- Email: [email protected]
Effect of residual stress induced by pulsed-laser irradiation on initiation of chloride stress corrosion cracking in stainless steel
Materials Science & Engineering A 590 (2014) 433–439
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of residual stress induced by pulsed-laser irradiation on initiation of chloride stress corrosion cracking in stainless steel Shuzo Eto n, Yasufumi Miura, Junichi Tani, Takashi Fujii Central Research Institute of Electric Power Industry, 2-6-1 Nagasaka, Yokosuka, Kanagawa 240-0196, Japan
art ic l e i nf o
a b s t r a c t
Article history: Received 6 August 2013 Received in revised form 21 October 2013 Accepted 23 October 2013 Available online 31 October 2013
The atmospheric corrosion test and residual stress measurement were performed to evaluate the effect of laser irradiation on stress corrosion cracking (SCC) initiation. Second-harmonic Nd:YAG laser pulses (pulse width: 10 ns) were irradiated on a type-304L stainless-steel plate. The specimens were placed in a chamber at 353 K with RH ¼ 35% for the corrosion test. When laser energies were 30 and 300 mJ, cracks caused by SCC or pitting were observed on the surface of the specimens. The cracks were classified into two types on the basis of cumulative probability distribution; one of the types is related to the laser irradiation condition. The mean maximum crack depths were about 27 and 52 μm when laser energies were 30 and 300 mJ, respectively. These values were the same as the depth at which the tensile residual stress was induced from the surface of the specimen by laser irradiation. These results suggest that the maximum stress corrosion crack depth was caused by the tensile residual stress induced by laser irradiation, and that the crack stopped propagating when the crack depth was larger than several dozen μm in this test set. When laser pulses of 300 mJ energy were irradiated on the surface of the specimen by shot peening, the tensile stress was induced up to 20 μm from the surface, and the compressive stress was observed at a larger depth. These results show that the laser irradiation is less effective in obtaining tensile residual stress of the specimen compared to when laser pulses are irradiated on the specimen treated by shot peening. The depth of tensile stress obtained by laser irradiation is much shorter than that of compressive stress obtained by shot peening. & 2013 Elsevier B.V. All rights reserved.
Keywords: Stress corrosion cracking X-ray stress measurement Stainless-steel Pitting Pulsed-laser Peening
1. Introduction Laser surface melting [1] and laser peening by irradiation are used for the surface fabrication or modification of metal materials. Pulsed-laser irradiation generates plasma and causes a plastic deformation by the impulsive pressure of the plasma. Although heat input by laser irradiation results in the occurrence of tensile residual stress on the surface of a metal in general, the use of coatings prevents the ablation and melting at the surface of the metal [2]. The technology of introducing compressive residual stress by irradiation is called laser peening, and this is used to prevent stress corrosion cracking (SCC). Such a surface treatment effect on the residual stress is observed when spark- or laserinduced breakdown spectroscopy [3] is applied to metal materials because the plasma is produced by spark ignition or laser irradiation [4,5]. The pulsed-laser irradiation effects on a metal are heat input and plastic deformation. When the laser pulse width, e.g., the Q-switch laser pulse width, is shorter than several nanoseconds,
n
Corresponding author. Tel.: þ 81 46 856 2121; fax: þ 81 46 856 3540. E-mail address: [email protected] (S. Eto).
0921-5093/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2013.10.066
the effect of plastic deformation is much larger than that of heat input because the laser-induced plasma can be produced with several milliJoules per pulse. Therefore, it is expected that the material composition does not change and the residual stress and morphology at the surface of the material change slightly when a laser pulse of low energy is irradiated on the metal materials. Peyre et al. reported that such a plastic deformation does not affect the occurrence of pitting [6]. On the other hand, Horikawa et al. reported that the crack propagation rate increased at the tungsten inert gas (TIG)-welded joint [7]. Tani et al. reported that chlorideinduced SCC of the austenitic stainless-steel (SS) occurred independently of the tensile residual stress value [8]. They have performed the SCC test on the austenitic SS using loading device and showed that the specimen of type 304L SS was fractured even when the applied stress was below 200 MPa which was less than 0.2% proof stress of type 304L SS. The previous studies indicate that the pulsed-laser irradiation on SS induces the tensile residual stress and results in the SCC initiation. Although pulsed-laser irradiation effects on the surface of metals were investigated experimentally and theoretically upon laser melting and laser peening, almost all the studies reported the relationship between the laser irradiation condition and the properties of metal (e.g., residual stress, morphology, and hardness).
434
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
The susceptibility of a metal to corrosion was also reported when the effect of plastic deformation was dominant, i.e., the input laser energy was small [6,9]. However, the quantitative relationship between the residual stress and SCC or pitting was unclear. The loading test results reveal the stress threshold at which SCC of active-pass corrosion type in austenite SS occurs [10]. The report shows that the stress threshold is 0.2% proof stress at the chlorideinduced SCC of sensitized SS and is lower than 0.2% proof stress at the chloride-induced SCC of unsensitized SS. Takemoto reported that a crack longer than 0.1 mm propagates itself, but a crack shorter than 0.1 mm does not propagate [11]. Although the loading test was examined under the constant loading condition, the crack propagation will be different when the residual stress is induced by laser irradiation owing to the change in the loading value at the tip of a propagating crack. In this study, the atmospheric corrosion test were performed using type 304L SS irradiated by a pulsed laser to evaluate quantitatively the feature of SCC occurred by irradiation. In addition, the relationships between the tensile residual stress induced by irradiation and the length of SCC were clarified by the residual stress measurement. The residual stress measurement results of the specimen treated by shot peening are also shown to discuss the laser irradiation effect on the residual stress with or without surface treatment.
