Induced magnetic field used to detect the sigma phase of a 2205 duplex stainless steel

! 1! Induced magnetic field used to detect the sigma phase of a 2205 duplex stainless steel Edgard de Macedo Silva1, Josinaldo Pereira Leite2, João Pereira Leite3, Walter Macedo Lins Fialho1, Victor Hugo C. de Albuquerque4, João Manuel R. S. Tavares5 1 Instituto Federal de Educação Ciência e Tecnologia da Paraíba, Av. Primeiro de Maio, 720 - Jaguaribe, João Pessoa - PB, 58015-435, Brazil. Email: [email protected], [email protected] 2 Universidade Federal da Paraíba, Jardim Universitário, s/n - Castelo Branco, João Pessoa - PB, 58051-900, Brazil. Email: [email protected] 3 Universidade Federal de Campina Grande, R. Aprígio Veloso, 882 - Universitário, Campina Grande - PB, 58429-900, Brazil. Email: [email protected] 4 Programa de Pós-Graduação em Informática Aplicada, Universidade de Fortaleza, Fortaleza, Ceará, Brazil. Email: [email protected] ! 2! 5 Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial, Departamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, Porto, Portugal, Rua Dr. Roberto Frias, S/N - 4200-465 Porto, Portugal. Email: [email protected] Corresponding author: Prof. João Manuel R. S. Tavares Faculdade de Engenharia da Universidade do Porto Rua Dr. Roberto Frias, s/n 4200-465 Porto, PORTUGAL email: [email protected]p.pt Phone: +351 22 5081487, Fax: +351 22 5081445, url: www.fe.up.pt/~tavares ! 3! Induced magnetic field used to detect the sigma phase of a 2205 duplex stainless steel ! Abstract Sigma phases are formed due to heat treatments and/or welding processes during the solidification stage, and they are responsible for embrittlement of duplex stainless steels. Only a small amount of this phase promotes unfavorable mechanical properties and liability to corrosion. In this work, a new affordable approach to detect and follow-up the kinetics of the sigma phase transformation is evaluated. The measurements are based on an induced magnetic field generated through the interaction between an external magnetic field and the microstructure under study. To validate this approach, the induced magnetic field values are compared with the values of the Charpy impact energy, and the sigma phase is assessed by optical microscopy. Moreover, surface fractures are analyzed by scanning electron microscopy and X-ray diffraction. The results from the 2205 duplex stainless steel used show that there is a direct relation among the impact energy, fracture mechanism and induced magnetic field. The method proved to be able to follow up the embrittlement of the DSS successfully. Moreover, the results confirm that the presence of a sigma phase can be studied based on ! 4! an induced magnetic field, even when in low amounts, and that a critical threshold value can be defined to monitor structures in service. ! Keywords: Non-destructive testing; Optical microscopy; Charpy test; Scanning electron microscopy; Sigma phase; X-ray diffraction. ! 5! 1 Introduction Duplex stainless steels (DSS) hold important properties of both the facecentered cubic (austenitic) and body-centered cubic (ferritic) phases in their microstructures. The ferrite and austenite phases are present in roughly equal volume fractions [1, 2]. Consequently, DSS have excellent strength and toughness, improved corrosion resistance (especially to localized corrosion) and exceptional resistance to halide stress-corrosion cracking [3, 4]. Due to these excellent properties, DSS are widely used in the marine and petrochemical industries, as well as for desalination services and in paper mills [5-7]. However, when these steels are heated above 600 °C, the formation of the brittle and non-magnetic sigma (σ) phase occurs [8-10]. This phase presents high hardness (around 900 HV) and is a phase rich in chromium. Also the sigma phase, which is usually formed from the ferrite phase, reduces the impact toughness and the resistance to localized corrosion due to the adjacent zones that become depleted of Cr and Mo [11-14]. Small amounts (4%) of the sigma phase are able to compromise the material toughness [6, 9, 15]. Even these low quantities promote a considerable decrease of toughness without any notable influence on the hardness. For instance, the precipitation of 1.3 % of this phase decreased the impact toughness from 320 J (solute treated) to 24 J (aged samples at 800 oC for 10 min) [6, 9, 16]. The sigma ! 