F had high chloride (Cl-) contents due to a low fresh water supply (< 6 m3/ t pulp). The sulphate (SO42-) level was exceptionally low in the Scandinavian mill D and in the Asian mill G, which has been the case with several other Asian mills. The reason for the low sulphate concentration might be the raw material. Remarkable high sulphite (SO32-) and thiosulphate (S2O32-) concentrations were determined at mill C, where no bleaching was used. The reason for this high thiosulphate concentration was suspected to be remnants from the pulp cooking. Table 5 shows the summary of the test results from the mill exposures. Especially at mill A and mill F, some weight loss was measured, which most certainly was due to mechanical impacts. The mechanical load on the test coupons was really heavy indicated by the fact that, at mill F only one coupon of grade 1.4432 remained after the test period and at mill G all the test coupons had fallen off the test test racks and disappeared in the process. The weight loss was maximum 226.5 mg for grade LDX 2101 (at mill A) and 646.6 mg for grade 1.4432 (at mill F), corresponding to a maximum corrosion rate of 0.011 mm/year and 0.025 mm/year respectively. A material with a corrosion rate below 0.1 mm/year is normally considered as corrosion proof.
Table 5. Summary of the mill exposure test results after approximately 2, 4 and 6.5 months exposure. Test Steel grade site Mill A LDX 2101 1.4432 Mill B LDX 2101 1.4432 Mill C LDX 2101 1.4432 Mill D LDX 2101 Corrosion rate [mm/y] 2 month 4 month 6.5 month 0.016 0.009 0.011 0.012 0.006 0.008 0.000 0.000 0.000 0.001 0.001 0.000 0.000 0.000 0.000 0.000 0.001 0.001 Remarks No local corrosion, minor mechanical marks No local corrosion, minor mechanical marks No local corrosion No local corrosion, minor mechanical marks No local corrosion, some weld sputter on the coupons No local corrosion, some weld sputter on the coupons Localized corrosion, max 200 m, crevice corrosion under deposits (2 screen support rods for 7 months) No local corrosion, minor mechanical marks, 3 coupons for 7 months No local corrosion, minor mechanical marks, 3 coupons for 7 months Lost during exposure No local corrosion, mechanical damage. 2 and 6.5 month coupons lost during exposure All coupons lost during 24 months exposure All coupons lost during 24 months exposure

Mill E

LDX 2101 1.4432 -

0.025 -

602

Mill F

LDX 2101 1.4432 -

Mill G LDX 2101 1.4432 F03-11

Laboratory erosion corrosion tests Erosion corrosion tests were performed in three different test solutions with or without an immersion period followed by a wearing cycle. Duplicate test samples were used in all tests. In the 24 h erosion corrosion test cycle (Test1) without any immersion pre-treatment, all the tested materials remained passive and the weight losses obtained were due to wearing purely. However, when the test materials were pre-treated by immersion for one week in the high-chloride test solution (1000 mg/l) followed by 5 h wearing (Test2Env2), the passive film broke down locally, resulting in both some pitting and wearing. In the 1 N H2SO4 solution all the test materials corroded actively and wearing took place very rapidly during the 5 h wearing cycle (Test3Env3), corresponding to a erosion corrosion rate of about 1 mm/year. The results are presented in Figure 2.
1200 Weight loss [mg/m2*h] 1000 800 600 400 200 0
Test1Env1 Test1Env2 Test2Env1 Test2Env2 Test3Env3

ACTIVE

PASSIVE

LOCALIZED

1.4301 1.4404 1.4432 LDX 2101

Figure 2. Erosion corrosion test results presented as a function of weight loss [mg/m2*h].

The erosion corrosion rate of 1.4432 was 24 30% higher than the erosion corrosion rate of LDX 2101 in the mildest environment containing 200 mg/l chloride (Env1). In 1000 mg/l chloride (Env2) containing environment the erosion corrosion rate difference was 16 18%. In the most aggressive environment (Env3) the erosion corrosion rate of 1.4432 was only 6% higher than the rate of LDX 2101.

Discussion
Sulphate ions are known for inhibiting the pitting corrosion of stainless steel in chloride containing media. The inhibitive effect of sulphate ions on the initiation of pitting, has been proposed to be based on the competitive adsorption with chloride ions [2]. So, with high chloride- and low sulphate levels, the molar ratio [Cl-]/ [SO42] becomes high, which in turn increases the risk of pitting and crevice corrosion. This is also indicated in the result of this study, mill D having the highest chloride to sulphate ratio, was the only mill showing signs of corrosion. In mills A, B, C and E having low chloride to sulphate ratios, there were no signs of corrosion on any of the investigated steel grades. Experience from the paper industry has shown that traces of thiosulphate can cause pitting on stainless steel grade 1.4301 (304) in what otherwise is rather benign and non-corrosive white water. Thiosulphate comes mainly from the hydrosulphite brightening, as a decomposition product, in the papermaking process. The most sensitive concentration range has been found to be when the molar ratio ([Cl-]+[SO42-])/ [S2O32-] is in the order of 10 to 30 [3]. The result from this study does neither confirm nor reject the theory of this corrosion sensitive molar ratio. Most
F03-11

603

of the mills had a molar ratio far above 30, only mill C had a ratio close to the sensitive area. However, there was no corrosion at all in mill C. Except from some minor pitting and crevice corrosion attacks on LDX 2101 at mill D, having the highest chloride to sulphate ratio, no measurable difference regarding corrosion resistance could be detected between the tested grades LDX 2101 and 1.4432. From corrosion point of view, recycled pulp is not a very aggressive environment. On the other hand, the recycled fibre contains a lot of abrasive impurities, e.g. plastics, paper-clip, staples, netting spines from books, which cause an abrasive condition. The fact that most of the coupons were lost at mill F and G is an indication of that the mechanical forces during the processing of the recycled pulp are high. In the absence of corrosion, the mechanical abrasion of the coupons was, especially for grade 1.4432, the most predominating feature. That is also supported by the fact that the surface roughness for grade 1.4432 increased slightly after exposure at the test sites. The laboratory erosion corrosion tests show that the mechanical strength of the material has a big impact on the overall performance as grade LDX 2101 experienced the lowest weight losses in all test environments. Duplex stainless steels are more and more taking over the role of austenitic stainless steels like 1.4301 and 1.4432 in papermaking equipments. Nitrogen as an alloy addition makes the duplex steels more pitting resistant and stronger. They also have the added advantage of resisting stress corrosion cracking that can occur on higher temperature components such as steam boxes.

Conclusions
The result from the exposure at the mills and the erosion corrosion tests shows that: The duplex stainless steel grade LDX 2101 can in most situations replace the conventional austenitic grade 1.4432 in recycled fibre applications. Some cautions should however be taken when the chloride level of the process water becomes high. The high strength of LDX 2101 gives an extra advantage of high resistance under abrasive conditions.

References
[1] [2] [3] J.P. Isoaho, personal communication 2007. H.P. Leckie: Environmental Factors Affecting the Critical Potential for Pitting in 18-8 Stainless Steels, J.Electrochem. Soc. 113, 12, 1966, pp. 1262 1267. R.C. Newman, W.P. Wong, H. Ezuber and A. Garner: Pitting of stainless steel by thiosulfate ions, Corrosion 45 , 4, 1989, pp. 282 287.

F03-11

604

A NEW LEAN DUPLEX STAINLESS STEEL WITH HIGH MECHANICAL AND CORROSION PROPERTIES: 1.4062

J. Peultier1, E. Chauveau2, S. Jacques1, M. Mantel2

Industeel (ArcelorMittal group), France, 2Ugitech (Schmolz + Bickenbach group), France

Abstract
Due to low Ni content, the price of duplex family is less sensitive to the price fluctuation of raw materials than the price of austenitic family. In numerous applications, a very cost effective duplex solution can nowadays be proposed as an alternative to austenitic material with at least a similar corrosion resistance and better mechanical properties. For instance, 1.4362 (UNS S32304) replaced 1.4404 (316L) material in evaporators of sea water desalination units or in pumps and valves for process and water industries; 1.4462 (UNS S32205) is used instead of 1.4439 (317LMN) in the absorbers of wet flue gas desulphurization systems. This paper presents a new lean duplex grade developed in close cooperation between Industeel and Ugitech with the aim to propose a cost effective alternative to 1.4307 (304L), coated or galvanized carbon steel or concrete in structural applications, potable water systems or pulp and paper industry. The reduction of Ni is obtained by a nitrogen addition in order to obtain microstructure containing approximately 50% of ferrite and 50% of austenite. After a preliminary study performed with laboratory heats, several industrial heats were produced with 22Cr%, 2Ni% and 0.2%N as typical composition. In this paper, the results of investigation performed on industrial bars, cold-drawn wires and hot rolled plates are presented and discussed. It appears that this new lean duplex grade (UNS S32202 / EN 1.4062) has a localized and uniform corrosion better than 1.4307 material with yield strength about twice.