2. Experimental setup 2.1. Materials Type-304L SS of half-inch thickness was used for the experiment. The surface finish was No. 1. Table 1 shows the chemical composition of the test materials. The shape of the specimen and the position irradiated by the laser are shown in Fig. 1. The specimens of 20 30 mm2 area were cleaned with acetone. Laser irradiation was carried out using second-harmonic Nd:YAG lasers (Hoya Continuum, Powerlite 8010) operating at a repetition rate of 10 Hz. The laser pulses were focused by a convex lens of 250 mm focal length and irradiated perpendicularly to the six points on the surface for each specimen. The diameter of the irradiated laser spot on the specimen was about 0.5 mm. 2.2. Stress measurement The residual stress was measured by the X-ray diffraction method (Bruker, D8 Discover). X-ray radiation source of Cr Kα with the applied voltage of 38 kV and the current of 90 mA was used. The 2D method was used for the residual stress analysis [12]. The X-ray beam diameters were set to 0.3 and 0.5 mm to measure the lateral and depth profiles of residual stress, respectively. The detector position was set at 1281. The angle between the normal to the diffracting lattice planes and the sample surface (ψ) and the rotation angle of the sample surface (φ) were set at (ψ, φ)¼ (0, 0), (15,0), (15, 45), (15, 90), (15, 135), (15,180), (30, 0), (30, 45), (30, 90), (30, 135), (30,180), (45, 0), (45, 45), (45, 90), (45, 135), (45,180), (60, 0), (60, 45), (60, 90), (60, 135), and (60,180) for each measurement point. An X-ray diffraction device (Rigaku, AutoMATE) was used to measure the circumferential specimen treated by shot peening. X-ray radiation source of Cr Kα with the applied Table 1 Chemical composition of tested material. Element
C
Si
Mn
P
S
Ni
Cr
Fe
wt%
0.010
0.060
1.19
0.032
0.002
10.33
18.37
Bal.
Fig. 1. Specimen and point of laser irradiation.
voltage of 40 kV and the current of 30 mA was used. The sin2 ψ method was used for the residual stress analysis. The X-ray beam diameter was set to 0.5 mm. ψ were set at 01, 18.41, 26.61, 33.21, 39.21, 45.01, 50.81 for each measurement point. In both devices, the X-ray diffracted from (γ-Fe 220) plane was measured, and the electrochemical polishing was performed to etch the specimen when the depth profile of residual stress was measured. In order to clarify the dependences of laser energy and the number of laser shots on residual stress at the surface of the specimen, the laser energy and the number of laser shots were set to 30, 150 and 300 mJ, and 1, 10 and 100, respectively. 2.3. Corrosion test Droplets (10 μl) of synthetic seawater were placed on the laserirradiated position of the specimens using a micropipette. The major components of the synthetic seawater are NaCl (2.45 wt%), MgCl2 (1.11 wt%), Na2SO4 (0.41 wt%), CaCl2 (0.15 wt%) and so forth. The specimens were placed in a chamber with the temperature and relative humidity controlled at 353 K and 35%, respectively. The experimental setup satisfies the environmental condition for the SCC initiation because the SCC initiation was reported using the type 304L SS sprayed with synthetic seawater under the environmental condition (353 K with RH ¼ 35%) [8]. The surfaces and cross sections of the specimens were observed after test durations of 1500 and 3000 h. In order to observe the cross section of the specimens, a wire cutting-off machine was used to cut the line passing through the center of the laser-irradiated position. The crack depth was defined as the length from the surface of the specimen to the tip of the crack observed at the cross section of the specimen.
3. Results and discussion 3.1. Feature of crack depth Typical images of the cross section and surface of the specimen after the corrosion test are shown in Fig. 2. Microscopic cracks observed around the laser-irradiated position were produced through the No. 1 surface finish. The other cracks were not observed at the laser-irradiated position owing to the surface melting of the specimen. SCC and pitting were observed at the point on which the droplets of synthetic seawater were placed. Results of field investigation of the SCC of austenite SS under the
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
Fig. 2. (a) Typical feature of cracks and schematics of (b) cross section and (c) surface of a specimen around the center of laser irradiation after exposure to test environment. Experimental conditions were the laser energy of 30 mJ, the number of pulses of 10 and the test duration of 1500 h.
atmospheric condition suggest that pitting plays an important role in the SCC initiation [13]. The corrosion tests of type 304L SS using synthetic seawater have been conducted; the results show that SCC occurred at the tip of pitting [14]. On the other hand, a heat input by laser irradiation improves the resistance to pitting and intergranular SCC [1]. The number of pittings that occurred at the surface of SS irradiated by a pulsed laser was the same as that of SS unirradiated by a pulsed laser at the corrosion area [6]. Under our experimental condition, pittings were observed at the point at which the droplets of synthetic seawater were placed, regardless of whether laser was irradiated. The cracks caused by SCC can be classified into 2 types. Some cracks caused by SCC (type 1) start from the surface of the specimen or from the tip of a small pit (typically less 20 μm); others (type 2) start from the tip of a large pit as shown in Fig. 2. In this study, the length of the pit was shorter than that of SCC for type 1, and that the length of the pit was longer than that of SCC for type 2. The histograms of the crack are shown in Fig. 3. The bin size was set as 5 μm for each test duration (1500 or 3000 h) and laserirradiation condition (30 mJ or 300 mJ). Almost all the cracks of type 1 were smaller than 40 μm, and those of type 2 were larger than 20 μm, and the variability of crack depth of type 2 is larger than that of type 1. These results suggest that the crack depth of type 2 is determined by the pit depth, which is weakly related to the laser irradiation, and that of type 1 can be determined on the basis of the depth of the tensile residual stress induced by laser irradiation as discussed in the results of residual stress measurement. The total number of cracks was approximately the same between the experiments when the test durations were 1500 (Fig. 3(a) and (b)) and 3000 h (Fig. 3(c) and (d)). However, the number of cracks was smaller with the laser energy of 30 mJ (Fig. 3 (a) and (c)) than with that of 300 mJ (Fig. 3(b) and (d)). The reason for this is that the melted surface area of SS became larger when the laser energy was larger owing to the increase in heat input. The number of cracks and crack depth were also different when the laser energies were 30 and 300 mJ. The cumulative probability
435
Fig. 3. Histogram of crack depth at a laser-irradiated point with laser energies of (a, c) 30 mJ and (b, d) 300 mJ. Test durations were (a, b) 1500 h and (c, d) 3000 h.
Fig. 4. Cumulative probability of crack depth with laser energies of 30 and 300 mJ.
Fig. 5. Test duration dependence of maximum depth of type 1 crack that occurred within a radius of 1 mm from the center of laser irradiation. The values are the maximum crack depths at the cross section of the specimens at each laserirradiated position.
436
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
of crack depth can be described using two normal probability distributions (F1 and F2) as shown in Fig. 4 because the cumulative distribution functions (CDFs) of the crack depth are proportional to Table 2 Mean and standard deviation of maximum crack depth.
