This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
This paper presents an overview of techniques for characterizing inhomogeneities in various forms of organic surface films including organic coatings and anti-corrosion oil and inhibitor films. Particular focus is on technological innovations that have been made over the past two decades for gaining better understanding of inhomogeneities in surface films and their effects on underfilm localized corrosion. These include scanning probe techniques such as the scanning Kelvin probe, scanning vibrating electrode technique, local electrochemical impedance spectroscopy, and the wire beam electrode method.
Keywords: Anti-corrosion coatings, rustproofing oil, inhibitor film, coating inhomogeneity, coating testing methods, the wire beam electrode
Inhomogeneities on metal surfaces covered with organic coating films have long been known to significantly affect underfilm corrosion . A common experience is that rusts on a coated metal surface normally initiate at the localized weaker areas of the coating film. Inhomogeneities in a coating film are believed to influence the permeability and the transport of aggressive species such as water, oxygen, and cations through the coating and along the coating-substrate interface, significantly affecting the anti-corrosion performance of a coating system . Nonuniform crosslink in a coating film was found to result in local 'D' and 'I' sites that lead to major variation in coating resistances and anticorrosion behavior . Pigments in a coating film could lead to local voids formation and behave as the initiation sites of coating failure, especially when the concentration is above the critical value . Residual solvent in coating films was found to promote the formation dark oxide spots under alkyd lacquers . The existence of localized internal stress in the coating may lead to loss of adhesion and local coating cracking , creating degradation-susceptible regions associated with inhomogeneities in the coating systems. These degradation-susceptible regions are microscopic in dimension and have properties that are different from the rest of the film. Although they occupy only a small fraction of the film volume, they control the corrosion-protection performance of a polymer coating. Underfilm corrosion is found to be directly beneath degradation-susceptible regions in the coating. In addition, coating inhomogeneity has also been found to significantly influence the reproducibility and reliability of electrochemical evaluation of organic coatings [7,8,9]. Knowledge of inhomogeneities in organic surface films is thus critical for understanding underfilm corrosion processes, for evaluating corrosion prevention by coatings and for providing insights into possible ways of improving anti-corrosion coatings and protective surface films.
The presence of inhomogeneities on coated metal surfaces is usually detected by visual observation or by employing imaging techniques such as optical microscopy, scanning electron microscopy (SEM), scanning tunneling microscopy (STM) and atomic force microscopy (AFM). Some forms of inhomogeneities such as rusts, air bubbles, pores and cracks could be visually or microscopically observable, many other forms of inhomogeneities such as dissolved salts, trapped solvents, internal stress, imperfect film formation and nonbonded areas are often invisible. In order to detect and measure the effects of inhomogeneities on the process and mechanism of underfilm corrosion, techniques that could measure underfilm chemical and electrochemical changes have been employed for characterizing the anticorrosion performance and the degradation mechanism of anticorrosive coatings and inhibitor films chemical. For instance localized pH electrodes have been used to investigate the effects of inhomogeneities on local pH changes in cathodic sites . Electrical resistance measurement and electrochemical methods such as electrochemical impedance spectroscopy (EIS) have also found widespread applications study the nature of underfilm corrosion on metal surfaces [11,12]. However inhomogeneity is still one of the less understood coating properties and is considered to be one of the hardest to predict accurately , primarily due to technological limitations in probing metal-solution interfaces.
Recent advances in research methods have enabled better understanding of inhomogeneities in surface films as a critical factor affecting underfilm corrosion processes. These include various forms of scanning probe techniques and an electrochemically integrated multi-electrode array namely the wire beam electrode (WBE). This paper presents an overview of these laboratory techniques for examining inhomogeneities in surface coatings and the subsequent localised corrosion processes. Particular focus is on techniques developed based on the WBE concept for visualizing and characterizing electrochemical inhomogeneity and underfilm localized corrosion.
2. Conventional methods for characterising surface inhomogeneities
The ability to detect the location of inhomogeneities in coating films in a quantitative manner would help identify the source of coating failure and provide insight into the mechanisms of coating degradation. Visual inspection of coated coupon surfaces can often determine the sizes and shapes of air bubbles and cracks in a coating film, while microvoids and contaminants can be observed through a more detailed examination of a coated surface using an optical microscope and SEM. The detection of weak areas in a coating film such as nonbonded and imperfect film areas often needs assistance of accelerated testing such as the immersion test and salt-fog test . These accelerated laboratory weathering tests intensify the effects from the environments so visible coating breakdown or corrosion sites develop more rapidly than in naturally occurring environments. Unfortunately, however these accelerated exposure tests still often cannot, within their exposure time, visually show the negative effects of inhomogeneity on intact coated surfaces.