Introduction
Austenitic stainless steels, such as 1.4301 (18Cr8Ni) and 1.4401 (17Cr10Ni2Mo) types account for 60% of stainless steels usage all over the world. This is certainly the result of their corrosion resistance properties but also of their versatility and ease of fabrication. Their major drawbacks are their low mechanical strength and their exposure to alloy cost variations. While ferritic grades (17Cr, 17CrTi) have found increasing applications in thinner gauges, they cannot easily replace austenitics in thicknesses over 3mm, due to their inherent tendency to grain coarsening (especially in heat affected zone of welds). Furthermore, 200 series Mn grades are limited to very low corrosive media due to their relatively low Cr content. In the duplex family and thanks to progress made in steel metallurgy since 70s, 1.4462 (2205) is now recognized as a cost effective and technically efficient solution1. For instance, 1.4462 replaced 1.4439 (317LMN) in air pollution control equipment2 and 1.4429 (316LN) or 1.4404 (316L) in chemical tankers3. Although 1.4362 (2304) was developed over 20 years ago, it never
F03-13

605

succeeded in challenging the supremacy of 1.4462 in period where the price of Ni and Mo remained under control. But, present raw materials prices have increased the price gap between austenitics and duplex. Consequently, Mo free 1.4362 grade constitutes at that time an excellent cost alternative to the austenitic 1.4404 solution, explaining 1.4362s growth acceleration since 2003, as for instance in desalination industry4, marine applications or production process5. The new lean duplex grade 1.4062 (UNS S32202), presented in this paper, was developed to match the corrosion resistance of 1.4301 or 1.4307 austenitic grades in most environment and with twice mechanical strength. The nominal chemical composition of this grade is 22Cr%, 2Ni% and 0.2%N with iron balance. It was designed not only to obtain mechanical properties and corrosion resistance, but also structural stability and good toughness properties in the heat affected zone of welded assemblies. The Cr content, element which is known to be beneficial in fighting against all the corrosion forms, was kept higher than 21.5%. Ni content was optimized to obtain crevice corrosion resistance and toughness properties without increasing the material price. N content was adjusted to obtain a microstructure containing approximately equal amounts of ferrite and austenite after an annealing treatment performed in the range 980 1100C. This grade is Mo free to offer better structural and cost stabilities. Finally Mn was kept below 2% in order to limit its detrimental effect on pitting corrosion resistance, due to the formation of manganese sulphides or manganese oxisulphides, but also on uniform corrosion resistance in sulphuric acid solution6. After a first test program with laboratory heats of 25kg, several industrial heats were produced in Industeel and Ugitech melting shops, then transformed in plates, bars, rebars or cold drawn wires. This paper presents the mechanical and corrosion properties obtained on these industrial products. Compositions of tested materials including the reference austenitic and duplex grades are given in Table 1.
Table 1. Typical chemical composition and PREN value for tested stainless steels (PREN = Cr% +3.3Mo% +16N%)

Euronorm 1.4301 1.4307 1.4404 1.4571 1.4429 1.4062 1.4362

AISI 304 304L 316L 316Ti 316LN 2202 2304

C < 0.070 < 0.030 < 0.030 < 0.080 < 0.030 < 0.030 < 0.030

Cr 18.5 18.5 17 17 17.5 22.5 23

Ni 9 10.5 11.5 11 11.5 2 4

Mo N

2.1 2.1 2.6 0.3 0.3

PREN 18.5 18.5 24 Ti5(C+N) 24 0.15 29 0.20 26 0.10 25

Others

Mechanical properties
Table 2 shows the ultimate tensile strength (U.T.S.) and the yield strength (Y.S.0.2%) measured on hot rolled plates with thickness in the range 7-20mm. From these results, 450 and 650MPa can be done as minimum values for Y.S.0.2% and U.T.S. respectively. The objective to obtain tensile properties twice than the conventional austenitic is reached by combining duplex microstructure with high N level.
Table 2. Room temperature tensile data of 1.4062 hot rolled plates

Thickness 7 12 20

Y.S. 0.2% (MPa) 548 532 475

U.T.S (MPa) 725 750 689

E (%) 36 36 40

F03-13

606

Corrosion resistance
Pitting corrosion resistance Cold-drawn wires pitting corrosion resistance in chloride environments Pitting corrosion is evaluated by an accelerated electrochemical test which determines the pitting potential: the higher the pitting potential, the better the pitting corrosion resistance (potentiodynamic testing with measure of pitting potential for a current density of 100 A/cm2). Two types of samples for 1.4062 and 1.4404 grades were tested: - Industrial rod-wire with a diameter of 5.5mm; the samples are tested after mechanical polishing (paper SiC 1200) and after air ageing for natural passivation (during 24 hours). - Industrial cold-drawn wire, with 2.3mm and 1mm diameters. The samples are tested with their industrial surface after only a short degreasing. We used a chlorides containing solution (NaCl 0.86M or 5%weight) at 35C and neutral pH; this medium is the solution of the salt test (ASTM B1177). The results are given in Figure 1. The lean duplex 1.4062 presents pitting potentials higher than the ones measured for the austenitic 1.4404 for rod and cold drawn wires. In addition, it should be noted that the difference of corrosion resistance between 1.4062 cold-drawn wires of diameters 2.3 and 1mm is due to the roughness difference (1mm diameter wires are smoother and roughness has an influence on pitting corrosion resistance). Rebar pitting corrosion resistance in severe concrete environments Pitting corrosion potentials have been measured in synthetic, alkaline, carbonated and chlorides containing solutions. The synthetic media were defined to take into account the evolution of the interstitial solution in the vicinity of reinforcement and concrete in time. The solution represents a concrete composition modified over time, with a high content of sodium chlorides. Indeed, the stainless reinforcements are often selected for aggressive environments such as marine conditions. The results obtained in a containing chlorides carbonated medium at pH = 8 are given in Figure 2. The values of pitting potential of stainless steels are definitely higher than those measured under the same conditions for traditional steel (-350 mV/ECS). Lean duplex stainless steels 1.4362 and 1.4062 which contain low level of Ni and Mo, present pitting potentials higher than the ones measured for the austenitics 1.4301, 1.4404 and 1.4571. In addition, it should be noted that a good correlation is obtained with the PREN (Pitting Resistance Equivalent Number) in these alkaline solutions.
500 Pitting potential (mV/SCE)
Pitting potential (mV/SCE) 350 300 250 200 150 100 50 0

1.4062 1.4362 1.4571 1.4301 1.4404

400 300 200 100 0 1.4404 Rod 1.4062 Rod 1.4404 Cold 1.4062 Cold 1.4062 Cold drawn drawn wire 5.5mm wire 5.5mm drawn wire 2.5mm wire 2.3mm wire 1mm

18

20

22

24

26

28

PREN = %Cr + 3.3Mo% + 16%N

Figure 1. Pitting potential for various grades in neutral medium with an addition of sodium chloride of 50 g/L at 35C.

Figure 2. Pitting potential for various grades in the synthetic medium carbonated at pH= 8 and with an addition of 21 g/L of chlorides.

F03-13

607

Plate pitting corrosion resistance in chloride containing solutions Pitting corrosion characterizations were performed on polished samples removed from 7mm thick hot rolled plate. Electrochemical tests began 24 hours after sample preparation in order that the passive film may be naturally formed as a result of electrochemical reactions with the atmosphere. After 5000s at the free potential, potentiodynamic curves were plotted at a scan rate of 900mV/hour from -50mV/free potential in the anodic direction until the current density reach 500A/cm. Pitting potential was measured at a current density of 100A/cm. After the completion of the electrochemical tests, stainless steel samples were observed by means of an optical microscope Firstly, pitting potentials were measured in a solution containing 250mg/L of chlorides (NaCl 7.10-3M) at pH 5.5 0.1 and 25 0.1C. The electrolyte was prepared from deionised water (R = 18.2M). These experimental conditions are the most aggressive, in respect of pH and chlorides content, encountered for fresh water. Indeed 250mg/L is the maximum concentration of several drinking water standards8,9,10 and 5.5 is the pH value taken by natural aerated fresh water. With these experimental conditions, 1.4307 and 1.4404 have pitting potential values around 800mV/SCE whereas no pits were observed for both duplex grades 1.4062 and 1.4362 after completion of the electrochemical test (see Figure 3). This indicates that the new duplex grade, with a pitting corrosion resistance higher than the one of 1.4307, will be suitable in environments containing limited chlorides content as drinking water. Then critical pitting temperature (CPT) was measured according to ASTM G150-99 standard11. The specimen is exposed to a 1M NaCl (35.5g/l chlorides) solution and heated from 1 1C to CPT at a rate of 1C/min. 60s before the start of the temperature scan, the specimen is anodically polarized at 700mV/SCE. The current is monitored during the temperature scan, and the CPT is defined as the temperature at which the current density exceeds 100 A/cm2 for 60s. Pitting on the specimen is confirmed by a visual examination performed at the end of the test. The CPT for the new duplex grade 1.4062 is higher than the one of conventional austenitic stainless steels 1.4307 and 1.4404 and near from the one of most alloyed austenitic grade 1.4429 (see Figure 4). In the austenitic family, beneficial effect of Cr, Mo and N on CPT value are highligthed. For duplex, the highest value is measured on 1.4362 sample that would indicate a beneficial effect of Ni too.
1,5 1,25 Pitting potential (mV/SCE) 1 0,75 0,5 0,25 0 1.4307 1.4404 1.4429 1.4062 1.4362

No pits

30 Critical pitting temperature (C) 25 20 15 10 5 0 1.4307 1.4404 1.4429 1.4062 1.4362

Figure 3. Pitting potential in 250g/L chlorides containing solution at pH 5.5 and 25C.

