Role of S2− ions on the microstructure change and the pitting behaviour of aluminum in saline solution

The electrochemical behaviour and the passive film microstructure of aluminum during its exposure to 3.5 wt% NaCl solution in the absence and presence of S2− ions are investigated using potentiodynamic polarization curves, electrochemical impedance spectroscopy measurements, XRD, XRF, SEM and AFM. Electrochemical measurements show that the presence of S2− ions enhances the uniform corrosion of aluminum in NaCl solution, but delay its susceptibility to the pitting corrosion. In addition, EIS analysis illustrate that the formation of more compact and protective passive layer in the presence of S2− ions compared to its rough surface in the absence of S2− ions as evidenced by the lower value of constant phase element (CPE) and higher value of phase shift (N). Cracks, non- homogenous and open large pits with high degree of roughness are clearly observed on the aluminum surface in the absence of S2− ions, compared to oriented grooves, elongated ridges with the accumulation of the corrosion products inside the pits in the presence of S2− ions. The inhibitory effect of S2− ions for the pitting corrosion of aluminum is interpreted on the basis of the change in its microstructure of the passive film in the absence and presence of S2−ions.

Aluminum and its alloys are widely used in industrial applications, because of their good mechanical and industrial properties as well as, their highly corrosion protection [1][2][3][4] . Aluminum when immersed in chloride solutions, it will expose to the pitting corrosion, that limits its industrial and marine applications 5 . The pitting corrosion of 7A60 aluminum alloy in 3.5 wt% NaCl solution was investigated using electrochemical impedance spectroscopy, electrochemical noise, scanning electron microscopy and energy dispersive spectrometer 6 . Results revealed that the intense pitting corrosion in 7A60 aluminum alloy is explained to the presence of electrochemical active MgZn 2 , Al 2 MgCu and Mg 2 Si, while the contribution of Al 7 Cu 2 Fe to the pitting corrosion is little. Zhang et al. 7 studied the relationship between the intergranular corrosion and the crystallographic pitting in AA2024-T351 aluminum alloy in 3.5 wt% NaCl solution. It was found that the intergranular corrosion occurs firstly, and the crystallographic pitting initiates from the crevice wall behind the intergranular corrosion front.
Sulphide polluted saline solutions are very aggressive medium for the corrosion of many metals, that used in many industrial installations processing sulfides. Corrosion behaviour of Cu-Al-5Ni alloy in 3.5 wt% NaCl solution containing different concentrations of sulphide ions was studied using electrochemical impedance spectroscopy measurements 8 . Data showed that, increasing the sulpide ion concentrations leads to increase the corrosion rate of this alloy that can be explained on the basis of the presence of Cu 2 S that decrease the protective efficiencies of Cu 2 O film. Dissolution of aluminum in 0.01 M NaOH solution as a function of sulphate, nitrate and sulphide ion concentrations was evaluated using electrochemical noise measurements 9 . It was determined that the corrosion rate of aluminum in the alkaline solution was occurred more seriousness in the presence of these additives, as concluded from the positive shift in the frequency to more highly frequency region. Sherar et al. 10 , studied the effect of the sulphide ions on the aerobic corrosion of steel in near neutral saline solution using scanning electron microscopy, energy dispersive X-ray analysis and Raman spectroscopy. They found that the addition of the sulphide ions to the aerobically-exposed surface doubles the corrosion rate compared to its value in the sulphide free solution. Electrochemical performance of Cu, Cu-10Al-10Ni, Cp-Ti and C-steel in 3.5 wt% NaCl containing 2 ppm Na 2 S was studied 11 . Results clarified that the addition of S 2− , shifted the corrosion potential for all alloys to more negative values except Cp-Ti that be shifted to more positive.
