- Email: [email protected]
Magnetic evaluation of the external surface in cast heat-resistant steel tubes with different aging states
Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
Review Articles
Magnetic evaluation of the external surface in cast heat-resistant steel tubes with different aging states Mónica P. Arenas a,b,⇑, Rosa M. Silveira a, Clara J. Pacheco b, Antonio C. Bruno c, Jefferson F.D.F. Araujo c, Carlos B. Eckstein d, Laudemiro Nogueira d, Luiz H. de Almeida a, João M.A. Rebello a, Gabriela R. Pereira a,b a
Department of Metallurgical and Materials Engineering, COPPE-UFRJ, Rio de Janeiro, RJ 21941-972, Brazil Laboratory of Non-destructive Testing, Corrosion and Welding (LNDC), UFRJ, Rio de Janeiro, RJ 21941-596, Brazil Department of Physics, Pontifícia Universidade Católica do Rio de Janeiro, Rio de Janeiro 22451-900, Brazil d Petrobras, Rio de Janeiro, RJ 21040-000, Brazil b c
a r t i c l e
i n f o
Article history: Received 16 January 2018 Received in revised form 15 February 2018 Accepted 16 February 2018 Available online 21 February 2018 Keywords: HP steel Aging External surface Oxide scale Eddy current testing Magnetic force microscopy
a b s t r a c t Heat-resistant austenitic stainless steels have become the principal alloys for use in steam reformer tubes in the petrochemical industry due to its mechanical properties. These tubes are typically exposed to severe operational conditions leading to microstructural transformations such as the aging phenomenon. The combination of high temperatures and moderate stresses causes creep damages, being necessary to monitor its structural condition by non-destructive techniques. The tube external wall is also subjected to oxidizing atmospheres, favoring the formation of an external surface, composed by an oxide scale and a chromium depleted zone. This external surface is usually not taken into account in the tube evaluation, which can lead to erroneous estimations of the service life of these components. In order to observe the magnetic influence of this layer, two samples, exposed to different operational temperatures, were characterized by non-destructive eddy current testing (ECT), scanning DC-susceptometer and magnetic force microscopy (MFM). It was found that the external surface thickness influences directly in the magnetic response of the samples. Ó 2018 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Eddy current testing (ECT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Scanning DC–susceptometer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. Magnetic force microscopy (MFM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Eddy current testing measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Contrast map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Impedance plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Scanning DC-susceptometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. SEM and MFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
⇑ Corresponding author at: Department of Metallurgical and Materials Engineering, COPPE-UFRJ, Rio de Janeiro, RJ 21941-972, Brazil. E-mail address: [email protected] (M.P. Arenas). https://doi.org/10.1016/j.jmmm.2018.02.051 0304-8853/Ó 2018 Elsevier B.V. All rights reserved.
347 347 347 347 347 348 348 348 348 348 350 351 351 351 351
347
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
1. Introduction Centrifugally cast-heat resistant stainless steels are suitable for work in severe operational conditions while maintaining their high mechanical performance and creep resistance to temperatures up to 1000 °C [1]. Modified HP alloys have been increasingly used in petrochemical industry as radiant tubes in reformer and pyrolysis furnaces [2]. Steam reformer furnaces are used for large-scale production of hydrogen by catalytic reactions [3]. Hydrogen production furnaces tubes are assembled in columns, averaging in length from 10 m to 14 m, outer diameter from 100 mm to 200 mm and thickness between 10 mm and 20 mm [1]. Although their lifetime is projected to 100,000 h [3–5], the remaining life depends on several variables, such as internal pressure, wall temperature, time in service and structural strengths that can originate creep damages, crack growth and corrosion. In normal operational conditions, the HP alloy exhibits microstructural changes according to temperatures, being this process known as aging. Besides the aging phenomenon, a complex oxide scale is usually formed on the tube external wall. The oxide scale composition and thickness is strongly dependent on the oxidizing environments and its kinetics variables [6,7]. Pyrolysis tubes, that present carburization damages in their internal walls, can also display an oxide scale in their external walls, because those walls are in contact with oxidizing environments. Some studies have detected that this oxide scale presents a ferromagnetic behavior [8–11] which could decrease or extinguish the signal obtained from the internal wall. The solution to this problem has been the use of an external DC-magnetizer in order to saturate this effect and promote the correct acquisition of the signal from the carburization effect in the internal wall. It has not been possible, so far, to acquire significant signals related to the aging process of this material in presence of the external surface. In this case, it would be necessary to understand exactly how the presence of this layer affects the signal. The present study aims to assess the effect of the external surface on the magnetic response of the material using eddy current testing (ECT), scanning DC-susceptometer and magnetic force microscopy (MFM) techniques. There is, as of yet, no literature correlating the external surface formation with the magnetic response in samples coming from steam reformer tubes.
VI were analyzed. The samples were taken from the same reformer tube, which was in service during 90,000 h. The eddy current inspection was performed on the tube external wall. On the other hand, the samples for DC-susceptometer and MFM measurements were extracted from the tube cross-section. Table 2 shows the characteristics of the samples.