Figure 4. Typical value of CPT according to ASTM G150 standard (vertical arrow indicates that CPT is lower than 5C)

Crevice corrosion resistance Crevice corrosion resistance of 7mm hot rolled plate was investigated by an electrochemical technique. Several potentiodynamic curves were plotted on samples, free of crevice promoting
F03-13

608

equipement, with pH values decreased from 3 to 0.5 (controlled by HCl) in order to measure the maximum current density on the active peak. These tests were performed at 20C in a 2M NaCl (70 g/l chlorides) solution, which corresponds to the expected chloride concentration range inside a crevice. One hour before the beginning of the test and during this test, the solution and the cell are deaerated with N2. After 15 minutes at free potential, a fixed potential of 750mV/SCE was applied for 2 minutes in order to reduce the surface species. Then, the potentiodynamic curve was plotted in the anodic direction at a scanning rate of 600mV/hour from 750mV/SCE until the current density reached 500A/cm2. From all the curves plotted for each grade, the maximum current density in the active domain was plotted in function of pH value and depassivation pH (pHd) determined at 10A/cm2 (see Figure 5). pHd corresponds to the onset of an active peak in the potentiodynamic curves. For the four tested grades, pHd values are very similar and equal about 1.6. Considering a mechanism of crevice initiation based on general breakdown of the passivity, that means that the high Cr content of the duplex grades allows obtaining a resistance to initiation similar to the one of a 2%Mo containing austenitic grade. On the other hand, for pH values lower than the depassivation pH, the current density decreases with the increase of the Ni and Mo contents. This confirms the beneficial effect of these alloying elements on the resistance to crevice propagation. Uniform corrosion resistance Coupons taken from hot rolled plates were immerged during several 48h periods in stagnant sulphuric acid. The corrosion rate was evaluated by weight loss measurements. The iso-corrosion curves plotted on Figure 6 for both duplex grades and Mo containing austenitic grade 1.4404 are very similar. This confirms the beneficial effect of high Cr content on the uniform corrosion resistance in diluted sulphuric acid.
600 Maximum current density in the active domain (A/cm) 1.4062 500 400
activity

90 80 70 Temperature (C) 60 50 40 30 20 10 1.4362 1.4062 1.4404 1.4307

1.4362 1.4307
passivity

1.4404

300 200 100 0 0,5 1 1,5 pH 2 2,5

0 0 5 10 [H2SO4] (%) 15 20

Figure 5. Maximum current density in the active domain versus pH at 20C in 2 M NaCl.

Figure 6. ico-corrosion curves for pure diluted sulphuric acid (corrosion rate = 0.2mm/y).

Atmospheric corrosion resistance A5 size coupons removed from 1.4062 hot rolled plates were welded and prepared with different mechanical surface treatments: shot blasted, sand blasted and polished. They were exposed in industrial-urban atmosphere and in rural atmosphere. The aggressiveness of these two locations is classified C2 according to corrosion rates measured on steel, zinc, copper and aluminium reference coupons (see ISO 9226 standard)12. Up to now, no signs of localised corrosion have been observed on these specimens after more than one year of exposure.

F03-13

609

Conclusion
By combining low Ni content with N addition and without Mn content increase, a new lean duplex grade EN 1.4062 (UNS S32202) was developed by Industeel and Ugitech. This low Ni, Mo free grade is less sensitive to fluctuations in raw material prices than the standard austenitic grades. Cr content higher than 21.5% gives pitting, crevice or uniform corrosion resistance better than 1.4301 or 1.4307 and sometimes similar to 1.4404 or 1.4571. Due to duplex microstructure and 0.2% N addition, the tensile properties are also very high and about twice the ones of standard austenitic grades. Finally, 1.4062 is today available on hot rolled plate, bar, rebar and cold drawn wire forms. It presents a very interesting cost/technical performance ratio and appears as a promising alternative to standard austenitic materials, cement, coated or galvanized carbon steels in construction (storage, architecture, bridge ) or transport applications.

References
[1] P. Soulignac and J.-C. Gagnepain: Why duplex usage will continue to grow, Duplex conference, Grado, Italy, 18 20 June 2007. [2] J. Peultier, F. Barrau, J.-C. Gagnepain and J. Grocki: Duplex and superduplex stainless steels for wet FGD , AIRPOL conference, Louisville, Kentucky, USA, June 26-28, 2007. [3] S. Jacques and G. Hagi: Tour Pomerol : eight year experience with duplex 1.4462, Duplex conference, Grado, Italy, June 18-20, 2007. [4] S. Jacques, J. Peultier, V. Baudu, B. Chareyre and J.-C. Gagnepain: Corrosion resistance of duplex stainless steels for thermal desalination plant, IDA conference, Maspalomas, Gran Canaria, Spain, October 21-26, 2007. [5] E. Chauveau, M. Mantel, B. Drab, S. Chedal; Stainless Steel World Magazin, July 2006, pp30-33. [6] J. Kerr, P. V. T. Sheers and R. Paton: A new lean duplex stainless steel with a low nickel content, 4th european Stainless Steel Science and Market Congress, Paris, 2002. [7] Standard practice for operating salt spray (fog) apparatus, ASTM B117, 2007. [8] Guidelines for Drinking Water Quality, World Health Organisation, 1993. [9] European Commission Directive on the quality of water intended for human consumption (98/83/EC), 1998. [10] United States Environmental Protection Agency (USEPA) requirements based on the National Primary Drinking Water Regulations as amended under the Safe Drinking Water Act of 1996. [11] Standard Test Method for Electrochemical Critical Pitting Temperature Testing of Stainless Steels, ASTM G150-99 standard, www.astm.org, 2004. [12] Corrosion of metals and alloys Corrosivity of atmospheres Determination of corrosion rate of standard specimens for the evaluation of corrosivity, ISO 9226 standard, www.iso.org, 1992.

F03-13

610

MECHANICAL PROPERTIES AND CORROSION RESISTANCE OF W BEARING SUPERDUPLEX STAINLESS STEELS

C. Muoz, A. Pal, A. Gallardo, J. A. Odriozola

Universidad de Sevilla-CSIC.C, Spain

Abstract
Superduplex stainless steels present an excellent combination of mechanical properties and localised corrosion resistance. Their chemical composition is based in high contents of expensive elements such as Ni (~ 7%) and Mo (~4%). Ni and Mo contents can be reduced by alloying with N, V and W while maintaining good corrosion resistance (PRE number above 40) and mechanical properties at room temperature. The addition of W will not only account for the reduction in the Mo content, it will also lead to enhanced mechanical behaviour, making these alloys good candidates in applications where high strength and high corrosion resistance are required. In this work we compare the pitting corrosion resistance in chloride media and mechanical properties at room temperature of several experimental alloys with the standard SAF 2507 superduplex stainless steels. The chemical composition of the experimental alloys has the following percents: Cr 25%, Ni 7%, Mo 1-3.8%, N 0.4-0.5% and W 0-6%. The other alloying elements are maintained with the same concentration as in the SAF 2507 stainless steel.

Introduction
Duplex and superduplex stainless steels have an excellent combination of mechanical properties and corrosion resistance. These properties rely on a microstructure formed by approximately equal parts of austenite and ferrite, its morphology and chemical composition. [1]. The first duplex stainless steel had a nominal chemical composition of 22% Cr, 5% Ni and high N content, up to 0.17%. Second generation of duplex alloys included Mo to a maximum of 3% and 0.2%N to increase the pitting corrosion resistance of 2205 type. 22Cr-5Ni-3Mo stainless steels are typically employed in the food industry and off-shore applications [2]. Third and, until now, last generation of duplex stainless steels, known as superduplex, have higher contents on Cr, Ni Mo and N whose typical composition is of 25Cr-7Ni-4Mo-0.3N. These have the best resistance to pitting corrosion than 2205 grades and good mechanical resistance with high elastic limit that allow for material costs reduction. These alloys are increasingly used in many applications as paper and chemical industry, chemical products transport and petroleum and gas manufacturing as well as in off-shore structures [3,4]. The corrosion resistance of duplex stainless steels is described by the Pitting Resistance Equivalent (PRE), which takes into account the influence of the alloying elements in the pitting potential of the stainless steels. Originally PRE = %Cr + 3.3%Mo
F03-14

(1)
611

Later, nitrogen was included in the PRE value; the pitting resistance equivalent accounting for nitrogen is designed as PREN and is given by PREN = %Cr + 3,3% Mo + x% N (2)

where x varies between 16 and 30 depending on the steel type, chemical composition and the heat treatment of the duplex stainless steels. As research on duplex stainless steels proceed, other elements needed to be included in the PREN expression, this is the case of tungsten, W. Some author name the new PRE including W as PREW, the influence of W in the PREW is related to Mo but weights half the value of that [5,6]. PREW = %Cr + 3,3(% Mo + %W) + x% N (3)

Duplex stainless steels can be regarded as superduplex if its PRE number (PREN or PREW) has a value above 40. Tungsten is usually added above 2% to enhance the pitting resistance increasing the passive potential range and the crevice corrosion resistance in hot chlorine solutions. This is due to the migration of W to the passive layer where it forms WO3 which is insoluble in water. In neutral chlorine solutions WO3 interacts with other oxides increasing the passive layer stability [7]. The main pitfall of duplex stainless steels is their tendency to form the hard and brittle phase. It has been confirmed that W levels between 1 and 3% difficults the formation of intergranular phase, but it still can precipitate inside the grains in intragranular form. It is thought that this is due to the diffusion of W and Mo to the grain boundaries. Usually, addition of W is accompanied by an equal reduction in Mo in such a way that W + Mo are bellow 5 to 6% [8]. In this work we have studied new superduplex stainless steels with reductions in Mo content by substitution of this element with increasing contents of W. Standard SAF 2507 (EN 1.4410), 25% Cr, 7% Ni, 4% Mo y 0.3% N, duplex alloy will serve as reference. Mechanical properties and pitting corrosion resistance will be determined by standard methods to compare the new alloys with the standard one.

