Responses of the corroded surface layer of austenitic stainless steel to different corrosive conditions under cavitation

Responses of the corroded surface layer of austenitic stainless steel to different corrosive conditions under cavitation

Materials Science & Engineering A 671 (2016) 118–126 Contents lists available at ScienceDirect Materials Science & Engineering A journal homepage: w...

2MB Sizes 1 Downloads 47 Views

Materials Science & Engineering A 671 (2016) 118–126

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Responses of the corroded surface layer of austenitic stainless steel to different corrosive conditions under cavitation Xingyue Yong n, Ning Xiao, Hanjie Shen, Yili Song State Key Laboratory of Organic-Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, PR China

art ic l e i nf o

a b s t r a c t

Article history: Received 18 March 2016 Received in revised form 8 June 2016 Accepted 9 June 2016 Available online 10 June 2016

Nanoindentation was used to measure the nano-mechanical properties of the corroded surface layer of austenitic stainless steel after cavitation corrosion tests. The phase structures and chemical compositions of the corroded surface layer were analysed using X-ray diffraction and X-ray Photoelectron Spectroscopy. The results show that corrosion caused a decrement of the nano-mechanical properties of the corroded surface layer. Once corrosion was weakened, the formation of a work-hardened layer resulted from the transformation of austenite into martensite. The synergistic effect caused the more rapid dissolution of Fe hydroxides, resulting in the enrichment of Cr and property changes of the corroded surface layer under cavitation. & 2016 Elsevier B.V. All rights reserved.

Keywords: Nanoindentation X-ray diffraction Austenite Martensite Interfaces Phase transformation

1. Introduction Cavitation corrosion, which has also been called cavitation erosion-corrosion in the literature [1], often occurs in a severe manner on metallic components in industrial services. Cavitation corrosion is caused by the synergistic effect between cavitation erosion (mechanical factor) and corrosion (electrochemical factor) [1–3]. Cavitation erosion was defined as the progressive loss of original material from a solid surface due to continued exposure to cavitation [4]. Therefore, the cavitation corrosion resistance of metals depends not only on their corrosion resistance but also on their mechanical properties, such as the yield properties (hardness and rate of straining hardening), elastic properties (elastic modulus, resilience and super elasticity) and surface topography [1,5]. For example, hardness, tensile strength, engineering strain and grain size are significant parameters for the cavitation corrosion resistance of duplex structure alloys [6]. The excellent cavitation erosion resistance of Fe-Mn-Si-Cr shape memory alloys was found to be mainly attributed to their excellent elasticity in the local micro-zone of the corroded surface layer. In addition, the local elasticity of the corroded surface layer was one of the predominant factors characterising the cavitation erosion resistance of stainless steels and shape memory alloys [7,8]. The total recovery/deformation ratio (Ft), which reflects both super-elasticity and n

Corresponding author. E-mail address: [email protected] (X. Yong).

http://dx.doi.org/10.1016/j.msea.2016.06.019 0921-5093/& 2016 Elsevier B.V. All rights reserved.

pseudo-plasticity, had also been found to correlate well with the cavitation erosion resistance of heat-treated NiTi [9]. Furthermore, C. Godoy et al. [10] found that a linear relationship between cavitation corrosion resistance and the ratio of the square of the hardness to the elastic modulus of the surface layer (H2/E) could be established. For the cavitation erosion resistance of a hard coating, the derived parameter was proportional to the plasticity index, the ratio of the hardness to the Young's modulus (H/E), the adhesion force (LC2), and the ratio of the thermal conductivity of the coating to that of the substrate. The derived parameter was inversely proportional to the number of phases in the coating phase composition, the ratio of the thermal expansion of coatings to that of the substrate, and the coating thickness [11–14]. Among these parameters, the mechanical parameters, for example H/E, had a close relationship to the cavitation corrosion resistance of the materials under cavitation [14,15]. Corrosion often causes a degradation of the mechanical properties of the corroded surface layer [16]; this degradation can accelerate corrosion on the corroded surface layer of metals under cavitation, resulting in the formation of a corrosion-induced soft layer [17] and a decrease in the thickness of the metals [15,18]. This type of degradation would also further lead to the degradation of the structural mechanical properties of metals [19], resulting in disastrous accidents and permanent failures of metallic components. In addition, when austenitic stainless steel in a 3.5% NaCl solution under cavitation was polarised at different potentials, the nano-mechanical properties of the corroded surface layer

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

List of symbols E the averaged value of nano-elastic modulus, GPa Enano nano-elastic modulus, GPa H the averaged value of nano-hardness, GPa Hnano nano-hardness, GPa H/E ratio of hardness to elastic modulus (H/E)nano corroded surface layer nano-mechanical property

were changed and the cavitation corrosion mechanism of the austenitic stainless steel was different [20]. To date, there are few articles available regarding the relationship of the properties of the corroded surface layer of metals with their corrosion resistance [7,8,18,21,22]. Thus, it is very important to study the corroded surface layer properties of metals, including the mechanical properties of the corroded surface layer, to understand the close relationship of these properties to the cavitation corrosion resistance of metals under cavitation. Nanoindentation technology is a powerful tool for quantitatively measuring the surface layer nano-mechanical properties [10,15,23–25]. In the present paper, nanoindentation technology was used to quantitatively measure the corroded surface layer nano-mechanical properties of austenitic stainless steel in corrosive media without cavitation and after cavitation corrosion exposures. The phase structures and chemical compositions of the corroded surface layer without cavitation and under cavitation were analysed using X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), respectively. The relationship of the nano-mechanical properties of the corroded surface layer to its phase structures and chemical compositions was studied to understand the relationship of the corroded surface layer properties to cavitation corrosion resistance.