30 mJ 300 mJ
1500 h
3000 h
27.2 7 9.24 46.9 7 12.4
26.5 7 9.72 59.3 7 23.7
Fig. 6. Residual stress dependence on laser energy and number of laser irradiations.
the crack depth. The crack depths of type 1 and type 2 are contained in F1 and F2, respectively. On the other hand, Hayashibara et al. reported that the cumulative probability was described using one normal distribution from the result of the atmospheric corrosion test using SS as the specimen [15]. These results show that the characteristics of type 1 and type 2 can be described on the basis of the CDF of the crack depth, and the difference between type 1 and type 2 is explained by the slope of CDF. The number of cracks with depths less than 15 μm cannot be described using the normal distribution. The reason for this is that the crack depths produced by the No. 1 finish were less than 10 μm, and the CDFs were attributed to the crack depth. The maximum depths of type 1 in each crack at the laser irradiation points are shown in Fig. 5. In addition, the means of the maximum crack depth for each exposure time and laser energy are shown in Table 2. There was no significant difference between the results obtained after exposure for 1500 and 3000 h when the laser energy was the same. Hayashibara et al. reported that the incubation time of SCC using SS was about 100 h [15], and Shirai et al. reported that the crack growth rate of SCC was 5.6 10 5 mm/h [16]. If a crack propagates continually with the crack growth rate during the test duration, the maximum crack depth should become over 190 μm ( 5.6 10 5 mm/h 2900 h), which is much larger than the maximum crack depth of type 1. These results suggest that the crack has stopped propagating to the depth direction with the test duration from 1500 to 3000 h. The maximum crack depth was larger at the laser energy of 300 mJ than at that of 30 mJ. The difference can be explained by the depth of the tensile residual stress induced by laser irradiation as discussed in the following section on stress measurement. 3.2. Residual stress induced by pulsed-laser irradiation The dependences of the surface residual stress on laser energy and number of pulses are shown in Fig. 6. Regardless of the
Fig. 7. Lateral profile of residual stress in (a), (c) longitudinal and (b), (d) lateral directions. (a), (b) 1 Shot and (c), (d) 10 shots.
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
number of pulses and direction of residual stress, the residual stress value became the same when the number of pulses was over 10 and the laser energy was over 150 mJ. The same tendency was also reported for the results of laser peening and was explained by the repetition of heat input and heat diffusion after laser irradiation as follows [17]. First, the surface of the specimen was heated by laser irradiation, and the laser-irradiated point was melted. Second, the melted point was cooled by heat diffusion, and tensile residual stress was induced by the contraction of the melted area. Third, the point was irradiated by the next laser pulse and melted again. Consequently, the cycle of the heat input by laser irradiation and heat diffusion occurred when laser pulses were successively irradiated on the same point, and the residual stress was determined by the energy of the final laser pulse. However, the laser energy dependences of residual stress differed only when one laser pulse was irradiated on the specimen, or the laser energy was 30 mJ. The reason for this can be that the heat input was too small to determine the residual stress after laser irradiation, and that the residual stress before laser irradiation also affected that after laser irradiation. The surface residual stress distributions of longitudinal and lateral directions are shown in Fig. 7. When no laser pulse was irradiated, the residual stresses of the longitudinal and lateral directions were about 50 and 150 MPa, respectively. Despite the difference in residual stress, the values of longitudinal and lateral directions were close when the laser pulse was irradiated at any laser energy and number of pulses. This means that the equibiaxial tensile stress was induced by the surface melting of SS. When the number of pulses was one, or the laser energy was 30 mJ, the tensile stress was induced within the area of laser irradiation, which corresponds to the laser spot ( 0.5 mm). On the other hand, the area in which the tensile stress was induced was within 0.9 mm in diameter, which was larger than the laser spot when the laser energy was 300 mJ and the number of pulses was 10. This is due to the enlargement of the heat-affected zone by the increase in the laser energy and number of pulses. The position of maximum tensile residual stress was the center of the laser spot, and the values did not correlate clearly with the laser energy. These results are similar to those shown in Fig. 6. The depth profiles of residual stress were measured from the surface of the specimen to a depth of 130 μm as shown in Fig. 8. When no laser was irradiated, the longitudinal residual stress remained constant at about 100 MPa from 10 to 130 μm, and the lateral residual stress increased versus depth from 10 to 130 μm. Since the surface finish affected the surface longitudinal and lateral residual stresses, these values came close to zero. Although the tensile residual stress was observed from the surface to 130 μm, the crack with a depth larger than 10 μm was not observed at the unirradiated area from the results of the corrosion test. Therefore, it is considered that the crack can occur when the residual stress is larger than 100 MPa, which is the typical value in the longitudinal direction under the experimental condition. The tensile residual stress was induced up to a deeper position with a higher laser energy and a larger number of pulses. For example, the tensile stress was induced from the surface of the specimen to a depth of 65 μm, when the laser energy was 300 mJ and the number of pulses was 10, which are nearly equal to the maximum crack depth under the same experimental condition as shown in Table 2. These results show that the crack propagates through the area where the tensile residual stress is induced and stops propagating at a depth of several dozen μm under the experimental condition of this study. However, the induced tensile residual stress is also related to the heat input; the depth of the heat-affected zone was considered to be much small than several dozen μm. In order to
437
Fig. 8. Depth profile of residual stress in (a), (c) longitudinal and (b), (d) lateral directions. (a), (b) 1 Shot and (c), (d) 10 shots.
438
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
consider the effect of the heating by laser irradiation on the residual stress, the heat diffusion length was estimated. For laser processing, the heat diffusion length lT is an important quantity to estimate the depth of the heat-affected zone and is written as lT 2(ατl)0.5, where α and τl are heat diffusivity and laser pulse width, respectively [18]. Although the equation is derived from a one-dimensional nonstationary heat diffusion equation, the temperature distribution in the direction of the laser pulse and lT were obtained from the one-dimensional heat equation. lT was about 0.36 μm under the experimental condition of this study, where α and τl are 3.19 10 6 m2/s [19] and 10 ns, respectively. The estimation shows that the tensile residual stress caused by laser irradiation is not determined by the heating by laser irradiation. Surface treatment techniques such as shot peening are effective for the prevention of SCC initiation because compressive residual stress is induced at the surface of a material by the treatment. However, a tensile residual stress will be induced when some inspection such as laser- or spark-induced breakdown spectroscopy is applied to the treated surface, by shot peening, owing to the small heat input. Stress measurement is useful for evaluating the effect of SCC initiation as discussed for the results of the depth profile of residual stress. Therefore, the depth profile of residual stress was measured using the specimen treated by shot peening, and the laser irradiation effect on the SCC initiation was evaluated. A type-304L circular plate with tungsten inert gas (TIG) welding [16] was used for the stress measurement as shown in Fig. 9. The TIG-welded plate with a diameter of 100 mm and a thickness of 13 mm was subjected to the shot peening of zirconium in a quarter areas. Laser pulses were irradiated near the TIG weld line. Subsequently, the residual stress was measured by the X-ray diffraction method. The area around a laser-irradiated position was masked by chloromethane tape to shield the diffracted X-ray from the area. Some 2θ diffraction peaks were not used for estimating the residual stress because the 2θ diffraction peaks could not be measured clearly owing to the existence of coarse grains. The depth profiles of residual stress are shown in Fig. 10. The compressive residual stress was observed from the surface of the plate to a depth of 140 μm at the point treated by shot peening, and the tensile residual stress was observed at the points at which laser pulses were irradiated. No remarkable difference was observed between the stress distributions in the radial and circumferential directions. The tensile residual stress at the surface of the plate irradiated by a laser energy of 300 mJ was higher than that at the
Fig. 9. Schematic of circular specimen welded by TIG welding. The laser was irradiated at the quarter area treated by shot peening of zirconium.