Experimental
The chemical composition must be such that the and volume fractions are about 50/50 and to avoid the precipitation of secondary phases that might negatively affect the alloy properties: , nitrides and carbides. Also C and S will be maintained as low as possible since they enhance the precipitation of undesirable phases. The design of the alloys has been carried out using the Schoeffer equations [9] which take into account the effect of W and N. Nieq = %Ni + 30%C + 0,5%Mn + 30%N Creq = %Cr + %Mo + 1,5%Si + 0,72%W Experimental 500 g ingots are fabricated using an induction centrifugal furnace (LECOMELT 6.6 p VAC) with controlled atmosphere of N2 at a pressure of 2 bar. As-cast alloy ingots are soaked at 1050C during 1 hour to homogenise the microstructure. Chemical composition was measured by glow discharge optical emission spectrometry

F03-14

612

(GD-OES) and elemental analysis for light elements, table 1 gives the chemical composition of the four experimental alloys. Volume fraction of austenite and ferrite where measured using standard metallographic techniques and both phases was determined by X-ray diffraction peaks. Also the PREW number was determined using the equation (3) with a coefficient for N equal to 16, so this will give us the lowest PREW number possible. The values of ferrite % and PREW number are also indicated in table 1.
Table 1. Chemical composition ferrite % and PREW number of experimental superduplex stainless steels fabricated in this work. C SD- 1 SD- 2 SD- 3 Si Mn P S Cr Ni Cu N Mo 3.69 2.9 2.2 1.2 W 0.06 2.15 4.06 6.49 Creq 23,45 28,14 29,89 30,48 Nieq 18,86 18,82 20,00 18,95 % 33 30 23 28 PREW 43.6 44.8 47.1 47.7

SAF 2507 0.015 0.56 0.24 0.02 0.0093 0.016 0.64 0.19 0.02 0.0059 0.018 0.69 0.24 0.02 0.0060 0.022 0.82 0.23 0.018 0.0060

25.0 7.55 0.32 0.39 24.8 7.01 0.30 0.41 25.2 6.92 0.60 0.46 24.7 6.97 0.37 0.41

A specimen is extracted from all the alloys and cold rolled (figure1) to a thickness of 0.5 mm. Cold rolled sheet where heat treated in the same conditions than the as-cast specimens to relieve the stress in the deformed microstructure. As can be seen in table 1 the ferrite content is low for all the steels prepared, this is probably due to the high Ni content and the low Mn content in all the alloys. Microstructure of ashomogenised and cold rolled specimens is shown in figure 1; these are representative of all the alloys. As-cast microstructure after homogenisation treatment is formed by island of austenite in ferrite grains with solidification structure. Grains are of irregular shape and show no equiaxiality although no preferred direction can be seen in the microstructures. In the microstructure of cold rolled specimens ferrite grains are deformed in the rolling direction giving a fine duplex microstructure with grains strongly elongated in the rolling direction. Numerous spherical Si-rich inclusions are seen in all the samples. These are caused by the small weight of the ingots which give a low surface to mass (volume) ratio and large interaction with crucible walls during the casting process. Si content is a little bit higher in alloy SD-3.

Figure 1. Microstructure of SAF 2507 alloys. Left, as-cast specimen and right, cold rolled alloy. Etching Vilellas

Mechanical properties at room temperature where determined by tension and compression tests. Tension test was performed on the cold rolled specimens while compression tests were done on the as homogenised ones. Pitting corrosion resistance was determined in as homogenised specimens by cyclic potentiodinamic experiments in 3.5 wt. % NaCl water solution.

F03-14

613

Results and Discussions

X-ray diffraction results, figure 2, show the characteristic peaks of the austenitic and ferritic phases in the alloys. No other crystalline phases were detected.
Phase Phase

3500

3000

2500

Intensity, a.u

2000

1500

SD3

1000

SD2 SD1 SAF2507

500

0 20 30 40 50 60 70 80

2 theta, deg

Figure 2. X-ray diffraction for the alloys.

Table 2 show the tension test results. Experimental superduplex alloys have a higher maximum tensile stress then the standard SAF 2507 alloy. Rmax value for SD3 is three times that of the 2507 type alloy.
Table 2. Results for tension test to the alloys. * These alloys fractured outside the calibrated zone

Alloy SAF2507 SD1* SD2* SD3

Rmax (MPa) 1151 2787 2740 3313

Elongation (%) 30.08 ND ND 12.20

Figure 3 is a bar diagram of the Rmax value for the alloys, it is evident from the graph that alloying with tungsten to 6.49% enhances the mechanical properties. Figure 4 is a comparison of the stress-deformation curves for reference and SD1 alloys. The increase in the mechanical resistance is accompanied by a proportional decrease in the elongation (the data for those specimens could not be calculated because fracture was outside the calibrated area).
3500

3000

2500

(MPa)

2000

1500

1000

500

saf2507

SD1

SD2

SD3

Figure 3. Bar diagram of Rmax for the experimental alloys in this work.
F03-14

614

3,0

SD1

2,5

SAF-2507 Stress (MPa)

2,0

1,5

1,0

0,5

0,0 0 2 4 6 8 10 12

Deformation (mm)

Figure 4. Stress-deformation curves for alloys SAF2507 and SD1 under tension load.

Compression test results are plotted in figure 5. Stress-strain curves indicate that there is a similar behaviour between alloys SAF-2507 and SD1 (their curves are parallel) while a different trend is encountered for SD2 and SD3 alloys. Those differences in trend during the compression tests can be due to the precipitation of W rich phases in the - grain boundary.

Figure 5. Stress-strain compression curves for the alloys studied in this work

Results indicate that decreasing Mo and increasing the W content enhances the mechanical resistance both in tension and compression tests. Tension tests indicates that there is an increase in the elastic limit data as the W content is increase. Also, the higher the W content the higher the maximum resistance of the specimen. However, the compression test does not indicate that differences in the elastic limits. That behaviour is due to the different microstructure of the specimens. Tension tests were made on cold rolled specimens in the rolling direction so that the microstructure was of elongated grains in the tension direction while the compression experiments were done on the as-cast microstructure that shows no preferential direction. The different microstructures give a different mechanical behaviours. Regarding the corrosion results, table 3 and figure 6, alloy SAF-2507 has a slightly higher pitting potential than the other samples but the differences are not relevant and all the alloys have similar pitting resistance as it was expected from the PREW numbers.
F03-14

615

Table 3. Pitting potential for the superduplex alloys studied in this work.

Muestra SAF2507 SD-1 SD-2 SD-3

Epic (V) 1,14 1,08 1,10 1,11

1,25

E/(V)

1,00

SD3 SD2 SD1 SAF-2507

0,75 0,000 0,002 0,004 0,006 0,008

2

0,010

0,012

I/(A/Cm )

Figure 6. Cyclic potentiodynamic polarization curves for the superduplex alloys studied in this work.

Conclusions
Result in this work indicate that substitution of Mo by W in superduplex stainless steel will lead to better mechanical resistance while maintaining similar resistance to pitting corrosion. The mechanical properties are related to the microstructure of the alloys and, thus, to their thermomechanical history. Cold rolled specimens have higher elastic limit as W increases in the alloy. The pitting potential in chlorine media in all the as-cast alloys is similar and differences between them are negligible.

References
[1] [2] [3] [4] [5] [6] [7] [8] [9] S. Jana, 1st European Stainless Steel Conference, v.3 p.343-348. Dionicio, E. Rev. Inst. investig. Fac. minas metal cienc. geogr, jul. 1999, vol.2, no.3, p.1121. ISSN 1561-0888. R. N. Gunn. Abington, Duplex Stainless Steels. Microstucture, Properties and Applications. Publishing. 1997. J. Foct, T. Magnin, P. Perrot, J.-B. Vogt, in: J. Charles, S. Bernhardsson (Eds.), Duplex Stainless Steels 91, Beaune, 1991, p. 49. J. E. Truman. Conf. Proc. UK Corrosion. Birmingham 2 (1987) 11. L.F. Garfias-Mesias, J.M. Sykes, S.S. TUC Corrosion Science 38 (1996) 1319-1330. C. J. Park, H. S. Kwon, Corrosion Sci. 44 (2002) 2817 B.W.Oh, J.I.Kim et al, 1st European Stainless Steel Conference, v.3 p.59-64 E.A. Schoefer, Welding J. 53 (1974) 10.

F03-14

616

HIGH TEMPERATURE FORMING OF A SUPERDUPLEX STEEL AND ITS SIMULATION BY TORSION TESTING. COMPARISON BETWEEN SUPERDUPLEX STEELS AT SIMILAR TEMPERATURES AND STRAIN RATES RANGES

M. Cars1, I. Rieiro2, F. Pealba3, J. Muoz2, J. Castellanos2, O. A. Ruano1

National Center for Metallurgical Research, Spain, 2University of Castilla-La Mancha, Spain, 3 INASMET Foundation, Spain

Abstract
The forming behaviour of a super duplex steel is investigated by means of high temperature torsion tests. The composition of the steel is 23.5 Cr, 5.56 Ni, 3.2 Mo, 0.18 N and balance iron and has a PREN value of 37.1. This type of steel has an application in the production of seamless steel pipes that are used for oil extraction and transport. The torsion was performed at temperatures in the range 850 to 1200C and strain rates in the range of 2 to 26 s-1 to characterize the mechanical behaviour of the steel. The torsion tests were used, in addition, to simulate the hot forming of pipes under comparable conditions of temperature, strain rate and strain. The parameters of the Garofalo equation were calculated from the experimental torsion data to describe the deformation behaviour of the alloy at various temperatures and strain rates. A nonlinear method, involving an algorithm specifically developed for the treatment of this equation, was used. The high temperature forming of the steel was analyzed by means of energy efficiency maps. In addition, a study of the maximum mechanical stability conditions was made. The intersection region for maximum stability defined by the Liapunov criteria together with the maximum efficiency region allowed determination of the best conditions for the forming process. It is concluded that these conditions were 1000C at the typical industrial forming strain rates of 10 s-1. The results for this steel are compared with those for other superduplex steel with a PREN value of 39.73.