2. Experimental materials and methods The material used in this study was S30400 austenitic stainless steel (wt%: C, 0.07; Cr, 18.8; Ni, 8.9; Si, 1.0; Mn, 2.0; S, 0.03; P, 0.035; Fe, balance) in the form of a bar with a diameter of 19 mm. All specimens, which were solution annealed at 1050 °C for 30 min and subsequently quenched in water, were manufactured in accordance with ASTM G32 [4], wet polished with SiC paper up to 800 grit, cleaned using the ultrasonic method, degreased with ethanol, and finally dried in desiccators. Cavitation corrosion exposures were performed using an ultrasonic cavitation corrosion apparatus [18,20]. The vibration frequency and vibration amplitude were 20 kHz and approximately 50 mm, respectively. The probe was mounted vertically with the specimen immersed in the test medium contained in an 800-ml glass beaker. To perform electrochemical tests, each specimen was first fixed on the top of the PTFE block. Next, the specimen was placed co-axially relative to the horn of the ultrasonic probe and was held still at a distance of 0.6–0.8 mm from the horn tip. The horn tip was immersed at a maximum depth of 10 mm, as shown Fig. 1. Specimens made of S30400 stainless steels were used as the working electrodes. Platinum wire and saturated calomel electrodes (SCE) were used as the counter and reference electrodes, respectively. Specimens were subjected to a series of cavitation corrosion exposures in a 3.5% (wt%) pH 6.0–8.0 NaCl solution at 30 °C for 1.0 h. The test media were saturated with air. At the same time, some specimens were subjected to cavitation corrosion

119

h indentation depth, nm h1 a maximum of indentation depth, nm Md30 transformation temperature of austenite Men þ metallic ion (Me)2On metallic oxides Me(OH)n metallic hydroxides MeOOH hydroxyl metallic hydroxides

Fig. 1. A diagram of the ultrasonic cavitation corrosion apparatus with the specified electrochemical test cell.

exposures in the same media for 1.0 h while they were polarised at a 0.80 V (SCE) potential. The other specimens were subjected to cavitation corrosion in pH 4.5–6.5 distilled water at 30 °C for 1.0 h. For comparing the specimens under cavitation, specimens were also only immersed in a 3.5% (wt%) pH 6.0–8.0 NaCl solution at 30 °C for 1.0 h without cavitation. The measurements of the nanohardness (Hnano) and nanoelastic modulus (Enano) of the corroded surface layer were made on specimens without cavitation and after cavitation corrosion exposure. Before taking the nanoindentation measurements, the corrosion products on the corroded surfaces of specimens were removed by ultrasonic cleaning to reduce the effects of corrosion product films [16]. Based on the nanoindentation continuous stiffness measurement technique, all of these measurements were performed using a MTS Nano Indenter XP in accordance with ISO 14577-1:2007 and GB/T22458-2008 [25,26]. A maximum indentation depth was 2.0 mm. The Berkovich indenter was a threesided pyramid [27]. Five indents were made per specimen. The nanohardness (Hnano) and the nanoelastic modulus (Enano) were determined using the initial unload slope, contact area and peak load according to the method of Oliver and Pharr [28,29]. Thus, the profiles of Hnano and Enano for the corroded surface layer with indentation depth (h) can be obtained first. The averages (H, E) of Hnano and Enano were then calculated using the following formulas:

1 h1

∫0

1 h1

∫0

H=

E=

h1

h1

Hnano ( h)⋅dh

Enano ( h)⋅dh

(1)

(2)

The nano-mechanical property of the corroded surface layer was comprehensively defined as (H/E)nano, which is a dimensionless function [18]. The profiles of (H/E) nano with indentation depth (h) were then obtained. Similarly, the average (H/E) of (H/E) nano can be calculated using the following formula:

120

H/E=

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

1 h1

∫0

h1

⎛ H⎞ ⎜ ⎟ ( h)⋅dh ⎝ E ⎠nano

(3)

where, h1, which denoted the upper limit of the definite integrations in formulas (1)–(3), was expressed as the maximum of the indentation depth for nanoindentation measurements. A Rigaku D/Max 2500 VB2/PC X-ray powder diffractometer with monochromatised Cu-Kα radiation from 2θ ¼5° up to 2θ ¼90° was used to characterise the phase structures of the corroded surface layer without cavitation and after cavitation corrosion exposure. The X-ray diffraction (XRD) spectra were analysed using the Jade software, version 5.0. C.T.Kwok found that austenite underwent cavitation, resulting in the transformation of austenite into martensite [30]. The peak areas of the (110) diffraction peak at 44.4° for the transformed martensite and the (111) diffraction peak at 43.5° for the austenite in stainless steel were first calculated on the basis of the respective XRD spectra. Next, the relative percentage of the transformed martensite in stainless steel after the cavitation corrosion tests can be determined. The chemical compositions of the corroded surface layer formed in corrosive media without cavitation and after cavitation corrosion exposures were investigated using X-ray photoelectron spectroscopy (XPS). X-ray photoelectron spectroscopy (XPS) analyses were performed using an ESCALAB 250 instrument with a monochromatic Al-Kα radiation source. An Al-Kα radiation with an acceleration voltage of 1486.6 eV was used at a band-pass energy of 40 eV. Fitting curves was performed using the commercial XPS peak version 4.1 software.