6! phase forms after aging at temperatures between 600 and 950 ºC and after cooling from temperatures above 900 ºC, as occurs in the heat-affected zone during welding. Appropriate toughness properties are achieved by applying proper annealing temperatures and cooling rates, which explains why the toughness properties in DSS are in general satisfactory [17, 18]. The ferrite phase is ferromagnetic while the austenite and sigma phase are paramagnetic. Therefore, magnetic-based testing approaches can be used for testing duplex stainless steels, since these tests are sensitive to the presence and amount of the ferromagnetic ferrite phase. Mohapatra et al. [19, 20] showed that the remanence measurement of the magnetic hysteresis loop is able to follow the formation of the σ phase even when presented in low volume fractions. Structure-sensitive magnetic properties like coercivity and remanence are affected by microstructural modifications. Therefore, considerable attention has been given to magnetic techniques to study ferromagnetic materials. Previous works have also shown the applicability of magnetic-based measurements to detect ferritic phase decomposition and the presence of the sigma phase in duplex stainless steels [9, 21- 23]. Magnetic susceptibility tests have been applied on samples with different levels of ferritic phase decomposition, and a decreasing of the magnetic susceptibility was observed as the time of thermal aging increased due to the formation of the sigma ! 7! phase. The non-magnetic sigma phase reduces the magnetic field permeability of the materials and therefore also the magnetic susceptibility of such materials [21, 22]. Other traditional non-destructive tests that have been used to characterize material microstructures are based on eddy current testing (ECT) [23-25], ultrasound [26-30] and magnetic Barkhausen noise [8, 31, 32], which have also been used to detect the sigma phase [8, 23-26]. For example, Normando et al., in [8], studied the sigma phase transformation in samples aged at 800 and 900 oC for times up to 2 hours using the eddy current technique. The authors observed that the impedance decreases as the heat treatment time increases. These authors noted a sharp decreasing in impedance in the first 15 min, which was not associated to sigma phase precipitation. Instead, the authors associated this behavior to some second austenite phase precipitation and also to a very sharp softening that was detected in the same time interval. A considerable decrease of impedance observed after 15 min is said to be due to the paramagnetic sigma phase precipitation. Camerini et al., in [24], evaluated conventional electromagnetic (ECT) and saturated low frequency eddy current (SLOFEC) techniques to characterize super duplex stainless steel samples. SLOFEC differs from ECT because it uses an external DC magnetic field that reaches the magnetic saturation of the sample under study. The depth penetration of the eddy ! 8! current field lines is increased once the local relative permeability is decreased by the DC magnetization. This is attractive as it means that larger volumes of materials can be analyzed due to the greater penetration of the eddy current. Both techniques were able to evaluate the presence of the sigma phase and to estimate the δ ferrite content. Other works discussed the advantages of using the eddy current technique for the characterization of different materials [33-35]; however, neither of these latter two studies presented the potential to characterize the sigma phase when it was in low amounts in SDSS, nor did they correlate the electromagnetic results to the δ content. However, a limitation of these two studies was the low number of samples tested which could jeopardize their results. Ultrasound testing has also been used to detect the sigma phase transformation in duplex stainless steels [8, 26-30, 36]. For example, Normando et al., [8, 26] analyzed the influence of the sound velocity to follow-up the σ phase formation at temperatures of 800 and 900 oC for times up to 2 h. The results showed that ultrasonic velocity increases with longer times of heat treatment. The sound velocity is influenced by the material density and the elastic modulus [37], therefore the changes observed for the sound velocity indicate changes of the material properties due to the generation of the sigma phase from the ferrite phase. The results also showed that the sound velocity measurements are more precise for aging times above ! 9! 