Fig. 10. Depth profile of residual stress in (a), (c) radial and (b), (d) circumferential directions of the circular welded specimen. (a), (b) 1 Shot and (c), (d) 10 shots.
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
surface of the plate irradiated with 30 mJ in the radial direction. This tendency was similar to that shown in Fig. 6. When the number of pulses was one, no tensile stress was observed except under the experimental condition with a laser energy of 300 mJ. The residual stresses at the irradiated and unirradiated points were the same at the depth of 30 μm, as shown in Fig. 10(d), when the laser energy was 300 mJ and the number of pulses was 10. The compressive residual stress was observed from the surface to a depth of 1 mm [16], which is much larger than the depth of 30 μm under the treatment condition. It is clear that the depth at which the compressive residual stress was induced and affected by shot peening is about 30 times deeper than that at which the tensile residual stress was induced by laser irradiation with the laser energy of 300 mJ and number of pulses of 10, which are the maximum heat input conditions in the experiment. The comparison between the results with and without the shot peening shows that the depth profiles of residual stress were less affected by laser irradiation when laser pulses were irradiated on the circular specimen treated by shot peening. In particular, no tensile residual stress was observed when the laser energy was 30 mJ and the number of pulses was 1, as shown in Fig. 10, despite the fact that the surface residual stress was over 100 MPa under the same laser irradiation conditions as shown in Fig. 8. It is considered that the energy spent on the plastic deformation increased owing to the increase in the hardness at the surface of the specimen by shot peening. 4. Conclusions The propagation of stress corrosion cracks and the change in residual stress by Nd:YAG laser irradiation were investigated through the atmospheric corrosion test and stress measurement. The propagation of such cracks terminated in the area of compressive stress when the corrosion test duration was from 1500 to 3000 h, and two cracks were observed; one was the crack from a large pit, and the other was that from a small pit. The tensile residual stress was induced more deeply with a higher laser energy and a larger number of pulses. The tensile stress was induced from the surface of the specimen to a depth of 65 μm when the laser energy was 300 mJ and the number of pulses was 10, which is almost equal to the
439
maximum crack depth under the same experimental condition. The depth of the tensile stress became smaller when laser pulses were irradiated on the treated surface of SS by shot peening. These results suggest that the maximum stress corrosion crack depth can be determined on the basis of the induced tensile residual stress by laser irradiation; the crack depth was smaller than the depth at which the compressive residual stress was induced by shot peening when the laser energy was smaller than 300 mJ and the number of pulses was smaller than 10. In addition, when laser pulses were irradiated on the specimen treated by shot peening, the depth profile of residual stress was less affected by laser irradiation.
References [1] T.R. Anthony, H.E. Cline, J. Appl. Phys. 49 (1978) 1248–1255. [2] C.S. Montross, T. Wei, L. Ye, G. Clark, Y.-W. Mai, Int. J. Fatigue 24 (2002) 1021–1036. [3] D.A. Cremers, L.J. Radziemski, Handbook of Laser-Induced Breakdown Spectroscopy, John Wiely & Sons, England, 2006(283 pp). [4] S. Palanco, L.M. Cabalin, D. Romero, J.J. Laserna, J. Anal. At. Spectrom. 14 (1999) 1883–1887. [5] S. Eto, J. Tani, K. Shirai, T. Fujii, Spectrochim. Acta B 87 (2013) 74–80. [6] P. Peyre, C. Carboni, P. Forget, G. Beranger, C. Lemaitre, D. Stuart, J. Mater. Sci. 42 (2007) 6866–6877. [7] K. Horikawa, A. Sakakibara, T. Mori, Trans. JWRI 12 (1983) 135–142. [8] J.I. Tani, M. Mayuzumi, Zaiyo to Kankyo 54 (2005) 582–586. [9] S. Eto, Y. Miura, J. Tani, T. Fujii, Development of Measurement Technology of Chlorine Attached on Canister Using Laser—Effect of Pulsed-Laser Irradiation on Initiation of Stress Corrosion Cracking, CRIPEI Report no. H12003, , 2013. [10] M. Takemoto, Study on the failure threshold stress criteria for the prevention and mechanism of stress corrosion cracking, Proceedings of the International Congress on Metallic Corrosion, (1984) 565–572. [11] M. Takemoto, Boshoku Gijutsu 25 (1986) 18–26. [12] B.B. He, K.L. Smith, in: Proceedings of the ICRS-5, Linkoping, Sweden, 1997, pp. 634–639. [13] T. Kawamoto, Boshoku Gijutsu 37 (1988) 30–33. [14] M. Mayuzumi, T. Arai, K. Hide, Zairyo-to-Kankyo 52 (2003) 166–170. [15] H. Hayashibara, M. Mayuzumi, Y. Mizutani, J.I. Tani, J. Jpn. Inst. Met. 70 (2006) 1012–1015. [16] K. Shirai, J.-I. Tani, M. Gotoh, M. Wataru, in: Proceedings of the PSAM 11 & ESREL 2012, June 25–29, 2012, Helsinki, Finland, 16BS-We2-3, 2012. [17] H. Tanaka, K. Akita, Y. Sano, N. Mukai, S. Ohya, Zairyo 55 (2006) 641–646. [18] D. Bauerle, Laser Processing and Chemistry, 2nd ed., Springer Verlag, Berlin, 1996 (21 pp). [19] R.S. Graves, T.G. Kollie, D.L. McElroy, K.E. Gilchrist, Int. J. Thermophys. 2 (1991) 409–415.
