Laser peening without coating induced phase transformation and thermal relaxation of residual stresses in AISI 321 steel

Laser peening without coating induced phase transformation and thermal relaxation of residual stresses in AISI 321 steel

Surface & Coatings Technology 291 (2016) 161–171 Contents lists available at ScienceDirect Surface & Coatings Technology journal homepage: www.elsev...

3MB Sizes 0 Downloads 5 Views

Surface & Coatings Technology 291 (2016) 161–171

Contents lists available at ScienceDirect

Surface & Coatings Technology journal homepage:

Laser peening without coating induced phase transformation and thermal relaxation of residual stresses in AISI 321 steel D. Karthik, S. Swaroop ⁎ Surface Modification Laboratory, School of Advanced Sciences, VIT University, Vellore 632014, India

a r t i c l e

i n f o

Article history: Received 29 October 2015 Revised 15 February 2016 Accepted in revised form 16 February 2016 Available online 17 February 2016 Keywords: Austenitic steel Laser peening Phase transformation Residual stress Thermal relaxation Hardness

a b s t r a c t Laser peening without coating (LPwC) was performed on austenitic stainless steel AISI 321 with a power density of 5.97 GW cm−2 and pulse densities of 1600, 2500 and 3900 pulses cm−2. A large compressive residual stress resulted for 1600 pulses cm−2. Surface roughness was found to have increased to a relatively smaller extent with lesser number of valleys or peaks. Micro-hardness increased after the process with the depth of hardened layer extending up to 900 μm. LPwC induced martensitic phase transformation (about 18% in volume fraction) was observed near treated surface. Microstructures confirmed the phase transformation effect. The grains were not refined as a result of single LPwC scan. Thermal relaxation of residual stresses at 700 °C for 2 h were 41% in longitudinal and 140% in transverse (relative to LPwC scanning) directions. The influential factors on stress relaxation were discussed. No significant reduction in hardness was seen after thermal treatment. Followed by the thermal treatment, fully austenitic phase was recovered. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The laser peening without protective coating (LPwC) is an innovative surface modification method being used widely on various industrial metals and alloys. LPwC is frequently performed with low energy laser and without surface sacrificial coatings in contrast to conventional laser peening (LP) process [1]. LPwC develops large and deep compressive residual stresses when laser–matter interaction induced plasma in confined mode generates shock waves of several GPa pressure exceeding dynamic yield strength of the material during its propagation [2]. Few LP or LPwC reports using the laser energy of greater than 1 J indicated the pronounced thermal damage and surface roughness [3–5]. On the other hand, reports with less than 0.5 J demonstrated the minimal thermal damage and smaller surface roughness [6–13]. These investigations point out that use of laser energy of b0.5 J would leave the surface with smaller thermal damage or roughening. It is well known that material surface with high compressive residual stresses induced up to few mm depth show following behavior: (1) show comparatively smaller degradation in corrosive environment, (2) results in enhanced wear resistance in sliding or rolling friction systems and (3) extended the fatigue live in fluctuating loading conditions. Such improvements are extensively investigated on various industrial metal alloys. For example, Sano et al. [6], examined improvement in fatigue

⁎ Corresponding author. E-mail addresses: [email protected] (D. Karthik), [email protected] (S. Swaroop). 0257-8972/© 2016 Elsevier B.V. All rights reserved.

life and stress corrosion cracking of AISI 316L, Ti–6Al–4V and AC4CH alloys after LPwC. In another study, Sano et al. [7] observed better fatigue performance in LPwC processed friction stir welded Al6061-T6 alloy. Further LPwC effect on fatigue properties of different alloys were examined [8–10]. Kumar et al. [15] showed enhanced wear resistance after LPwC in Ti–6Al–4V alloy. Trdan et al. [16–17] indicated that corrosion resistance of Al6082–T651 alloy significantly improved with lowered dissolution rate, pitting and inter-granular attack. They also examined the effect of ultrahigh strain rate plastic deformation of LPwC on Ag–Mg–Si alloy and reported grain refinement and increased dislocation density [4]. Deformation induced martensitic phase transformation is one of the strengthening mechanisms in austenitic stainless steels. There are few LP studies that report the phase transformation in austenitic stainless steels. For example, Ye et al. [18] observed martensitic phase transformation in AISI 304 after LP at shock wave pressure of 5.56 GPa. Their report indicated that higher phase transformation is favorable at cryogenic temperature than room temperature where besides the shock wave pressure, stacking fault energy (SFE) at low temperatures reported to be dominant. Gerland and Hallouin [19] suggested shock pressure range of 15–25 GPa and laser pulse duration of 0.6 ns to induce phase transformation. Turski et al. [20] have found phase transformation in AISI 304 steel using LP. Zhou et al. [21] reported the influence of number LP impacts on deformation induced martensitic phase transformation on AISI 304 steel and surface grain refinement mechanisms. Mordyuk et al. [22] have showed no phase transformation in AISI 321 steel after LPwC with shock wave pressure of 2.08 GPa and no confinement medium.


D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

Among these steels, AISI 321 is titanium stabilized grade with higher metastability for martensitic transformation and widely considered for petroleum refining unit, water–water energetic type nuclear reactors, and in offshore oil platform [23–26,36]. This alloy in service is subjected to higher temperatures (≥ 700 °C) where it shows poor mechanical properties [23,26]. Since it is a low SFE (which promotes the larger number of martensitic nucleation sites) and high metastable (which enhances the rate of martensitic transformation) alloy, use of LPwC would result a better surface integrity and mechanical strength. However, previous LPwC report by Mordyuk et al. [22] showed that no phase transformation or residual stresses were induced. It was observed in that study that used shock wave pressure of 2.08 GPa was much smaller than the threshold value of 5 GPa suggested by Ye et al. [18] on AISI 304 steel (a comparable grade). So, LPwC with appropriate parameters would result in a better surface and mechanical properties in AISI 321 steel. In view of the above considerations, LPwC using power density of 5.97 GW cm−2 (corresponding shock wave pressure of 8.41 GPa) was performed on AISI 321 steel. Different pulse densities were used to investigate the LPwC effect on residual stress, surface roughness, phase transformation, microstructures and micro-hardness. Further, influence of LPwC on thermal treatment at 700 °C for 2 h was investigated and correlated with other results.