30 min [8, 26]. Silva et al. [27, 28] applied ultrasound to microstructure characterization. These authors studied the ferrite decomposition at temperatures of 425 and 475 oC and showed that the changes of the sound speed measurements are directly proportional to the variation of the material hardness, presenting sensitivity to the phase transformations. Thus, the sound speed is an important nondestructive parameter to follow-up the hardening kinetics of duplex stainless steels. In this study, induced magnetic field measurements obtained in the reversibility region of magnetic domains were used to follow the formation of the undesirable sigma phase. Samples of the 2205 duplex stainless steel were aged in order to obtain different amounts of the sigma phase up to 18 %. The aim was to compare this approach with others like the ones based on ultrasound and eddy currents, when applied to regions containing low amounts of the sigma phase of the duplex stainless steel under study. The toughness of these samples had already been reduced due to the presence of the sigma phase. The proposed approach revealed to be promising to follow the decomposition of small particles of ferrite at temperatures below 550 oC for SAF 2205 duplex stainless steels [36] and this approach was used here to evaluate the sigma phase. The findings obtained show that the proposed approach has an interval of error of around 2 x 10-5 tesla considering an error of 5 %. In order to verify the dispersion of the signals acquired due to ! 16! Figure 5 shows the absorbed impact energy in function of the induced field for the samples aged at 800 and 900 oC and times up to 2 hours. As already reported [9, 10], the thermal aging at 800 and 900 oC presented the highest rates of sigma phase precipitation. Figure 5 demonstrates that the induced magnetic field is able to follow the embrittlement process of the analyzed steel. The aged sample at 800 oC for 1 h presented a decrease of around 82 % in terms of toughness. Sigma phases greater than 4 % reduce the impact strength of the steel by 70 % [47]. According to Fargas et al., [47], as the precipitation of the sigma phase increases, cracks occur preferentially in the contours of the particles of this phase that are oriented in the rolling direction. Thermal aging at temperatures of 800 and 900 oC for 0.25 h generate losses in toughness of 31.3 and 54.5 %, respectively. Hence, in order to monitor duplex stainless steel structures, it is important to assess their microstructure even for times inferior to 0.5 h. Figure 6 shows the fractured surfaces obtained by SEM of the samples aged at 800 and 900 °C for 0.25 h. A change in the fracture mechanism from ductile to brittle is apparent in this Figure. For the 800 oC sample the microstructure shows a ductile fracture mode with a large number of deep dimples visible, and with some cracks already perceptible due to the presence of the sigma phase. These cracks are even more noticeable in the 900 oC sample. ! 17! The variations of the induced magnetic field and the impact energy in function of the amount of the sigma phase are presented in Figure 7. Here the absorbed energy decreases rapidly and tends to a plateau as the material becomes brittle. Small amounts of sigma phase promote a significant decrease in toughness: 3 % causes a reduction of 78 % in the absorption of the impact energy. The correlation between this phase and the impact energy has already been reported [9, 48]. Figure 7 also shows that the induced magnetic field follows the formation of the sigma phase and that there is a critical value that corresponds to the brittle condition of the material. This can be a useful tool to monitor structures in service, since it can detect the presence of the sigma phase even when in low amounts. As previously mentioned, other techniques like the ones based on eddy currents and ultrasound are not so precise for short aging times of around 0.25 h at temperatures of 800 and 900 oC [8, 26]. Based on the experimental findings, the measurement of induced magnetic fields can be used to detect, in service, material embrittlement due to the sigma phase. For example, welded joints can be tested by performing analysis before the welding, or in regions distance from the welded regions, and after the welding; and, if the reduction in the induced magnetic field corresponds to a content of 4 % of sigma phase, then it should be interpreted ! 18! as this result is a strong indication that the structure under analysis needs to be repaired. Within the scope of this work, an outlier analysis was also performed on the values obtained for the induced field. Twenty-two external fields were applied for each aging condition until the saturation of the Hall sensor. Outliers were only found in the data acquired for the samples aged at 800 oC for 2 h with an applied external field of 102.01 A/m and at 900 oC for 2 h with an applied external field of 133.26 A/m, as shown in Figure 8. In the first case, the outlier is not considered a new value as it is stills inside the error bar. In the second case, the outliner is outside the error bar, but the difference was not enough to affect the results obtained. These results therefore demonstrate that the proposed approach to test for the sigma phase is accurate and robust. 