2.2. Eddy current testing (ECT) The eddy current inspection was performed using the OmniScan Olympus equipment. It was used an absolute probe, composed by a ferrite core in which 1000 turns of cooper wire were wound [15]. The probe frequency was set in 5 kHz, reaching a depth of penetration of approximately 6.65 mm in as-cast samples. Olympus equipment provides the results in the impedance plane. The eddy current signals were obtained by balancing the coil in air and then placing on the test material (liftoff). In order to classify the magnetic response of the samples, a standard block of calibration, with eight different materials, was used to differentiate ferromagnetic and non-ferromagnetic materials [16]. The calibration block was also used for setting up the equipment. The materials classified as ferromagnetic exhibited a signal with phase angle between 90° and 180°. In contrast, materials classified as nonferromagnetic presented a phase angle between 180° and 270°. The eddy current measurements were conducted along the tube external wall. These results were presented in two forms, through the contrast map and impedance plane. For the contrast map, the external tube surface was divided in regions evenly distributed along the perimeter and length, as shown in Fig. 1. Thus, the eddy current measurements were performed in each region acquiring the phase angle, which is calculated by the inverse tangent of the inductive reactance and resistance ratio. After the data acquisition, the external surface was milled off to a depth of 1 mm in an area of approximately 50 mm 50 mm, in order to evaluate the influence of the external surface on the magnetic response. After this procedure, the impedance plane signals were obtained with and without this external surface. The curves were obtained when the probe to sample distance (liftoff) is changed
2.3. Scanning DC–susceptometer 2. Experimental procedure 2.1. Samples The chemical analysis was performed in an as-cast sample using a carbon and sulfur analyzer and the X-ray fluorescence and plasma atomic emission spectrometers. As shown in Table 1, the elements are within the tolerance stablished by the ASTM-A608 standard [12]. However, it is observed the presence of Nb and Ti, indicating that is a modified HP alloy. The addition of these elements improve the mechanical properties when the alloy is exposed to high temperatures [2,13,14]. The steam reformer tubes are subjected to different heating along their length, receiving an aging classification varying from I to VI [4]. This criterion has been adopted by other authors and also used in this work. Two HP-NbTi samples with state of aging III and
The basic principle of the scanning DC-susceptometer consists in applying a uniform DC magnetic field to the sample perpendicularly to the scanning plane, and measuring the induced or remanent magnetic response in the same direction. The susceptometer is equipped with a Hall axial gradiometer that virtually eliminates the applied field, thus the resulting output field (Bz) is the magnetic response from the sample [17]. It has a 200 mm spatial resolution and is able to detect magnetic moments down to 10 10 Am2. Samples from the cross section of the tube, with dimension of ca. 13 mm 10 mm 1.8 mm, were exposed to a 400 mT magnetic field for a few seconds. The field is subsequently turned off and the remanent response was mapped at a 140 lm liftoff over a 10 mm 15 mm area. Later on, a line drawn along the tube thickness (red dashed line), as shown in Fig. 2, is extracted from the complete map and visualized.
Table 1 Chemical composition of an as-cast austenitic HP-NbTi alloy (weigh%). Cr
Ni
C
Mn
Si
P
S
Nb
Ti
25.5
35
0.54
1.3
1.6
0.02
0.006
1.13
0.083
Balance: %Fe.
348
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
Table 2 Operational temperature and dimensions of analyzed samples. Tube
Aging state
Operational temperature (°C)
Tube Thickness (mm)
Length (mm)
Outer diameter (mm)
A B
III VI
700 1000
13
80
110
Fig. 1. Schematic draw of the procedure for the sample characterization. The eddy current inspection was performed on the tube external wall. Metallographic samples were extracted from the tube cross-section for the MFM and the susceptometer analyzes.
Initially, the topography of the sample is measured in intermittent contact mode. Then, the probe is lifted to a constant distance of 100 nm above the sample surface (lift mode), where magnetic forces dominate. At this point, the phase-shift induced by the magnetic force gradient between the probe and the sample is acquired [18]. The MFM measurements were performed in presence of an external magnetic field (in attractive mode with the probe), in order to highlight the regions with ferromagnetic response [19]. Gwyddion software was used for image processing. 3. Results 3.1. Eddy current testing measurements
Fig. 2. Schematic draw of the tube cross-section embedded into a resin sample holder. The remanent magnetic field (BZ) was mapped along the tube thickness, later on a line drawn along the tube thickness (red dashed line), is taken from the complete map. The left side of the sample corresponds to the tube external wall. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2.3.1. Magnetic force microscopy (MFM) The samples were extracted from the tube cross-section with dimension of approximately 10 mm 13 mm 5 mm. In order to improve the surface finishing, the samples were sanded with 1500-grit sandpaper and polished using 1 lm diameter diamond paste. Scanning electron microscopy (SEM) was performed in backscattered electron mode in order to identify the regions that will be measured by MFM. MFM measurements were performed in the external surface. Thus, by combining this information with SEM micrographs, it is possible to measure the external surface thickness with ferromagnetic response. MFM measurements were performed using the nanoIR2-s Anasys Instruments, conducted at room temperature. A cobalt-coated silicon probe (Nanosensors) was used for the measurements. The MFM data is acquired in two steps (double pass).
3.1.1. Contrast map The eddy current testing was conducted on the tube external wall of the samples A and B. As observed in the contrast map, Fig. 3, the horizontal and vertical axes correspond to the perimeter and length of the tube, respectively; the color bar indicates the variation of the phase angle. The tube A exhibited a phase angle variation between 97° and 220°, showing a heterogeneous behavior, which may be related to oxide spallation caused by thermal stresses during service [20]. According to the criterion established on Section 2.2, this sample can be classified in two regions, one with ferromagnetic response (A-R1) and, the other one, with non-ferromagnetic behavior (A-R2). The huge variation is mainly observed along the perimeter, as shown in Fig. 3a. In contrast, the sample B showed a homogeneous response which phase angle was located between 93° and 100°, being classified as ferromagnetic, Fig. 3b. The difference in the magnetic response is mainly related to different operational temperatures and oxidizing environments [10,11,17]. 3.1.2. Impedance plane As described in Section 2.2, the impedance plane was obtained with and without the tube external surface. Initially, measurements performed on the external surface showed that phase angle of the samples A-R1, A-R2 and B was 109° ± 15.3°, 215° ± 12.3° and 95° ± 2.3°, respectively, Fig. 4a, being these results in accordance with the contrast map of Fig. 3. After the external surface has been removed, the bulk exhibited a phase angle of approx. 200°,
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
349
Fig. 3. Contrast maps of the phase angle obtained by eddy current testing. (a) Sample A (aging III) presented a ferromagnetic (A-R1) and non-ferromagnetic region (A-R2). (b) Sample B (aging VI) exhibited a uniform ferromagnetic response.
Fig. 4. Impedance plane conducted on samples A and B. (a) The inspection was performed on the external surface showing a huge variation of the magnetic behavior. (b) After the external surface has been removed, the signals presented a non-ferromagnetic behavior.
Fig. 5. a) Remanent field maps 140 lm over the three samples after applying 400 mT field for a few seconds. (b) Remanent field along a single line over the center of the samples.