Table 2 represents the list of experimental parameters and specimen designations. 2.3. Residual stress The residual stresses were obtained using (3 1 1) plane from X'pert Pro system (PANalytical, Netherlands). The characteristic CuKα1 ray (wavelength = 1.5406 Å) was irradiated on the of 2 mm2 area of centre LPwC processed region in the measurement. The residual stresses along L and T directions were estimated by XRD sin2Ψ method [14]. 2.4. Surface roughness In contact mode, a stylus profiler (MarSurf, Germany) was used to traverse 5.6 mm within LPwC processed region where filter cut-off of 0.8 mm and vertical resolution of 50 nm were employed to analyze surface roughness. On the other hand, an atomic force microscope (AFM) (Nanosurf easyScan 2, Switzerland) was used in tapping mode to examine the 3D surface topography in 25 μm2 of LPwC treated area with cantilever of length 450 μm, width of 45 μm, thickness of 1.5 μm, tip height of 12 μm, spring constant of 0.15 Nm−1 and vertical resolution of 0.2 nm [14]. 2.5. XRD and microstructure

2. Materials and methods 2.1. AISI 321 steel The austenitic stainless steel AISI 321 was procured commercially for this investigation. The result of elemental composition analysis with Spark analyzer (Thermoelectron, USA) is presented in Table 1. Electric Discharge Machining (EDM) cutting was used to prepare the specimens of dimension 20 mm × 20 mm × 5 mm. Further, these specimens were subjected to stress relief heat treatment at 700 °C for 2 h, air cooled and mechanically ground using SiC abrasive sheets up to #2000 grit size. 2.2. LPwC treatment The LPwC was performed with a Q-switched Nd:YAG laser (Litron, UK) which delivered laser beam of wavelength 1064 nm. The pointing stability of the laser beam was less than ±70 μrad and its divergence of output was ≤ 0.5 mrad. Pulses of 10 ns duration (measured at full width at half maximum) with repetition rate of 10 Hz were focused on the specimen surface to be treated using a biconvex lens having focal length of 300 mm. The spot diameter of the focused beam at specimen surface was 0.8 mm. A water layer of 1–3 mm thickness was maintained on specimen surface as confinement medium using treated water through water jet setup. This confinement medium was used to increase the intensity of shock wave generated in LPwC and also to replace ablated fine particles so to insure pure laser–matter interaction. For mentioned conditions, power density of 5.97 GW cm−2 was attained with laser energy of 300 mJ. In a similar way to our previous study [14], pressure of the shock wave generated during LPwC was evaluated to be 8.41 GPa. Further, three different pulse densities with corresponding overlapping rates were obtained by varying feed rate of specimen fixed on 2D translation stage (SVP Lasers, India). For each condition, peening duration per cm−2 (in minutes) was noted. As indicated schematically in Fig. 1a, a zigzag type LPwC scan was employed with longitudinal (L) and transverse (T) movements corresponding to specimen surface. Table 1 Chemical composition of AISI 321 steel. Element




















A high resolution X-ray diffractometer (D8 Brucker, USA) with characteristic CuKα1 ray (wavelength = 1.5406 Å) was used to investigate the structural changes and peak broadening effects in the LPwC specimens. In the analysis, 2θ range of 40 to 100° with 0.02° step size was used. Microstructural variations by LPwC were analyzed with an optical microscope (ZEISS, Germany) and scanning electron microscope (FEI — Quanta, USA). 2.6. Hardness and thermal treatment Depth wise micro-hardness profile was measured with Vickers hardness tester (Mitutoyo, Japan) using a load of 300 g and hold time of 10 s. LPwC specimen was subjected to the heat treatment at 700 °C for 2 h, air cooled, where heating and cooling rate involved was 5 °C min− 1. 3. Results and discussion 3.1. Optimization of LPwC parameters Optimization of the processing parameter was made for better surface integrity with an expectation of larger compressive residual stresses showing smaller anisotropy and surface roughness. 3.1.1. Residual stresses Residual stresses (RS) analyzed on untreated and LPwC specimens are shown in Fig. 1b. The RS of untreated specimen were relatively smaller having compressive stress (−164 ± 4 MPa) in L direction and tensile stress (48 ± 6 MPa) in T direction. Followed by LPwC, RS on as peened surface were tensile of greater magnitudes. Larger tensile RS were observed in T direction for LPwC #3 (890 ± 10 MPa) compared with LPwC #1 (139 ± 6 MPa) and LPwC #2 (54 ± 4 MPa) specimens. On the other hand, LPwC #1 indicated larger tensile RS (483 ± 6 MPa) in L direction compared to LPwC #3 (386 ± 10 MPa) and LPwC #2 (279 ± 6 MPa) specimens. This larger tensile RS on LPwC processed surface can be explained by the factors (1) pulse density (or overlapping rate) and (2) peening duration. The former is crucial as the localized surface melting and resolidification effects are predominantly influenced by pulse density in LPwC process. The increase in pulse density drastically enhances the melting and at the end of the process within short duration a quenching phenomenon takes places which leads to

D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171


Fig. 1. (a) Schematic of zigzag type LPwC scan showing longitudinal (L), transverse (T) sweep directions and their overlapping effects. (b) Residual stresses of untreated and LPwC specimens. (c) Averaged and difference of residual stresses in L and T directions of LPwC specimens at a depth of 50 μm.

tensile RS. The latter (peening duration) was varied with respect to pulse density in this investigation. The peening duration of LPwC #3 was (6 min) longer than that of LPwC #2 (4 min) and LPwC #1 (3 min). This means that same area (1 cm2) was processed for longer duration with higher overlapping rate in LPwC #3 compared to other conditions. It is obvious that if there is longer duration for localized melting over the same area, there would be large tensile RS after resolidification. From those, the larger tensile RS that resulted in LPwC #3 can be explained. Similar results were observed in few previous LPwC reports. For instance, Gill et al. [5] have noticed tensile RS in LPwC treated inconel 718 alloy. Shadangi et al. [27] recently demonstrated that for LP or LPwC duration of ≥5 min, localized melting and resolidification are predominant. Further, Trdan et al. [3–4] have noted the compressive RS at a depth of 33 μm after LPwC which indicates the same effect at treated surface. Our previous report [14] suggested that to separate out pure mechanical effect, at least 50 μm beneath the LPwC treated surface is required, as the surface undergoes both thermal and mechanical effect, with the latter predominating at larger depths. In accordance with some of the LPwC reports [3–4,14], considering thermal effect at treated surface and to investigate the pure mechanical effect of LPwC, a 50 μm layer was removed by electro-polishing using 80% methanol and 20% perchloric acid solution at 18 V for duration of

Table 2 LPwC conditions. Specimen designations

Pulse density (pulses cm−2)

Overlapping rate (%)

Peening duration per cm−2 (minutes)