4 Conclusions Embrittlement in a 2205 duplex stainless steel due to the presence of the sigma phase was studied in this work. The phase was detected based on an induced magnetic field that was generated by the interaction of an external magnetic field and the microstructure under analysis. ! 19! The findings obtained in the various tests executed conclude the following: (1) There is an ideal external magnetic field that should be applied in order to obtain the best amplitude for the phase assessment. In the present work, a value of 211.5 A/m was used as the ideal field. (2) The presence of the paramagnetic sigma phase reduces the induced magnetic field even when the phase is in reduced quantities. Other testing techniques such as the ones based on eddy currents and ultrasound have assessment limits of 0.25 h for temperatures above 800 oC. (3) The induced magnetic field is directly proportional to the absorbed energy. Based on the experimental findings, we can conclude that the induced magnetic field can be used to continually monitor in-service structures that are susceptible to embrittlement. The method can follow the presence of low amounts of the sigma phase, which cannot be easily detected by the common non-destructive testing techniques. However, low sigma phases are enough to promote significant reductions in toughness. (4) The induced magnetic field in the region of reversibility of magnetic domain is directly proportional to the material permeability and this means that the induced magnetic field obtained by the ideal external magnetic field is not affected by changes in the geometry of the samples. (5) Additionally, in comparison to other non-destructive testing techniques, another attractive characteristic of this technique is the ease of use and ! 20! interpretation, besides being more affordable than ultrasound testing, for example. To conclude, the testing approach proposed can predict accurately and efficiently the best moment to carry out maintenance services, which leads to reduced costs and maintenance time. As to future work, we intend to use computational classifiers in order to identify the material phases based on the data collected by the proposed testing approach. Other interesting works could be studies about the flux density within the samples under study considering the thickness of the samples and the applied magnetic fields. Acknowledgments The sponsorship from the Coordination for the Improvement of Higher Education Personnel (CAPES), in Brazil, through the scholarship process BEX 2634 /15-5 at Faculdade de Engenharia da Universidade do Porto (FEUP), in Portugal, and from the Federal Institution of Paraiba (IFPB), in Brazil. Victor Hugo C. de Albuquerque acknowledges the sponsorship from the Brazilian National Council for Research and Development (CNPq) through the Grants 470501/2013-8 and 301928/2014-2. ! 21! References [1] Martins M, Casteletti LC (2009) Microstructural characterization and corrosion behavior of a super duplex stainless steel casting. Mater Charact 60:150-155. [2] Badjia R, Bouabdallah M, Bacroix B, Kahloun C, Bettahar K, Kherrouba N. (2008) Effect of solution treatment temperature on the precipitation kinetic of r-phase in 2205 duplex stainless steel welds. Mater Sci Eng A 496(1-2):447-54. [3] Deng B, Jiang Y, Gong J, Zhong C, Gao J, Li J (2008) Critical pitting and repassivation temperatures for duplex stainless steel in chloride solutions. Electrochim Acta 53(16):5220-5225. [4] Tavares SSM, Terra VF, Neto PL, Matos DE (2005) Corrosion resistance evaluation of the UNS S31803 duplex stainless steels aged at low temperatures (350–550 oC) using DLEPR tests. J Mater Sci 40(15):4025- 4028. [5] Jiang ZHL, Chen XY, Huang H, Liu XY (2003) Grain refinement of Cr25Ni5Mo1.5 duplex stainless steel by heat treatment. Mater Sci Eng A 363(1-2):263-267. ! 22! [6] Chen TH, Weng KL, Yang JR (2002) The effect of high temperature exposure on the microstructural stability and toughness property in a 2205 duplex stainless steel. Mater Sci Eng A 338(1-2):259-270. [7] Pohl M, et al (2007) Effect of intermetallic precipitation on the properties of duplex stainless steel. Mater Character 58:65-71. [8] Normando PG, Moura EP, Souza JA, Tavares SSM, Padovese LR (2010) Ultrasound, eddy current and magnetic Barkhausen noise as tools for sigma phase detection on a UNS S31803 duplex stainless steel. Mater Sci Eng A 527:2886-2891. [9] Tavares SSM, Pardal JM, Guerreiro JL, Gomes AM, Da Silva MR (2010) Magnetic detection of sigma phase in duplex stainless steel UNS S31803. J Magn Magn Mater 322(17):L29-L33. [10] Sieurin H, Sandström R (2007) Sigma phase precipitation in duplex stainless steel 2205. Mater Sci Eng A 444(1-2):271-276. [11] Souza CM, Abreu HFG, Tavares SSM, Rebello JMA (2008) The σ phase formation in annealed UNS S31803 duplex stainless steel: Texture aspects. Mater Charact 59(9):1301-1306. [12] Sathirachinda N, et al (2009) Depletion effects at phase boundaries in 2205 duplex stainless steel characterized with SKPFM and TEM/EDS. Corros Sci 51:1259-1266. ! 