Morphology and microstructure of the passive film play an important role for its mechanical and electrochemical properties 12 . Liu et al. 13 studied the effect of the abrasion on the microstructural and pitting corrosion of 2297Al-Li alloy in borate-buffered 0.001 M NaCl. Data revealed that the pitting corrosion of this alloy was strongly affected by the size, the population density and the area fraction of intermetallic particles as well as its smoother surface has lower pitting ability and better corrosion resistance compared to the rougher surface. In addition, the crystallographic orientations within a microstructure was observed to be considerable effect on pitting corrosion. It was observed the relation between the polarization and pitting behaviors with the surface orientation in pure aluminum and aluminum alloys [14][15][16][17][18][19][20] .  www.nature.com/scientificreports www.nature.com/scientificreports/ Effect of the microstructure on the corrosion behaviour of cast Mg-Al alloys in 5 wt. % NaCl solution saturated with Mg(OH) 2 was studied 21 . Data showed that the corrosion protection of homogeneous α-phase increases with increasing Al-content, due to the higher chemical stability of α-phase with higher Al contents compared to its stability with lower Al contents. The effect of heat treatment on the electrochemical corrosion and microstructural behaviour of API X70 line pipe steel in sea water containing thiosulphate solutions was evaluated 22 . Data implies that the presence of different phases of ferrite during heat treatment. Polygonal ferrite and fine grained ferrite microstructures formed at 600 °C inhibits the corrosion rate compared to the tempered martensitic microstructure formed at 300 °C. The inhibition effect was explained on to the difference in the diffusion rate of the corrosion product across the fine and coarse grained microstructure. Corrosion protection of PEO coating on extruded Al6Cu alloy in 3.5% wt NaCl was evaluated 23 . The corrosion protection of Al6Cu alloy was enhanced in the presence of PEO coating compared to as-cast alloy due to the change in its microstructure.
Yang et al. 24 studied the effect of sulphide ions on the passivation behaviour of titanium TA2 alloy in simulated seawater solutions. They found that the passive film was composed of TiO 2 top layer with TiO sub-oxide layer and the highly resistance of Ti substrate was correlated to the presence of TiO 2 top layer. In addition, the presence of low concentrations of highly amount of TiO 2 . While at highly concentration of S 2− ions (>2 mM/L) decreased the corrosion resistance due to the highly fraction of TiOS and TiS 2 compared to TiO 2 . The influence of sulphide ions on the passive behaviour of super 13Cr martensitic stainless steel in borate/NaCl solution was investigated 25 . Results showed that the addition of sulphide ions increased the values of the corrosion current and shifted the pitting potentials to more noble values. Moreover, the presence of sulphide ions changed the microstructure and the crystallinity of the passive layer.
By far, to our knowledge, little work reported the effect of the microstructure of the passive film on the pitting behaviour of aluminum in sulphide polluted saline solution. Therefore, the goal of our present work is to investigate the influence of the microstructure of the oxide passive film on the uniform and pitting corrosion behaviour of aluminum in 3.5 wt% NaCl solution in the absence and presence of different concentrations of Na 2 S using atomic force microscopy, scanning electron microscopy, X-ray fluorescence, X-ray diffractometer, potentiodynamic polarization and electrochemical impedance spectroscopy techniques.

Experimental
Materials. Sodium chloride, sodium sulphide and acetone are provided from Merck Chemical Co.
(Germany). Working electrode is made from aluminum rod with the following chemical composition (wt%): 99.57% Al, 0.31% Fe, 0.07% Si, 0.015% Ti%, 0.0016% Zn, 0.0003% Cr, 0.0019% Mg, 0.0021% Mn and 0.0007 Cu.  Table 2. XRF analysis of aluminum surface immersed in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S solutions for 24 hours at 30 °C.  www.nature.com/scientificreports www.nature.com/scientificreports/ performed with soft cantilevers: n + -doped Si cantilever from Nanosensors (PPP-CONT-10), k n = 0.09 N m −1 . Scanning electron microscopy is achieved using JSM-6510LA (JEOL, Tokyo, Japan). Surface composition, crystallinity and phase identification of aluminum samples are investigated using X-ray fluorescence (ARL TM Perform'x Sequential XRF spectrometer, Thermo Scientific) and X-ray diffractometer (PANalytical Empyrean, Netherlands). electrochemical measurements. All the electrochemical measurements are carried out using the Potentiostat/Galvanostat (AUTOLAB PGSTAT 128 N), using a standard three-electrode cell with Al (1.0 cm 2 ) working, saturated Ag/AgCl reference and Pt sheet (1.0 cm 2 ) counter electrodes. NOVA 1.10 software is used to records and fits the electrochemical measurements. Potentiodynamic polarization measurements are achieved in the potential range between −100 to 200 mV vs. E OCP values at 30 °C with the scan rate of 1.0 mV s −1 . Electrochemical impedance spectra at the respective E OCP values are recorded using AC signals of amplitude 5 mV peak to peak in the frequency range of 10 kHz to10 MHz. Prior to each experiment, working electrode is polished successively with fine grade emery papers, cleaned with acetone, washed with bi-distilled water and finally dried.