Mayne and co-workers [3,11] found that most inhomogeneity of coatings is not due to pores or cracks but instead due to the inhomogeneious bonding within the polymer film. This inhomogeneity cannot be observed even using a SEM but can be detected using electric resistance measurement. They found that there is a significant difference in electric resistance between different areas of an organic coating. This was done by cutting a large coating sample into smaller pieces (e.g.1 cm ´ 1 cm in size) and measuring the DC resistance of each individual piece . Some pieces of the coating sample had very low DC resistances whereas others showed much higher resistances. They named areas of high and low resistance as "I" (indirect) and "D" (direct) type films respectively. Normally the film resistance for an I-type film is around 1010 ~ 1012 Wcm2 and for a D-type film is around 106 ~ 108 Wcm2. They assumed that the 'D' type areas are about 75 ~ 250 mm in diameter and are randomly distributed across the coating surface according to Poisson's law. They also found that the metal surface under "D" type film is very sensitive to corrosion [1,3,11]. However, Mayne's technique can only be used to determine whether a coating sample (e.g.1 cm ´ 1 cm in size) has at least one 'D' area.
Although DC coating resistance measurement may not necessarily be suitable for evaluating corrosion over an inherently non-linear electrochemical interface, the measurement of coating resistance after a period of immersion has been employed as a traditional method of assessing the performance of anticorrosive coatings. Coatings that are unable to maintain a high electrical resistance are usually those with pin-holes, low coating thickness, and other defects that allow oxygen, water and ions to penetrate the polymer film [15-17].
Electrochemical methods have been widely used for characterizing the anticorrosion performance of anticorrosive coatings and inhibitor films because electrochemical methods are believed to be able to detect changes in the insulating structure and electrochemical corrosion characteristics of surface films. An advantage of electrochemical methods is considered to be their ability to obtain information regarding the degradation of both coating and substrate before the degradation can be visually observed. The most widely applied electrochemical methods for characterization of anticorrosive coatings is probably EIS [12,18-22].
The usefulness of EIS in characterizing the anticorrosion performance of organic coatings and other surface films lies in its ability to distinguish the individual components of a coated electrode-electrolyte interface. The analysis of EIS data using electrical equivalent circuits is able to determine coating resistances, coating and electrochemical double layer capacitors, and other parameters related to electrode-electrolyte interfacial components. The values of these interfacial parameters and their changes with coating degradation are useful for understanding the behaviour and performance of coating systems. For this reason EIS has become a very important technique which has broadened the range of corrosion phenomena which can be studied using electrochemical techniques [18-22]. For instance EIS has been extensively applied for the investigation of water and ion transport in organic coatings and the subsequent corrosion processes. A typical example of using EIS data and modelling for investigating the effects of environmental factors on inhibitor filmed electrode-solution interface has been discussed in reference . It should be noted that although EIS has been widely applied for the investigation of organic coatings it is applied mostly in a qualitative manner. There have been difficulties in relating EIS measurements directly and quantitatively to lifetime prediction of coatings .
Electrochemical noise analysis (ENA) is another electrochemical method that has been found useful in evaluating anticorrosion coatings [24,25] and inhibitor films . ENA is based on the measurement of the natural voltage and current fluctuations generated from coated electrodes in corrosion cells. The most useful parameter has been considered to be the noise resistance derived as the standard deviation of the voltage noise divided by the standard deviation of the current noise [26-30]. The noise resistance measurement has been found to correlate with the DC resistance measurements for coated specimens, and also the polarisation resistance measurements for bare metal [31-32].
When applied in parallel to coated electrodes, EIS and ENA often produce similar results [33,34]. An interesting study on the application of the ENA for monitoring coating detergation can be found in references [35,36]. An embedded two-electrode configuration has been adopted in ENA measurements of coating performance on exposure in a cyclic salt fog test chamber . In a typical experiment, the degradation of urethane topcoat/epoxy primer systems were monitored using the electrochemical noise method and measured with embedded electrodes. The trend in the noise resistance parameter was found to be consistent with the trend in the low frequency impedance modulus obtained from EIS experiments .
In addition to evaluating organic coatings, ENA has also been proposed as a means of distinguishing various types of corrosion ; however this application remains a controversial subject.
It should be noted however that in principle conventional electrochemical methods are applicable only to homogeneously coated electrode surfaces because they can only measure averaged elechemical values. For instance EIS is not able to perform direct measurements of local ac impedance. This limitation can be illustrated by examining traditional methods of measuring the impedance of an electrode covered with a surface film or coating. Conventional EIS measurements using a coated electrode with a large area only detect an impedance that is a mixture of contributions from many local impedance values, none of which we can evaluate independently.
Obviously there is a need of techniques that are able to measure local electrical or electrochemical parameters such as local coating film resistance and impedance. This is an important requirement since corrosion failure in organic coating/substrate systems is often found to initiate at a chemical or physical inhomogeneity on the coating/electrode interface.
