The contribution to pitting corrosion from chloride is a well-accepted agreement, however, the mechanism and the relationship between the presence of chloride and the generation of pitting corrosion in different cases is still remains to be studied. Likewise, the study of the inhibition mechanism of corrosion inhibitors is also deficient. Spectroscopic methods such as Raman and ellipsometry have the advantage of being surface sensitive and non-destructive characterization processes. It gives one a chance to investigate the dynamics of localized corrosion on steel in high alkaline solution (pH=12.5) in the presence of chloride ions. An industrial example of such an environment is reinforced concrete.

In this study, these spectroscopic techniques are combined with electrochemical measurements such as open circuit potential (OCP; Figure 1) and electrochemical impedance spectroscopy (EIS; Figure 2) to analyze the corrosion tendency of attack by chloride at different concentrations at pH12.5. It is found that for chloride concentrations higher than 50mM the OCP spontaneously decreases, indicating an irreversible destruction of the passive film formed in the alkaline solution. The EIS data could be fitted to a simple one time constant equivalent circuit in either the Bode or Nyquist formats (Figure 3). The constant phase element CPE is linked the capacitance by:

Z_CPE=[C_ox(jω)ⁿ]^-1

It is shown that the capacitance, C_ox, increased with the increasing chloride concentration, possibly indicating a thicker oxide layer on the carbon steel sample surface. EIS also revealed that the corrosion rate increased with increasing chloride content, as indicated by a reduction in the charge transfer resistance, R_ct (Figure 3).

SEM, Raman spectroscopy and ellipsommetry were applied for surface characterization, with the in-situ Raman and ellipsometry cells shown in the Figure 4. SEM revealed that when NaCl was introduced to the alkaline solution, the sample surface obtained a higher roughness after 24 h immersion test. For Raman, because of the thinness of the corrosion product layer, it was first necessary to electrochemically deposit to induce the surface enhanced Raman (SERS) in order to collect spectra from the carbon steel in pH=12.5 alkaline solution with or without NaCl (Figures 5 & 6). The locations of Raman peaks i of selected iron oxides, hydroxides and oxyhydroxides are listed in the table (Ref 1 & 2), with the most intense peak being bold.

Component		Raman shift (cm-1)
α-FeOOH	290
β-FeOOH	400	720
γ-FeOOH	252	380	526	650
δ-FeOOH	663(braod)
Fe(OH)2	460	545
Fe(OH)3	692(broad)
Fe3O4	550	670

It was found that in the absence of chloride the main components of the passive are Fe(OH)₂, Fe₃O₄, γ & δ – FeOOH. However, on addition of chloride, the γ - FeOOH in the oxide layer is converted to β-FeOOH. The in-situ ellipsometry data allowed the thicknesses of the films to be determined and it was found that the oxide layer thicken by around 20% on the addition of chloride to the alkaline solution. It those concluded that as the chloride ions convert γ - FeOOH converted to β-FeOOH this both thickens and destabilizes the passive film.