LPwC #1 LPwC #2 LPwC #3

1600 2500 3900

70 75 80

3 4 6

130–150 s and RS was analyzed. The RS observed in this depth was completely compressive irrespective of processing conditions. The specimen LPwC #1 resulted in larger compressive RS (−854 ± 7 MPa in L and − 827 ± 7 MPa in T directions) compared to LPwC #2 (− 372 ± 3 MPa in L and − 262 ± 3 MPa in T directions) and LPwC #3 (−407 ± 4 MPa in L and −554 ± 4 MPa in T directions) specimens. Further, effect of processing conditions can be evaluated by the induced compression and anisotropy of RS in L and T directions. Fig. 1c represents the averaged RS and difference in RS between L and T directions. It is clear that LPwC #1 specimen indicates larger compression (averaged RS) of 841 ± 7 MPa and smaller anisotropy (difference in RS between L and T directions) of 27 ± 7 MPa compared to those of LPwC #2 (316 ± 3 MPa and 110 ± 3 MPa) and LPwC #3 (481 ± 4 MPa and 147 ± 3 MPa) conditions. Although there was slight increase in compression from LPwC #2 to LPwC #3, anisotropy was larger for higher pulse density. An important point to discuss here is that the appearance of smaller compressive RS in higher pulse densities compared to lower pulse density. Since pulse density (or overlapping rate) and peening duration were higher for LPwC #2 and #3 compared to LPwC #1, thermal relaxation RS due to the combined effect of plasma and shock waves could be expected. Although the overall trend of depth wise distribution of RS is needed for a better understanding, most applications such as corrosion and fatigue depend on the details of RS within few tens of micrometers. Therefore the present study restricts RS distribution within 50 μm. Trdan et al. [3] and Sathyajith et al. [11,13] have observed similar results for higher pulse density. Further, according to Correa et al. [29] larger anisotropy in RS could result if zigzag type scan is used. Furthermore, Trdan et al. [3] discussed that for smaller pulse diameter, larger compressive RS near treated surface (within 50 μm) will be induced. Similarly, our recent LPwC report on 17–4 PH steel with 0.8 mm laser spot


D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

diameter indicated higher compressive RS at 50 μm below treated surface. Hu et al. [37] showed the rapid attenuation of shock wave pressure for larger spot diameter (≥ 1 mm). From the observed results in this study and according to the reports [3,14], it can be concluded that large compressive RS is induced at 50 μm below LPwC processed surface

with the pulse density of 1600 pulses cm−2 and pulse diameter of 0.8 mm. Further, irrespective tensile RS in treated surface, LPwC #1 among other conditions can be argued to be favorable for better surface integrity. Some of the previous LPwC studies [11,13,16–17] suggest that large compressive RS near treated surface can effectively undermine the

Fig. 2. Surface roughness results of untreated and LPwC specimens measured by stylus contact profilometer (a) and AFM (b). 3D surface topography of untreated (c), LPwC #1 (d), LPwC # (e) and LPwC #3 (f) specimens.

D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171


deleterious failure mechanisms, such as corrosion and fatigue. Hence, RS induced near surface for LPwC #1 can be argued to be sufficient to prevent the material surface from detrimental failures. 3.1.2. Surface roughness Fig. 2a shows the surface roughness values (average roughness — Ra, highest peak — Rp, and deepest valley — Rv) of untreated and LPwC specimens analyzed with stylus profiler in contact mode for traverse length of 5.6 mm. The Ra, Rp and Rv values for untreated specimen were (respectively 0.48 μm, 1.39 μm and 2.03 μm) relatively smaller, indicating a smoother surface. After LPwC, these values were increased with respect to pulse density or overlapping rate. Larger surface roughness values recorded in LPwC #3 having Ra of 1.12 μm, Rp of 3.55 μm and Ra of 4.36 μm in comparison with LPwC #2 (Ra — 1.04 μm, Rp — 3.22 μm and Ra — 4.09 μm) and LPwC #1 (Ra — 0.89 μm, Rp — 2.86 μm and Ra — 3.86 μm). These results indicated that increase in pulse density or overlapping rate increases roughness. Many LPwC reports previously investigated have drawn analogous conclusions. For instance, Trdan et al. [3–4,16–17] in their LPwC studies (with laser energy of ˃1 J) noted that surface roughness was larger for higher pulse density or overlapping rate with larger craters or waviness on the treated surface. Gill et al. [5] have observed Ra of 19 μm in inconel 718 alloy after LPwC with the laser energy of 4 J. Luo et al. [28] noticed the better surface profile in LP processed LY2 Al alloy using the laser energy of 5 J, for overlapping rate of 50–70%. It should be noted that besides common influencing factors such as pulse density and overlapping rate, laser energy of less than 0.5 J in LPwC have indeed resulted in smaller surface roughness. For example, Sano et al. [6] reported the Ra of 2 μm after LPwC with energy of 0.25 J. Sathyajith et al. [11] reported Ra of 1.4 μm after LPwC with 0.2 J energy. Recently, in our study [14] with 0.3 J and Shadangi [27] with 0.17–0.34 J, it was confirmed that Ra can be minimized to ≤1 μm. Hence, these roughness values in the present study can be concluded to be relatively smaller and less deleterious. The surface topography was recorded using AFM in tapping mode over 25 μm2 area of untreated and LPwC specimens to resolve the surface feature in a better way. The 3D surface topographical images are presented in Fig. 2c, d, e and f. Their average roughness (Sa), highest peak (Sp) and deepest valley (Sv) results are plotted in Fig. 2b. Specimen surface in untreated condition was smoother having Sa of 0.04 μm, Sp of 1.03 μm and Sv of −0.48 μm. On the other hand, LPwC conditions have increased their magnitudes. Larger values were observed in LPwC #3 (Sa of 0.45 μm, Sp of 1.89 μm and Sv of −1.46 μm) compared to LPwC #2 (Sa of 0.4 μm, Sp of 1.6 μm and Sv of −1.45 μm) and LPwC #1 (Sa of 0.37 μm, Sp of 1.43 μm and Sv of −1.31 μm) conditions. It further indicates that roughness in LPwC process strongly depends on pulse density or overlapping rate showing larger roughness for higher pulse density or overlapping rate. From the topographical images, it is evident that LPwC #1 show relatively smaller number of valleys or peaks than those seen in other conditions. Few LPwC studies reported [3,14] that for smaller pulse density relatively smaller roughness can result. Further, the surface with sharp peaks can readily induce failures in fluctuating loads as they act as stress concentration sites [10]. Lee et al. [30] indicated that a surface with more number of deep valleys can drastically enhance the corrosion rate. In this view, it can be seen that LPwC #1 is favorable condition among others as far corrosion and fatigue improvements are concerned. Since the trend in roughness data is similar both in AFM and profilometer measurements, it is clear that the roughness on the sample is evenly distributed over a large scale (μm to mm). In summary LPwC #1 is considered to be favorable and optimized LPwC parameter. Further analyses therefore included only untreated and LPwC #1 conditions. 3.2. Phase transformation Fig. 3 presents the XRD spectra of untreated and LPwC #1 conditions. Predominantly austenitic (γ) and less intensity martensitic (αʹ) peaks