3. Results and discussion 3.1. The nano-mechanical properties of the corroded surface layer It is known that the synergistic effect between cavitation erosion and corrosion plays an important role in cavitation corrosion [2,3]. It was also recognised that the nano-mechanical responses of the corroded surface layer to different levels of synergistic effects were different [20]. Furthermore, note that it is extremely difficult to use conventional tests to measure the nano-mechanical properties (Hnano and Enano) of the corroded surface layer at the nanoscale, including the Vickers hardness measurement [12]. In this present work, nanoindentation technology was used to measure Hnano and Enano of the corroded surface layer of austenitic stainless

steel under different corrosion conditions to understand the effect of corrosion on the nano-mechanical properties of the corroded surface layer. These results are shown in Fig. 2, and the calculated averages (H, E, H/E) of Hnano, Enano and (H/E)nano are listed in Table 1. If austenitic stainless steel in a 3.5% NaCl solution was not subjected to cavitation and only immersed for 1.0 h, then its Hnano displayed an initial slight increase with indentation depth (h), followed by a gradual decrease, as shown in Fig. 2. The Enano increased initially and then was almost independent of the indentation depth. The averages (H, E) of Hnano and Enano were 5.47 GPa and 206.89 GPa, respectively, and they were the same as the received austenitic stainless steel overall, as presented in Table 1. After austenitic stainless steel in a 3.5% NaCl solution was subjected to cavitation for 1.0 h, as shown in Fig. 2, both Hnano and Enano were obviously decreased and became much lower than those without cavitation. When the indentation depth increased further, Hnano and Enano approached those of the substrate. As listed in Table 1, the averages (H, E) of Hnano and Enano decreased to 1.55 GPa and 131.84 GPa, respectively. It is well known that hardness depends on the chemical compositions and microstructures of metals [31], and that the elastic modulus denotes the capability of a metal to resist deformation under compression. In nature, hardness and the elastic modulus depend on the separation, access and shear movement of atoms in lattices [32]. During cavitation corrosion, nonuniform dissolution would occur, generating a large number of vacancies in the corroded surface layer. Thus, the interatomic bonds in the corroded surface layer could be attenuated. Such changes will deteriorate the mechanical properties of metals [15] because the vacancies are attracted to dislocations and increase their mobility. Accordingly, the resistance of metal to plastic deformation is reduced, resulting in a lower hardness. In this case, the surface of the specimen became rougher, but no localised corrosion occurred [33]. The corrosion-induced soft layer was over 2000 nm. However, when specimens, which were polarised in 3.5% NaCl solution at a  0.80 V (SCE) potential, had been subjected to cavitation for 1.0 h, as shown in Fig. 2, both Hnano and Enano initially increased with indentation depth and reached their maxima. Afterwards, they decreased slowly with indentation depth and finally reached the same value as that of the substrate. An approximately 400-nm thick hardened surface layer formed because of the plastic strain deformation induced by the high temperature,

Fig. 2. Hnano and Enano of the corroded surface layer as a function of the indentation depth (h) (a, nanohardness, Hnano; b, nanoelastic modulus, Enano).

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

121

Table 1 H, E and H/E of the corroded surface layer of austenitic stainless steel in different corrosive media without cavitation and under cavitation for 1.0 h. Corrosion conditions

As-received In a 3.5% NaCl solution for 1.0 h without cavitation

In a 3.5% NaCl solution under cavitation for 1.0 h

In a 3.5% NaCl, polarised at  0.80 V (SCE) under cavitation for 1.0 h

In distilled water under cavitation for 1.0 h

H (GPa) E (GPa) H/E

5.50 206.95 0.0266

1.55 131.84 0.0118

5.55 212.70 0.0261

6.18 219.67 0.0281

5.47 206.89 0.0264

Fig. 3. XRD spectra of the corroded surface layers of austenitic stainless steel in different corrosive media without cavitation and after cavitation corrosion exposure (a, in a 3.5% NaCl solution for 1.0 h. without cavitation; b, in a 3.5% NaCl solution under cavitation for 1.0 h; c, in a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h; d, in distilled water under cavitation for 1.0 h).

pressure and impact velocity under cavitation [1,11]. This type of deformation can induce the transformation of austenite into martensite [5,30,34–36]. Furthermore, there was still a thinner corrosion-induced soft layer outside of the hardened layer. Thus, the corroded specimen consists of three sequential layers, an outer corrosion-induced soft layer, a work-hardened layer in the middle and an unaffected layer (the substrate). This result is consistent with the literature [5,15], which reported that a soft layer and hardened layer exists from the surface to the substrate through the microhardness of cross-section of a tested stainless steel specimens after cavitation exposure. Under this condition, the averages (H, E) of Hnano and Enano reached 5.55 GPa and 212.70 GPa, respectively, as listed in Table 1. If austenitic stainless steel in distilled water was subjected to cavitation for 1.0 h, as shown in Fig. 2, the profiles of Hnano and Enano with the indentation depth were the same as those polarised at a  0.80 V (SCE) potential under cavitation. A corroded surface layer with a thickness of approximately 900 nm was hardened, which was thicker than that under cathodic polarisation. Similarly, there was still a thinner corrosion-induced soft layer outside of

