Corrosion of dissimilar alloys: Electrochemical noise

A new approach for interpreting the electrochemical noise generated by freely corroding electrodes is presented. For a single electrode, with well-defined anodic and cathodic regions, it is proposed that the instantaneous values of the corrosion current and open circuit potential are functions of the time-invariant potential difference between the anodic and cathodic reactions and the time-dependent impedances associated with the anodic and cathodic reactions. The fluctuations in the values of anodic and cathodic impedances are determined by the interaction between the aggressive environment and the oxide, hydroxide or adsorbed surface layer covering the anodic and cathodic regions; the instantaneous values of impedances modulate the corrosion current. The mathematical implications of these assumptions are discussed and the results are compared with the traditional approach, based on time-dependent current sources acting on constant impedances. Subsequently, the findings for a single electrode are transferred to two dissimilar electrodes and an experimental technique that enables the estimation of the time evolution of anodic and cathodic resistances for each of the dissimilar electrodes is presented.     

The significance of electrochemical noise measurements on asymmetric electrodes

The technique of electrochemical noise measurement has now reached a relatively mature state, in that the methods required to make reliable measurements are well-understood, and some aspects of the analysis of such measurements have achieved general acceptance. Thus, the analysis of the resultant data in terms of the noise resistance, R_n, is widely accepted, and methods for the extraction of information about corrosion type are starting to become available.One requirement of most of the analysis methods is the assumption that the two electrodes used to determine the current noise are similar. This assumption is always something of a concern, and a number of methods have been proposed to enable the simultaneous or sequential determination of both potential and current noise associated with a single electrode, so that the assumption of similarity is not necessary. This paper is concerned with methods that use deliberately asymmetric electrodes.The normal justification for using deliberately asymmetric electrodes is to permit the study of what is happening on one of the two electrodes. The situation that applies when two asymmetric working electrodes are coupled together and the current and potential noise determined has been analysed previously. Unfortunately, however, the significance of this work has not generally been appreciated by those groups using asymmetric electrodes. In part this may be because the prior analysis did not explicitly consider the theoretical and experimental justification claimed for this type of experiment, and this paper aims to review this analysis, and to rationalise the claims and observations of previous workers.     

Electrochemical noise and impedance study of aluminium in weakly acid chloride solution

The electrochemical noise (EN) characteristics of pure aluminium in unbuffered potassium chloride solution and with acetic acid/sodium acetate buffer at pH 5.4 and 4.3 have been analysed to throw light on the influence of pH and of the presence of buffer at the aluminium surface on chloride ion-induced corrosion. Comparison has been made with results obtained by electrochemical impedance spectroscopy (EIS) and quantitative deductions made concerning the values of the noise resistance and the magnitude of the electrochemical impedance. Deviations between results obtained by the two experimental techniques are discussed.     

Electrochemical characterization of different screen-printed gold electrodes

Voltammetric characterization of electrochemical systems is very important with the aim of optimising a required application. In this case, different gold screen-printed electrodes have been evaluated. Although the use of carbon has widespread, gold is still in the beginning. Two different substrates: polycarbonate and alumina have been evaluated. Differences in morphology are clear when observed by SEM. Potassium ferricyanide, p-aminophenol, indigo carmine, silver nitrate and ferrocene have been chosen as model analytes. Potential scans can be made at scan rates as high as 2000 mV s⁻¹ which could result very advantageous. Proportionality between peak current and concentration maintains at this scan rate as demonstrated with ferricyanide. Sensitivity is increased and linear dynamic ranges are widened when alternative electrochemical techniques (SWV and ACV) are employed. In the silver system in basic media, catalysis-mediated reduction is only possible on alumina electrodes since the potential window is very narrow in polymeric electrodes. The precision (intra and interelectrode) has been thoroughly evaluated. Reusability is possible since adequate RSDs are obtained when measurements are performed in different drops without cleaning in-between. On the other hand, low cost of fabrication makes disposability of electrodes an advantageous characteristic that avoids tedious cleaning treatments.     