Fig. 3. XRD of untreated and LPwC #1 specimens. Appearance of (110) peak on surface of LPwC #1 specimen indicates the austenitic (γ) to martensitic (αʹ) transformation.

irrespective of the specimen conditions can be seen. Untreated and LPwC #1 at 50 μm depth conditions indicate a relatively low αʹ phase. On the other hand, αʹ peak on LPwC treated surface appeared with appreciable intensity. It shows that deformation induced γ → α’ transformation was occurred on LPwC surface alone. Hence, quantification of this effect was done by evaluation of volume fraction (%) of these phases by the relation [31],

Vi ¼

j 1 Xn Ii j¼1 j n R i

j 1 Xn I γ 1 Xn Iαj 0 1 Xn Iεj þ þ j¼1 j¼1 j j¼1 j n Rγj n Rε R 0 n



where, Vi is the volume fraction of the ith phase, Iγ, Iα, and Iε and Rγ , Rα , and Rε are integrated intensities and material scattering factors respectively for γ,α',and ε phases; j represents the plane index. According to the reports on AISI 304 steel [31–32], the ε hcp martensitic phase presents if the temperature is less than 0 °C during the deformation process. Since the LPwC process was carried out in room temperature (25 °C), the ε part of Eq. (1) was ignored in the calculation. The results show relatively smaller volume fraction of αʹ phase in untreated (90% of γ and 10% of αʹ) and LPwC at 50 μm depth (91% of γ and 9% of αʹ) conditions. It is quite interesting to see that 72% of γ and 28% of αʹ phases resulted in LPwC treated surface. This implies that about 18% of γ phase was transformed to αʹ at LPwC treated surface. This transformation in AISI 321 is one of the strengthening mechanisms. To understand this transformation in a better way, an influence of various factors, as summarized in Table 3 for few austenitic stainless steels treated by LP or LPwC in the literature were considered. 3.2.1. (I) Alloying elements It can be seen from the table that there was no phase transformation in AISI 316L steel by LP [34] or LPwC [11] compared to AISI 304 steel in [18] and AISI 321 reported in this study. This indicates that alloying element may influence the phase transformation. Lebedev and Kosarchuk [33] have suggested that the alloying elements like Ni, Cr, Mo, Mn, Si and C can influence the rate of austenitic to martensitic transformation.


D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

Table 3 LP/LPwC effect on phase transformation in austenitic stainless steels in the literature. Stainless steel grade

AISI 316L AISI 316L AISI 304 AISI 304 AISI 304 AISI 304 AISI 304 AISI 321 AISI 321 a b

Niequ (%)

27.03 27.03 26.96 26.96 26.96 26.96 26.96 23.09 23.09



Processing conditions Temperature (°C)

Pressure (GPa)a

25–30b 25–30b 25–30b −269 25–30b 25–30b 25–30b 25–30b 25–30b

5 5.45 5.56 5.56 15–25 11 7.14 2.08 8.41

Surface residual stress (MPa)

Volume fraction of αʹ (%)


−500 −273 Not given Not given Not given −273 −435 −80 +483

0 0 35.3 45.5 Unknown Relatively low Not given 0 18

Peyre et al. [34] Sathyajith et al. [11] Ye et al. [18] Ye et al. [18] Gerland and Hallouin [19] Turski et al. [20] Zhou et al. [21] Mordyuk et al. [22] Present study

Evaluated using Fabbro et al. [1] model. Room temperature.

They further proposed a quantity called Niequ. A smaller value of Niequ indicates the higher metastability and higher rate of austenitic to martensitic transformation. According to them, Niequ can be obtained by the relation [33], Niequ ¼ %Ni þ 0:65%Cr þ 0:98%Mo þ 1:05%Mn þ 0:35%Si þ 12:6%C:


From the percentage of alloying elements in AISI 321 shown in Table 1, it was found to be 23.09%, which is smaller than 26.96% of AISI 304 (estimated from the report of Ye et al. [18]) and 27.03% of AISI 316L (estimated from the report of Peyre et al. [34]) steels. Hence, it can be stated that alloying elements in AISI 321 steel have influenced the increasing austenitic to martensitic transformation rate in this study. 3.2.2. (II) Temperature during deformation According to Ye et al. [33], LP at cryogenic temperatures (−269 °C) resulted in 45.5% of αʹ’ volume fraction than at room temperature (25–30 °C) having 35.3% on AISI 304 steel. It was observed from their study that γ → αʹ transformation readily occurs at cryogenic temperatures than room temperature under same shock wave pressure of 5.56 GPa, because (1) at cryogenic temperatures the mechanical driving force (ΔGmech) required for phase transformation is small due to the lessened gap between critical driving force (ΔGcrit — threshold force needed for γ → αʹ transformation ) and chemical driving force (ΔGchem. — free energy force between γ and αʹ phases) and on the other hand (2) at room temperatures, larger ΔGmech should be applied for γ → αʹ transformation to take place. Lowered SFE of AISI 304 steel at cryogenic temperature which promotes more number of nucleation sites for martensitic phase than at room temperature. It is clear that, as the present investigation was performed at room temperature, the influence of temperature on phase transformation induced can be ruled out. 3.2.3. (III) Shock pressure of LPwC It is obvious that LP or LPwC by the ultrahigh strain rate plastic deformation mechanism can result in strain induced αʹ phase transformation in austenitic stainless steel through (1) formation of nucleation sites for αʹ phase (by the formation of shear bands and their intersection) and (2) their growth under certain shock pressure range. Ye et al. [18] suggested that in low SFE austenitic steels, if shock wave pressure above 5 GPa is used, γ → αʹ transformation can readily occur even at room temperature. They reported an αʹ phase volume fraction of 35.3% by LP at room temperature on AISI 304 steel. Gerland and Hallouin [19] have suggested the range of 15–25 GPa for 0.6 ns pulse duration. However, their report [19] did not estimate αʹ phase transformation in (in volume fraction) AISI 304 steel. Other studies [11,20–21,34] were also reported no phase transformation in AISI 316L or AISI 304 steels. Particularly, in AISI 321 steel, Moryduk et al. [22] have shown no