this hardened layer. It can also be observed from Table 1 that the averages (H, E) of Hnano and Enano further increased to 6.18 GPa and 219.67 GPa, respectively, indicating that it was easy to form a work-hardened layer when there were no chloride ions in the corrosive media. As listed in Table 1, the averages (H, E and H/E) of Hnano , Enano and (H/E) nano were the highest in distilled water and the lowest in a 3.5% NaCl solution under cavitation. The results further demonstrated that the nano-mechanical responses of the corroded surface layer to different levels of synergistic effects were different. Once corrosion is retarded or weakened, the synergistic effect between cavitation erosion and corrosion will decrease. The workhardened layer forms, resulting in an increment of H, E and H/E of the corroded surface layer for austenitic stainless steel in corrosive media under cavitation. This process can be explained by the generation of non-equilibrium vacancies in the corroded surface layer during the corrosion process [16]. Therefore, it was corrosion that caused the decrement of H and E of the corroded surface layer for austenitic stainless steel in corrosive media under cavitation.

122

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

Fig. 4. Fe 2p3 XPS spectra of the corroded surface layers of austenitic stainless steel in different corrosive media without cavitation and after cavitation corrosion exposure (a, in a 3.5% NaCl solution for 1.0 h. without cavitation; b, in a 3.5% NaCl solution under cavitation for 1.0 h; c, in a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h; d, in distilled water under cavitation for 1.0 h).

3.2. Phase structures in the corroded surface layer Regarding austenitic stainless steel, the Ni equivalent is a parameter that indicates the stability of an austenite phase. In accordance with the equation (Nieq.¼Ni þ0.65Cr þ0.98Mo þ 1.05Mn þ 0.35Si þ12.6C, %) for calculating the Ni equivalent of austenitic stainless steel [17], the Ni equivalent of the austenitic stainless steel used in this study is 24.58, which is in-between 16 and 26, indicating that the austenite of stainless steel used in this study can susceptibly transform into martensite. In this study, XRD analyses of the corroded surface layer for austenitic stainless steel in corrosive media without cavitation and after cavitation corrosion exposure were performed to further study the transformation of the phase structure in the corroded surface layer. The XRD spectra are shown in Fig. 3. If austenitic stainless steel in a 3.5% NaCl solution was not subjected to cavitation and only immersed for 1.0 h, as shown in Fig. 3(a), there were three typical diffraction peaks, 2θ ¼43.5°, 50.6° and 74.5°, that corresponded to lattice plane indices of (111), (200) and (220), respectively. These diffraction peaks represented face-centred cubic (fcc (111)) austenite in the corroded surface layer in accordance with Bragg’s equation [37]. The diffraction peak of (fcc (111)) austenite at 2θ ¼43.5° was very strong, and there was no diffraction peak of (bcc (110)) martensite. In this case, the H/E of the corroded surface layer was 0.0264, basically remaining the same as that of the received austenitic stainless steel, see Table 1.

After austenitic stainless steel in a 3.5% NaCl solution was subjected to cavitation for 1.0 h, austenite remained in the corroded surface layer, as shown in Fig. 3(b). However, XRD peaks at 2θ ¼44.4°, 46.7°, 64.5° and 81.7° appeared, which represented (bcc (110)) martensite, showing that a new phase structure – martensite formed and existed in the corroded surface layer. In this case, the relative percentage of transformed martensite was only 14.0%. Therefore, the H and E of the corroded surface layer were significantly deceased and its H/E was decreased to 0.0118, see Table 1. When specimens in distilled water or in a 3.5% NaCl solution under cathodic polarisation at  0.80 V (SCE) were subjected to cavitation for 1.0 h, their XRD spectra were the same, see Fig. 3 (c) and (d). Similarly, the austenite still existed in the corroded surface layers, but XRD peaks of martensite at 2θ ¼44.4°, 46.7°, 64.5° and 81.7° were obviously observed. The intensity of the (110) diffraction peak at 2θ ¼44.4° for transformed martensite increased, and the intensity of the (111) diffraction peak at 2θ ¼43.5° for the austenite decreased. In addition, the intensity of the (110) diffraction peak at 2θ ¼44.4° for transformed martensite was higher for the sample in distilled water than for the sample in a 3.5% NaCl solution under cathodic polarisation. The relative percentages of the transformed martensite were 56.0% in distilled water and 54.0% in a 3.5% NaCl solution under cathodic polarisation. Therefore, as shown in Table 1, the H/E reached 0.0281 in distilled water and 0.0261 in a 3.5% NaCl solution under cathodic polarisation.

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

123

Fig. 5. Cr 2p3 XPS spectra of the corroded surface layers of austenitic stainless steel in different corrosive media without cavitation and after cavitation corrosion exposure (a, in a 3.5% NaCl solution for 1.0 h without cavitation; b, in a 3.5% NaCl solution under cavitation for 1.0 h; c, in a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h; d, in distilled water under cavitation for 1.0 h).