Introduction
An increase in popularity has been observed since the introduction of the first generation of duplex and super duplex steels. These kinds of steels are now used, for instance, for tubes, pipe fittings and valves in the oil extraction and transport. Among other reasons, this is due to a better intergranular and pitting corrosion resistance than other stainless steels. The high content of Cr, Ni and Mo ensures this high protection against corrosion. The accurate study of the forming behaviour and forming stability of these steels is of great importance since the cost of production can be significantly reduced and its safety can be improved for the applications previously described. In this work, the hot forming behaviour of a super duplex steel is investigated by means of high temperature torsion tests. The results obtained are compared with those obtained for another superduplex steel tudied in a previous work, which has a coarser initial grain size and variations in the relative content of Ni, Mo and Mn [1]. For this purpose, the Garofalo equation is used as a constitutive relation and its
617

parameters are employed to obtain the controlling creep mechanism and the most stable forming regions by means of efficiency stability maps. In addition, the ductility of both steels are evaluated and compared in order to determine the influence of the chemical composition.

Materials and experimental method

The two steels used in this investigation are commercial grade of type S32760. They were received as a semiproduct in the form of bars 300 mm in diameter with the usual quality conditions and had the following composition:
Table 1. Chemical composition of the superduplex steels. Composition Steel 1 Steel 2 %C 0.025 0.03 %Si 0.46 0.44 %Mn 1.64 0.5 %P 0.014 --%Cr 23.45 24.8 %Ni 5.56 7.0 %Mo 3.25 3.7 %Cu 0.17 --N 0.18 0.17

The major composition differences between the two steel are the content of Mn, Mo and Ni. Steel 1, can be considered as a superduplex steel since its PREN index (PREN = %Cr + 3.3%Mo+16%N) is 37.1. Steel 2 is also a superduplex steel with a PREN index of 39.73. The simulation of the hot forming behavior of this steel was studied in a previous work by means of torsion and tensile tests [1]. Simulation of the forming process for steel 1 was carried out by means of torsion tests. An induction furnace heats the test sample until the test temperature. This variable is continuously measured by means of a two-color pyrometer. A silica tube with argon atmosphere ensures protection against oxidation. The torsion samples have an effective gage length of L= 17 mm and a radius of R= 3 mm. The samples were deformed in a SETARAM high temperature torsion machine at CENIM. Strain rates varied between 2 and 26 s-1 and the temperature between 850 and 1200C. The mean initial grain size of the as-received steels was 30 m for steel 1 and 60 m for steel 2.

Results and Discussion

Fitting of the Garofalo equation Figure 1 a) and b) shows the torsion data for steel 1 and 2 respectively.

Figure 1. Logarithm of the strain rate vs logarithm of true stress at peak for a) steel 1 and b) steel 2

F03-15

618

The Garofalo equation is commonly used to unify the creep data in the entire stress range. It is known that this constitutive relation is capable of heuristically interpreting the creep behavior of polycrystalline materials [2]. This equation is given as follows:

= Ae

Q RT

[sinh( )]

(1)

where is the strain rate, T is the absolute temperature, is the stress, R is the universal gas constant, Q is the activation energy for deformation, and , n and A are material constants. An important characteristic of this equation is that it allows extrapolating the torsion data in order to approach the industrial conditions. The fitting of the Garofalo equation consists in determining the A, n, Q and parameters that best reproduce the torsion data. A non-linear method involving an algorithm specifically developed for the treatment of this equation was used in order to make the parameter identification [3]. The method grants an evaluation of the conditioning of the tests, by means of the F function of Snedecor [3]. The Garofalo equation is usually fitted at the maximum of the stress-strain curves (peak value) [4]. For this case, the optimal solutions of the parameters of the Garofalo equation obtained by the algorithm previously described are the following: Steel 1: Steel 2:

= (5.481011 s 1 )e

285 kJ / mol RT 447 kJ / mol RT

sinh(0.0083( MPa 1 ) ) sinh(0.0115( MPa 1 ) )

2.99

(2)
3.77

= (2.681016 s 1 )e

(3)

A comparison of equations (2) and (3) reveals that steel 1 shows lower activation energy than steel 2. This can be attributed to the different initial grain size of the steels evolving differently at the various strain rates during testing [5]. This is also the origin of the large values of the activation energy that are usually observed in these steels, much higher than that for iron selfdiffusion [6]. Furthermore, the value of n for steel 1 is close to that found in fine grained materials [7]. In contrast the n value for steel 2 is 3.77. This value is close to that associated to a creep mechanics controlled by the climb of dislocations at dislocation pile-ups [8]. Comparison of hot ductility of the Superduplex Steels The evolution of the number of turns to failure Nf, can be considered to be a measurement of the ductility of the material. The evolution of Nf with temperature and strain rate for steel 1 and 2 are shown in Figure 2 a) and b) respectively. The low stress exponent of the Garofalo equation for steel 1 suggests that this steel should have better ductility than steel 2 which has a higher stress exponent. This is true at the lower temperatures and strain rates. However, steel 2 shows a jump in ductility at the highest strain rates and temperatures. This surprising behavior needs to be investigated since it could be even associated to an experimental artefact.

F03-15

619

Figure 2. Number of turns to failure with temperature and strain rate for a) steel 1 and b) steel 2.

Forming stability of the studied steels

The hot forming behaviour of the two steels has been investigated. The hot forming of a metallic material is normally limited by the generation of different processes such as flow localization, cavitations, shear band formation, etc. All these phenomena are a function of the forming variables, T, and . The forming stability maps searches the forming regions where these phenomena are minimized. According to [9] it is possible to consider the workpiece material under hot working conditions as a dissipater of power. For this situation, the power P, absorbed by the workpiece during plastic flow can be expressed as:

P = = G + J = d + d
0 0

(4)

Where G is the dissipater content and represent the power spent in the deformation without changing the internal structure, and J is the dissipater co-content which is the power spent in the deformation with a change of the internal structure. From this point of view, the Lyapunov stability theorems for the dynamical systems can be applied to the plastic flow in hot deformation [10]: m 0 < m 1 (5) DMLE = <0 ln( ) s s 1 DSLE = (6) <0 ln( ) where m is the strain rate sensitivity and s is the entropy of the system. These parameters follow the expressions: ln( ) 1 ln( ) m= s= ln ( ) T (1 T ) (7) Working with these equations and using the Garofalo equation as the constitutive relation, the previous stability criteria can be expressed as [11]:
1

1 0<m= n

n
1 + sinh
2 n 1 1 n

DMLE =

1 <0 n
2

(8)

F03-15

620

Q s= 2 n RT

n
1 + sinh 1
2 n 1 n

DSLE =

Q <0 n 2 RT

(9)

n sinh 1 n n 1 + n
= 1 sinh 1 +
1 n 2 n 2 3 2

e RT

(10)

By means of expression (8) to (10) it is possible to construct stability maps that consist on a two dimensional representation of the previous variables, m, s, DMLE and DSLE with T and . The most stable region is that with the most negative values of the variables DMLE and DSLE. Figure 3 a) and b) show the contour map for the DMLE functions at strain at peak and for the two superduplex steels.

Figure 3. DMLE( ,T) for strain at peak for a) steel 1, and b) steel 2.

Figure 3 a) shows that the maximum stability region for steel 1 is about 1000C for the typical industrial strain rate of 10 s-1.(2.3 in the y-axis). For steel 2, figure 3 b) shows that this region is observed at a higher temperature, 1175C. Therefore, a higher temperature should be applied to steel 2 for a safe industrial forming process.

Conclusions
Steel 1 is less creep resistant than steel 2. The stress exponent and the activation energy in the Garofalo equation for steel 1 are lower than for steel 2. This may be attributed to the finer grain size of steel 1. The ductility of both steels even at the most favorable conditions is relatively low. Steel 1 is clearly more ductile than steel 2 at low strain rates and temperatures. At industrial working conditions, 10 s-1, the temperature for maximum stability for steel 1 is 175C lower than for steel 2.

Acknowledgement
The work was carried out through the Project PBC-05-010-1 from JCCM (Castilla-La Mancha, Spain).
F03-15

621

References
M. Cars, J.A. Jimnez, I. Rieiro, O.A. Ruano, Simulation of the hot forming behavior of a superduplex steel by means of torsion testing, Proceedings of the III Seminar on Metallurgical Technologies, Barcelona, 2001. [2] M.Y. Wu, O.D. Sherby, Unification of Harper-Dorn and power law creep through consideration of internal stress, Acta Metallurgica, 32, 1984, pp. 15611572. [3] I. Rieiro, O.A. Ruano, M. Eddahbi, M. Cars, Integral method from initial values to obtain the best fit of the Garofalo creep equation, J. Mater. Processing Technol., 78,1-3, 1998, pp. 177-183. [4] J. Adamczyk, M. Cars, R. Kozik, R. Wusatowski, Structure forming processes during hot deformation of a C-Mn-V-N steel, Steel Research, 66, 7, 1995, pp. 279-324. [5] H.J. McQueen, N.D. Ryan, Constitutive analysis in hot working, Mater. Sci. Eng. A 322, 2002, pp. 43-63. [6] I. Tamura, C. Ouchi, T. Tanaka, H. Sekine Thermomechanical processing of high strength low alloy steels, pp. 17-48; 1988, London, Butterworth. [7] O.D. Sherby, J. Wadsworth, Superplasticity: recent advances and future directions, Progress in Material Science, 33, 1989, pp. 169-221. [8] J.A. Jimnez, S. Klaus, M. Cars, O.A. Ruano, G. Frommayer Microstructure and High Temperature mechanical behavior of the NiAl27 at.% Cr intermetallic composite, Acta Mater., 47, 13, 1999, pp. 3655-3662. [9] Y.V.R.K. Prasad, H.L. Gegel, S.M. Doraivelu, J.C. Malas, J.T. Morgan, K.A. Lark, D.R. Barker, Modeling of dynamic material behavior in hot deformation: Forging of Ti-6242, Metall. Mater. Trans. A, 15, 1984, pp. 1883-1892. [10] J.C. Malas, V. Seetharaman, Using material behavior models to develop process control strategies, The Journal of the Minerals, Metals and Materials Society, 44, 1992, pp. 8-13 [11] M. Cars, I. Rieiro, J. A. Jimnez, F. Pealba, O. A. Ruano, Forming stability of an Al TiMo intermetallic compound and its dependence on microstructure, J. Mater. Processing Technol., 143-144, 2003, pp. 416-419. [1]