phase transformation after LPwC. It should be noted that in contrast to [22], the present study have indicated about 18% of induced αʹ phase transformation AISI 321 steel. This contradiction can be resolved by considering the larger shock wave pressure used in the current investigation (8.41 GPa) compared to Mordyuk et al. [22] report (2.08 GPa) which was a major influencing factor for the phase transformation. Hence, it is evident that if there is an appreciable shock pressure in LPwC; even at room temperature strain induced martensitic phase transformation in austenitic stainless steels can be induced. 3.2.4. (IV) Residual stresses From Figs. 3 and 1b, a correlation between martensitic phase transformation and tensile residual stresses in LPwC #1 seems to be evident. For instance, the untreated specimen has smaller averaged compressive residual stresses (between L and T directions) of −117 ± 5 MPa with relatively smaller (10%) αʹ phase as confirmed from Fig. 3. Similar effect is seen for LPwC #1 at a depth of 50 μm, where, it has averaged compressive residual stresses of −841 ± 7 MPa with αʹ phase volume fraction of 9%. By contrast, surface of LPwC #1 showed averaged tensile residual stresses of + 261 ± 6 MPa with a larger martensitic volume fraction of about 28%. From these results, it seems that αʹ phase transformation is probably associated with tensile residual stresses while the reverse effect is notable for austenite phase, that is, compressive residual stresses are favorable to transform austenite phase. A basic idea of tensile deformation promoted austenite to martensitic phase transformation with significant volume fraction was reported in [33]. Results indicated in Table 3 also seem to support this effect. However, one of the previous studies [51] indicate that larger transformation induced αʹ phase of volume fraction of 100% is favorable with larger compressive residual stresses of −1400 MPa near treated surface of AISI 304 steel subjected to ultrasonic nano-crystal surface modification process. In contrast the volume fraction of the induced martensitic phase transformation in the present study is relatively smaller compared to the previous reports [33,51]. Hence, it can be stated that the induced phase transformation could be attributed to the ultrahigh strain plastic deformation and larger shock wave pressure of LPwC alone; indicating weak correlation with tensile residual stresses in general. It can be concluded that as a result of larger shock wave pressure and ultrahigh strain rate effects, strain induced martensitic phase transformation occurred in the higher metastable (due to alloying elements) AISI 321 steel. 3.3. Microstructure Fig. 4 represents SEM results of untreated and LPwC treated specimens. From these images, it is clear that there is no grain refinement in peened specimens (Fig. 4b and c) compared to untreated specimen (Fig. 4a). Further, average grain size was estimated to be 16–18 μm.

D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171


Fig. 4. Scanning electron micrographs of (a) untreated, (b) as peened surface of LPwC #1, (c) at 50 μm depth of LPwC #1 and (d) single martensitic grain in image (c). Higher magnification of single martensitic grain (d) shows ε platelets or shear bands like structures. Optical micrographs of untreated, as peened surface and at 50 μm depth of LPwC #1 (e) showing unaltered grains.

The optical micrographs shown in Fig. 4e further support the no grain refinement effect. According to Lu et al. [44], in a LY2 Al alloy, multiple LP can result in a severely plastically deformed (SPD) layer up to 239 μm from processed surface, where grain refinement is more significant. In another report, Lu et al. [45] have seen the grain refinement in AISI 304 after 3 times LP scan. However, the use of single time LP or LPwC was found to be insufficient to refine the grain. Recently we observed unaltered grains in LPwC processed 17–4 PH steel [14]. Telang et al. [46] have shown no grain refinement in alloy 600 after LP. In contrast, Kalainathan et al. [41] reported random grain refinement near

peened surface in AISI 316L steel but a clear investigation as reported with multiple LP [44–45] was not carried out. Mordyuk et al. [22] using LPwC on AISI 321 steel showed no refined grains with the shock wave pressure of 2.08 GPa. Although a larger shock wave pressure of 8.41 GPa was used in the current study, it was found that grains were unaltered. Alternatively, in correlation with XRD results as discussed in Section 3.2, microstructure on untreated surface (Fig. 4a) and LPwC processed surface at a depth of 50 μm (Fig. 4c) show prominently γ grains and while as peened surface (Fig. 4b) shows few αʹ grains. A


D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

single αʹ grain shown in Fig. 4d indicates the formation of what appears like shear bands as a result of probable dislocation pile-ups or ε phase like structure as reported earlier in literature [18]. In this context it is interesting to note that, Ye et al. [18] demonstrated that during the LP martensitic nuclei sites like shear bands appear and intersect with others, grow to form the martensitic phase and Zhou et al. [21] discussed the multidirectional intersection of deformation induced martensitic phase using multiple LP.

3.4. Thermal relaxation AISI 321 steel is used in conditions like petrochemical plants [23, 26] where the high temperatures (≥ 700 °C) for long time exposure commonly exist. With the aim of studying the LPwC influence on thermal stress relaxation, processed specimen was subjected to a higher temperature of 700 °C for 2 h of exposure time in the present investigation.

Fig. 5. Residual stresses (a) and micro-hardness (b) results in untreated, LPwC #1 and LPwC # + TR specimens.