Martensitic transformation plays a very important role in the cavitation erosion resistance of austenitic stainless steel because martensitic transformation absorbs the energy produced by cavitation and thus reduces cavitation damage [30]. The martensitic transformability of austenite in stainless steels under repetitive cyclic strain by cavitation depends on the temperature. The transformation temperature is an index of the stability and is denoted by Md30. A higher value of Md30 indicates that the austenite of stainless steels is more susceptible to martensitic transformation. For S30400, its Md30 is high and close to room temperature [30]; as a result, it has a high transformability. The above results provided evidence of the transformation of austenite (face-centred cubic (fcc)) into martensite (body-centred cubic (bcc)). This transformation resulted in the formation of a work-hardened layer. Eventually, the H, E and H/E of the corroded surface layer for austenitic stainless steel increased when corrosion was retarded or weakened under cavitation, see Table 1. Therefore, the higher cavitation corrosion resistance of austenitic stainless steel under cathodic polarisation was credited to the improvement of the mechanical properties of the corroded surface layer and not to gas (mainly hydrogen) development [38]. The production of hydrogen did not occur because the cathodic polarised potential was at V (SCE). In this case, the reduction reaction of dissolved oxygen mainly occurred on the surface of the specimen: H2O þO2 þ4e ¼ 4OH  .

3.3. Chemical compositions of the corroded surface layer Based on studying the phase structures of the corroded surface layer of austenitic stainless steel in corrosive media without cavitation and after cavitation corrosion exposure, the chemical compositions of the corroded surface layers were investigated using XPS technology to understand the chemical compositions in the corroded surface layers under cavitation, as shown in Figs. 4–6, where both the metallic and oxidised states of Fe 2p3 and Cr 2p3, including O 1s, are shown. The analyses based on XPS spectra are listed in Tables 2–4. The Fe 2p3 XPS spectra show three types of Fe elements in the corroded surface layers formed in corrosive media without cavitation and under cavitation for 1.0 h (see Fig. 4): Fe (706.5 eV), FeO (709.9 eV) and FeOOH (711.5 eV) . Similarly, Figs. 5 and 6 show the peaks related to Cr 2p3 and O 1s, respectively, in the corroded surface layer formed under the same conditions. Cr 2p3 was detected under three different states: Cr (574.3 eV), Cr2O3 (575.6 eV) and Cr(OH)3 (576.6 eV). O 1s was detected under three different oxidised states: O2  (529.8 eV), OH  (531.8 eV) and H2O (532.9 eV). The quantitative analyses of the elements constituting the corroded surface layer are listed in Table 2. The average chemical compositions were obtained from the integrated peak intensities of O, Fe and Cr. Only the oxidised forms of Fe and Cr elements were considered. A very low amount of Ni element was detected in the specimens [39], indicating that oxides of Fe and Cr are the main components of the passive film layer (outer layer) of the corroded

124

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

Fig. 6. O 1s XPS spectra of the corroded surface layers of austenitic stainless steel in different corrosive media without cavitation and after cavitation corrosion exposure (a, in a 3.5% NaCl solution for 1.0 h without cavitation; b, in a 3.5% NaCl solution under cavitation for 1.0 h; c, in a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h; d, in distilled water under cavitation for 1.0 h).

Table 2 Elementary chemical compositions of the corroded surface layer of austenitic stainless steel in different media without cavitation and under cavitation for 1.0 h (atom %). Conditions

O1s

Fe2p3 Cr2p3 Ni2p

In a 3.5% NaCl solution for 1.0 h without cavitation In a 3.5% NaCl solution under cavitation for 1.0 h In a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h In distilled water under cavitation for 1.0 h

20.2 2.1 18.6 1.8 18.8 1.1

0.9 1.1 1.2

0.0 0.0 0.0

19.5

1.1

0.0

1.9

surface layer. Under cavitation, the amount of Fe decreased and Cr increased, showing that cavitation induces changes of the contents of the metallic elements in the corroded surface layer [39]. Table 3 shows the contents of the metals and their oxides in the corroded surface layer. The chemical compositions of the corroded surface layer formed in a 3.5% NaCl solution without cavitation were mainly FeO, FeOOH, Cr2O3 and Cr(OH)3. Here, the contents of the chemical compositions of the corroded surface layer without cavitation were regarded as the base line, and it was assumed that their relative percentages were zero. It was found that the ratio of Fe oxides to Cr oxides was 5.33 and that the ratio of oxides to hydroxides was 0.76 (see Table 4), further indicating that there were mainly Fe oxides and hydroxides in the corroded surface layer. The anions in the corroded surface layer should be O2  and OH  [40], which was confirmed by the O 1s XPS spectra, see Fig. 6

Table 3 Change rates of the metals and the oxides in the corroded surface layers of austenitic stainless steel in different media under cavitation for 1.0 h compared with that without cavitation (%). Compounds In a 3.5% NaCl solution for 1.0 h without cavitationa

In a 3.5% NaCl In a 3.5% NaCl solution, posolution under cavitation larised at  0.80 V (SCE) for 1.0 h under cavitation for 1.0 h

In distilled water under cavitation for 1.0 h

Fe FeO FeOOH Cr Cr2O3 Cr(OH)3 O2  OH  H2O

 0.5  11.9 60.5  21.1 11.2 7.4 25.7  14.1  8.4

 0.1  16.8 81.6  0.4  4.1 2.3 13.5  11.4 1.8

0 0 0 0 0 0 0 0 0

 28.0 5.9 78.4  21.6  5.1 18.0  13.2  0.1 15.9

a The volumes of chemical compositions of the corroded surface layer without cavitation were regarded as the base line and it was assumed that their relatively changed rates are zero.