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of residual stress induced by pulsed-laser irradiation on initiation of chloride stress corrosion cracking in stainless steel Shuzo Eto n, Yasufumi Miura, Junichi Tani, Takashi Fujii Central Research Institute of Electric Power Industry, 2-6-1 Nagasaka, Yokosuka, Kanagawa 240-0196, Japan
art ic l e i nf o
a b s t r a c t
Article history: Received 6 August 2013 Received in revised form 21 October 2013 Accepted 23 October 2013 Available online 31 October 2013
The atmospheric corrosion test and residual stress measurement were performed to evaluate the effect of laser irradiation on stress corrosion cracking (SCC) initiation. Second-harmonic Nd:YAG laser pulses (pulse width: 10 ns) were irradiated on a type-304L stainless-steel plate. The specimens were placed in a chamber at 353 K with RH ¼ 35% for the corrosion test. When laser energies were 30 and 300 mJ, cracks caused by SCC or pitting were observed on the surface of the specimens. The cracks were classified into two types on the basis of cumulative probability distribution; one of the types is related to the laser irradiation condition. The mean maximum crack depths were about 27 and 52 μm when laser energies were 30 and 300 mJ, respectively. These values were the same as the depth at which the tensile residual stress was induced from the surface of the specimen by laser irradiation. These results suggest that the maximum stress corrosion crack depth was caused by the tensile residual stress induced by laser irradiation, and that the crack stopped propagating when the crack depth was larger than several dozen μm in this test set. When laser pulses of 300 mJ energy were irradiated on the surface of the specimen by shot peening, the tensile stress was induced up to 20 μm from the surface, and the compressive stress was observed at a larger depth. These results show that the laser irradiation is less effective in obtaining tensile residual stress of the specimen compared to when laser pulses are irradiated on the specimen treated by shot peening. The depth of tensile stress obtained by laser irradiation is much shorter than that of compressive stress obtained by shot peening. & 2013 Elsevier B.V. All rights reserved.
Keywords: Stress corrosion cracking X-ray stress measurement Stainless-steel Pitting Pulsed-laser Peening
1. Introduction Laser surface melting [1] and laser peening by irradiation are used for the surface fabrication or modification of metal materials. Pulsed-laser irradiation generates plasma and causes a plastic deformation by the impulsive pressure of the plasma. Although heat input by laser irradiation results in the occurrence of tensile residual stress on the surface of a metal in general, the use of coatings prevents the ablation and melting at the surface of the metal [2]. The technology of introducing compressive residual stress by irradiation is called laser peening, and this is used to prevent stress corrosion cracking (SCC). Such a surface treatment effect on the residual stress is observed when spark- or laserinduced breakdown spectroscopy [3] is applied to metal materials because the plasma is produced by spark ignition or laser irradiation [4,5]. The pulsed-laser irradiation effects on a metal are heat input and plastic deformation. When the laser pulse width, e.g., the Q-switch laser pulse width, is shorter than several nanoseconds,
n
Corresponding author. Tel.: þ 81 46 856 2121; fax: þ 81 46 856 3540. E-mail address: [email protected] (S. Eto).
0921-5093/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2013.10.066
the effect of plastic deformation is much larger than that of heat input because the laser-induced plasma can be produced with several milliJoules per pulse. Therefore, it is expected that the material composition does not change and the residual stress and morphology at the surface of the material change slightly when a laser pulse of low energy is irradiated on the metal materials. Peyre et al. reported that such a plastic deformation does not affect the occurrence of pitting [6]. On the other hand, Horikawa et al. reported that the crack propagation rate increased at the tungsten inert gas (TIG)-welded joint [7]. Tani et al. reported that chlorideinduced SCC of the austenitic stainless-steel (SS) occurred independently of the tensile residual stress value [8]. They have performed the SCC test on the austenitic SS using loading device and showed that the specimen of type 304L SS was fractured even when the applied stress was below 200 MPa which was less than 0.2% proof stress of type 304L SS. The previous studies indicate that the pulsed-laser irradiation on SS induces the tensile residual stress and results in the SCC initiation. Although pulsed-laser irradiation effects on the surface of metals were investigated experimentally and theoretically upon laser melting and laser peening, almost all the studies reported the relationship between the laser irradiation condition and the properties of metal (e.g., residual stress, morphology, and hardness).
434
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
The susceptibility of a metal to corrosion was also reported when the effect of plastic deformation was dominant, i.e., the input laser energy was small [6,9]. However, the quantitative relationship between the residual stress and SCC or pitting was unclear. The loading test results reveal the stress threshold at which SCC of active-pass corrosion type in austenite SS occurs [10]. The report shows that the stress threshold is 0.2% proof stress at the chlorideinduced SCC of sensitized SS and is lower than 0.2% proof stress at the chloride-induced SCC of unsensitized SS. Takemoto reported that a crack longer than 0.1 mm propagates itself, but a crack shorter than 0.1 mm does not propagate [11]. Although the loading test was examined under the constant loading condition, the crack propagation will be different when the residual stress is induced by laser irradiation owing to the change in the loading value at the tip of a propagating crack. In this study, the atmospheric corrosion test were performed using type 304L SS irradiated by a pulsed laser to evaluate quantitatively the feature of SCC occurred by irradiation. In addition, the relationships between the tensile residual stress induced by irradiation and the length of SCC were clarified by the residual stress measurement. The residual stress measurement results of the specimen treated by shot peening are also shown to discuss the laser irradiation effect on the residual stress with or without surface treatment.