350
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
showing a non-ferromagnetic behavior, Fig. 4b. The signal of sample A is slightly different from sample B, indicating that microstructural variations in the bulk also lead in electromagnetic properties changes [21]. However, it is clear that the main variable influencing the eddy current results is the external surface formation [17]. 3.2. Scanning DC-susceptometer One of the characteristics of ferromagnetic materials is their remanent magnetization. In order to check if the external surface at a particular stage of aging is ferromagnetic, a remanent field
map (BZ) was measured perpendicularly to the cross section of the tube, as shown in Fig. 5a. It is possible to observe a remanent field over the external surface in all three samples presenting different intensities. It can be seen that the bulk region and the sample holder, as expected, did not present any remanent field. The maps show that the remanent field is higher, at the tube external wall of sample B, certainly due to the presence of the oxide scale and the chromium carbide depleted zone, as reported in [17]. The samples A-R1 and A-R2 presented also a remanent magnetic field, indicating that even having the same state of aging, the remanence over the external surface can be different along the tube perimeter. Fig. 5b displays measurements made along one line
Fig. 6. SEM micrographs on backscattered electron mode and MFM images performed in the magnetic transition regions of samples: (a) A-R1, (b) A-R2 and (c) B.
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
over the center of the three samples. The external surface of the sample B exhibited a maximum remanent magnetic field of approximately 1.1 mT. Samples A-R1 and A-R2 displayed a maximum remanent field of 0.78 mT and 0.33 mT respectively. These findings reinforce the previous statement that the main variable influencing the eddy current results is the external surface formation [17]. In order to estimate the thickness of the external surface with ferromagnetic response, the MFM was used for build magnetic maps (phase-shift) on that region.
3.3. SEM and MFM As previously described, the external surface is composed by an oxide scale and a chromium depleted zone. It has been reported [10,11] that the oxide scale presents a ferromagnetic response. However, it has not been found literature that relates the ferromagnetic response on the chromium depleted zone. Thus, in order to estimate the external surface thickness with ferromagnetic response, the MFM was used for build magnetic maps (phaseshift) on that region. It was observed in Fig. 5b that a region of small thickness placed between the external surface and the bulk exhibits a lower remanent magnetic field. Therefore, the MFM analysis was focused in this region in order to clarify this behavior. The MFM maps were build over an area of 80 mm 10 mm. Fig. 6a presents the SEM micrograph of sample A-R1, in which is observed three regions: an oxide scale (dark region), a chromium carbide depleted zone (Cr-depleted zone) and the bulk (in which is observed niobium and chromium carbides). The MFM measurements were performed on the chromium carbides depleted zone, outlined by the red-box. The MFM image exhibits well defined magnetic domains in the region close to the tube external wall, Fig. 6a, being also observed a region with darkest contrast, which does not correspond to topographic artifacts since the SEM image does not present carbides on that region. Thus, this contrast is related to a magnetic response. On the other hand, the sample A-R2 exhibits a slight magnetic response, as shown in the MFM image, Fig. 6b. Finally, the sample B exhibited very well-defined magnetic domains, with high intensity in the region close to the external wall, Fig. 6c. The magnetic response is almost spread across the chromium carbides depleted zone. However, the intensity of the magnetic domains decrease gradually as a function of the external wall distance. These results demonstrate that the chromium depleted zone does not exhibit ferromagnetic response in all its extension and it was also observed that the magnetic intensity decrease insofar as it approaches the bulk. Indicating that, in addition to the selective oxidation of the subsurface chromium carbides, the austenitic matrix also underwent a severe chromium depletion in the region adjacent to the oxide scale, causing a more intense magnetic response on that region. The correlation between the chromium depletion and the ferromagnetic response has been previously reported in HP steel samples that presented carburization damages [8,9,22–26]. Table 3 Mean (m) and standard deviation (r) of the external surface thickness. Sample
External surface thickness (mm)
External surface with ferromagnetic response (mm)
A-R1
l r
102.61 8.33
52.42 4.7
A-R2
l r
68.9 4.49
31.34 0.68
B
l r
229.34 12.98
130.35 1.07
351
As depicted in Table 3, the thickness of external surface (oxide scale plus Cr-depleted zone) does not exhibit ferromagnetic response in all its extension. It is observed that samples with state of aging III exhibited an external surface with different thickness, which may be due to the flow rate variation, to the catalyst activity losses or even to the oxide spallation due to the thermal stresses caused during service [20,27]. For instance, sample A-R1 presented a thicker ferromagnetic region (52.42 ± 4.7 lm) in comparison with the sample A-R2 (31.34 ± 0.68 lm), as depicted in Table 3. On the other hand, sample B, with state of aging VI, exhibited the thickest external surface with ferromagnetic behavior, estimated in 130.35 ± 1.07 mm. 4. Conclusions In summary, it was observed that the operational conditions of the steam reformer tubes influence directly on the external surface formation. It was demonstrated, through different and complementary magnetic techniques, that the external surface exerts a strong influence on the magnetic response. By using the DC-susceptometer a remanent field due to the external surface of all samples were measured, thus confirming their ferromagnetic characteristic. The MFM results have demonstrated that samples exposed to higher temperatures and oxidizing environments present a thicker external surface with ferromagnetic response. Thus, it can be concluded that samples with no oxide or even with a slight oxide formation exhibit a non-ferromagnetic behavior, indicating that the external surface thickness is directly proportional to the magnetic response intensity. Therefore, if the non-destructive eddy current inspection of the bulk is desirable, it is crucial to take into account the ferromagnetic characteristic of the external surface for the correct evaluation of the reformer tubes. Acknowledgments The authors would like to acknowledge Petrobras for the samples supply, to the Brazilian research agencies CNPq, CAPES, FAPERJ and FINEP for the financial support and LNNano/CNPEM for cooperation and assistance with MFM experiments. References [1] T.L. da Silveira, I. Le May, Reformer furnaces: materials, damage mechanisms and assessment, Arab. J. Sci. Eng. 31 (2006) 99–119. [2] F.C. Nunes, L.H. de Almeida, J. Dille, J.L. Delplancke, I. Le May, Microstructural changes caused by yttrium addition to NbTi-modified centrifugally cast HPtype stainless steels, Mater. Charact. 58 (2007) 132–142. [3] A. Alvino, D. Lega, F. Giacobbe, V. Mazzocchi, A. Rinaldi, Damage characterization in two reformer heater tubes after nearly 10 years of service at different operative and maintenance conditions, Eng. Fail. Anal. 17 (2010) 1526–1541. [4] I. Le May, T.L. da Silveira, C.H. Vianna, Criteria for the evaluation of damage and remaining life in reformer furnace tubes, Int. J. Press. Vessel. Pip. 66 (1996) 233–241. [5] J. Swaminathan, K. Guguloth, M. Gunjan, P. Roy, R. Ghosh, Failure analysis and remaining life assessment of service exposed primary reformer heater tubes, Eng. Fail. Anal. 15 (2008) 311–331, https://doi.org/10.1016/j. engfailanal.2007.02.004. [6] N. Xu, D. Monceau, D. Young, J. Furtado, High temperature corrosion of cast heat resisting steels in CO+CO2 gas mixtures, Corros. Sci. 50 (2008) 2398– 2406, https://doi.org/10.1016/j.corsci.2008.06.001. [7] A. Hayashi, N. Hiraide, Y. Inoue, Spallation behavior of oxide scale on stainless steels, Oxid. Met. 85 (2016) 87–101, https://doi.org/10.1007/s11085-0159582-z. [8] K.J. Stevens, W.J. Trompetter, Calibration of eddy current carburization measurements in ethylene production tubes using ion beam analysis, J. Phys. D. Appl. Phys. 37 (2004) 501–509. [9] K. Stevens, A. Tack, C. Thomas, D. Stewart, Through-wall carburization detection in ethylene pyrolysis tubes, J. Phys. D. Appl. Phys. 34 (2001) 814– 822.