23! [13] Moura VS, Lima LD, Pardal JM, Kina AY, Corte RRA, Tavares SSM (2008) Influence of the microstructure on the corrosion resistance of the duplex stainless steel UNS S31803. Mater Charact 59(8):1127-1132. [14] Zou D, Han Y, Zhang W, Fan G (2010) Phase transformation and its effects on mechanical properties and pitting corrosion resistance of 2205 duplex stainless steel. J of Iron and Steel Research Inter 17(11):67s-72s. [15] Wei Z, Laizhu J, Jincheng H, Hongmei S (2009) Effect of ageing on precipitation and impact energy of 2101 economical duplex stainless steel. Mater Charact 6(1);50–55. [16] Nilssom JO, Kangas P, Karlsson T, Wilson A (2000) Mechanical properties, microstructural stability and kinetics of sigma phase formation in a 29Cr–6Ni–2Mo–0.38N super duplex stainless steel. Metall Mater Tran A 31(1):35-45. [17] Muthupandi V, Srinivasan BP, Seshadri SK, Sundaresan S (2003) Effect of Weld Metal Chemistry and Heat Input on the Structure and Properties of Duplex Stainless Steel Welds. Mater Sci Eng A 58:9-16. [18] Pardal JM, Tavares SSM, Fonseca MPC, DA Silva MR, M. Ferreira LR (2012) Study of deleterious phase precipitation under continuous cooling of superduplex stainless steel UNS S32750. Mater Sci Tech 28(3):295-302. ! 24! [19] Mohapatra JN, Kamada Y, Kikuchi H, Kobayashi S, Echigoya J, Park DG, Cheong YM (2011) Evaluation of embrittlement in isochronal aged Fe- Cr alloys by magnetic hysteresis loop technique. J Magn 16(2):173-176. [20] Mohapatra JN, Kamada Y, Murakami T, Echigoya J, Kikuchi H, Kobayashi S (2013) Magnetic hysteresis loop technique as a tool for the evaluation of σ phase embrittlement in Fe–Cr alloys. J Magn Magn Mater 327:71–75. [21] Lo KH, Lai JKL, Shek CH, Li DJ (2007) Magnetic and transformation behaviour of duplex stainless steels under non-isothermal conditions and temperature-fluctuation monitoring. Mater Sci Eng A 452-453:149-160. [22] Lo KH, Lai JKL (2011) Microstructural characterisation and change in a.c. magnetic susceptibility of duplex stainless steel during spinodal decomposition. J Nucl Mater 401(1–3):143-148. [23] Ghanei S, Kashefi M, Mazinani M (2013) Eddy current nondestructive evaluation of dual phase steel. Mater Des 50:491-496. [24] Camerini C, Sacramento R, Areiza MC, Rocha A, Santos R, Rebello JM, Pereira G (2015) Eddy current techniques for super duplex stainless steel characterization. J Magn Magn Mater 388:96-100. [25] Ghanei S, Kashefi M, Mazinani M (2014) Comparative study of eddy current and Barkhousen noise nondestructive testing methods in ! 25! microstructural examination of ferrita-martensite dual-phase steel. J Magn Magn Mater 356:103-110. [26] Nunes TM, de Albuquerque VHC, Papa JP, Silva CC, Normando PG, Moura EP, Tavares JMRS (2013) Automatic microstructural characterization and classification using artificial intelligence techniques on ultrasound signals. Expert Syst Appl 40(8):3096-3105. [27] Albuquerque VHC, Silva EM, Leite JP, Moura EP, de Araújo Freitas VL, Tavares JMRS (2010) Spinodal decomposition mechanism study on the duplex stainless steel UNS S31803 using ultrasonic speed measurements. Mater Des 31(4):2147-2150. [28] Albuquerque VHC, Silva CC, Normando PG, Moura EP, Tavares JMRS (2012) Thermal aging effects on the microstructure of Nb-bearing nickel based superalloy weld overlays using ultrasound techniques. Mater Des 36:337–347. [29] Silva EM, Albuquerque VHC, Leite JP, Varela ACG, Moura EP, Tavares JMRS (2009) Phase transformations evaluation on a UNS S31803 duplex stainless steel based on nondestructive testing. Mater Sci Eng A 516:126-130. [30] Albuquerque VHC, Melo TAA, Gomes FORM, Tavares JMRS (2010) Evaluation of grain refiners influence on the mechanical properties in a ! 32! FIGURES Figure 1 Figure 2a ! 33! Figure 2b Figure 2c ! 34! Figure 2d Figure 3 Figure 4 ! 35! Figure 5a Figure 5b Figure 6a ! 36! Figure 6b Figure 7 Figure 8a ! 37! Figure 8b ! 38! TABLES Table 1 Temperature (oC) Time (h) Number of samples Tests As-received 0 5 CI OM SEM IMFM 700 0.25 1 2 15 OM IMFM 750 0.25 1 2 15 OM IMFM 800 0.25 1 2 15 CI OM SEM IMFM 850 0.25 1 2 15 OM IMFM 900 0.25 1 2 15 CI OM SEM IMFM 1000! 0.25 1 2 15 OM IMFM CI - Charpy Impact MO - Optical Microscopy SEM - Scanning Electron Microscopy IMFM - Induced Magnetic Field Measurement