3. Microelectrochemical techniques for probing localized coating failure
Inhomogeneities in a coating or on a coating/electrode interface often lead to various forms of coating failure, such as blisters and disbondment. Microelectrochemical techniques such as microelectrodes and scanning electrochemical probes have been employed, often in conjunction with conventional methods, to achieve a better understanding of the processes and mechanisms behind localized coating failures. In a typical experiment described in reference , conventional EIS and a microelectrode have been applied to understand the formation and growth of blister on coating/aluminium interfaces and to understand the influence of the environmental factors. Impedance spectroscopy with and without microelectrodes was applied to study localised coating defects on an aluminium alloy AA 2024-T3 because features in regular impedance spectroscopy could be related to local coating failure phenomena and thereby allowed validation of the equivalent circuits used for data interpretation . Evidence for a strong local decrease in the coating resistance on the top of the blister was found. The coating resistance on areas not corroded appeared to decrease even after the end of the initial water uptake, although the impedance was still several orders,of magnitude higher than on top of the coating blister .
In recent years, scanning probe techniques such as AFM has been employed to map polymer heterogeneity. In a typical experiment described in reference , a new approach to organic coating condition evaluation at micrometer scale was carried out using a AFM in contact mode for localized impedance measurements. Impedance was measured between conductive AFM tip and metal substrate covered with an organic coating film. A single-frequency voltage perturbation signal was applied between the electrodes and current response signal is registered. As the AFM tip is scanned over the surface of the specimen a localized impedance characteristics of the material was obtained. Such impedance data could be correlated with surface features mapped via classical AFM measurements such as height profile . In a comprehensive study , AFM phase imaging and nanoscale indentation has been used to detect heterogeneous regions in polymer coatings that are believed to range from nano- to micrometers. This overcome limitations associated with micro- and spectroscopic techniques such as scanning electron microscopy and X-ray photoelectron spectroscopy. In an experiment, AFM was used to study heterogeneity in thin film of approximately 250 nm of polystyrene and polybutadiene blends . Pits were observed to reach the film/substrate interface, creating pathways that lead to corrosion of the substrate.
Many other scanning probe techniques such as the scanning Kelvin probe, the scanning Kelvin probe force microscopy (SKP), the scanning reference electrode technique (SRET), scanning vibrating electrode technique (SVET), local electrochemical impedance spectroscopy (LEIS) and scanning electrochemical microscopy (SECM) have also been applied in research aimed at understanding electrode inhomogeneity in surface films and its effects on localized electrode processes. These techniques are often used as complementary tools to understand localized corrosion processes and mechanisms in defects and underneath coatings, in particular the mechanisms of cathodic delamination and filiform corrosion. For instance, the standard SKP has played a major role in gaining a deeper understanding of cathodic delamination, for instance Stratmann et al applied SKP for measuring the interfacial potential between a defect and a random location at the coating-steel interface, permitting the rate of cathodic delamination to be the measured non-destructively [41-44].
SKP is a noninvasive, no-contact vibrating capacitor technique that has been performed to investigate localized defects in organic coatings by mapping Volta potential differences at buried metal/polymer interfaces between a vibrating microelectrode and a sample with a high resolution [41-46]. It was reported that coating defect site and delamination front form anode and cathode of a galvanic cell that is discernible in the SKP potential maps by a steep increase of the potential. In a typical experiment, the potential drop at the delamination front was measured to be 200 mV, with the sharpness of its gradient being 30 mV Î¼mâˆ’1, while a more gradual slope in the electrode potential signifies the already delaminated area . In a typical experiment, SKP measurements were performed before and after the immersion of tin plated mild steel food-can protective coatings, a TiO2 enriched melaminic coating and TiO2 and a carbon black enriched phenolic coating in a 0.35 wt% NaCl solution at pH 4 for 120 hours . Some defects were observed on the surface represented by high and localized work function variation, compared with the average value over the surface. SKP measurements have also been performed on silane treated copper panels and a reactive sputtered TiN coated mirror polished steel surface . It was reported that there is a different mean work function value for the coated and the bare substrates, indicating the dissimilar electrochemical activity of the different surfaces. SKP has also been applied in conjunction with EIS and other scanning probe techniques such as scanning vibrating electrode because these three techniques present a very complementary approach to understand the ensemble of coating degradation, processes in defects and corrosion underneath coatings such as cathodic delamination and filiform corrosion, respectively .
However standard SKP does not allow a high enough resolution necessary to probe the submicroscopic coating defects, in order to gain more information about the microscopic and submicroscopic processes at the delamination front, the scanning Kelvin probe force microscopy (SKPFM) that combines AFM in the KPM mode has been applied to probe Volta potentials of delaminating electrode/coating interfaces with submicron resolution . SKPFM has been shown to be an in situ technique for gaining a more detailed understanding of localised delamination processes in the microscopic and submicroscopic range. However a practical difficulty in the application of the SKP and SKPFM to practical coated electrodes under localized corrosion is that for all SKP and SKPFM measurements the resolution is strongly dependent on the distance between tip and the coating/metal interface. This requires the preparation of special model samples that are characterized by ultrathin polymer coatings and specially prepared defects that show a very sharp borderline to the intact coating . In addition, the SKP and SKPFM scanning tips are only pseudo-references since their Volta potential may vary from tip to tip due to slight differences in the oxide covering them, or contaminants deposited on the tip during scanning. Furthermore, studies suggest that the nature of the polymer film has a pronounced effect on the resulting image . Another limit of the SKP is considered to be the difficulty of the interpretation of the experimental data .