D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

3.4.1. Relaxation of residual stresses Fig. 5a presents the residual stresses of untreated, LPwC #1 before and after thermal relaxation (TR) at 700 °C for 2 h (designated as LPwC #1 + TR). The condition LPwC #1 + TR shows smaller compressive RS in L (−40 ± 2 MPa) and large tensile RS in T (+320 ± 2 MPa) directions on the surface. On the other hand, at 50 μm depth, the large compressive RS of –420 ± 2 MPa in L direction and tensile RS of higher magnitude (+342 ± 2 MPa) in T directions can be seen. These results represent that at treated surface, after heat treatment, large tensile stresses in L direction observed in LPwC #1 were reduced to smaller compressive type while, tensile RS increased in T direction. A large stress relaxation in T direction at 50 μm with final tensile RS and average relaxation in L direction having compressive RS were noticed in LPwC #1 + TR. The percentage of stress relaxation in L was 49% whereas in T it approached to be 141% which is much greater. This large RS relaxation in T direction from compression to tension state after TR may be attributed by, firstly, large difference in thermal expansion coefficient between austenitic (18 × 10−6 K−1) and martensitic (12.4 × 10−6 K−1) phases, where, upon the thermal treatment at elevated temperature, the former would have resulted in a tensile RS while latter might have in resulted compressive RS [52]. Secondly, as suggested in the previous reports, the obvious effect of inducing large compressive RS in T direction compared to L direction [3–4,14] which could have resulted the more RS relaxation after TR for long time exposure [47]. However, the reason for resulting to large RS anisotropy along T direction in LPwC #1 + TR specimen having the shift of RS from compression to tension is unclear at this point. On the other hand, the effect of LPwC showing compressive RS in L direction in LPwC #1 + TR is significant to address. The mechanism influencing this relaxation is higher relaxation rate at 700 °C for exposure time of 2 h. This could be attributed to the increased relaxation rate as the temperature and exposure duration were notably larger. It can be explained as follows, (a) the LPwC transforms vast energy to deform the alloy where this energy is stored by increased dislocation density; (b) the deformed alloy will be at thermodynamically unstable state; (c) during isothermal process, the stored energy will be released as dislocation glide and climb occur and decrease its density; (d) for a long exposure, if the initial deformed alloy has refined grains, then those grains will grow to result to larger grains. An important point is that, even after exposure of 700 °C for 2 h, there exists a large compressive residual stress in L direction. An existing compressive RS in LPwC #1 + TR could be attributed by the similar mechanism. Further, researchers [38–43,47–50] have found that, stress relaxation is rapid during initial exposure (3 to 30 min) and after which relaxation rate lowered and stabled. Furthermore, stress relaxation was more observed when there was a large compressive residual stress due to their thermodynamically unstable state. Most notably, LP or LPwC effected RS are highly stable under thermal exposure compared to shot peening [40–41] because of smaller percentage of cold working. Hence, it can be concluded that, larger thermal relaxation occurred at the treated surface whereas appreciable relaxation did not occur at least in one direction at a depth of 50 μm after LPwC in AISI 321 steel. 3.4.2. Effect on micro-hardness Micro-hardness measured as function of depth is shown in Fig. 5b. The average hardness value of untreated specimen was 127 HV. Hardness of LPwC #1 increased gradually up to 100 μm having the maximum of 173 HV and further decrease, reached the magnitude of untreated sample at 900 μm. This effect was due to the mechanism of shock wave energy dissipation during the propagation as it propagates deeper into the material. Our previous LPwC reports indicate a similar mechanism [11,14]. A commonly known fact is that the presence of martensitic phase induces larger hardness. On the other hand, existence of large tensile stresses can lower the hardness. In these views, our present result as shown in Fig. 5b the increase in hardness at treated surface in LPwC #1 can be explained by induced appreciable martensitic phase volume fraction


which undermined the effect of tensile residual stresses. While, pure mechanical effect was influenced the hardness up to depth of 900 μm. For LPwC #1 + TR, a 20% reduction in hardness was observed on treated surface while in depths it is relatively low (can be seen in inset of Fig. 5b). The maximum hardness was 160 HV at 100 μm and further gradually decreases identical to LPwC #1. This result shows that thermal treatment did not influence the hardness. Further, compared with RS of LPwC #1 + TR, insignificant reduction in hardness explains the beneficial effect of LPwC. The percentage of cold working can be inferred from the peak broadening on each condition [20,34]. The smaller percentage of cold working means that the degree of work hardening is smaller. In this context, evaluated integrated FWHM of X-ray diffraction peaks from Fig. 3 (for untreated and LPwC #1 conditions) and Fig. 6a (for LPwC #1 + TR condition) indicated a relatively lower value for LPwC #1 (1.25°) and LPwC #1 + TR (1.03°) specimens compared to untreated specimen (1.35°). This implies that work hardening was not influenced by the process alone or together with thermal treatment. Our previous report on 17–4 PH steel indicated the same effect [14]. 3.4.3. Effect on phase transformation and microstructure The XRD spectra and microstructure at surface of LPwC #1 + TR are shown in Fig. 6. It shows reverse phase transformation, that is, martensitic to austenitic, resulting in full austenitic phase from surface LPwC #1 after heat treatment which can be inferred from Fig. 6a. It further indicates that the shear bands like structure seen in martensitic grain (Fig. 4d) disappeared which are evident in Fig. 6b. These results imply

Fig. 6. XRD (a) and microstructure (b) of LPwC #1 + TR specimen.


D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171

that as a result of isothermal process at 700 °C for 2 h, lessened defects and further recovery as discussed in Section 3.4.1 could be attributed to the fully austenitic phases in LPwC #1 + TR. Further, from Figs. 4 and 6b, no grain refinement effect can be seen. The state of residual stresses at LPwC #1 + TR surface represent reduced tensile RS which can be correlated with its microstructure indicated in Fig. 6b. Grosse et al. [35] reported that in AISI 321 steel the martensitic content decrease exponentially with increasing temperatures. Hence, the austenitic phase as represented in Fig. 6 can be said to be recovered after thermal treatment. The hardness measurement shown in Fig. 5b relates this effect where after thermal treatment hardness at surface dropped below the untreated conditions. 3.5. Conclusion The LPwC induced a better surface integrity of AISI 321 steel used. The following conclusions can be drawn from this study. 1) A large compressive residual stress resulted in the alloy on the effect of LPwC with 1600 pulses cm−2. Surface roughness was found to be increased to a relatively smaller extent with lesser number of valleys or peaks. 2) LPwC induced martensitic phase transformation, about 18% volume fraction near treated surface. Microstructures confirmed the phase transformation effect. The grains were not refined as a result of single time LPwC scan. 3) Thermal relaxation residual stresses at 700 °C for 2 h, were 41% in longitudinal and 140% in transverse of LPwC scanning directions. The influential factors for stress relaxation were discussed. 4) Micro-hardness increased after process with depth of hardened layer extended up to 900 μm. No significant reduction in hardness was seen after thermal treatment. Followed by the thermal treatment, the processed surface showed fully austenitic phase retransformed from martensitic phase. Acknowledgment The AISI 321 steel was provided by Dr. G. Sairam. LPwC, hardness and optical microscope facilities were provided by Dr. S. Kalainathan. The authors acknowledge VIT University, India for providing the financial support to file a part of this work as an Indian patent titled Method for improving compressive residual stress and hardness in AISI 321 stainless steel using laser peening (App. No.: 1365/CHE/2015). Residual stress measurement was carried out at IIT — Bombay. References [1] R. Fabbro, J. Fournier, P. Ballard, D. Devaux, J. Virmont, Physical study of laserproduced plasma in confined geometry, J. Appl. Phys. 68 (1990) 775, http:// [2] Y. Sano, N. Mukai, K. Okazaki, M. Obata, Residual stress improvement in metal surface by underwater laser irradiation, Nucl. Instrum. Methods Phys. Res., Sect. B 21 (1997) 432–436, [3] U. Trdan, J.A. Porro, J.L. Ocaña, J. Grum, Laser shock peening without absorbent coating (LSPwC) effect on 3D surface topography and mechanical properties of 6082T651 Al alloy, Surf. Coat. Technol. 208 (2012) 109–116, j.surfcoat.2012.08.048. [4] U. Trdan, M. Skarba, J. Grum, Laser shock peening effect on the dislocation transitions and grain re fi nement of Al–Mg–Si alloy, Mater. Charact. 97 (2014) 57–68, [5] A.S. Gill, A. Telang, V.K. Vasudevan, Characteristics of surface layers formed on inconel 718 by laser shock peening with and without a protective coating, J. Mater. Process. Technol. 225 (2015) 463–472, jmatprotec.2015.06.026. [6] Y. Sano, K. Akita, K. Masaki, Y. Ochi, I. Altenberger, B. Scholtes, Laser peening without coating as a surface enhancement technology, J. Laser Micro/Nanoeng. 1 (2006) 161–166, [7] Y. Sano, K. Masaki, T. Gushi, T. Sano, Improvement in fatigue performance of friction stir welded A6061-T6 aluminum alloy by laser peening without coating, Mater. Des. 36 (2012) 809–814,