352
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
[10] K. Hasegawa, T. Oikawa, N. Kasai, Development of an eddy current inspection technique with surface magnetization to evaluate the carburization thickness of ethylene pyrolysis furnace tubes, J. Nondestruct. Eval. 31 (2012) 349–356. [11] N. Kasai, S. Ogawa, T. Oikawa, K. Sekine, K. Hasegawa, Detection of carburization in ethylene pyrolysis furnace tubes by a C core probe with magnetization, J. Nondestruct. Eval. 29 (2010) 175–180. [12] ASTM-A608/A608M-14, Standard Specification for Centrifugally Cast IronChromium-Nickel High-Alloy Tubing for Pressure Application at High Temperatures, 2014. doi:10.1520/A0608. [13] G.D. Barbabela, L.H. de Almeida, T.L. da Silveira, I. Le May, Role of Nb in modifying the microstructure of heat-resistant cast HP steel, Mater. Charact. 26 (1991) 193–197. [14] F.C. Nunes, J. Dille, J.L. Delplancke, L.H. De Almeida, Yttrium addition to heatresistant cast stainless steel, Scr. Mater. 54 (2006) 1553–1556, https://doi.org/ 10.1016/j.scriptamat.2006.01.024. [15] J.M.A. Rebello, R. Sacramento, M.C.L. Areiza, K.S.D.E. Assis, Quantification of sigma phase precipitation by magnetic non destructive testing, Stud. Appl. Electromagn. Mech. 35 (2011) 311–321, https://doi.org/10.3233/978-160750-750-5-311. [16] J. García-Martín, J. Gómez-Gil, E. Vázquez-Sánchez, Non-destructive techniques based on eddy current testing, Sensors 11 (2011) 2525–2565. [17] J.M.B. Pereira, C.J. Pacheco, M.P. Arenas, J.F.D.F. Araujo, G.R. Pereira, A.C. Bruno, Novel scanning dc-susceptometer for characterization of heat-resistant steels with different states of aging, J. Magn. Magn. Mater. 442 (2017) 311–318, doi: 10.1016/j.jmmm.2017.07.004. [18] Y. Martin, H.K. Wickramasinghe, Magnetic imaging by ‘‘force microscopy” with 1000, A resolution, Appl. Phys. Lett. 50 (1987) 1455–1457.
[19] M.P. Arenas, E.M. Lanzoni, C.J. Pacheco, C.A.R. Costa, C.B. Eckstein, L.H. De Almeida, J.M.A. Rebello, C.F. Deneke, G.R. Pereira, Separating the influence of electric charges in magnetic force microscopy images of inhomogeneous metal samples, J. Magn. Magn. Mater. 446 (2018) 239–244, https://doi.org/10.1016/j. jmmm.2017.09.041. [20] S. Baleix, G. Bernhart, P. Lours, Oxidation and oxide spallation of heat resistant cast steels for superplastic forming dies, Mater. Sci. Eng. A. 327 (2002) 155– 166, https://doi.org/10.1016/S0921-5093(01)01529-5. [21] D.M. Firth, C.W. Thomas, M. Thakur, R. Caporn, Reformer component management after an overheating incident that resulted in tube failures, in: Conf. 58th AIChE, Germany, 2013: pp. 187–197. [22] K.J. Stevens, A. Parbhu, J. Soltis, D. Stewart, Magnetic force microscopy of a carburized ethylene pyrolysis tube, J. Phys. D. Appl. Phys. 36 (2003) 164–168. [23] A.C. McLeod, C.M. Bishop, K.J. Stevens, M.V. Kral, Microstructural characterization and image analysis in Ex-Service HP alloy stainless steel tubes for ethylene pyrolysis, Metallogr. Microstruct. Anal. 5 (2016) 178–187, https://doi.org/10.1007/s13632-016-0274-0. [24] K.J. Stevens, A. Parbhu, J. Soltis, Magnetic force microscopy and cross-sectional transmission electron microscopy of carburised surfaces, Curr. Appl. Phys. 4 (2004) 304–307, https://doi.org/10.1016/j.cap.2003.11.034. [25] I.C. da Silva, R.S. da Silva, J.M.A. Rebello, A.C. Bruno, T. Silveira, Characterization of carburization of HP steels by non destructive magnetic testing, NDT E Int. 39 (2006) 569–577, https://doi.org/10.1016/j.ndteint.2006.03.004. [26] I.C. Silva, J.M.A. Rebello, A.C. Bruno, P.J. Jacques, B. Nysten, J. Dille, Structural and magnetic characterization of a carburized cast austenitic steel, Scr. Mater. 59 (2008) 1010–1013. [27] H.E. Evans, Cracking and spalling of protective oxide layers, Mater. Sci. Eng. A. 120–121 (1989) 139–146, https://doi.org/10.1016/0921-5093(89)90731-4.