Results and Discussion
electrochemical measurements. Polarization measurements. Figure 1a represents the potentiodynamic polarization measurements of aluminum electrode in x wt% NaCl solution (x = 0.5-3.5), while its electrochemical parameters derived from the polarization curves (E corr , E pit , I corr , β a , β c ) are listed in Table 1. Results show that the presence of active/passive/pitting as the electrochemical behaviour of aluminum for all studied concentrations. It is observed that, increasing the concentrations of Cl − ions, shifts both the corrosion potential (E corr ) and the pitting potential (E pit ) to more negative values as well as increases the values of the corrosion current densities (I corr ) from 0.6 μA cm −2 in case of 0.5 wt% NaCl solution to 2.1 μA cm −2 for 3.5 wt% NaCl solution. This finding explains that the rate of both uniform and pitting corrosion of aluminum increases as a results of the aggressive attack of Cl − ions. Mechanism of the anodic and cathodic reactions associated with the uniform corrosion of aluminum in Cl − ion solutions based on the previously reported is explained as follows 26-28 : 1 Cathodic reaction:   Moreover, the pitting corrosion of aluminum is explained on the basis of the reaction between the adsorbed Cl − ions with Al 3+ in the crystal lattice of the dual nature Al 2 O 3 passive film, that consists of an adherent, compact and stable inner film covered with a porous, less stable outer film to form soluble oxychloride complex as shown in Eq. (5) 29,30 . The presence of the adsorbed chloride ions on the Al 2 O 3 passive film is confirmed using XRF analysis as shown in Table 2. Our results are in a good agreement with Soltis 20 and Tang et al. 31 , who reported that the pitting potential (E pit ) was shifted to more negative values with increasing the concentration of the aggressive ions according to the following relation: pit where A and B are constants, whilst A measures the aggressiveness of the chloride ions at a given concentration. Explanation of this trend can be illustrated as follows: At low Cl − concentration, Al 3+ ions in the crystal lattice of the passive layer prefers to hydrolyze with OH − rather than reacts with the Cl − ions, therefore the growth and the propagation of the pits are decreased 27 . At high Cl − concentration, Al 3+ ions react with Cl − ions producing soluble complex, that accelerates both the nucleation and the propagation of the pits 29,31 . Figure 1b represents the effect of Na 2 S concentrations on the anodic and cathodic polarization curves of aluminum and the detailed electrochemical parameters are listed in Table 3. The results imply that the uniform corrosion of aluminum increases with increasing the S 2− ion concentrations without any sign for pitting corrosion within the studied polarization range. Mechanism of the uniform corrosion of aluminum in S 2− solution can be summarized as previously reported 8,32-34 , S 2− ions is hydrogenated to give HS − , that reacts with aluminum to produces Al 2 S 3 and finally Al 2 O 3 according to the following equations:    The presence of Al 2 O 3 as a passive film instead of Al 2 S 3 is confirmed using XRF analysis (c.f. Table 2). On the other hand, the production of OH − during the hydrogenation of S 2− ions as shown in Eq. (7) increases the solution pH, that explains the enhancement of the uniform corrosion rate of aluminum with increasing the concentration of S 2− ions.
Addition of different concentrations of Na 2 S (1-5 ppm) to 3.5 wt% NaCl solution, enhances the uniform corrosion of aluminum, compared to 3.5 wt% NaCl and 5 ppm Na 2 S solutions as indicated from the values of both E corr and I corr as shown in Tables 1, 3 and 4. In contrast, the presence of Na 2 S in NaCl solution inhibits the pitting corrosion of aluminum as can be seen from Fig. 1c. The inhibition effect of S 2− ions on the pitting corrosion of aluminum can be seen from the decreasing in the values of the I pit and increasing the difference between the values of E corr and E pit (ΔE = E corr − E pit ) with increasing S 2− ion concentrations (c.f. Fig. 1c).