The SRET and SVET have also been used to detect inhomogeneities in coating films and associated localized corrosion damages. In a typical experiment, SVET was used to scan localized electrochemical events over a coated surface area immersed in a 0.005 M NaCl solution . SVET was able to detect defects in a pigment-free coating. The growth of the local anode area on the paint film was statistically calculated and determined by the difference of the potential gradient values between the local anode (defective area of painted film) and the local cathode part corresponding to a non-defective area . The SRET has been employed to the study of polyaniline coatings on carbon steel . SRET results demonstrate that conductive polyaniline "passivates" pinhole defects in coatings on carbon steel and thus the principal potential advantage offered by the polyaniline coating is toleration of pinholes and minor scratches. A model is proposed which entails passivation of the metal surface through anodization of the metal by polyaniline and formation of an insoluble iron-dopant salt at the metal surface .
The SRET and SVET are frequently applied in conjunction with other electrochemical and analytical techniques of micrometer spatial resolution. In a study of the self-repair ability of coatings modified with submicron containers loaded with corrosion inhibitors, the SVET has been applied in combination with the scanning ion-selective electrode technique and SEM. Complementary studies were carried out by EIS to assess the effect of the containers filled with corrosion inhibitors on the barrier properties of the coatings. The electrochemical results highlight the importance of the combined use of integral and localized electrochemical techniques to extract information for a better understanding of the corrosion processes and corresponding repair of active microscopic defects formed on thin coatings containing inhibitor filled containers . Similar methods have been applied to study corrosion inhibition in microdefects of protective coatings on magnesium alloy . The combination of SVET and the scanning ion-selective electrode technique demonstrated to be a useful approach to investigate the inhibition of corrosion processes in microdefects on coated magnesium alloy AZ31. Results show that 1,2,4-triazole showed the highest inhibition efficiency among the studied inhibitors and was able to prevent the increase of pH in the corroding defects, by keeping the corrosion activity on a very low level during the immersion period .
In another study, the SVET was used to discriminate the corrosion protection performance of selected sol-gel based coating systems that were developed as part of an environmentally compliant coating system alternative to the currently used chromate-based systems . The SVET results, as an early performance discriminator for newly developed coating systems, were compared with data obtained from chromium inhibition coating systems and EIS measurements . The SVET was also used to investigate the effects of a remote stainless steel cathodes on the corrosion of polyvinyl chloride (PVC) coated galvanized steels .
LEIS is another scanning electrochemical probe that has been used to study the degradation of an organic coating with defects. The LEIS results clearly demonstrate that it is possible to separate the impedance response of the intact coating from that of the defect and that single frequency impedance mapping of the surface can provide complementary data supporting the physical interpretations of the impedance response . In a typical study reported in reference , both LEIS and conventional EIS were used to investigate the degradation of polyester coil-coated galvanized steel on the same specimen. Specimens containing a central 250 mm laser-ablated defect in the organic coating layer were immersed in a 10 mM NaCl solution for up to 30 days. The local multifrequency impedance was determined by placing a novel impedance probe, either directly above the coating defect or above an area of intact coating. In addition, single frequency impedance mapping of the specimen surface was carried out at 1 kHz and compared with optical microscopy of the surface. The results demonstrate clearly that macroscopic electrochemical impedance provides a surface-averaged measurement of the properties of the coating, plus any defects. Thus, macroscopic impedance spectra convolute the separate responses of the coating and defect together. However, local electrochemical impedance can effectively separate the local properties of the organic coating from the local electrochemical behavior at a coating defect .
LEIS was also used for the detection and mapping of defects and local corrosion events in organic coatings . Various types of intentional local heterogeneities including chemical defects within the coating such as absorbed oil and physical defects such as subsurface bubbles, underfilm salt deposits, pinholes, and underfilm corrosion were successfully detected with a five-electrode LEIS system that utilizes a split microreference electrode .
The LEIS technique was further used to investigate localized corrosion of steel at defect of coating and, furthermore, to determine the effects of cathodic protection on local electrochemical environment and the resultant corrosion reaction at the base of coating defect . The results demonstrated that conventional EIS measurements on a macroscopic-coated electrode reflect the "averaged" impedance results from both coating and defect. Corrosion of coated steel is dependent on cathodic protection potential and the defect geometry [57,58].