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
Review Articles
Magnetic evaluation of the external surface in cast heat-resistant steel tubes with different aging states Mónica P. Arenas a,b,⇑, Rosa M. Silveira a, Clara J. Pacheco b, Antonio C. Bruno c, Jefferson F.D.F. Araujo c, Carlos B. Eckstein d, Laudemiro Nogueira d, Luiz H. de Almeida a, João M.A. Rebello a, Gabriela R. Pereira a,b a
Department of Metallurgical and Materials Engineering, COPPE-UFRJ, Rio de Janeiro, RJ 21941-972, Brazil Laboratory of Non-destructive Testing, Corrosion and Welding (LNDC), UFRJ, Rio de Janeiro, RJ 21941-596, Brazil Department of Physics, Pontifícia Universidade Católica do Rio de Janeiro, Rio de Janeiro 22451-900, Brazil d Petrobras, Rio de Janeiro, RJ 21040-000, Brazil b c
a r t i c l e
i n f o
Article history: Received 16 January 2018 Received in revised form 15 February 2018 Accepted 16 February 2018 Available online 21 February 2018 Keywords: HP steel Aging External surface Oxide scale Eddy current testing Magnetic force microscopy
a b s t r a c t Heat-resistant austenitic stainless steels have become the principal alloys for use in steam reformer tubes in the petrochemical industry due to its mechanical properties. These tubes are typically exposed to severe operational conditions leading to microstructural transformations such as the aging phenomenon. The combination of high temperatures and moderate stresses causes creep damages, being necessary to monitor its structural condition by non-destructive techniques. The tube external wall is also subjected to oxidizing atmospheres, favoring the formation of an external surface, composed by an oxide scale and a chromium depleted zone. This external surface is usually not taken into account in the tube evaluation, which can lead to erroneous estimations of the service life of these components. In order to observe the magnetic influence of this layer, two samples, exposed to different operational temperatures, were characterized by non-destructive eddy current testing (ECT), scanning DC-susceptometer and magnetic force microscopy (MFM). It was found that the external surface thickness influences directly in the magnetic response of the samples. Ó 2018 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Eddy current testing (ECT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Scanning DC–susceptometer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. Magnetic force microscopy (MFM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Eddy current testing measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Contrast map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Impedance plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Scanning DC-susceptometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. SEM and MFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
⇑ Corresponding author at: Department of Metallurgical and Materials Engineering, COPPE-UFRJ, Rio de Janeiro, RJ 21941-972, Brazil. E-mail address: [email protected] (M.P. Arenas). https://doi.org/10.1016/j.jmmm.2018.02.051 0304-8853/Ó 2018 Elsevier B.V. All rights reserved.
347 347 347 347 347 348 348 348 348 348 350 351 351 351 351
347
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
1. Introduction Centrifugally cast-heat resistant stainless steels are suitable for work in severe operational conditions while maintaining their high mechanical performance and creep resistance to temperatures up to 1000 °C [1]. Modified HP alloys have been increasingly used in petrochemical industry as radiant tubes in reformer and pyrolysis furnaces [2]. Steam reformer furnaces are used for large-scale production of hydrogen by catalytic reactions [3]. Hydrogen production furnaces tubes are assembled in columns, averaging in length from 10 m to 14 m, outer diameter from 100 mm to 200 mm and thickness between 10 mm and 20 mm [1]. Although their lifetime is projected to 100,000 h [3–5], the remaining life depends on several variables, such as internal pressure, wall temperature, time in service and structural strengths that can originate creep damages, crack growth and corrosion. In normal operational conditions, the HP alloy exhibits microstructural changes according to temperatures, being this process known as aging. Besides the aging phenomenon, a complex oxide scale is usually formed on the tube external wall. The oxide scale composition and thickness is strongly dependent on the oxidizing environments and its kinetics variables [6,7]. Pyrolysis tubes, that present carburization damages in their internal walls, can also display an oxide scale in their external walls, because those walls are in contact with oxidizing environments. Some studies have detected that this oxide scale presents a ferromagnetic behavior [8–11] which could decrease or extinguish the signal obtained from the internal wall. The solution to this problem has been the use of an external DC-magnetizer in order to saturate this effect and promote the correct acquisition of the signal from the carburization effect in the internal wall. It has not been possible, so far, to acquire significant signals related to the aging process of this material in presence of the external surface. In this case, it would be necessary to understand exactly how the presence of this layer affects the signal. The present study aims to assess the effect of the external surface on the magnetic response of the material using eddy current testing (ECT), scanning DC-susceptometer and magnetic force microscopy (MFM) techniques. There is, as of yet, no literature correlating the external surface formation with the magnetic response in samples coming from steam reformer tubes.
VI were analyzed. The samples were taken from the same reformer tube, which was in service during 90,000 h. The eddy current inspection was performed on the tube external wall. On the other hand, the samples for DC-susceptometer and MFM measurements were extracted from the tube cross-section. Table 2 shows the characteristics of the samples.