2. Experimental setup 2.1. Materials Type-304L SS of half-inch thickness was used for the experiment. The surface finish was No. 1. Table 1 shows the chemical composition of the test materials. The shape of the specimen and the position irradiated by the laser are shown in Fig. 1. The specimens of 20 30 mm2 area were cleaned with acetone. Laser irradiation was carried out using second-harmonic Nd:YAG lasers (Hoya Continuum, Powerlite 8010) operating at a repetition rate of 10 Hz. The laser pulses were focused by a convex lens of 250 mm focal length and irradiated perpendicularly to the six points on the surface for each specimen. The diameter of the irradiated laser spot on the specimen was about 0.5 mm. 2.2. Stress measurement The residual stress was measured by the X-ray diffraction method (Bruker, D8 Discover). X-ray radiation source of Cr Kα with the applied voltage of 38 kV and the current of 90 mA was used. The 2D method was used for the residual stress analysis [12]. The X-ray beam diameters were set to 0.3 and 0.5 mm to measure the lateral and depth profiles of residual stress, respectively. The detector position was set at 1281. The angle between the normal to the diffracting lattice planes and the sample surface (ψ) and the rotation angle of the sample surface (φ) were set at (ψ, φ)¼ (0, 0), (15,0), (15, 45), (15, 90), (15, 135), (15,180), (30, 0), (30, 45), (30, 90), (30, 135), (30,180), (45, 0), (45, 45), (45, 90), (45, 135), (45,180), (60, 0), (60, 45), (60, 90), (60, 135), and (60,180) for each measurement point. An X-ray diffraction device (Rigaku, AutoMATE) was used to measure the circumferential specimen treated by shot peening. X-ray radiation source of Cr Kα with the applied Table 1 Chemical composition of tested material. Element
C
Si
Mn
P
S
Ni
Cr
Fe
wt%
0.010
0.060
1.19
0.032
0.002
10.33
18.37
Bal.
Fig. 1. Specimen and point of laser irradiation.
voltage of 40 kV and the current of 30 mA was used. The sin2 ψ method was used for the residual stress analysis. The X-ray beam diameter was set to 0.5 mm. ψ were set at 01, 18.41, 26.61, 33.21, 39.21, 45.01, 50.81 for each measurement point. In both devices, the X-ray diffracted from (γ-Fe 220) plane was measured, and the electrochemical polishing was performed to etch the specimen when the depth profile of residual stress was measured. In order to clarify the dependences of laser energy and the number of laser shots on residual stress at the surface of the specimen, the laser energy and the number of laser shots were set to 30, 150 and 300 mJ, and 1, 10 and 100, respectively. 2.3. Corrosion test Droplets (10 μl) of synthetic seawater were placed on the laserirradiated position of the specimens using a micropipette. The major components of the synthetic seawater are NaCl (2.45 wt%), MgCl2 (1.11 wt%), Na2SO4 (0.41 wt%), CaCl2 (0.15 wt%) and so forth. The specimens were placed in a chamber with the temperature and relative humidity controlled at 353 K and 35%, respectively. The experimental setup satisfies the environmental condition for the SCC initiation because the SCC initiation was reported using the type 304L SS sprayed with synthetic seawater under the environmental condition (353 K with RH ¼ 35%) [8]. The surfaces and cross sections of the specimens were observed after test durations of 1500 and 3000 h. In order to observe the cross section of the specimens, a wire cutting-off machine was used to cut the line passing through the center of the laser-irradiated position. The crack depth was defined as the length from the surface of the specimen to the tip of the crack observed at the cross section of the specimen.
3. Results and discussion 3.1. Feature of crack depth Typical images of the cross section and surface of the specimen after the corrosion test are shown in Fig. 2. Microscopic cracks observed around the laser-irradiated position were produced through the No. 1 surface finish. The other cracks were not observed at the laser-irradiated position owing to the surface melting of the specimen. SCC and pitting were observed at the point on which the droplets of synthetic seawater were placed. Results of field investigation of the SCC of austenite SS under the
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
Fig. 2. (a) Typical feature of cracks and schematics of (b) cross section and (c) surface of a specimen around the center of laser irradiation after exposure to test environment. Experimental conditions were the laser energy of 30 mJ, the number of pulses of 10 and the test duration of 1500 h.
atmospheric condition suggest that pitting plays an important role in the SCC initiation [13]. The corrosion tests of type 304L SS using synthetic seawater have been conducted; the results show that SCC occurred at the tip of pitting [14]. On the other hand, a heat input by laser irradiation improves the resistance to pitting and intergranular SCC [1]. The number of pittings that occurred at the surface of SS irradiated by a pulsed laser was the same as that of SS unirradiated by a pulsed laser at the corrosion area [6]. Under our experimental condition, pittings were observed at the point at which the droplets of synthetic seawater were placed, regardless of whether laser was irradiated. The cracks caused by SCC can be classified into 2 types. Some cracks caused by SCC (type 1) start from the surface of the specimen or from the tip of a small pit (typically less 20 μm); others (type 2) start from the tip of a large pit as shown in Fig. 2. In this study, the length of the pit was shorter than that of SCC for type 1, and that the length of the pit was longer than that of SCC for type 2. The histograms of the crack are shown in Fig. 3. The bin size was set as 5 μm for each test duration (1500 or 3000 h) and laserirradiation condition (30 mJ or 300 mJ). Almost all the cracks of type 1 were smaller than 40 μm, and those of type 2 were larger than 20 μm, and the variability of crack depth of type 2 is larger than that of type 1. These results suggest that the crack depth of type 2 is determined by the pit depth, which is weakly related to the laser irradiation, and that of type 1 can be determined on the basis of the depth of the tensile residual stress induced by laser irradiation as discussed in the results of residual stress measurement. The total number of cracks was approximately the same between the experiments when the test durations were 1500 (Fig. 3(a) and (b)) and 3000 h (Fig. 3(c) and (d)). However, the number of cracks was smaller with the laser energy of 30 mJ (Fig. 3 (a) and (c)) than with that of 300 mJ (Fig. 3(b) and (d)). The reason for this is that the melted surface area of SS became larger when the laser energy was larger owing to the increase in heat input. The number of cracks and crack depth were also different when the laser energies were 30 and 300 mJ. The cumulative probability
435
Fig. 3. Histogram of crack depth at a laser-irradiated point with laser energies of (a, c) 30 mJ and (b, d) 300 mJ. Test durations were (a, b) 1500 h and (c, d) 3000 h.
Fig. 4. Cumulative probability of crack depth with laser energies of 30 and 300 mJ.
Fig. 5. Test duration dependence of maximum depth of type 1 crack that occurred within a radius of 1 mm from the center of laser irradiation. The values are the maximum crack depths at the cross section of the specimens at each laserirradiated position.
436
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
of crack depth can be described using two normal probability distributions (F1 and F2) as shown in Fig. 4 because the cumulative distribution functions (CDFs) of the crack depth are proportional to Table 2 Mean and standard deviation of maximum crack depth.