(a) and Table 3. In this case, the corroded surface layer formed homogeneously and integrally, and this layer had better mechanical properties, see Table 1. After the specimen in a 3.5% NaCl solution was subjected to cavitation for 1.0 h, the amounts of Fe and Cr metal in the corroded

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

Table 4 Contents of the oxides and hydroxides in the corroded surface layer of austenitic stainless steel in different media without cavitation and under cavitation for 1.0 h. Conditions

Ratio Feox/Crox Ratio ox/ hydrox

In a 3.5% NaCl solution for 1.0 h without cavitation In a 3.5% NaCl solution under cavitation for 1.0 h In a 3.5% NaCl solution, polarised at  0.80 V (SCE) under cavitation for 1.0 h In distilled water under cavitation for 1.0 h

5.33

0.76

3.13

0.97

2.16

0.65

2.54

1.10

surface layer were decreased, see Table 3, indicating that cavitation corrosion occurred. It was also found that FeO decreased, and that FeOOH and both Cr2O3 and Cr(OH)3 increased, demonstrating that under cavitation the transformation of FeO into FeOOH occurred, and that Cr metal initially transformed into Cr2O3 and then into Cr(OH)3. In this case, the corroded surface layer dissolved continuously, and then, the new corroded surface layer formed simultaneously. Thus, 2Men þ nO2  ¼(Me)2On occurred on the surface of substrate. At the same time, (Me)2On reacted with OH  , which came from the corroded surface layer/solution interface, thereby forming Me(OH) n or MeOOH. Next, Me(OH)n or MeOOH further reacted with the absorbed Cl  , forming dissolvable complexes [41]. Finally, as a result of the mechano-chemical effect induced by cavitation, the dissolution of the corroded surface layer was accelerated, resulting in the occurrence of heavy cavitation corrosion. The data in Table 4 reveal that the ratio of Fe oxides to Cr oxides was decreased from 5.33 to 3.13 and that the ratio of oxides to hydroxides was increased from 0.76 to 0.97, indicating that the Cr oxides in the corroded surface layer increased and that the total hydroxides slightly decreased. As a result, the O2  concentration in the corroded surface layer was higher than that of OH  , see Table 3. All of these results indicate that FeOOH dissolved faster than Cr(OH)3 as a result of the complexation of FeOOH with the absorbed Cl  under cavitation. Finally, Cr oxides were enriched in the corroded surface layer. During the cavitation corrosion process, Fe oxides were rapidly formed and then FeOOH was rapidly dissolved, generating a large number of vacancies. The corroded surface layer became inhomogeneous and the grains in the corroded surface layer lost their binders, resulting in the loss of the integrity of the corroded surface layer. The vacancies in the corroded surface layer blocked plastic deformation, resulting in the lower transformed martensite, see Fig. 3(b). Finally, the nanomechanical properties of the corroded surface layer significantly decreased, see Table 1 and Fig. 2. For the corroded surface layer formed in a 3.5% NaCl solution at 0.80 V (SCE) under cavitation for 1.0 h, the amounts of Fe and Cr metal decreased as well(see Table 2), indicating that cavitation corrosion still occurred, even if the specimen was controlled at a cathodic potential ( 0.80 V (SCE)). Both FeO and FeOOH increased, but Cr2O3 decreased and Cr(OH) 3 increased. As listed in Table 4, the ratio of the Fe oxides to the Cr oxides decreased from 5.33 to 2.16 and the ratio of the oxides to the hydroxides decreased from 0.76 to 0.65, indicating that the contents of the Cr oxides and hydroxides in the corroded surface layer increased. The O2  concentrations in the corroded surface layer were lower than that of OH  , see Table 3, further indicating that there was a higher amount of Cr(OH)3 in the corroded surface layer under cathodic polarisation. In this case, cathodic polarisation accelerated the reduction reaction of dissolved oxygen: H2Oþ O2 þ4e ¼ 4OH  , resulting in the enrichment of OH  at the corroded surface-layer/ solution interface, promoting the formation of hydroxides. The

125

metallic oxides formed firstly, and then these metallic oxides transformed into metallic hydroxides. Therefore, a lower cavitation corrosion rate was observed. The corroded surface layer was homogeneous and integrated. As shown in Fig. 3(c), the transformation of austenite (face-centred cubic (fcc)) into martensite (body-centred cubic (bcc)) easily occurred as a result of plastic deformation, resulting in the better mechanical properties of the corroded surface layer, see Table 1 and Fig. 2. Regarding the corroded surface layer formed in distilled water under cavitation for 1.0 h, the amounts of Fe and Cr metal almost remained the same as that in the 3.5% NaCl solution without cavitation, see Table 2, indicating that cavitation corrosion was minimal in distilled water. As listed in Table 3, FeO and Cr2O3 decreased, while FeOOH and Cr (OH)3 increased. Moreover, it can be observed from Table 4 that the ratio of Fe oxides to Cr oxides decreased from 5.33 to 2.54 and that the ratio of oxides to hydroxides increased from 0.76 to 1.10, similarly indicating that Cr was significantly enriched in the corroded surface layer. The main components of the corroded surface layer were FeOOH and Cr(OH)3 when there were no Cl  ions in the media. Under this condition, the corroded surface layer was homogeneous and integrated, similar to the case of the sample under cathodic polarisation. As a result, the mechanical properties of the corroded surface layer performed very well. In summary, under cavitation, the increment of the ratio of Cr oxides to Fe oxides was observed in corroded surface layers formed in different media, including under cathodic polarisation. This observation may explain the beneficial effects of compressive stresses [21], which may further explain the higher dissolution rates of the Fe hydroxides resulting from the synergistic effect between cavitation erosion and Cl  complexation. Under cavitation, chemical compositions change, including the microstructures in the corroded surface layer, can result in the alteration of the nano-mechanical properties of the corroded surface layer, which have a close relationship to the cavitation corrosion resistance of austenitic stainless steel in corrosive media.