Electrochemical noise analysis of the corrosion of AZ91D magnesium alloy in alkaline chloride solution

The corrosion behavior of AZ91D magnesium alloy in alkaline chloride solution was investigated by electrochemical noise (EN). The wavelet transform, as well as noise resistance (R_n) and power spectral density (PSD), had been employed to analyze the EN data. It was revealed that there exist three different stages of corrosion for AZ91D magnesium alloy in alkaline chloride solution, including an anodic dissolution process companying with the growth, absorption and desorption of hydrogen bubbles, a development of pitting corrosion, an inhibition process by protective MgH₂ film. The results demonstrated that energy distribution plot (EDP) was a powerful tool to provide useful information about the dominant process for the different corrosion stages.     

Electrochemical noise analysis of O2 evolution on PbO2 and PbO2-matrix composites containing Co or Ru oxides

Electrochemical noise measurements have been performed on various electrode materials during oxygen evolution reaction. Fluctuations of both electrode potential and electrolyte resistance, recorded under galvanostatic polarization, have been analysed in order to separate ohmic and non-ohmic contributions. The anode materials compared in this study were PbO₂ and composite oxides consisting of a PbO₂ matrix and variable amounts of catalytic dispersed phases (Co₃O₄, RuO₂ and hydrous Co oxides); the composite oxide layers were generally obtained by anodic codeposition of fine particles (typically 0.1-1 μm in size) with a PbO₂ matrix electrochemically grown by Pb²⁺ oxidation. The power spectral density (PSD) of the electrolyte resistance fluctuations is very similar for PbO₂ and composite electrodes, suggesting a similar gas evolution profile. On the contrary, the potential PSD is markedly different for PbO₂ and composites containing catalytic particles. In the latter case, the potential fluctuations are entirely due to ohmic effects, while non-ohmic components are clearly dominant in the case of O₂ evolution on pure PbO₂, probably because of the much higher activation overpotential. Furthermore, on PbO₂ electrodes a plateau was obtained at intermediate frequencies (0.1-10 Hz), which was tentatively explained by bubble coalescence phenomena.     

Electrochemical potential noise of 321 stainless steel stressed under constant strain rate testing conditions

An investigation was carried out on the features of potential electrochemical potential noise of stressed AISI321 stainless steel in solution 0.5 mol/L Na₂SO₄ + 5 × 10⁻³ mol/L H₂SO₄ under constant-strain-rate-testing (CSRT) conditions. The results showed that the strained steel exhibited a white noise feature at low frequencies and the amplitude of potential fluctuation depended on elongation of the steel. Power spectral density (PSD) of the noise increased with increasing strain level. The noise level in the elastic region of the steel was relatively low, which increased with elongation. After the steel yielded, the electrochemical noise level became higher but it increased less significantly with increasing strain. In the fracture region, the potential noise reached the maximum level. In addition to the dependence of electrochemical noise on strain level, it was also found that the electrochemical noise level increased with increasing strain rate. To interpret the generation mechanism of electrochemical noise, a simple model was proposed based on an assumption that strain results in breakdown and repairing of the passive film on the steel. With this model, the dependence of electrochemical potential noise and its power spectral density on strain level and strain rate can be successfully explained.     

Electrochemical noise analysis of cathodically polarised AISI 4140 steel. I. Characterisation of hydrogen evolution on vertical unstressed electrodes

For investigating stressed carbon-steel electrode (AISI 4140) cracking induced by hydrogen embrittlement with the electrochemical noise (EN) technique, the contribution of the evolving hydrogen bubbles to the EN has to be determined. Under cathodic current control, various reasons may explain the fluctuations of the electrode potential. In the first paper of this series the electrolyte resistance (ER) fluctuations, which yield ohmic-drop fluctuations, were analysed for unstressed vertical tensile specimens. A simplified model was proposed to tentatively derive the characteristic parameters of the gas evolution, such as the bubble mean size and evolution mean rate, from the power spectral density (PSD) of the ER fluctuations. However, results in only qualitative agreement with optical observations were obtained, indicating that the complicated screening and dragging effects of rising bubbles have to be taken into account in the modelling for vertical electrodes. The electrode potential fluctuations of unstressed tensile specimens were investigated in the second paper of this series and the influence of a stress was examined in the third paper.     