SECM [59,60] is a tool that enables us to perform difficult tasks of detecting localised chemistry changes by means of variously designed scanning probes. SECM is a scanning electrochemical probe that detects amperometrically surface-generated electroactive ions or molecules in the solution phase as a function of spatial location with an electrochemically sensitive or ion-selective ultramicroelectrode tip. It has been extensively applied for topography mapping, surface modification and redox reactivity imaging [59,60]. SECM has been used in studying corrosion of coated metals. In a typical experiment reported in reference , negative-feedback SECM was successfully applied to study the effects of lixiviation from a nickel foil coated with plasticized PVC by visualizing spatially resolved differences in the topography of coated metal samples upon exposure to aqueous electrolyte solutions of different compositions. This method allowed the investigation of the uptake of reactants from the electrolyte phase through the polymeric matrix to the metal/polymer interface to be performed even at early exposures. Yet, the method must be carefully checked to discard transport processes from the organic matrix into the solution phase, such as those related to lixiviation. In this later case, the topography of the polymer layer may evolve with time accordingly, not longer exclusively responding to the uptake by the polymer matrix of components from the electrolyte phase. Furthermore, lixiviated species may also react with the SECM tip, eventually leading to the continuous modification of the active surface area of the electrode during the measurements .
Each scanning probe technique has its advantages and limitations, for this reason, different techniques are often combined and applied in a synergistic manner. For instance, traditional optical microscope, Microscope and SEM are often applied with scanning probe techniques to provide topographical information that are often critical for anti-corrosion coating research. Analysis of the electroctrochemically active sites can be carried out using SEM, energy dispersion spectroscopy system (EDS) and X-ray photoelectron spectroscopy (XPS).
It should be noted that scanning probe techniques including SRET, SVET, LEIS and SECM can detect ionic currents, carried by ions in the electrolyte phase, flowing over a corroding metal surface, however they are unable to measure the currents flowing exactly at the metal-coating interface. For this reason, they may not be able to accurately detect all ionic currents, especially those flow at the metal-coating interface. Scanning probe techniques commonly operate in a relatively specific and localized area, and thus, in many circumstances, the scan image does not necessarily represent the full details of an electrode process that involves different reactions occurring simultaneously over distinctively separated electrode areas.
4. Characterising coating inhomogeneities using coupled electrode arrays
Another approach of understanding inhomogeneities in surface films and localized underfilm corrosion is using an electrochemically integrated multi-electrode array namely the wire beam electrode (WBE) . The WBE is a non-scanning probe technique that is able to visualize the processes of localized corrosion under a coating or an inhibitor film by measuring parameters from local areas of a working electrode surface, such as local resistance, corrosion potential and galvanic current, providing spatial and temporal information on underfilm localized corrosion.
The WBE was firstly proposed as a means of detecting and quantifying inhomogeneities in organic coating films by measuring electrical resistance distribution over a coated electrode surface [8,9,63,64]. The WBE method involves subdividing an area of coated surface (e.g. 1 cm2) into many small sections and measuring the electrochemical properties of each part by means of individual sensors. Using a simple experimental setup shown in Figure 1, nonuniform distribution of electrical resistances over a coated WBE surface was mapped. A typical example of nonuniform distribution of coating electrical resistance is shown in Figure 2.
Figure 1. Schematic diagram of measuring the distribution of electrical resistances in coating film using a WBE [8-9].
Figure 2. The distribution of DC resistance over a coated WBE surface [8-9].
Wu et al.  investigated electrochemical inhomogeneities in organic coatings, in particular the so-called 'D' and 'I' areas, using a high resistance measurement technique under strict experimental condition control. The existence of 'D' and 'I' areas in coating films is a significant coating characteristic first reported by Mayne et al. [1,3]. An experimental setup, as shown in Figure 3, was used in the work . In a series experiments the inhomogeneities in three organic coatings: phenolic resin, alkyd resin and polyurethane varnish, were quantified by measuring the distributions of DC resistances over various surface areas of coated WBEs exposed to a 3% NaCl brine . A WBE with 121 iron wires of 1.0 mm diameter was used to measure coating electrical resistance over a large resistance range (102~1014 Wcm2). Careful moisture control, electrostatic shielding, cable insulation, careful electrode surface preparation and equipment calibration were applied. Electrical resistance measurements were carried out by applying a voltage between a cathode (a wire in the coated WBE) and an anode (an iron wire made from the same material as those in the WBE). The terminals of the wires in the WBE were individually connected sequentially and manually to the ammeter, in order to measure the currents induced by the applied voltage. Measurements were repeated after various immersion periods in the electrolyte solution. The current measured by the ammeter was used to calculate coating film resistance using Ohm's law .
Figure 3. Schematic diagram of the improved experimental apparatus for mapping coating resistance distribution .