2.2. Eddy current testing (ECT) The eddy current inspection was performed using the OmniScan Olympus equipment. It was used an absolute probe, composed by a ferrite core in which 1000 turns of cooper wire were wound [15]. The probe frequency was set in 5 kHz, reaching a depth of penetration of approximately 6.65 mm in as-cast samples. Olympus equipment provides the results in the impedance plane. The eddy current signals were obtained by balancing the coil in air and then placing on the test material (liftoff). In order to classify the magnetic response of the samples, a standard block of calibration, with eight different materials, was used to differentiate ferromagnetic and non-ferromagnetic materials [16]. The calibration block was also used for setting up the equipment. The materials classified as ferromagnetic exhibited a signal with phase angle between 90° and 180°. In contrast, materials classified as nonferromagnetic presented a phase angle between 180° and 270°. The eddy current measurements were conducted along the tube external wall. These results were presented in two forms, through the contrast map and impedance plane. For the contrast map, the external tube surface was divided in regions evenly distributed along the perimeter and length, as shown in Fig. 1. Thus, the eddy current measurements were performed in each region acquiring the phase angle, which is calculated by the inverse tangent of the inductive reactance and resistance ratio. After the data acquisition, the external surface was milled off to a depth of 1 mm in an area of approximately 50 mm 50 mm, in order to evaluate the influence of the external surface on the magnetic response. After this procedure, the impedance plane signals were obtained with and without this external surface. The curves were obtained when the probe to sample distance (liftoff) is changed
2.3. Scanning DC–susceptometer 2. Experimental procedure 2.1. Samples The chemical analysis was performed in an as-cast sample using a carbon and sulfur analyzer and the X-ray fluorescence and plasma atomic emission spectrometers. As shown in Table 1, the elements are within the tolerance stablished by the ASTM-A608 standard [12]. However, it is observed the presence of Nb and Ti, indicating that is a modified HP alloy. The addition of these elements improve the mechanical properties when the alloy is exposed to high temperatures [2,13,14]. The steam reformer tubes are subjected to different heating along their length, receiving an aging classification varying from I to VI [4]. This criterion has been adopted by other authors and also used in this work. Two HP-NbTi samples with state of aging III and
The basic principle of the scanning DC-susceptometer consists in applying a uniform DC magnetic field to the sample perpendicularly to the scanning plane, and measuring the induced or remanent magnetic response in the same direction. The susceptometer is equipped with a Hall axial gradiometer that virtually eliminates the applied field, thus the resulting output field (Bz) is the magnetic response from the sample [17]. It has a 200 mm spatial resolution and is able to detect magnetic moments down to 10 10 Am2. Samples from the cross section of the tube, with dimension of ca. 13 mm 10 mm 1.8 mm, were exposed to a 400 mT magnetic field for a few seconds. The field is subsequently turned off and the remanent response was mapped at a 140 lm liftoff over a 10 mm 15 mm area. Later on, a line drawn along the tube thickness (red dashed line), as shown in Fig. 2, is extracted from the complete map and visualized.
Table 1 Chemical composition of an as-cast austenitic HP-NbTi alloy (weigh%). Cr
Ni
C
Mn
Si
P
S
Nb
Ti
25.5
35
0.54
1.3
1.6
0.02
0.006
1.13
0.083
Balance: %Fe.
348
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
Table 2 Operational temperature and dimensions of analyzed samples. Tube
Aging state
Operational temperature (°C)
Tube Thickness (mm)
Length (mm)
Outer diameter (mm)
A B
III VI
700 1000
13
80
110
Fig. 1. Schematic draw of the procedure for the sample characterization. The eddy current inspection was performed on the tube external wall. Metallographic samples were extracted from the tube cross-section for the MFM and the susceptometer analyzes.
Initially, the topography of the sample is measured in intermittent contact mode. Then, the probe is lifted to a constant distance of 100 nm above the sample surface (lift mode), where magnetic forces dominate. At this point, the phase-shift induced by the magnetic force gradient between the probe and the sample is acquired [18]. The MFM measurements were performed in presence of an external magnetic field (in attractive mode with the probe), in order to highlight the regions with ferromagnetic response [19]. Gwyddion software was used for image processing. 3. Results 3.1. Eddy current testing measurements
Fig. 2. Schematic draw of the tube cross-section embedded into a resin sample holder. The remanent magnetic field (BZ) was mapped along the tube thickness, later on a line drawn along the tube thickness (red dashed line), is taken from the complete map. The left side of the sample corresponds to the tube external wall. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2.3.1. Magnetic force microscopy (MFM) The samples were extracted from the tube cross-section with dimension of approximately 10 mm 13 mm 5 mm. In order to improve the surface finishing, the samples were sanded with 1500-grit sandpaper and polished using 1 lm diameter diamond paste. Scanning electron microscopy (SEM) was performed in backscattered electron mode in order to identify the regions that will be measured by MFM. MFM measurements were performed in the external surface. Thus, by combining this information with SEM micrographs, it is possible to measure the external surface thickness with ferromagnetic response. MFM measurements were performed using the nanoIR2-s Anasys Instruments, conducted at room temperature. A cobalt-coated silicon probe (Nanosensors) was used for the measurements. The MFM data is acquired in two steps (double pass).
3.1.1. Contrast map The eddy current testing was conducted on the tube external wall of the samples A and B. As observed in the contrast map, Fig. 3, the horizontal and vertical axes correspond to the perimeter and length of the tube, respectively; the color bar indicates the variation of the phase angle. The tube A exhibited a phase angle variation between 97° and 220°, showing a heterogeneous behavior, which may be related to oxide spallation caused by thermal stresses during service [20]. According to the criterion established on Section 2.2, this sample can be classified in two regions, one with ferromagnetic response (A-R1) and, the other one, with non-ferromagnetic behavior (A-R2). The huge variation is mainly observed along the perimeter, as shown in Fig. 3a. In contrast, the sample B showed a homogeneous response which phase angle was located between 93° and 100°, being classified as ferromagnetic, Fig. 3b. The difference in the magnetic response is mainly related to different operational temperatures and oxidizing environments [10,11,17]. 3.1.2. Impedance plane As described in Section 2.2, the impedance plane was obtained with and without the tube external surface. Initially, measurements performed on the external surface showed that phase angle of the samples A-R1, A-R2 and B was 109° ± 15.3°, 215° ± 12.3° and 95° ± 2.3°, respectively, Fig. 4a, being these results in accordance with the contrast map of Fig. 3. After the external surface has been removed, the bulk exhibited a phase angle of approx. 200°,
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
349
Fig. 3. Contrast maps of the phase angle obtained by eddy current testing. (a) Sample A (aging III) presented a ferromagnetic (A-R1) and non-ferromagnetic region (A-R2). (b) Sample B (aging VI) exhibited a uniform ferromagnetic response.
Fig. 4. Impedance plane conducted on samples A and B. (a) The inspection was performed on the external surface showing a huge variation of the magnetic behavior. (b) After the external surface has been removed, the signals presented a non-ferromagnetic behavior.
Fig. 5. a) Remanent field maps 140 lm over the three samples after applying 400 mT field for a few seconds. (b) Remanent field along a single line over the center of the samples.