30 mJ 300 mJ
1500 h
3000 h
27.2 7 9.24 46.9 7 12.4
26.5 7 9.72 59.3 7 23.7
Fig. 6. Residual stress dependence on laser energy and number of laser irradiations.
the crack depth. The crack depths of type 1 and type 2 are contained in F1 and F2, respectively. On the other hand, Hayashibara et al. reported that the cumulative probability was described using one normal distribution from the result of the atmospheric corrosion test using SS as the specimen [15]. These results show that the characteristics of type 1 and type 2 can be described on the basis of the CDF of the crack depth, and the difference between type 1 and type 2 is explained by the slope of CDF. The number of cracks with depths less than 15 μm cannot be described using the normal distribution. The reason for this is that the crack depths produced by the No. 1 finish were less than 10 μm, and the CDFs were attributed to the crack depth. The maximum depths of type 1 in each crack at the laser irradiation points are shown in Fig. 5. In addition, the means of the maximum crack depth for each exposure time and laser energy are shown in Table 2. There was no significant difference between the results obtained after exposure for 1500 and 3000 h when the laser energy was the same. Hayashibara et al. reported that the incubation time of SCC using SS was about 100 h [15], and Shirai et al. reported that the crack growth rate of SCC was 5.6 10 5 mm/h [16]. If a crack propagates continually with the crack growth rate during the test duration, the maximum crack depth should become over 190 μm ( 5.6 10 5 mm/h 2900 h), which is much larger than the maximum crack depth of type 1. These results suggest that the crack has stopped propagating to the depth direction with the test duration from 1500 to 3000 h. The maximum crack depth was larger at the laser energy of 300 mJ than at that of 30 mJ. The difference can be explained by the depth of the tensile residual stress induced by laser irradiation as discussed in the following section on stress measurement. 3.2. Residual stress induced by pulsed-laser irradiation The dependences of the surface residual stress on laser energy and number of pulses are shown in Fig. 6. Regardless of the
Fig. 7. Lateral profile of residual stress in (a), (c) longitudinal and (b), (d) lateral directions. (a), (b) 1 Shot and (c), (d) 10 shots.
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
number of pulses and direction of residual stress, the residual stress value became the same when the number of pulses was over 10 and the laser energy was over 150 mJ. The same tendency was also reported for the results of laser peening and was explained by the repetition of heat input and heat diffusion after laser irradiation as follows [17]. First, the surface of the specimen was heated by laser irradiation, and the laser-irradiated point was melted. Second, the melted point was cooled by heat diffusion, and tensile residual stress was induced by the contraction of the melted area. Third, the point was irradiated by the next laser pulse and melted again. Consequently, the cycle of the heat input by laser irradiation and heat diffusion occurred when laser pulses were successively irradiated on the same point, and the residual stress was determined by the energy of the final laser pulse. However, the laser energy dependences of residual stress differed only when one laser pulse was irradiated on the specimen, or the laser energy was 30 mJ. The reason for this can be that the heat input was too small to determine the residual stress after laser irradiation, and that the residual stress before laser irradiation also affected that after laser irradiation. The surface residual stress distributions of longitudinal and lateral directions are shown in Fig. 7. When no laser pulse was irradiated, the residual stresses of the longitudinal and lateral directions were about 50 and 150 MPa, respectively. Despite the difference in residual stress, the values of longitudinal and lateral directions were close when the laser pulse was irradiated at any laser energy and number of pulses. This means that the equibiaxial tensile stress was induced by the surface melting of SS. When the number of pulses was one, or the laser energy was 30 mJ, the tensile stress was induced within the area of laser irradiation, which corresponds to the laser spot ( 0.5 mm). On the other hand, the area in which the tensile stress was induced was within 0.9 mm in diameter, which was larger than the laser spot when the laser energy was 300 mJ and the number of pulses was 10. This is due to the enlargement of the heat-affected zone by the increase in the laser energy and number of pulses. The position of maximum tensile residual stress was the center of the laser spot, and the values did not correlate clearly with the laser energy. These results are similar to those shown in Fig. 6. The depth profiles of residual stress were measured from the surface of the specimen to a depth of 130 μm as shown in Fig. 8. When no laser was irradiated, the longitudinal residual stress remained constant at about 100 MPa from 10 to 130 μm, and the lateral residual stress increased versus depth from 10 to 130 μm. Since the surface finish affected the surface longitudinal and lateral residual stresses, these values came close to zero. Although the tensile residual stress was observed from the surface to 130 μm, the crack with a depth larger than 10 μm was not observed at the unirradiated area from the results of the corrosion test. Therefore, it is considered that the crack can occur when the residual stress is larger than 100 MPa, which is the typical value in the longitudinal direction under the experimental condition. The tensile residual stress was induced up to a deeper position with a higher laser energy and a larger number of pulses. For example, the tensile stress was induced from the surface of the specimen to a depth of 65 μm, when the laser energy was 300 mJ and the number of pulses was 10, which are nearly equal to the maximum crack depth under the same experimental condition as shown in Table 2. These results show that the crack propagates through the area where the tensile residual stress is induced and stops propagating at a depth of several dozen μm under the experimental condition of this study. However, the induced tensile residual stress is also related to the heat input; the depth of the heat-affected zone was considered to be much small than several dozen μm. In order to
437
Fig. 8. Depth profile of residual stress in (a), (c) longitudinal and (b), (d) lateral directions. (a), (b) 1 Shot and (c), (d) 10 shots.
438
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
consider the effect of the heating by laser irradiation on the residual stress, the heat diffusion length was estimated. For laser processing, the heat diffusion length lT is an important quantity to estimate the depth of the heat-affected zone and is written as lT 2(ατl)0.5, where α and τl are heat diffusivity and laser pulse width, respectively [18]. Although the equation is derived from a one-dimensional nonstationary heat diffusion equation, the temperature distribution in the direction of the laser pulse and lT were obtained from the one-dimensional heat equation. lT was about 0.36 μm under the experimental condition of this study, where α and τl are 3.19 10 6 m2/s [19] and 10 ns, respectively. The estimation shows that the tensile residual stress caused by laser irradiation is not determined by the heating by laser irradiation. Surface treatment techniques such as shot peening are effective for the prevention of SCC initiation because compressive residual stress is induced at the surface of a material by the treatment. However, a tensile residual stress will be induced when some inspection such as laser- or spark-induced breakdown spectroscopy is applied to the treated surface, by shot peening, owing to the small heat input. Stress measurement is useful for evaluating the effect of SCC initiation as discussed for the results of the depth profile of residual stress. Therefore, the depth profile of residual stress was measured using the specimen treated by shot peening, and the laser irradiation effect on the SCC initiation was evaluated. A type-304L circular plate with tungsten inert gas (TIG) welding [16] was used for the stress measurement as shown in Fig. 9. The TIG-welded plate with a diameter of 100 mm and a thickness of 13 mm was subjected to the shot peening of zirconium in a quarter areas. Laser pulses were irradiated near the TIG weld line. Subsequently, the residual stress was measured by the X-ray diffraction method. The area around a laser-irradiated position was masked by chloromethane tape to shield the diffracted X-ray from the area. Some 2θ diffraction peaks were not used for estimating the residual stress because the 2θ diffraction peaks could not be measured clearly owing to the existence of coarse grains. The depth profiles of residual stress are shown in Fig. 10. The compressive residual stress was observed from the surface of the plate to a depth of 140 μm at the point treated by shot peening, and the tensile residual stress was observed at the points at which laser pulses were irradiated. No remarkable difference was observed between the stress distributions in the radial and circumferential directions. The tensile residual stress at the surface of the plate irradiated by a laser energy of 300 mJ was higher than that at the
Fig. 9. Schematic of circular specimen welded by TIG welding. The laser was irradiated at the quarter area treated by shot peening of zirconium.