Significant differences in DC currents were recorded from different areas of coated WBE surfaces. The measured current values often show major differences between neighbouring wires of only 2 mm separation. For instance the maximum current measured from a typical WBE wire was 6´10-7 A, while the minimum current measured from another wire of the same WBE was 3´10-13 A. This indicates the existence of a more than 1 million times of difference in electrical resistance over different coating areas. Two typical types of areas were identified that showed a significant difference in their DC resistance. Figure 4 shows a typical pattern of the inhomogeneious DC resistance distribution. The two 'peaks' discontinuous bimodal distribution, rather than a normal distribution, suggests the presence of two types of coating areas. One type of coating area has higher resistance and another has lower resistance and there is an obvious boundary between them. This is a direct evidence for the existence of "I" and "D" areas that were proposed by Mayne et al [1,3]. Table 1 summarizes the resistance ranges and percentages of low and high resistance of three organic coating films. The lower resistance areas should be covered with 'D' type films whereas higher resistance areas should be covered with 'I' type films.
Figure 4. Typical DC resistance distribution plot of a coated WBE .
Table 1. An estimation of the percentage of high resistance and low resistance films .
(dry film thickness 16mm)
(dry film thickness 13mm)
(dry film thickness 13mm)
Resistance range of high resistance film ('I')
Resistance range of low resistance film ('D').
% of low resistance film ('D').
The thickness of coating film was found to significantly affect the inhomogeneity of coating films. Table 2 summarizes coating resistance data from a coating with different thicknesses. It can be seen that the increase in coating thickness lead to a major increase in the percentages of high resistance coating area ('I' zones, 2% ® 25% ®58%).
Table 2. A comparison of coatings with different thicknesses .
Thickness of coating films
'I'/'D' coating area boundary resistance
% of high resistance Film ('I')
% of low resistance Film ('D')
The method of coating application was also found to influence coating inhomogeneity. Table 3 shows results from a single layer and a double layer phenolic resin coating of 16 mm thickness. The resistance distribution of these coatings were obviously different. The percentage of high resistance coating area for double layer coating (38%) was larger than that of single layer coating (25%). The boundary resistance for the double layers coating film (1011ohms) was larger than that of single layer coating film (1010 ohms). This suggests that more layers can improve the corrosion protective ability of organic coating with certain thickness. Indeed some rust points were visually observed on the single layer coated WBE surface after 3 days' immersion in 3% NaCl brine, while no obvious rust was observed on the double layers coated WBE surface. This is in line with industrial practice that most corrosion control coating systems need at least two-coats, sometimes three or more coats, since it is well known that multiple coats of corrosion protective coatings protect better than a single coating. Multiple coatings could help reduce major weaknesses in a coating film because imperfections in the first layer could be covered by the upper layers, as it is unlikely that one imperfection in a given layer will exactly overlay another imperfection .
Table 3. A comparison of double layers and single layer phenolic resin coated WBE .
Thickness of coatings
Double Layers (16 mm in total)
Single Layer (16 mm)
'I'/'D' boundary resistance
% of high resistance film ('I')
% of low resistance film ('D')
The multi-electrode concept was also used by Tan et al in the evaluation of rustproofing oils  and the evaluation of the effectiveness of crevice corrosion inhibitors . Wu et al employed the WBE in a series of experiments to understand the electrochemical inhomogeneity on oil painted metal [67-68]. Their results showed that the distributions of corrosion potential and DC resistance of an oil film were inhomogeneous on oil painted metal. With the extension of exposure to corrosive media, the corrosion potential on substrate would shift to positive direction, low DC resistance area could be eliminated by adding oil soluble inhibitors . They found that repeatability and reliability of electrochemical measurements can be improved greatly by using the WBE. The protective property of organic coatings can be evaluated rapidly and quantitatively based on the distribution and the probability of weak areas in a coating film . Using similar experimental techniques, Zhong et al investigated electrochemical inhomogeneity in temporarily protective oil coatings by sensing the potential variation over a WBE surface coated with preventive oil films [69-71]. It was found that the distribution of corrosion potential on the surface of oil-coated WBE was heterogeneous. When the degradation of the oil film occurs, the distribution of corrosion potential was found to change from normal probability distribution to discontinuous bimodal distribution . The WBE was also used to investigate self-repairing ability of temporarily protective oil coating. It was shown that inhibited oil coatings had the ability of self-repairing, and oil-soluble inhibitors had direct effect on the self-repairing ability of oil coating . The method was also used to investigate the anti-contamination performance of temporarily protective oil coatings. It was shown that salt contamination on the metal substrate had influence on the heterogeneous distributions of corrosion potential and polarization resistance. With salt contamination, the corrosion potentials distribution of oil coatings followed a discontinuous bimodal probability distribution, whereas the anodic polarization resistance distribution of oil coatings transformed from a log-normal probability distribution to an exponential probability distribution and then to a discontinuous bimodal probability distribution, the cathodic polarization resistance distribution of oil coatings followed a log-normal probability distribution .