[8] Y. Sakino, Y. Sano, Y. Kim, Application of laser peening without coating on steel welded joints, Int. J. Struct. Integr. 2 (2011) 332–344, 17579861111162923. [9] I. Altenberger, Y. Sano, M.A. Cherif, I. Nikitin, B. Scholtes, Residual stress state and fatigue behaviour of laser shock peened titanium alloys, Mater. Sci. Forum 524-525 (2006) 129–134, [10] Y. Sano, M. Obata, T. Kubo, N. Mukai, M. Yoda, K. Masaki, et al., Retardation of crack initiation and growth in austenitic stainless steels by laser peening without protective coating, Mater. Sci. Eng. A 417 (2006) 334–340, msea.2005.11.017. [11] S. Kalainathan, S. Sathyajith, S. Swaroop, Effect of laser shot peening without coating on the surface properties and corrosion behavior of 316L steel, Opt. Lasers Eng. 50 (2012) 1740–1745, [12] S. Sathyajith, S. Kalainathan, S. Swaroop, Laser peening without coating on aluminum alloy Al-6061-T6 using low energy Nd:YAG laser, Opt. Laser Technol. 45 (2013) 389–394, [13] S. Sathyajith, S. Kalainathan, Effect of laser peening without coating on 316L austenitic stainless steel, IOP Conf. Ser., Mater. Sci. Eng. 73 (2015) 012152, 1088/1757-899X/73/1/012152. [14] D. Karthik, S. Kalainathan, S. Swaroop, Surface modification of 17-4 PH stainless steel by laser peening without protective coating process, Surf. Coat. Technol. 278 (2015) 138–145, [15] D. Kumar, S. Nadeem Akhtar, A. Kumar Patel, J. Ramkumar, K. Balani, Tribological performance of laser peened Ti–6Al–4V, Wear 322-323 (2015) 203–217, http:// [16] U. Trdan, J. Grum, SEM/EDS characterization of laser shock peening effect on localized corrosion of Al alloy in a near natural chloride environment, Corros. Sci. 82 (2014) 1–11, [17] U. Trdan, J. Grum, Evaluation of corrosion resistance of AA6082-T651 aluminium alloy after laser shock peening by means of cyclic polarisation and ElS methods, Corros. Sci. 59 (2012) 324–333, [18] C. Ye, S. Suslov, D. Lin, G.J. Cheng, Deformation-induced martensite and nanotwins by cryogenic laser shock peening of AISI 304 stainless steel and the effects on mechanical properties, Philos. Mag. 92 (2012) 1369–1389, 14786435.2011.645899. [19] M. Gerland, M. Hallouin, Effect of pressure on the microstructure of an austenitic stainless steel shock-loaded by very short laser pulses, J. Mater. Sci. 29 (1994) 345–351, [20] M. Turski, S. Clitheroe, A.D. Evans, C. Rodopoulos, D.J. Hughes, P.J. Withers, Engineering the residual stress state and microstructure of stainless steel with mechanical surface treatments, Appl. Phys. A Mater. Sci. Process. 99 (2010) 549–556, http:// [21] L. Zhou, W. He, S. Luo, C. Long, C. Wang, X. Nie, et al., Laser shock peening induced surface nanocrystallization and martensite transformation in austenitic stainless steel, J. Alloys Compd. 655 (2016) 66–70, j.jallcom.2015.06.268. [22] B.N. Mordyuk, Y.V. Milman, M.O. Iefimov, G.I. Prokopenko, V.V. Silberschmidt, M.I. Danylenko, et al., Characterization of ultrasonically peened and laser-shock peened surface layers of AISI 321 stainless steel, Surf. Coat. Technol. 202 (2008) 4875–4883, [23] M.J.G. Silva, A.A. Souza, A.V.C. Sobral, P. De Lima-Neto, H.F.G. Abreu, Microstructural and electrochemical characterization of the low temperature sensitization of AISI 321stainless steel tube used in petroleum refining plants, J. Mater. Sci. 38 (2003) 1007–1011, [24] B.A. Gurovich, E.A. Kuleshova, A.S. Frolov, D.A. Maltsev, K.E. Prikhodko, S.V. Fedotova, et al., Investigation of high temperature annealing effectiveness for recovery of radiation-induced structural changes and properties of 18Cr–10Ni–Ti austenitic stainless steels, J. Nucl. Mater. 465 (2015) 565–581, jnucmat.2015.06.045. [25] K. Guan, X. Xu, H. Xu, Z. Wang, Effect of aging at 700 °C on precipitation and toughness of AISI 321 and AISI 347 austenitic stainless steel welds, Nucl. Eng. Des. 235 (2005) 2485–2494, [26] K.S. Guan, X.D. Xu, Y.Y. Zhang, Z.W. Wang, Cracks and precipitate phases in 321 stainless steel weld of flue gas pipe, Eng. Fail. Anal. 12 (2005) 623–633, http://dx. [27] Y. Shadangi, K. Chattopadhyay, S.B. Rai, V. Singh, Effect of LASER shock peening on microstructure, mechanical properties and corrosion behavior of interstitial free steel, Surf. Coat. Technol. 280 (2015) 216–224, [28] K.Y. Luo, T. Lin, F.Z. Dai, X.M. Luo, J.Z. Lu, Effects of overlapping rate on the uniformities of surface profile of LY2 Al alloy during massive laser shock peening impacts, Surf. Coat. Technol. 266 (2015) 49–56, surfcoat.2015.02.017. [29] C. Correa, D. Peral, J.A. Porro, M. Díaz, L. Ruiz de Lara, a. García-Beltrán, et al., Random-type scanning patterns in laser shock peening without absorbing coating in 2024-T351 Al alloy: a solution to reduce residual stress anisotropy, Opt. Laser Technol. 73 (2015) 179–187, [30] S.M. Lee, W.G. Lee, Y.H. Kim, H. Jang, Surface roughness and the corrosion resistance of 21Cr ferritic stainless steel, Corros. Sci. 63 (2012) 404–409, 10.1016/j.corsci.2012.06.031. [31] E. Nagy, V. Mertinger, F. Tranta, J. Sólyom, Deformation induced martensitic transformation in stainless steels, Mater. Sci. Eng. A 378 (2004) 308–313, http://dx.doi. org/10.1016/j.msea.2003.11.074. [32] A.K. De, D.C. Murdock, M.C. Mataya, J.G. Speer, D.K. Matlock, Quantitative measurement of deformation-induced martensite in 304 stainless steel by X-ray diffraction, Scr. Mater. 50 (2004) 1445–1449, j.scriptamat.2004.03.011.