350
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
showing a non-ferromagnetic behavior, Fig. 4b. The signal of sample A is slightly different from sample B, indicating that microstructural variations in the bulk also lead in electromagnetic properties changes [21]. However, it is clear that the main variable influencing the eddy current results is the external surface formation [17]. 3.2. Scanning DC-susceptometer One of the characteristics of ferromagnetic materials is their remanent magnetization. In order to check if the external surface at a particular stage of aging is ferromagnetic, a remanent field
map (BZ) was measured perpendicularly to the cross section of the tube, as shown in Fig. 5a. It is possible to observe a remanent field over the external surface in all three samples presenting different intensities. It can be seen that the bulk region and the sample holder, as expected, did not present any remanent field. The maps show that the remanent field is higher, at the tube external wall of sample B, certainly due to the presence of the oxide scale and the chromium carbide depleted zone, as reported in [17]. The samples A-R1 and A-R2 presented also a remanent magnetic field, indicating that even having the same state of aging, the remanence over the external surface can be different along the tube perimeter. Fig. 5b displays measurements made along one line
Fig. 6. SEM micrographs on backscattered electron mode and MFM images performed in the magnetic transition regions of samples: (a) A-R1, (b) A-R2 and (c) B.
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
over the center of the three samples. The external surface of the sample B exhibited a maximum remanent magnetic field of approximately 1.1 mT. Samples A-R1 and A-R2 displayed a maximum remanent field of 0.78 mT and 0.33 mT respectively. These findings reinforce the previous statement that the main variable influencing the eddy current results is the external surface formation [17]. In order to estimate the thickness of the external surface with ferromagnetic response, the MFM was used for build magnetic maps (phase-shift) on that region.
3.3. SEM and MFM As previously described, the external surface is composed by an oxide scale and a chromium depleted zone. It has been reported [10,11] that the oxide scale presents a ferromagnetic response. However, it has not been found literature that relates the ferromagnetic response on the chromium depleted zone. Thus, in order to estimate the external surface thickness with ferromagnetic response, the MFM was used for build magnetic maps (phaseshift) on that region. It was observed in Fig. 5b that a region of small thickness placed between the external surface and the bulk exhibits a lower remanent magnetic field. Therefore, the MFM analysis was focused in this region in order to clarify this behavior. The MFM maps were build over an area of 80 mm 10 mm. Fig. 6a presents the SEM micrograph of sample A-R1, in which is observed three regions: an oxide scale (dark region), a chromium carbide depleted zone (Cr-depleted zone) and the bulk (in which is observed niobium and chromium carbides). The MFM measurements were performed on the chromium carbides depleted zone, outlined by the red-box. The MFM image exhibits well defined magnetic domains in the region close to the tube external wall, Fig. 6a, being also observed a region with darkest contrast, which does not correspond to topographic artifacts since the SEM image does not present carbides on that region. Thus, this contrast is related to a magnetic response. On the other hand, the sample A-R2 exhibits a slight magnetic response, as shown in the MFM image, Fig. 6b. Finally, the sample B exhibited very well-defined magnetic domains, with high intensity in the region close to the external wall, Fig. 6c. The magnetic response is almost spread across the chromium carbides depleted zone. However, the intensity of the magnetic domains decrease gradually as a function of the external wall distance. These results demonstrate that the chromium depleted zone does not exhibit ferromagnetic response in all its extension and it was also observed that the magnetic intensity decrease insofar as it approaches the bulk. Indicating that, in addition to the selective oxidation of the subsurface chromium carbides, the austenitic matrix also underwent a severe chromium depletion in the region adjacent to the oxide scale, causing a more intense magnetic response on that region. The correlation between the chromium depletion and the ferromagnetic response has been previously reported in HP steel samples that presented carburization damages [8,9,22–26]. Table 3 Mean (m) and standard deviation (r) of the external surface thickness. Sample
External surface thickness (mm)
External surface with ferromagnetic response (mm)
A-R1
l r
102.61 8.33
52.42 4.7
A-R2
l r
68.9 4.49
31.34 0.68
B
l r
229.34 12.98
130.35 1.07
351
As depicted in Table 3, the thickness of external surface (oxide scale plus Cr-depleted zone) does not exhibit ferromagnetic response in all its extension. It is observed that samples with state of aging III exhibited an external surface with different thickness, which may be due to the flow rate variation, to the catalyst activity losses or even to the oxide spallation due to the thermal stresses caused during service [20,27]. For instance, sample A-R1 presented a thicker ferromagnetic region (52.42 ± 4.7 lm) in comparison with the sample A-R2 (31.34 ± 0.68 lm), as depicted in Table 3. On the other hand, sample B, with state of aging VI, exhibited the thickest external surface with ferromagnetic behavior, estimated in 130.35 ± 1.07 mm. 4. Conclusions In summary, it was observed that the operational conditions of the steam reformer tubes influence directly on the external surface formation. It was demonstrated, through different and complementary magnetic techniques, that the external surface exerts a strong influence on the magnetic response. By using the DC-susceptometer a remanent field due to the external surface of all samples were measured, thus confirming their ferromagnetic characteristic. The MFM results have demonstrated that samples exposed to higher temperatures and oxidizing environments present a thicker external surface with ferromagnetic response. Thus, it can be concluded that samples with no oxide or even with a slight oxide formation exhibit a non-ferromagnetic behavior, indicating that the external surface thickness is directly proportional to the magnetic response intensity. Therefore, if the non-destructive eddy current inspection of the bulk is desirable, it is crucial to take into account the ferromagnetic characteristic of the external surface for the correct evaluation of the reformer tubes. Acknowledgments The authors would like to acknowledge Petrobras for the samples supply, to the Brazilian research agencies CNPq, CAPES, FAPERJ and FINEP for the financial support and LNNano/CNPEM for cooperation and assistance with MFM experiments. References [1] T.L. da Silveira, I. Le May, Reformer furnaces: materials, damage mechanisms and assessment, Arab. J. Sci. Eng. 31 (2006) 99–119. [2] F.C. Nunes, L.H. de Almeida, J. Dille, J.L. Delplancke, I. Le May, Microstructural changes caused by yttrium addition to NbTi-modified centrifugally cast HPtype stainless steels, Mater. Charact. 58 (2007) 132–142. [3] A. Alvino, D. Lega, F. Giacobbe, V. Mazzocchi, A. Rinaldi, Damage characterization in two reformer heater tubes after nearly 10 years of service at different operative and maintenance conditions, Eng. Fail. Anal. 17 (2010) 1526–1541. [4] I. Le May, T.L. da Silveira, C.H. Vianna, Criteria for the evaluation of damage and remaining life in reformer furnace tubes, Int. J. Press. Vessel. Pip. 66 (1996) 233–241. [5] J. Swaminathan, K. Guguloth, M. Gunjan, P. Roy, R. Ghosh, Failure analysis and remaining life assessment of service exposed primary reformer heater tubes, Eng. Fail. Anal. 15 (2008) 311–331, https://doi.org/10.1016/j. engfailanal.2007.02.004. [6] N. Xu, D. Monceau, D. Young, J. Furtado, High temperature corrosion of cast heat resisting steels in CO+CO2 gas mixtures, Corros. Sci. 50 (2008) 2398– 2406, https://doi.org/10.1016/j.corsci.2008.06.001. [7] A. Hayashi, N. Hiraide, Y. Inoue, Spallation behavior of oxide scale on stainless steels, Oxid. Met. 85 (2016) 87–101, https://doi.org/10.1007/s11085-0159582-z. [8] K.J. Stevens, W.J. Trompetter, Calibration of eddy current carburization measurements in ethylene production tubes using ion beam analysis, J. Phys. D. Appl. Phys. 37 (2004) 501–509. [9] K. Stevens, A. Tack, C. Thomas, D. Stewart, Through-wall carburization detection in ethylene pyrolysis tubes, J. Phys. D. Appl. Phys. 34 (2001) 814– 822.