F03-15

622

INVESTIGATION ON LOW NI DUPLEX STAINLESS STEEL GRADES

I. Calliari1, J. Dobranszky2, E. Ramous1, G. Straffelini3, G. Rebuffi1

DIMEG University of Padua, Italy, 2Hungarian Academy of Sciences, Hungary, 3DIMTI University of Trento, Italy

Abstract
Different stainless steels (SS) can provide a very wide range of mechanical properties with the advantage of no need for surface protection. Duplex stainless steels (DSS), in particular, with twice the mechanical strength of conventional austenitic and ferritic SS, have a potential use in constructions. The DSS grades have been improved and, parallel with development towards higher grades for corrosive conditions, there is a great interest in leaner composition. A useful way to reduce the cost is to reduce the Ni content and to compensate with manganese and nitrogen additions. In the present paper the structural and mechanical properties of two low Ni duplex grades are compared in order to investigate the structural stability of the austenite as it may convert to martensite and the secondary phases precipitation. The detailed characterization has been performed with SEM-EDS and Charpy test on as received and on thermally treated (600-850C) specimens. A few precipitates of chromium carbides and nitrides at the grain boundaries have been detected in both grades. The martensite structure has been noted only in 2101 type DSS. As concerns the impact toughness the behaviour of 2101 grade is quite similar to that of other DSS, while the 2304 has no drastic drop of toughness. Their corrosion properties in aggressive chloride environments are quite similar to that of austenitic AISI 304 grade.

Introduction
The good corrosion resistance of duplex stainless steels justify their applications in very aggressive environments, typical of chemical, offshore, oil and gas industries while their mechanical strength enables interesting applications both for constructions and transport vehicles, offering the advantage of reduced maintenance costs, being unnecessary the surface protection. The base cost of the alloy is becoming a conditioning feature in diffuse and quantitatively important applications, in medium level aggressive environments. A great interest to develop lean grades of DSS aimed at the reduction of the cost, maintaining the basic good properties: mechanical strength, weldability, formability and good even if not extreme, corrosion resistance is growing. An apparent way to reduce the alloys cost is to reduce the content of the most expensive alloy components: nickel and molybdenum. Such reduction could be compensated by the increase of the manganese a nitrogen contents, to maintain the typical balanced microstructure of DSS, with both ferrite and austenite. In the last decade research [1-4] has been carried out with the aim to define the composition of steel following the above criteria. Several different magnetic measurements were applied successfully to characterize the properties and the microstructure of duplex stainless steels [5, 6]. The main matters arise in maintaining the correct balance between ferrite and austenite contents and in the stability of the austenite against its transformation to martensite during cold forming
F-1 P

623

as a result of plastic deformation [1, 2]. For some years duplex grades with low Ni or Mn-N substitution for Ni have been proposed and are currently used, but a lack of information exist on structural stability after deformation or thermal treating. The present paper is aimed to analyse the microstructure of two typical lean DSS.

Experimental
The as received materials were wrought rods (30 mm) previously solubilised (1050C, 30 min), with chemical compositions lying in the ranges of Table 1.
Table 1. Chemical composition (%wt.).

Grade 2101 2304

C 0.030 0.03

Mn Cr Ni 0.60-0.80 4.0-5.0 22.0-0.5 1.0-1.5 0.56 1.4 23.2 4.3

Si

Mo 0.50 0.18

P 0.035 0.027

S N 0.005 0.19-0.22 0.001 0.10

Isothermal short ageing treatments of specimens, were carried out at temperatures 550-850C for 15-90 min and long treatment were carried out at 670 C for 15-200 h. The volume fractions of ferrite and austenite in a solution treated sample have been measured on 3 longitudinal and 3 transversal sections (20 fields for each section) by image analysis on light micrographs at 200, after etching with the Berahas reagent (reaction time, 10s). The martensite which has been detected by OM and SEM, after etching with Berahas reagent and by X ray diffraction (CrK radiation). Different phases have been observed by SEM examination of polished samples, using the backscattered electron (BSE) signal, on the basis of atomic number contrast effect: the ferrite appears slightly darker than austenite, while the secondary phases would appears lighter. The SEM operated at 25 kV; the BSE detector was set to maximize the atomic number contrast, allowing ferrite, austenite and other phases to be identified. Instrumented CharpyV impact specimens were prepared in the standard form of 101055 mm. Impact test was carried out at room temperature, on samples treated at 550-650-750 and 850C for 15-45-90 and 120 min.

Results and discussion

Solution treated material The banded structure of elongated islands is observed in the longitudinal section, while the isotropic structure of ferrite and austenite grains is displayed on the transverse section. No secondary phases were detected. The values of volume fractions of ferrite and austenite, measured on longitudinal and transverse sections (200), are reported in Table 2.
Table 2. Austenite ( ) and ferrite () % vol. in longitudinal and transverse sections.

%2101 %2304 %2101 %2304 56 2 50 2 53 3 Longitudinal 50 2 443 543 475 Transverse 463 Table 3 reports chemical composition of ferrite and austenite measured with EDS-analysis, expressed as partition coefficients. The Ni and Mn austenite enrichment and Cr ferrite enrichment are evident, the partition coefficients are quite similar to that observed in the common Cr-Ni-Mo grades.

F-1 P

624

Table 3. Austenite and ferrite compositions. (Wt %, EDS)

Cr Mn Ni Heat-treated samples

/ 2304 1.23 0.75 0.59

/ 2101 1,14 0,84 0,62

Microstructure of 2101 grade DSS The microstructure has been investigated mainly on not-etched specimens by SEM-BSE; the ferrite is darker than austenite. At the temperature of 600C, for treatment time < 40 min no precipitation of secondary phases has been detected, for longer times some small dark particles were detected at the ferrite grain boundaries. They were analyzed by SEM-EDS (close to the resolution limit) and an enrichment of Cr was observed at the grain boundaries so the precipitates were identified as chromium nitrides. The same grain boundaries precipitation was observed after soaking times longer than 40min at 650C, while at 750C the first grain boundary precipitation has been detected after a 20 min treatment (Figure1a) and can still be observed after 20 h (Figure1b). Increasing the temperature, particles became larger and the precipitation occurs also at the - boundaries (Figure 1b).

(a)
Figure 1. SEM-BSE micrographs of sample treated at 750 C for 45 min (a) and 20 h (b)

(b)

The shortest times for grain boundary carbide precipitation lies in the temperature range 650-750C, as already observed [3]. No and phases have been detected, neither for very long thermal treatments in the 650-900 temperature range. This could be related to the low Ni and Mo contents. In addition the Ni content may induce the instability of the austenite, as suggested in previous researches, which report of a probable transformation to martensite during cold forming (1). Moreover the martensite formation has been confirmed (2) in some low-Ni DSS after cold rolling and annealing (1040C, air quenched). We have detected different amount of martensite laths (Figures 2 and 3) in treated and rapidly quenched from 750-850C samples. Different cooling rates have no significant effect on the amount of final martensite. The X-ray diffraction spectra evidenced that the ferrite peaks increase as the amount of martensite increases.

F-1 P

625

Figure 2. SEM-BSE and OM micrograph (750 C, 25 min, WQ): martensite laths.

Microstructure of 2304 grade DSS The microstructure of the 2304 grade DSS is not affected by the heat treatment at 600C and no secondary phases or alpha-ferrite spinodal decomposition have been noted. A moderate precipitation of carbo-nitrides has been detected after long treatments (100 hours) at 670C and after 45 min at 750C. In Figure 3 the SEM-BSE micrographs of the specimen treated for 90 min and 20 hours are reported. The carbo-nitride precipitation is evident just below the austenite grain boundary, as it has moved towards the austenite (ferrite) giving the precipitation inside the austenite grains. A similar grain boundary precipitation was observed after long times (10 hours) treatment at 850C but the kinetics are slower than at 750C. On the other hand the main effect of the heat treatment in the range 600-850C is the increasing of the austenite volume fraction, with values ranging from 441% of the as received sample to 622% of the sample treated at 850C for 15 hours, accompanied by a decreasing of the Cr content in the austenite and by its increasing in the ferrite. This Cr enrichment and the very low amount of Mo seem to stabilize the ferrite and the secondary phases formation is not favoured. The austenite of this grade appears to be stabile, indeed no austenite to martensite transformation has been detected after heat treatment. Probably the Ni (4.3%) and N (0.18%) contents are high enough to stabilize the austenite avoiding the structural transformation evidenced in the 2101 grade DSS.