Typical experiments described above clearly demonstrate the applicability of the WBE method in mapping inhomogeneities over coated metal surfaces by detecting coating electrical resistances and potential differences. It is possible to correlate WBE coating resistance or potential distribution maps of the type shown in Figure 2 with Volta potential profile measurable using the Scanning Kelvin Probe. More detailed research is needed in these areas.
The effects of coating inhomogeneity on electrochemical measurement
Organic coating films and inhibitor films are inhomogenous in nature and this inhomogeneity could significantly affect the reproducibility and reliability of conventional electrochemical measurement of corrosion under organic coatings. It has been suspected for a long time that a small weakness area such as a pore in a coating film could significantly affect the results of electrochemical measurement such as EIS data .
Recently Zhang et al  carried out a study on the corrosion of steel under defective coatings in 3.5% NaCl solution by the WBE and EIS techniques. They found that EIS diagrams measured during the entire coating deterioration process were dominated by the substrate corrosion process under the defect areas, while electrochemical processes under the whole coated electrode were 'averaged' out. According to the current distribution maps plotted using the WBE and EIS responses; they found that the initial high anodic and cathodic current densities were generated only at the defect areas . This result suggests that nonuniform distributions of reaction and polarisation currents over the electrode would affect EIS measurement.
Lee et al carried out a sophisticated study on inhomogeneity over chemically modified electrodes and its effects on electrochemical measurement . A multielectrode array consisting of 100 nominally identical and individually addressable gold disk electrodes, each with a radius of 127 µm, was used (Figure 5) to mimic a single macrodisk electrode in order to detect and analyse the effects of electrode inhomogeneity on voltammetric responses. The fabricated individual electrodes are sufficiently large that they exhibit close to linear diffusion, and each is sufficiently separated so that, with a suitable scan rate, overlap of diffusion layers can be essentially avoided. Furthermore, the individual electrodes are sufficiently small so that ohmic (iR) drop is minimal in studies in aqueous media. A series of experiments was performed to examine the deviation in behaviour of each individual electrode relative to the summed response obtained when all electrodes are simultaneously used in an experiment .
In principle, under these circumstances, the sum of each individual response should equal that produced when all elements in the array electrode are operational. In their investigation, the heterogeneity effect of a thiol monolayer modified electrode surface is probed with respect to the diffusion controlled electrochemistry of cytochrome c. The array configuration was initially employed with the reversible and hence relatively surface insensitive [Ru(NH3)6]3+/2+ reaction and then with the more highly surface sensitive quasi-reversible [Fe(CN)6]3âˆ’/4âˆ’ process. In both these cases the reactants and products are solution soluble and, at a scan rate of 50 mV sâˆ’1, each electrode in the array is assumed to behave independently, since no evidence of overlapping of the diffusion layers was detected.
As would be expected, the variability of the individual electrodes' responses was significantly larger than found for the summed electrode behavior. In the case of cytochrome c voltammetry at a 4,4â€²-dipyridyl disulfide modified electrode, a far greater dependence on electrode history and electrode inhomogeneity was detected. In this case, voltammograms derived from individual electrodes in the gold array electrode exhibit shape variations ranging from peak to sigmoidal (Figure 6). However, the total response was always found to be well-defined. These results imply that random levels of inhomogeneity in gold electrode surfaces may contribute to the overall voltammetric response obtained from a gold electrode. In most cases, the influence of electrode inhomogeneity will be subtle, although in the case of a chemically modified electrode surface, inhomogeneity may drastically influence even the wave shape .
Figure Schematic representation of the experimental arrangement used to study the effects of electrode inhomogeneity on voltammetric responses .
Figure 6. Cyclic voltammograms obtained from each individual 4, 4¢-dipyridyl disulfide modified, 127 mm radius gold element (total of 98) of a gold multielectrode array, at a scan rate of 50 mV s-1 in 400 mM cytochrome c (0.1M NaCl in 20mM phosphate buffer) .
This voltammetry is consistent with a microscopic model of inhomogeneity where some parts of each chemically modified electrode surface are electroactive while other parts are less active. The findings are consistent with the common existence of electrode inhomogeneity in cyclic voltammetric responses at gold electrodes, that are normally difficult to detect, but fundamentally important, as electrode nonuniformity can give rise to subtle forms of kinetic and other forms of dispersion. These results imply that random levels of inhomogeneities in gold electrode surfaces may contribute to the overall voltammetric response. In most cases, the influence caused by electrode heterogeneity will be subtle, although in the case of a chemically modified electrode surface, heterogeneity may drastically influence even the wave shape. This study is in agreement with studies by Compton et al.  on effects of heterogeneity at carbon electrodes and imply that electrochemists may need to more widely recognize the influence of surface inhomogeneities as a factor that introduces nonideal behaviors relative to those predicted on the basis of a uniform surface .