Electrochemical impedance spectroscopy measurements. Electrochemical impedance spectroscopy measurements are performed to evaluate the Al/electrolyte interface with different electrolyte composition. Figure 2 represents the Nyquest plots of aluminum surface recorded at E OCP in the presence of 3.5 wt% NaCl, 5 ppm Na 2 S and mixture of 3.5 wt% NaCl + 5 ppm Na 2 S solutions. Results point out that the presence of two consecutive capacitive semicircles in 5 ppm Na 2 S solutions, that have been interpreted to the dual nature of Al 2 O 3 passive film including an inner compact layer covered with outer porous layer 35 . Whereas a single semicircle is observed in both 3.5 wt% NaCl and 3.5 wt% NaCl + 5 ppm Na 2 S solutions. This observation has been explained as a results of the adsorption of Cl − ions on the Al 2 O 3 passive layer, which reacts with Al 3+ in its crystal lattice, forming soluble www.nature.com/scientificreports www.nature.com/scientificreports/ oxychloride complex as discussed before, leads to the dissolution of the outer porous layer. Also, the deviations of Nyquest plots in cases of 3.5 wt% NaCl and 3.5 wt% NaCl + 5 ppm Na 2 S solutions from a perfect circular shape refers to frequency dispersion of interfacial impedance arising from the inhomogeneity of the electrode surface due to roughness phenomena 36 . On the other hand, the enhancement in the uniform corrosion of aluminum in 3.5 wt% NaCl + 5 ppm Na 2 S solution compared to 3.5 wt% NaCl and 5 ppm Na 2 S solutions is observed from the semicircle diameters as shown in the Fig. 2. Figure 3 represents the fitted electrochemical equivalent circuit as a function in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S solutions. It can be seen that the Nequest plot of aluminum in 5 ppm Na 2   www.nature.com/scientificreports www.nature.com/scientificreports/ electrolyte interface in 3.5 wt% NaCl + 5 ppm Na 2 S solution behaves as a more ideal capacitive rather than that in 3.5 wt% NaCl, which illustrates the role of the S 2− ions in decreasing the roughness and increasing the homogeneity of the aluminum surface.
It is concluded from the electrochemical measurements that, even the presence of S 2− ions in NaCl solution enhances the uniform corrosion of aluminum but delay its susceptibility to pitting corrosion. This finding can explained on the bases of the accumulation of the corrosion products on the aluminum surface that increases the stability and the protectionist of Al 2 O 3 passive layer and consequently, inhibits the rate of the pitting corrosion.
Microstructures of the aluminum surface. Figure 4 and Table 2 represents the surface composition, crystallinity and phase identification of as received aluminum surface immersed in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S using X-ray fluorescence and X-Ray Diffraction analysis. Data shows that the phase composition of the passive layer is Al 2 O 3 in all studied solutions, with 3.23, 0.154 and 2.127 µm average crystal size formed in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S solutions respectively. SEM images for as received aluminum surface immersed in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S solutions are represented in Fig. 5. Cracks, non-homogenous and open large pits with high degree of roughness are clearly observed on the aluminum surface exposed to 3.5 wt% NaCl solution in the absence of S 2− ions, compared to oriented grooves, elongated ridges with the accumulation of the corrosion products inside the pits in the presence of S 2− ions. Whilst, an intact surface with no cracks or pits are observed in case of 5 ppm Na 2 S free Cl − ions solution.
Topography of the aluminum surface immersed in 3.5 wt% NaCl, 5 ppm Na 2 S and 3.5 wt% NaCl + 5 ppm Na 2 S solutions is represented in Fig. 6, while its morphological parameters (RMS and R a ) are tabulated in Table 5. Data shows that the extensive pitting corrosion throughout the aluminum surface exposed to 3.5 wt% NaCl solution www.nature.com/scientificreports www.nature.com/scientificreports/ in the absence of S 2− ions is indicated by the large values of both root mean square roughness (RMS) and average roughness (R a ) compared to their values in the presence of S 2− ions. This behaviour is in a good agreement with the electrochemical measurements.
The enhancement in both the roughness and the homogeneity of aluminum surface in the presence of S 2− ions is due to the formation of smoother, fine crystalline sized Al 2 O 3 passive layer that accumulates inside the pits and inhibits their propagation, therefore decreases the susceptibility of the aluminum surface towards aggressive pitting corrosion in NaCl solution as schematically represented in Fig. 7. conclusion It can be concluded that the electrochemical behaviour and the microstructure of the passive film of aluminum during its exposure to 3.5 wt% NaCl solution is strongly affected by the presence of S 2− ions. Increasing the concentrations of S 2− ions in 3.5 wt% NaCl solution increases the pH values of the saline solution, thus increasing both the anodic reaction of Al dissolution and the dissolution kinetics of Al 2 O 3 passive layer resulting in an increase of the uniform corrosion. At the same time, the lower kinetic formation of smoother, compact, fine crystalline sized Al 2 O 3 inhibits the pitting corrosion. The inhibitory effect is interpreted on the basis that the smooth, fine crystalline surface of Al 2 O 3 increases its stability and decreases the aggressiveness of the Cl − ions. Consequently, decreases the rate of the initiation and the growth of the stable pits.

Data Availability
The dataset generated or analyzed during the current study are available from the corresponding author upon request.