Figure 3. SEM-BSE micrographs: 750C for 90min(left) and 20 hours(right)

F-1 P

626

Impact toughness The effect of heat treatments on toughness of both the grades was studied by Charpy impact tests carried out at room temperature. As shown in Figures 4 and 5, the steels have different impact toughness properties: the 2101 grade has the ductile and fragile behaviour, while the 2304 has no the fragile behaviour. The 2101 is ductile until 20-40 minutes of isothermal treatment at 600-650C, corresponding to first stages of the carbides-nitrides precipitation, where the impact energy drops down at about 50 J. The critical times for precipitation lie around 750C, in good agreement with (3). However the impact energy is never lower than 30 J, also after very long soaking times, of several hours. The presence of some laths of martensite on the impact toughness has not yet been investigated. The 2304 grade DSS is always ductile, with the impact energy values never lower than 200 J. At this stage of the research we may conclude that the presence of nitrides at the austenite grain boundaries has no remarkable effects on the toughness of the 2304 steel. The sample was treated at 550C (1), 650C (2), 750C (3), 850C (4), for 15 min (A), 45 min (B), 90 min (C), 120 min (D).
DUPLEX 2101
350 Impact Energy [J] (Geometric Test) 300 250 200 150 100 50 0 1A 1B 1C 1D 2A 2B 2C 2D 3A 3B 3C 3D 4A 4B 4C 4D Heat Treatment

Figure 4. Impact energy of 2101 versus time/temperature of treatment

DUPLEX 2304
350 Impact Energy [J] (Geometric Test) 300 250 200 150 100 50 0 1A 1B 1C 1D 2A 2B 2C 2D 3A 3B 3C 3D 4A 4B 4C 4D Heat Treatment

Figure 5. Impact energy for 2304 grade DSS versus time/temperature of heat treatment

Conclusions
Some results about the study of two duplex stainless steels with different low nickel contents were presented: The relatively low nickel and molybdenum contents make the precipitation of intermetallic phases more sluggish than in conventional duplex stainless steels, and no sigma related phases precipitation has been detected, also after long time isothermal aging treatments, in both the grades

F-1 P

627

Precipitation at the grain boundaries of chromium nitrides has been observed after isothermal treatment in the temperature range 600-750C, with different kinetics The austenite of the 2101 type DSS is quite instable, and a diffuse transformation austenite-martensite has been evidenced, while the austenite of the 2304 DSS is more stable and no martensite has been detected. The impact toughness after solution treatment is very good in both grades, The impact energy after isothermal treatment in the 2101 grade is never lower than 30 J, while in the 2304 is never lower than 200 J, General corrosion properties in chloride environments are quite similar to that of austenitic AISI 304 grades.

References
[1] [2] [3] [4] [5] [6] C.D. Van Lelyveld, A. Van Benekom, Mater. Sci. Eng., A 204 (1996) p. 229. J.M. Hauser et al., Stainless Steel 99, Science and Market, Sardinia, Italy, (1999), p. 85. P. Johansson, M. Liljas, AvestaPolarit Corrosion Management and Application Engineering, 24 (2002) p. 17. H. Sieurin, R. Sandstrom, E.L. Westin, Met. Mat. Trans. A, 37A (2006) p. 2975 I. Mszros, Physica B, 372 (2006) p. 181 I. Mszros, P.J. Szab:, Journal of NDT&E International, 38 (2005) 7, p. 517

F-1 P

628

DUPLEX STAINLESS STEEL WELDS: RESIDUAL STRESS DETERMINATION, MAGNETIC FORCE MICROSCOPY AND SUSCEPTIBILITY TO INTERGRANULAR CORROSION

B. Gideon1, L. Ward2, D.G. Carr3, O. Muransky3

ARV Offshore, Thailand, 2RMIT University, Australia, 3Australian Nuclear Science and Technology Organization (ANSTO), Australia

Abstract
A section of a Duplex Stainless Steel (DSS) pipeline girth weld (single vee joint configuration) was systematically analyzed to determine the stress / strain levels and correlation with susceptibility to IGC within austenite and ferrite phases in the weld cap, fill and root region. Stress / strain levels were determined by means of Neutron Diffraction techniques. Magnetic Force Microscopy (MFM) were used to determine the size, shape and distribution of the austenite and ferrite within the various regions. ASTM A262 and a modified Double Loop Electrochemical Potentiokinetic Reactivation (DL-EPR) Test methods were used to assess the susceptibility to IGC. A clear variation of stress/strain was evident between the austenite and ferrite in the base material, HAZ and weld from the neutron diffraction results obtained. The results of the weld metal from MFM shows the formation of both a finer and coarse structure within the weld metal, which is dependent on the level of undercooling. The values for Ir/Ia and Qr/Qa in the DL-ERP test results revealed that the fill area had the highest level of susceptibility to IGC. Keywords: Duplex Stainless Steel, Intergranular Corrosion, Double Loop Electrochemical Potentiokinetic Reactivation, Magnetic Force Microscopy, Neutron Diffraction

Introduction
Welding of Duplex Stainless Steel (DSS) is particularly difficult with respect to maintaining a ferriteaustenite ratio close to 50:50. Rapid cooling effects associated with weld thermal cycles, often results in ferrite contents in the weld metal in excess of 50% may result in the loss of strength and increased susceptibility to IGC. The weld structure and the austenite / ferrite phase ratio are largely influenced by weld heat inputs and the cooling rates. The aim of this study is to conduct a detailed analysis of a girth welded sections of a DSS pipeline, as a function of heat input and type of weld, in terms of the residual stress by neutron diffraction, metallurgical analysis by means of magnetic force microscopy, and to assess the susceptibility to IGC by means of ASTM A262 and a modified Double Loop Electrochemical Potentiokinetic Reactivation (DL-EPR) test.

F-2 P

629

Experimental Procedures
Welding Conditions The DSS linepipe wall thickness was 10mm with a 200mm diameter of UNS 31803. Full details are listed in Table 1.
Table 1. Chemical composition of pipe and filler material.
C Pipe Filler Material 0.030 0.016 Mn 2.0 1.69 P 0.025 S 0.015 Si 1.0 0.42 Ni 6.50 8.60 Cr 23.00 23.07 Mo 5.50 3.20 N 0.20 0.160 Cu 0.16 0.16 Creq 32.04 28.09 Nieq 10.78 11.90

Note; Creq = Cr+1.37Mo+1.5Si+2Nb+3Ti and Nieg = Ni+22C+0.31Mn+14.2N+Cu

Manual Gas Tungsten Arc Welding (GTAW) technique was performed as detailed in Table 2. Upon completion of welding, the weld was subjected to Non Destructive Testing.
Table 2. Weld Condition Single Vee joint configuration
Weld Pass 1 (weld root) 2 (weld fill) 3 (weld fill) 4 (weld fill) 5 (weld cap) Travel Speed mm/min 51.00 123.00 66.00 64.00 64.00 Average Heat Input J/min 1474.71 883.12 1745.45 1788.00 1685.63 1515.38

Residual Stress Measurements Residual strain measurements were made using neutron diffraction with a wavelength of 1.40 on TASS (The Australian Strain Scanner) at the Australian Nuclear Science Technology Organization (ANSTO), Strains were measured in the three directions - longitudinal, transverse and normal (L,T and N) to the welding direction. These were the assumed principal stress directions. The measurement of residual elastic strain by monochromatic neutron diffraction relies on the use of Braggs law to relate the lattice spacings, dhkl, to the angle of diffraction 2 hkl associated with the diffraction peak labeled by Miller indices hkl at a fixed wavelength. Strain was calculated from the selected planar atomic spacing for ferrite and austenite at discrete locations in the weldment using Eq. 1.
0 0 hkl = (d hkl d hkl ) / d hkl

(1)

The calculation of the residual strains requires the knowledge of an appropriate reference lattice 0 spacing d hkl . This is problematic in welds where there is a possibility of redistribution of alloying elements, and secondly, inhomogeneous plastic deformation across the weld will generate relatively strong intergranular stresses in DSS. This problem was addressed by cutting a companion slice 2mm thick from the weld and cutting slits every 2mm across its length in order to relieve the 0 macroscopic residual stress field. Thus, the reference measurements d hkl represented the lattice spacing as a function of position relative to the weld centre and included any effects of alloy diffusion and intergranular stresses.

F-2 P

630

The average phase stress was calculated in the L,T and N directions for ferrite and austenite using the generalized Hookes law:
_ phase

L
_ phase

= = =

E hkl phase (1 hkl ) Lphase + hkl Tphase + N (1 + hkl )(1 2 hkl )

[ [

)]
(2)

T
_ phase

E hkl

(1 + hkl )(1 2 hkl )

[(1

hkl

phase ) Tphase + hkl ( Lphase + N )]

E hkl phase (1 hkl ) N + hkl Tphase + Lphase (1 + hkl )(1 2 hkl )

)]

Where hkl and vhkl and the diffraction elastic constants for each phase. The macroscopic residual stress field was then calculated by weighing the contribution of the respective phase stresses according to Eq. 3,
Macro L,T , N = (1 V f ) L,T , N + V f L,T , N

(3)

The volume fraction Vf of ferrite was determined from the ASTM E562 point count method. Intergranular Corrosion Tests (IGC) A modified ASTM A262 [1] was adopted in conjunction with a modified quantitative test method namely The Double Loop Electrochemical Potentiokinetic Reactivation (DL-ERP) test. Modified ASTM A262 Standard Practices Ecoppercopper sulfate sulfuric acid test for detecting susceptibility to intergranular attack was used. The specimen was covered with copper shot and grindings and immersed in a solution of 16 wt% sulfuric acid with 6 wt% copper sulfate. The solution was then heated to its boiling point and maintained at this temperature for 48 hours. On removal from solution, the specimen was bent through 180 over a rod with a diameter equivalent to twice the thickness of the specimen instead of four times the thickness to ensure, if cracks appeared, they would be apparent by the more restrictive bending radius. The bent surface of the specimen was then examined for cracks at low magnifications in the range X5 to X20. A modified Double Loop Electrochemical Potentiokinetic Reactivation (DL-ERP) Tests, was used as conducted by Schultz et al. [2,3,4]. The solution used was 0.5M H2SO4 + 0.001M TA (thioacetamide). TA is added to reduce the extent of ferrite dissolution. The test was conducted at 60 C. The polarization scan was started 5 minutes after immersion of the specimen. The potential was scanned from -500 mV (SCE) to +200 mV (SCE) and back to -500 mV (SCE) at a rate of 1.67 mV/s. The ratio of the reactivation charge to the passivation charge was calculated and is shown in the results section. Magnetic Force Microscopy Analysis Magnetic force microscopy studies were conducted on metallographically prepared cross-sections of the welds, after grinding and polishing using 3 m diamond paste. The Scanning Probe Microscopy from Digital Instruments at ANSTO, operating in tapping and lift modes was employed to study the topographic and magnetic features of the DSS samples. Topographic and magnetic force data were taken in the same scan. In order to produce reliable images, repeated scans in different directions were done to ensure reproducibility of the features. Various scan sizes and speeds were tested to enhance height and magnetic induced signals, thus minimizing tip hysteresis and the delay between line scans.