Electrochemical reduction of trinitrotoluene on core-shell tin-carbon electrodes

In this work, we studied the electrochemical process of 2,4,6-trinitrotoluene (TNT) reduction on a new type of electrodes based on a core-shell tin-carbon Sn(C) structure. The Sn(C) composite was prepared from the precursor tetramethyl-tin Sn(CH₃)₄, and the product contained a core of submicron-sized tin particles uniformly enveloped with carbon shells. Cyclic voltammograms of Sn(C) electrodes in aqueous sodium chloride solutions containing TNT show three well-pronounced reduction waves in the potential range of −0.50 to −0.80 V (vs. an Ag/AgCl/Cl⁻ reference electrode) that correspond to the multistep process of TNT reduction. Electrodes containing Sn(C) particles annealed at 800 °C under argon develop higher voltammetric currents of TNT reduction (comparing to the as-prepared tin-carbon material) due to stabilization of the carbon shell. It is suggested that the reduction of TNT on core-shell tin-carbon electrodes is an electrochemically irreversible process. A partial oxidation of the TNT reduction products occurred at around −0.20 V. The electrochemical response of TNT reduction shows that it is not controlled by the diffusion of the active species to/from the electrodes but rather by interfacial charge transfer and possible adsorption phenomena. The tin-carbon electrodes demonstrate significantly stable behavior for TNT reduction in NaCl solutions and provide sufficient reproducibility with no surface fouling through prolonged voltammetric cycling. It is presumed that tin nanoparticles, which constitute the core, are electrochemically inactive towards TNT reduction, but Sn or SnO₂ formed on the electrodes during TNT reduction may participate in this reaction as catalysts or carbon-modifying agents. The nitro-groups of TNT can be reduced irreversibly (via two possible paths) by three six-electron transfers, to 2,4,6-triaminotoluene, as follows from mass-spectrometric studies. The tin-carbon electrodes described herein may serve as amperometric sensors for the detection of trace TNT.     

Electrochemical noise analysis of cathodically polarised AISI 4140 steel. III. Influence of hydrogen absorption for stressed electrodes

In this third part of the study of the electrochemical noise generated by hydrogen evolution on cathodically-polarised AISI 4140 steel specimens in 0.5 M sulphuric acid, a tensile stress was applied to the specimens. The two components of the measured potential fluctuations ΔV, namely ohmic-drop fluctuations ΔR_eI and faradaic potential fluctuations ΔE, were analysed. After bubble size homogenisation, the effect of an applied stress was an increase in the level of the power spectral densities (PSD) Ψ_V and Ψ_E of the ΔV and ΔE fluctuations, indicating a change in metal-hydrogen-interactions related to damages in the metal induced by the combined actions of stress and hydrogen embrittlement. In the meanwhile, the PSD Ψ_ReI of the ohmic-drop fluctuations did not exhibit any change, revealing that the departure rate and size of hydrogen bubbles were not modified by the internal damages in the specimen. The time evolution of Ψ_E up to fracture could be explained by the enhancement of hydrogen penetration into the metal induced by the increase in the density of microdefects or crack advances inside the metal.     

Electrochemical noise analysis of cathodically polarised AISI 4140 steel. II. Identification of potential fluctuation sources for unstressed electrodes

In this second part of the study of electrochemical noise (EN) generated by hydrogen evolution on a vertical cylindrical AISI 4140 steel electrode under galvanostatic control in 0.5 M sulphuric acid, the potential fluctuations induced by the growth and detachment of hydrogen bubbles at the electrode surface were analysed. They could be related to fluctuations of various quantities: electrode active surface due to bubble screening effects, concentration of dissolved hydrogen in the electrolyte close to the electrode surface, and metal-hydrogen interactions (MHI) on or beneath the electrode surface. The existence of MHI and their influence on faradaic potential fluctuations could be revealed by comparing the noise features on steel and platinum. The influence on EN of the charging cathodic current density, the presence of dissolved oxygen in the solution, and the electrode rotation speed was investigated in the absence of stress applied to the electrode. In the third paper of this series, the effect of hydrogen embrittlement on potential fluctuations of stressed electrodes will be examined.     