D. Karthik, S. Swaroop / Surface & Coatings Technology 291 (2016) 161–171 [33] A.A. Lebedev, V.V. Kosarchuk, Influence of phase transformations on the mechanical properties of austenitic stainless steels, Int. J. Plast. 16 (2000) 749–767, http://dx. [34] P. Peyre, X. Scherpereel, L. Berthe, C. Carboni, R. Fabbro, G. Be, Surface modifications induced in 316L steel by laser peening and shot-peening. Influence on pitting corrosion resistance, Mater. Sci. Eng. A. 280 (2000) 294–302, S0921-5093(99)00698-X. [35] M. Grosse, D. Kalkhof, M. Niffenegger, L. Keller, Influencing parameters on martensite transformation during low cycle fatigue for steel AISI 321, Mater. Sci. Eng. A 437 (2006) 109–113, [36] J.S. Corte, J.M.A. Rebello, M.C.L. Areiza, S.S.M. Tavares, M.D. Araujo, Failure analysis of AISI 321 tubes of heat exchanger, Eng. Fail. Anal. 56 (2015) 170–176, http://dx.doi. org/10.1016/j.engfailanal.2015.03.008. [37] Y. Hu, Z. Yao, Overlapping rate effect on laser shock processing of 1045 steel by small spots with Nd:YAG pulsed laser, Surf. Coat. Technol. 202 (2008) 1517–1525, http:// [38] E. Maawad, H.-G. Brokmeier, L. Wagner, Y. Sano, C. Genzel, Investigation on the surface and near-surface characteristics of Ti–2.5Cu after various mechanical surface treatments, Surf. Coat. Technol. 205 (2011) 3644–3650, j.surfcoat.2011.01.001. [39] W. Zhuang, B. Wicks, Mechanical surface treatment technologies for gas turbine engine components, J. Eng. Gas Turbines Power 125 (2003) 1021, 1115/1.1610011. [40] M.-C. Berger, J. Gregory, Residual stress relaxation in shot peened timetal 21 s, Mater. Sci. Eng. A 263 (1999) 200–204, [41] Z. Wang, Y. Chen, C. Jiang, Thermal relaxation behavior of residual stress in laser hardened 17-4PH steel after shot peening treatment, Appl. Surf. Sci. 257 (2011) 9830–9835, [42] I. Altenberger, R.K. Nalla, Y. Sano, L. Wagner, R.O. Ritchie, On the effect of deep-rolling and laser-peening on the stress-controlled low- and high-cycle fatigue behavior of Ti–6Al–4V at elevated temperatures up to 550 °C, Int. J. Fatigue 44 (2012) 292–302, [43] I. Altenberger, E.A. Stach, G. Liu, R.K. Nalla, R.O. Ritchie, An in situ transmission electron microscope study of the thermal stability of near-surface microstructures











induced by deep rolling and laser-shock peening, Scr. Mater. 48 (2003) 1593–1598, J.Z. Lu, K.Y. Luo, Y.K. Zhang, C.Y. Cui, G.F. Sun, J.Z. Zhou, et al., Grain refinement of LY2 aluminum alloy induced by ultra-high plastic strain during multiple laser shock processing impacts, Acta Mater. 58 (2010) 3984–3994, j.actamat.2010.03.026. J.Z. Lu, K.Y. Luo, Y.K. Zhang, G.F. Sun, Y.Y. Gu, J.Z. Zhou, et al., Grain refinement mechanism of multiple laser shock processing impacts on ANSI 304 stainless steel, Acta Mater. 58 (2010) 5354–5362, A. Telang, A.S. Gill, S. Teysseyre, S.R. Mannava, D. Qian, V.K. Vasudevan, Effects of laser shock peening on SCC behavior of alloy 600 in tetrathionate solution, Corros. Sci. 90 (2015) 434–444, X.D. Ren, W.F. Zhou, S.D. Xu, S.Q. Yuan, N.F. Ren, Y. Wang, et al., Iron GH2036 alloy residual stress thermal relaxation behavior in laser shock processing, Opt. Laser Technol. 74 (2015) 29–35, Z. Zhou, S. Bhamare, G. Ramakrishnan, S.R. Mannava, K. Langer, Y. Wen, et al., Thermal relaxation of residual stress in laser shock peened Ti–6Al–4V alloy, Surf. Coat. Technol. 206 (2012) 4619–4627, X.D. Ren, Q.B. Zhan, S.Q. Yuan, J.Z. Zhou, Y. Wang, N.F. Ren, et al., A finite element analysis of thermal relaxation of residual stress in laser shock processing Ni-based alloy GH4169, Mater. Des. 54 (2014) 708–711, 2013.08.054. Z. Zhou, A.S. Gill, D. Qian, S.R. Mannava, K. Langer, Y. Wen, et al., A finite element study of thermal relaxation of residual stress in laser shock peened IN718 superalloy, Int. J. Impact Eng. 38 (2011) 590–596, 2011.02.006. C. Ye, A. Telang, A.S. Gill, S. Suslov, Y. Idell, K. Zweiacker, et al., Gradient nanostructure and residual stresses induced by ultrasonic nano-crystal surface modification in 304 austenitic stainless steel for high strength and high ductility, Mater. Sci. Eng. A 613 (2014) 274–288, Y.D. Wang, R. Lin Peng, X.L. Wang, R.L. McGreevy, Grain-orientation-dependent residual stress and the effect of annealing in cold-rolled stainless steel, Acta Mater. 50 (2002) 1717–1734,