F-2 P

631

Results and Discussion

The residual stress, microstructure, resulting phase transformation, mechanical properties and degree of susceptibility to IGC are discussed in detail in this section. Table 3 summarises the results of the IGC tests carried out on the welded duplex stainless steels in this study.
Table 3. Summary of the DLERP results of the of the welded duplex stainless steels
Weld Pass 1 (weld root) 2 (weld fill) 3 (weld fill) 4 (weld fill) 5 (weld cap)

DL-ERP Test Qr/Qa 0.04 0.09 0.04 Ir/Ia 0.06 0.12 0.07

Residual Stress Measurements In the transverse direction, austenite exhibits tensile strains in the weld while the ferrite has contracted lattice spacing. As the distance from the weld centerline increases out to the HAZ (~5mm), an inversion occurs where upon ferrite strains are tensile and austenite is compressive. In the longitudinal direction, the strains for both phases are initially tensile in the weld, although the magnitudes are inverted for each phase in comparison to the transverse direction. In this direction the macroscopic residual stress field is at a maximum, due to constraint impeding contraction of the weld bead during cooling, and it is likely that this is the dominating effect. Moving out from the weld, the HAZ can be clearly distinguished from the weld as both average phase strains become uniformly tensile. This is interesting, in that this area of the weld undergoes transformation back to a completely austenitic structure before transforming partially back to ferrite [5]. In order to convert phase strain to stress (Eq.2) the diffraction elastic constants hkl and vhkl for each phase must be known, and these in turn depend on the crystallographic texture of the weldment which varies with position from parent to weld. Given the demanding experimental requirement for the texture at each location in the weld, a best approximation of the diffraction elastic constants was chosen using the self-consistent scheme proposed by Krner [6] for random texture. Such that, E 211 = 225.5, E 311 = 183.5 GPa, and 211 = 0.28 , 311 = 0.31 for the ferrite and austenite phases. The calculated phase and macro stresses (Eq.3). In the normal and transverse direction, the HAZ is strongly tensile for ferrite and compressive for austenite. These results suggest tensile ferrite regions could be susceptible to cracking in the HAZ. It is interesting to note that each phase is under very different stress states throughout the weldment. Considering the samples studied do not have the additive operational stresses normally superimposed on the residual stress field, it is quite likely that ferrite could be subjected to large tensile stresses in practice. Very high compressive stresses were estimated in the austenite phase for the transverse and normal directions, however, these stresses appear to balance by observation of the macroscopic stress field. It is a requirement for stress balance that the macroscopic stress in the normal direction tend towards zero at the surface and this generally true, however, the magnitude of the compressive stress in the austenite 2mm form the ID surface is questionable. A systematic error in the stress free reference may be a likely source of error. In the weld, the results show both austenite and ferrite to be under tensile stress in the transverse and longitudinal directions. Observation of the macroscopic stress field shows the highest stress to
F-2 P

632

occur in the longitudinal (welding) direction as expected. In the transverse direction, where cracking is most problematic in welds, the highest ferrite phase stresses occur in the mid-thickness of the plate. Microstructural Evaluation and Magnetic Force Microscopy Microstructural analysis for both GTAW weld conditions as shown by the magnetic force microscopy (MFM) images in Figure 1 reveals the presence of a two-phase banded structure, typical of such materials. In general, the austenite regions observed in the DSS weld metal is formed from ferrite in three modes, viz., as allotriomorphs at the prior-ferrite grain boundaries, as Widmansttten side-plates growing into the grains from these allotriomorphs and as intragranular precipitates [7]. In the micrographs, the grain boundary allotriomorphs and Widmansttten austenite are clearly seen. However, the austenite seen within the grain could be either intragranular precipitates or Widmansttten austenite intercepted transverse to the long axis. These microstructures, in addition to the presence of discontinuous grain boundary austenite layers (Figure 1a) and intragranular acicular ferrite are thought to be associated with variations in transformation rates and the degree of undercooling [8]. In summary, these observed microstructures are typical of those formed under such welding conditions. The topographic image of (Figure 1d) showed a very flat surface where the only distinguishable features were some contamination particles and a few grinding scratches. From this image, it was not possible to distinguish the distribution of the ferritic and austenitic phases over the surface. On the other hand, the magnetic domain distribution presented in Figures 1a, 1b and 1c are thought to be associated with the microstructures, typical of the various DSS weld regions. The MFM technique was capable of clearly imaging the magnetic domain structure of the ferrite phase that surrounds the islands of austenite, appearing flat and uniform due to their paramagnetic properties. Clear bands of ferrite could be easily distinguished, but a closer look revealed other regions of ferrite that did not exhibit the more typical striped magnetic domain configuration, similar to the ferrite regions.

F-2 P

633

Figure 1. MFM Image of DSS; a) Root region, b) Fill region, c) Cap region, d) Topographical image of weld.

Intergranular Corrosion Tests Modified ASTM A262 Standard Practices E Test The absence of cracks on the surface of the bent specimens, even under restricted and reduced bending radius, in accordance with ASTM A262 Standard Practices E, indicates no evidence of sensitization in all weld conditions.

F-2 P

634

Modified DL-ERP test The test efficiency was measured by means of a response test, which was characterized by weak values of the current density ratio (Ir/Ia <1%) and the charge ratio (Qr/Qa < 1%) for nonsensitized materials, and relatively high ratio values (Ir/Ia 1%) and (Qr/Qa 1%) for high-sensitized materials. The reverse polarization from the passive to the active region gave rise to a reactivation peak, the magnitude of which is sensitive to the degree of alloy element depletion. The susceptibility to corrosion was characterized in terms of both the ratio of the reactivation-current peak to the activation current peak as well as the ratio of the reactivation charge to the activation charge [9]. Analysis of the results shows that the fill region had a higher degree of sensitization (DOS) compared to the root and cap region of the weld. This correlates with the neutron diffraction measurements of the average phase stress, where tensile ferrite and austenite stresses were observed to be at a maximum in the fill region of the weld in the transverse direction.

Conclusions
Microstructure of the weld metal as detailed in Figure 1 shows a typical as weld structure, resulting in the formation of both fine and coarse structure within the weld metal. This was thought to be associated with variations in transformation rates and the degree of undercooling. It was shown the MFM is a powerful tool to use for differentiating the austenite and ferrite phase in duplex stainless steel. The DL-ERP test results revealed that the fill area had the highest values for Ir/Ia and Qr/Qa. Residual stress measurements by neutron diffraction revealed that the ferrite phase stress was tensile in the HAZ and weld and appeared to be balanced by a local compressive austenite phase stresses in the normal and transverse directions. The results showed that for ferrite and austenite, a maximum tensile stress is formed in the fill section of the weld and decreases in the root and cap regions for the transverse direction. A correlation was observed between the stress / strain distribution in the DSS weld regions and the degree of susceptibility to IGC.

References
[1] [2] [3] [4] [5] [6] [7]
F-2 P

ASTM A262, Standard Practices for Detecting Susceptibility to Intergranular Attack in Austenitic Stainless Steels. S. Schultze, J. Gollner, K. Eick, P. Veit, I. Garz, The modified EPR test: A new tool for examination of corrosion susceptibility of duplex stainless steel, Duplex 97, Paper D97-067, KCI Publishing, Zutphen, The Netherlands (1997), p. 639. A. Turnbull, P.E. Francis, A.J. Griffiths, E. Bennett, W. Nimmo, Measurement of Corrosion Resistance of Super-Duplex Stainless Steel Welds by Electrochemical Techniques, Eurocorr 2000 (London, U.K.: Institute of Materials, 2000). E. Otero, C. Merino, C. Fosca, P. Fernandez, Electrochemical Characterization of Secondary Phases in a Duplex Stainless Steel by EPR Test, Duplex 94, paper no. 56 (Cambridge, U.K.: TWI, 1994). B. Gideon, L. Ward, D. G. Carr Strain Measurements by Neutron Diffraction and Characterization of Duplex Stainless Steel Welds. Duplex 2007 Conference, paper 49 (Aquileia and Grado Italy) E. Krner, Berechnung der elastischen Konstanten des Vielkristalls aus den Konstanten des Einkristalls, Z. Physik. 151, 1958, p. 404-418. B. Gideon, L. Ward, G. Biddle, Testing and Characterization of Duplex Stainless Steel Welds and their Susceptibility to Intergranular Corrosion, Eurocorr06, (Maastricht,
635

[8] [9]

Netherlands, September 2006). L. Karlsson, Welding in the World, Vol. 43, no. 5, 1999. T. Amadou, C. Braham, and H. Sidhom, Metallurgical and Materials Transactions A, Vol. 35A, Nov. 2004, p. 3499.

F-2 P

636