4. Conclusions In this work, corrosion was found to dominate the decrement of the H and E of the corroded surface layer for austenitic stainless steel in a 3.5% NaCl solution under cavitation. Once the corrosion was weakened by the replacement of the 3.5% NaCl solution with distilled water or retarded by cathodic polarisation under cavitation, the formation of a work-hardened layer in the corroded surface layer resulted from the transformation of austenite into martensite, which was beneficial for improving the cavitation corrosion resistance of austenitic stainless steel under cavitation. A synergistic effect between cavitation erosion and Cl  complexation accelerates the dissolution rates of metallic hydroxides in the corroded surface layer under cavitation, resulting in the enrichment of Cr in the corroded surface layer. Eventually, the synergistic effect changes the properties of the corroded surface layer, which have a close relationship to the cavitation corrosion resistance of austenitic stainless steel in a 3.5% NaCl solution. Under cavitation, the ways by which corrosion is weakened or retarded are beneficial to protect austenitic stainless steel from cavitation corrosion.

Acknowledgements This work was supported by the National Natural Science Foundation of China (50871011). We acknowledge the support for

126

X. Yong et al. / Materials Science & Engineering A 671 (2016) 118–126

the measurements of the nano-mechanical properties from Prof. Taihua Zhang, State Key Laboratory of Nonlinear Mechanics (LNM), Institute of Mechanics, Chinese Academy of Sciences.

References [1] R.C. Barik, J.A. Wharton, R.J. Wood, K.R. Stokes, Electro-mechanical interactions during erosion – corrosion, Wear 267 (2009) 1900–1908. [2] X. Yong, C. Hou, J. Wu, Z. Zhang, D. Li, Cavitation corrosion behavior of anodized aluminum alloy, Corrosion 67 (9) (2011) 095003-1–095003-5. [3] C.T. Kwok, F.T. Cheng, H.C. Man, Synergistic effect of cavitation erosion and corrosion of various engineering alloys in 3.5% NaCl solution, Mater. Sci. Eng. A 290 (2000) 145–154. [4] ASTM G32-10, Standard Test Method for Cavitation Erosion Using Vibratory Apparatus, ASTM International, West Conshohochen, PA, 2010. [5] C.T. Kwok, H.C. Man, F.T. Cheng, Cavitation erosion and damage mechanism of alloys with duplex structures, Mater. Sci. Eng. A 242 (1998) 108–120. [6] Wang Zaiyou, Zhu Jinhua, Effect of primary factors on cavitation erosion resistance of some metastable austenitic stainless steel, ACTA Metall. Sin. 39 (3) (2003) 273–277. [7] Wang Zaiyou, Zhu Jinhua, Wang Zhangzhong, Analysis on predominant factors characterizing erosion resistance of some ferrous alloys, ACTA Metall. Sin. 43 (6) (2007) 648–652. [8] F.T. Cheng, P. Shi, H.C. Man, Using a modified Knoop indentation technique to estimate the cavitation erosion resistance of NiTi, Mater. Charact. 52 (2004) 129–134. [9] C. Godoy, R.D. Mancosu, R.R. Machado, P.J. Moodenesi, J.C. Avelar – Bastista, Which hardness (nano or macro-hardness) should be evaluated in cavitation? Tribol. Int. 42 (2009) 1021–1028. [10] Alicja K. Kerella, The new parameter to assess cavitation erosion resistance of hard PVD coatings, Eng. Fail. Anal. 18 (2011) 855–867. [11] Alicja Kerella, An experimental parameter of cavitation erosion resistance for TiN coatings, Wear 270 (2011) 252–257. [12] Alicja K. Kerella, An approach to evaluate the resistance of hard coatings to shock loading, Surf. Coat. Technol. 205 (2010) 2687–2695. [13] Alicja Kerella, The influence of TiN coatings properties on cavitation erosion resistance, Surf. Coat. Technol. 204 (2009) 263–270. [14] Yong Xingyue, Ji Jing, Zhang Yaqin, Li Dongliang, Zhang Zhanjia, Quantitative determination of mechanical properties for cavitation corrosion surface layer of metal by micro/nano mechanics measurement technology, J. Chin. Soc. Corros. Prot. 31 (1) (2011) 40–45. [15] H.X. Guo, B.T. Lu, J.L. Luo, Response of surface mechanical properties to electrochemical dissolution determined by in situ nano-indentation technique, Response of surface mechanical properties to electrochemical dissolution determined by in situ nano-indentation technique, Electrochem. Commun. 8 (2006) 1092–1098. [16] Shuji hattori, Ryohei Ishikura, Revision of cavitation erosion database and analysis of stainless steel data, Wear 268 (2010) 109–116. [17] Xingyue Yong, Dongliang Li, Hanjie Shen, Electrochemical responses to degradation of the surface layer nano-mechanical properties of stainless steels under cavitation, Mater. Chem. Phys. 139 (2013) 290–297. [18] M.P. Papadopoulos, C.A. Apostolopoulos, A.D. Zervaki, G.N. Haidemenopoulos, Corrosion of exposed rebars, associated mechanical degradation and correlation with accelerated corrosion tests, Constr. Build. Mater. 25 (8) (2011) 3367–3374.