Electrochemical noise applied to the study of the inhibition effect of CeCl3 on the corrosion behaviour of Al-Mg alloy AA5083 in seawater

Electrochemical noise measurements have been used to study the corrosion behaviour of aluminium alloy AA5083 in 3.5% NaCl solution with and without doping with CeCl₃ at 500 ppm. Information on the evolution of corrosion activity over a period of 4 days and on the corrosion mechanism itself has been obtained for the two systems. Experimental data obtained have been analysed using robust statistical parameters and wavelet transform; transient shapes were also studied. The various mathematical techniques applied to analyse the electrochemical noise data have been proposed recently. This paper thus not only characterises the corrosion systems studied but also illustrates the usefulness of these new methods.     

Using wavelets transform in the analysis of electrochemical noise data

Various concepts developed in the framework of wavelet transforms have been adapted to the analysis of electrochemical noise measurements (ENM). Wavelet transforms are proposed as an alternative tool to overcome the limitations of FFT in the analysis of ENM data. In this context, the theoretical basis of wavelet analysis is briefly reviewed. Then, both methods are applied to the study of various different corrosion systems covering a wide range of ENM signals. The results reported here demonstrate, firstly, that wavelet analysis is applicable to those systems in which FFT techniques also work. Moreover, in cases where FFT fails, wavelet analysis provides valuable knowledge about the behaviour of the system. Using wavelets, the different ENM components contributing to the original signal can be characterised. Each component is defined by a set of wavelet coefficients, which contain information about the time scale characteristic of the associated corrosion event.     

Electrochemical noise analysis of LY12-T3 in EXCO solution by discrete wavelet transform technique

Electrochemical noise technique coupled with SEM was used to study the corrosion process of LY12-T3 in the modulated EXCO test solution, and the EN data were analyzed by wavelet technique based on orthogonal db4 wavelet. The results show that, under the experimental conditions, the characteristics of energy distribution plot (EDP) obtained from wavelet analysis technique can be used as “fingerprints” of EN signal and can be used to differentiate the corrosion type. With the change of corrosion type of LY12-T3 from pitting corrosion through intergranular corrosion to exfoliation corrosion, the maximum relative energy defined in EDP will change from the region of crystal series coefficients with smaller scales through middle scales to larger scales.     

Characterization of corrosion processes by current noise wavelet-based fractal and correlation analysis

Electrochemical noise data in the presence of pitting, general corrosion and passivity were analyzed using the discrete wavelet transform. The registered current noise was decomposed into a set of band-limited details, which contain information about corrosion events occurring at a determined time-scale. It has been observed that the signal variance and variances of details depend on the intensity of processes. Distribution of the signal energy among different details was characteristic for the particular type of corrosion. The characterization of corrosion processes on the basis of in the wavelet domain calculated Hurst parameter H and fractal dimension, D, of electrochemical noise signals has been established. It is concluded that general corrosion is a stationary random process with a weak persistence and D = 2.14, whereas pitting corrosion is a non-stationary process with a long memory effect and D = 1.07. Passivity is a non-stationary process near to the Brownian motion with D = 1.56. The persistence features of electrochemical noise signals were explained also by correlation coefficients calculated between signals obtained by discrete wavelet multiresolution decomposition.     

Electrochemical characterization of SnO2 electrodes doped with Ru and Pt

Antimony-platinum doped tin dioxide electrodes supported on titanium have been prepared by thermal decomposition. The effect of the progressive replacement of Sb with Ru (x = 0.00; 3.25; 6.50; 13.00 at.%) on their electrochemical response in acid medium has been analysed by cyclic voltammetry. The morphology of the coatings was observed by scanning electron microscopy. Ti/SnO₂-Sb-Pt electrodes without Ru presented a cracked-mud structure, typical of oxide electrodes prepared by thermal decomposition. The introduction of Ru in the oxide layer modified the coating morphology. The roughness increased and passed through a maximum with the increase of Ru content. A relation between the surface morphology, the roughness factor, voltammetric charge and the electrochemical activity has been established. The mechanism and electrocatalytic activity towards the oxygen evolution reaction has been studied from Tafel measurements. The progressive introduction of Ru in the electrodes increased their electrocatalytic activity for the oxygen evolution reaction with a change on the mechanism from non-active to active electrodes. The electrocatalytic activity mainly depends on electronic factors.     