352
M.P. Arenas et al. / Journal of Magnetism and Magnetic Materials 456 (2018) 346–352
[10] K. Hasegawa, T. Oikawa, N. Kasai, Development of an eddy current inspection technique with surface magnetization to evaluate the carburization thickness of ethylene pyrolysis furnace tubes, J. Nondestruct. Eval. 31 (2012) 349–356. [11] N. Kasai, S. Ogawa, T. Oikawa, K. Sekine, K. Hasegawa, Detection of carburization in ethylene pyrolysis furnace tubes by a C core probe with magnetization, J. Nondestruct. Eval. 29 (2010) 175–180. [12] ASTM-A608/A608M-14, Standard Specification for Centrifugally Cast IronChromium-Nickel High-Alloy Tubing for Pressure Application at High Temperatures, 2014. doi:10.1520/A0608. [13] G.D. Barbabela, L.H. de Almeida, T.L. da Silveira, I. Le May, Role of Nb in modifying the microstructure of heat-resistant cast HP steel, Mater. Charact. 26 (1991) 193–197. [14] F.C. Nunes, J. Dille, J.L. Delplancke, L.H. De Almeida, Yttrium addition to heatresistant cast stainless steel, Scr. Mater. 54 (2006) 1553–1556, https://doi.org/ 10.1016/j.scriptamat.2006.01.024. [15] J.M.A. Rebello, R. Sacramento, M.C.L. Areiza, K.S.D.E. Assis, Quantification of sigma phase precipitation by magnetic non destructive testing, Stud. Appl. Electromagn. Mech. 35 (2011) 311–321, https://doi.org/10.3233/978-160750-750-5-311. [16] J. García-Martín, J. Gómez-Gil, E. Vázquez-Sánchez, Non-destructive techniques based on eddy current testing, Sensors 11 (2011) 2525–2565. [17] J.M.B. Pereira, C.J. Pacheco, M.P. Arenas, J.F.D.F. Araujo, G.R. Pereira, A.C. Bruno, Novel scanning dc-susceptometer for characterization of heat-resistant steels with different states of aging, J. Magn. Magn. Mater. 442 (2017) 311–318, doi: 10.1016/j.jmmm.2017.07.004. [18] Y. Martin, H.K. Wickramasinghe, Magnetic imaging by ‘‘force microscopy” with 1000, A resolution, Appl. Phys. Lett. 50 (1987) 1455–1457.
[19] M.P. Arenas, E.M. Lanzoni, C.J. Pacheco, C.A.R. Costa, C.B. Eckstein, L.H. De Almeida, J.M.A. Rebello, C.F. Deneke, G.R. Pereira, Separating the influence of electric charges in magnetic force microscopy images of inhomogeneous metal samples, J. Magn. Magn. Mater. 446 (2018) 239–244, https://doi.org/10.1016/j. jmmm.2017.09.041. [20] S. Baleix, G. Bernhart, P. Lours, Oxidation and oxide spallation of heat resistant cast steels for superplastic forming dies, Mater. Sci. Eng. A. 327 (2002) 155– 166, https://doi.org/10.1016/S0921-5093(01)01529-5. [21] D.M. Firth, C.W. Thomas, M. Thakur, R. Caporn, Reformer component management after an overheating incident that resulted in tube failures, in: Conf. 58th AIChE, Germany, 2013: pp. 187–197. [22] K.J. Stevens, A. Parbhu, J. Soltis, D. Stewart, Magnetic force microscopy of a carburized ethylene pyrolysis tube, J. Phys. D. Appl. Phys. 36 (2003) 164–168. [23] A.C. McLeod, C.M. Bishop, K.J. Stevens, M.V. Kral, Microstructural characterization and image analysis in Ex-Service HP alloy stainless steel tubes for ethylene pyrolysis, Metallogr. Microstruct. Anal. 5 (2016) 178–187, https://doi.org/10.1007/s13632-016-0274-0. [24] K.J. Stevens, A. Parbhu, J. Soltis, Magnetic force microscopy and cross-sectional transmission electron microscopy of carburised surfaces, Curr. Appl. Phys. 4 (2004) 304–307, https://doi.org/10.1016/j.cap.2003.11.034. [25] I.C. da Silva, R.S. da Silva, J.M.A. Rebello, A.C. Bruno, T. Silveira, Characterization of carburization of HP steels by non destructive magnetic testing, NDT E Int. 39 (2006) 569–577, https://doi.org/10.1016/j.ndteint.2006.03.004. [26] I.C. Silva, J.M.A. Rebello, A.C. Bruno, P.J. Jacques, B. Nysten, J. Dille, Structural and magnetic characterization of a carburized cast austenitic steel, Scr. Mater. 59 (2008) 1010–1013. [27] H.E. Evans, Cracking and spalling of protective oxide layers, Mater. Sci. Eng. A. 120–121 (1989) 139–146, https://doi.org/10.1016/0921-5093(89)90731-4.