[19] Zhang Ru, Shen Hanjie, Zhang Yaqin, Li Dongliang, Li Yanjia, Yong Xingyue, Surface layer nano-mechanical responses to interaction between cavitation and electrochemical corrosion, Acta Metall. Sin. 49 (5) (2013) 614–620. [20] V. Vignal, O. Delrue, O. Heintz, J. Peultier, Influence of the passive film properties and residual stresses on the micro-electrochemical behavior of duplex stainless steel, Electrochim. Acta 55 (2010) 7118–7125. [21] L.Q. Guo, M.C. Lin, L.J. Qiao, Alex A. Volinsky, Duplex stainless steel passive film electrical properties studied by in-situ current sensing atomic force microscopy, Corros. Sci. (2013), http://dx.doi.org/10.1016/j.corosci.2013.031. [22] Tingting Zhao, Yan Li, Yan Xiang, Xinqing Zhao, Tao Zhang, Surface characteristics, nano-indentation and corrosion behavior of Nb implant NiTi alloy, Surf. Coat. Technol. 205 (2011) 4404–4410. [23] Yuxi Zhao, Hong Dai, Weiliang Jin, A study of the elastic moduli of corrosion products using nano-indentation techniques, Corros. Sci. 65 (2012) 163–168. [24] Y. Kusano, I.M. Hutchings, Analysis of nano-indentation measurements on carbon nitride films, Surf. Coat. Technol. 169–170 (2003) 739–742. [25] ISO 14577-1:2007, Metallic Materials _ Instrumented Indentation Test for Hardness and Materials Parameters – Part 1: Test Method, International Standard Organisation, 1214 Vernier, Geneva,Switzerland, 2007. [26] GB/T 22458-2008, 2008. General Rules of Instrumented Nano-indentation Test. General Administration of Quality Supervision, Inspection and Quarantine of P.R. China. [27] Zhang Taihua, Testing Technologies & Applications of Micro/Nano-Mechanics, China Machine Press, Beijing, 2005. [28] W.C. Oliver, G.M. Pharr, Improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (6) (1992) 1564–1583. [29] X. Li, B. Bhushan, A review of nanoindentation continuous stiffness measurement technique and its applications, Mater. Charact. 48 (1) (2002) 11–36. [30] C.T. Kwok, H.C. Man, F.T. Cheng, Cavitation erosion of duplex and super duplex stainless steels, Scr. Mater. 39 (9) (1998) 1229–1236. [31] Zhou Yichun, Zheng Xuejun, The Macro/Micro Mechanical Properties of Materials, first ed., Higher Education Press, Beijing, 2009. [32] Zhao Xinbing, Ling Guoping, Qian Guodong, Material Properties, first ed., Higher Education Press, Beijing, 2009. [33] Yong Xingyue, J.I. Jing, Zhang Yaqin, L.I. Dongliang, Zhang Zhanjia, AFM morphology and cavitation corrosion process of austenitic stainless steel by cavitation, Corros. Sci. Prot. Technol. 23 (2) (2011) 116–120. [34] J.F. Santa, J.A. Balnco, J.E. Giraldo, A. Toro, Cavitation erosion of martensitic and austenitic stainless steel welded coatings, Wear 271 (2011) 1445–1453. [35] Diana Lopez, Neusa Alonso Falleiros, Andre Paulo Tschiptschin, Effect of nitrogen on the corrosion-erosion synergism in an austenitic stainless steel, Tribol. Int. 44 (2011) 610–616. [36] Lei, Yu-cheng., Feng, Liang-hou., Zhao, Xiao-jun., 2006. Cavitation erosion behaviour of an austenitic stainless steel. J. Jiangsu University, Natural Sci. Edition. Vol. 27, pp. 241–244. [37] Y.X. Yang, R. Qi, X-ray Diffraction Analysis, first ed., Shanghai Jiaotong University Press, Shanghai, 1989. [38] J.G. Auret, O.F.R.A. Damm, G.J. Wright, F.P.A. Robinson, Influence of cathodic and anodic currents on cavitation erosion, Corrosion 49 (11) (1993) 910–920. [39] O. Lavigne, Y. Takeda, T. Shoji, K. Sakaguchi, Water irradiation by high-frequency ultrasonic wave: effects on properties of passive film formed on stainless steel, Ultrason. Sonochem. 18 (2011) 1287–1294. [40] Yong Xingyue, Liu Jingjun, Lin Yuzhen, EIS of duplex stainless steel in flowing corrosive media, J. Chem. Ind. Eng. 54 (12) (2003) 1713–1718. [41] Lin Yuzhen, Yang Dejun, Fundamental of Corrosion & Control, first ed., Sinopek Press, Beijing, 2007.