Fig. 10. Depth profile of residual stress in (a), (c) radial and (b), (d) circumferential directions of the circular welded specimen. (a), (b) 1 Shot and (c), (d) 10 shots.
S. Eto et al. / Materials Science & Engineering A 590 (2014) 433–439
surface of the plate irradiated with 30 mJ in the radial direction. This tendency was similar to that shown in Fig. 6. When the number of pulses was one, no tensile stress was observed except under the experimental condition with a laser energy of 300 mJ. The residual stresses at the irradiated and unirradiated points were the same at the depth of 30 μm, as shown in Fig. 10(d), when the laser energy was 300 mJ and the number of pulses was 10. The compressive residual stress was observed from the surface to a depth of 1 mm [16], which is much larger than the depth of 30 μm under the treatment condition. It is clear that the depth at which the compressive residual stress was induced and affected by shot peening is about 30 times deeper than that at which the tensile residual stress was induced by laser irradiation with the laser energy of 300 mJ and number of pulses of 10, which are the maximum heat input conditions in the experiment. The comparison between the results with and without the shot peening shows that the depth profiles of residual stress were less affected by laser irradiation when laser pulses were irradiated on the circular specimen treated by shot peening. In particular, no tensile residual stress was observed when the laser energy was 30 mJ and the number of pulses was 1, as shown in Fig. 10, despite the fact that the surface residual stress was over 100 MPa under the same laser irradiation conditions as shown in Fig. 8. It is considered that the energy spent on the plastic deformation increased owing to the increase in the hardness at the surface of the specimen by shot peening. 4. Conclusions The propagation of stress corrosion cracks and the change in residual stress by Nd:YAG laser irradiation were investigated through the atmospheric corrosion test and stress measurement. The propagation of such cracks terminated in the area of compressive stress when the corrosion test duration was from 1500 to 3000 h, and two cracks were observed; one was the crack from a large pit, and the other was that from a small pit. The tensile residual stress was induced more deeply with a higher laser energy and a larger number of pulses. The tensile stress was induced from the surface of the specimen to a depth of 65 μm when the laser energy was 300 mJ and the number of pulses was 10, which is almost equal to the
439
maximum crack depth under the same experimental condition. The depth of the tensile stress became smaller when laser pulses were irradiated on the treated surface of SS by shot peening. These results suggest that the maximum stress corrosion crack depth can be determined on the basis of the induced tensile residual stress by laser irradiation; the crack depth was smaller than the depth at which the compressive residual stress was induced by shot peening when the laser energy was smaller than 300 mJ and the number of pulses was smaller than 10. In addition, when laser pulses were irradiated on the specimen treated by shot peening, the depth profile of residual stress was less affected by laser irradiation.
References [1] T.R. Anthony, H.E. Cline, J. Appl. Phys. 49 (1978) 1248–1255. [2] C.S. Montross, T. Wei, L. Ye, G. Clark, Y.-W. Mai, Int. J. Fatigue 24 (2002) 1021–1036. [3] D.A. Cremers, L.J. Radziemski, Handbook of Laser-Induced Breakdown Spectroscopy, John Wiely & Sons, England, 2006(283 pp). [4] S. Palanco, L.M. Cabalin, D. Romero, J.J. Laserna, J. Anal. At. Spectrom. 14 (1999) 1883–1887. [5] S. Eto, J. Tani, K. Shirai, T. Fujii, Spectrochim. Acta B 87 (2013) 74–80. [6] P. Peyre, C. Carboni, P. Forget, G. Beranger, C. Lemaitre, D. Stuart, J. Mater. Sci. 42 (2007) 6866–6877. [7] K. Horikawa, A. Sakakibara, T. Mori, Trans. JWRI 12 (1983) 135–142. [8] J.I. Tani, M. Mayuzumi, Zaiyo to Kankyo 54 (2005) 582–586. [9] S. Eto, Y. Miura, J. Tani, T. Fujii, Development of Measurement Technology of Chlorine Attached on Canister Using Laser—Effect of Pulsed-Laser Irradiation on Initiation of Stress Corrosion Cracking, CRIPEI Report no. H12003, , 2013. [10] M. Takemoto, Study on the failure threshold stress criteria for the prevention and mechanism of stress corrosion cracking, Proceedings of the International Congress on Metallic Corrosion, (1984) 565–572. [11] M. Takemoto, Boshoku Gijutsu 25 (1986) 18–26. [12] B.B. He, K.L. Smith, in: Proceedings of the ICRS-5, Linkoping, Sweden, 1997, pp. 634–639. [13] T. Kawamoto, Boshoku Gijutsu 37 (1988) 30–33. [14] M. Mayuzumi, T. Arai, K. Hide, Zairyo-to-Kankyo 52 (2003) 166–170. [15] H. Hayashibara, M. Mayuzumi, Y. Mizutani, J.I. Tani, J. Jpn. Inst. Met. 70 (2006) 1012–1015. [16] K. Shirai, J.-I. Tani, M. Gotoh, M. Wataru, in: Proceedings of the PSAM 11 & ESREL 2012, June 25–29, 2012, Helsinki, Finland, 16BS-We2-3, 2012. [17] H. Tanaka, K. Akita, Y. Sano, N. Mukai, S. Ohya, Zairyo 55 (2006) 641–646. [18] D. Bauerle, Laser Processing and Chemistry, 2nd ed., Springer Verlag, Berlin, 1996 (21 pp). [19] R.S. Graves, T.G. Kollie, D.L. McElroy, K.E. Gilchrist, Int. J. Thermophys. 2 (1991) 409–415.