Evaluation of the protection against corrosion of a thick polyurethane film by electrochemical noise

This paper focuses on the study of protection against corrosion by means of a thick polyurethane film. Noise resistance, R_n, calculated from a series of potential and current noise measurements, shows that electrochemical noise (EN) technique was able to monitor corrosion under coated metals even when applied at high thicknesses. By using spectral analysis, the nature of the attack on the metallic surface was determined; indeed, the characteristics of the corrosion process were determined even from the beginning of the phenomenon when the surface was exposed in a Salt Fog Chamber. It was possible to determine both how the continuous protection was given by the polyurethane film as a barrier and when the corrosion process was controlled by the appearance of pores on the barrier film. This paper seeks to establish a methodology to study highly resistant coatings.     

Corrosion behaviour of multiwall carbon nanotube/magnesium composites in 3.5% NaCl

Corrosion behaviour of multiwall carbon nanotube (MWNT) modified pure magnesium (Mg) composites (MWNT/Mg) prepared by melt stirring technique was investigated by weight loss, H₂ gas collection and pH measurements, as well as electrochemical methods such as open circuit potential (OCP) monitoring, polarisation curves and electrochemical impedance spectroscopy (EIS). Results show that the corrosion rate of MWNT/Mg composites is significantly higher than the corrosion rate of pure Mg. The corrosion rate of the composites also depends on the degree of dispersion of MWNTs during melt stirring process. The role of the MWNTs in increasing the corrosion rate clearly can be attributed to their high cathodic activity. Characterization of the corrosion product layers indicates that dispersion of MWNTs in the matrix leads to a more homogeneous coverage of the surface by corrosion products, and the lowest thickness of the corrosion product layer. The corrosion product layers in all cases provide poor protection of the surface. The corrosion rate of pure Mg as well as Mg/MWNT composites significantly increases during 1-week immersion; this effect is the strongest for the composite with dispersed nanotubes and can be attributed to the increased cathodic kinetics with time.     

Study on the corrosion behavior of reinforcing steel in cement mortar by electrochemical noise measurements

The corrosion behavior of reinforcing steel in cement mortar has been studied by electrochemical noise (EN) compared with the electrochemical impedance spectroscopy (EIS). The wavelet transform, as well as the statistical methods including the standard deviation of current noise (σ_I) and noise resistance (R_n), has been employed to analyze the EN data of reinforcing steel in mortar. It is revealed that there exist three different corrosion stages of reinforcing steel in cement mortar, including a competition process between breakdown and repassivation of passive film, a pitting development and an active corrosion during the 20 cyclic immersion and drying tests. The energy distribution plot (EDP) is able to provide useful information about the dominant process for the different corrosion stages.     

Fluctuations of concentration overpotential generated at gas-evolving electrodes

The electrode overpotential is usually expressed in terms of partial components related to ohmic, activation and concentration effects. For a gas-evolving electrode, the overpotential fluctuates because of bubble evolution but only the global overpotential fluctuations and, more recently, their ohmic part are easily accessible by routine experiments. In the present paper, high-amplitude fluctuations of concentration overpotential have been experimentally revealed by an anomalous behaviour of the current fluctuations induced by hydrogen evolution on platinum electrodes under potential control in alkaline media with or without the presence of a second mass-transport-controlled reaction. Inverted current jumps, reproducibly observed owing to an artificial bubble-nucleation site created on the edge of the rotating electrode, which provoked periodic bubble eruption, were ascribed to local concentration changes drastically affecting the current-lines distribution. These experimental results validate the theoretical models that predicted those anomalies but remained unverified up to now. A quantitative estimation of each overpotential component at bubble eruption is addressed by means of a simplified analytical approach supported by the simultaneous measurement of the electrochemical current noise and electrolyte resistance fluctuations.