Visualising underfilm corrosion using the WBE and scanning probes
The WBE was used to measure electrochemical parameters from local areas under an organic protective film, including galvanic corrosion current density and corrosion potential and their distributions. These electrochemical parameters were used for studying non-uniform corrosion of an electrode covered with organic coatings or films and for evaluating the corrosion protective ability of rustproof oil films. Electrochemical measurement and analysis using the WBE enables evaluation of organic coatings on a statistical basis and this statistical analysis could improve the reliability and reproducibility of coating evaluation and has avoided serious influences from random factors such as pores in a coating film on electrochemical evaluation of organic coatings [8-9].
In a typical experiment, as shown in Figure 7, a steel WBE was pre-filmed with a rustproof oil film and exposed to a water-drop . Water drops with various sizes often form on coated metal surface and that causes localised corrosion damage. This experiment used similar experimental designs to those shown in Figure 7, the only difference is that an organic film was pre-painted on the working surface of the WBE before it was exposed to water-drop corrosion conditions. Two rustproof oil films were used: The first was a thin film of a very widely used rustproof oil WD-40 and the second was a thin film of engine oil Mobil SAE 20W-50. The thickness of the oil films was approximately 10 mm.
Figure 7. A schematic diagram showing measurements of galvanic corrosion current density distribution .
Using an experimental design shown in Figure 7, galvanic current density distributions over a WBE surface, filmed with a thin layer of rustproof oil WD-40, were measured at various stages of the exposure period. At the beginning of the exposure, as shown in Figure 8(a), there was a small area which exhibited a large anodic current density peak (0.046 mA/cm2). This peak may correspond to a weak area in the oil film. However, when exposure is extended, this large anodic current density peak disappeared and was replaced with much smaller anodic peaks (Figure 8(b)). This phenomenon may be related to the self-repair processes of this rustproof oil film, although the exact reason is not clear. Generally only very small galvanic currents were recorded in this system although there was a clear separation of anodic and cathodic zones under the rustproof oil film. This corresponds well with the good rustproof ability of this widely used rustproof oil.
Figure 8. Galvanic current density distributions over a WBE surface, with a thin layer of rustproof oil WD-40, exposed to a drop of 0.05 N NaCl solution (approximately 12 mm in diameter).
Using the same experimental design, galvanic current density distributions over a WBE surface, filmed with a thin layer of less protective engine oil (Mobil SAE 20W-50), were measured at various stages of the exposure period. At the beginning of the exposure, as shown in Figure 9(a), a large anodic current density peak (0.016 mA/cm2) was recorded from a small electrode area. This anodic current density peak, however, did not disappear with the extension of exposure; instead, it increased with exposure time (Figures 5.9(b) and (c)). At the end of this exposure test, brown corrosion products were observed at the anodic current density peak location, which obviously corresponded to a weak area in the oil film. This oil film was not able to self-repair this weak area. This result correlated well with the lower rustproof ability of this engine oil.
Figure 9. Galvanic corrosion current density distributions over a WBE surface, with a thin layer of engine oil, exposed to a drop of 0.05 N NaCl solution (approximately 12 mm in diameter).
The WBE can also be used in conjunction with other techniques such as scanning probes to visualize underfilm corrosion processes from both the metallic and electrolyte phases in order to achieve a better understanding of corrosion mechanisms. In a typical experiment, a WBE was sprayed with a layer of WD-40 oil film and exposed to the Evans solution in an experimental arrangement as shown in Figure 10. The Evans solution was prepared by dissolving 0.017 moles NaCl and 0.008 moles Na2CO3 in 1000 ml deionised water. Experiments were carried out under static conditions at about 20 °C to allow corrosion to occur .
Figure 10. Schematic diagram showing experimental set-up of a filmed WBE
in combination with SRET
Figure 11. WBE current density (mA/cm2) and SRET maps measured from a mild steel WBE surface coated with a WD-40 oil layer and exposed to the Evans solution for various periods.
Figure 12. WBE galvanic current density (mA/cm2) and potential (V vs SCE) distribution maps measured from a mild steel WBE surface coated with a WD-40 oil layer and exposed to the Evans solution for 38 hours.
Figure 11(a) shows WBE and SRET maps measured immediately after the specimen was immersed in the solution. Both WBE and SRET maps successfully detected anodic sites that clearly correlated with each other, although cathodic zones in SRET maps were affected by scanning tip movement . At the beginning of exposure, as shown in Figure 11(a), there were 37 wires that behaved as anodes although the galvanic current density values detected by WBE method were very small (maximum anodic current density was 0.068 mA/cm2). These anodic sites are believed to be the weakness sites in the WD-40 oil layer, which can be attributed to the electrochemical inhomogeneity of organic coatings. An interesting observation in the experiment was that the locations of major anodic sites remained almost unchanged, but the number of anodic sites decreased considerably with the extension of experiment. After 2 hours immersion, only 14 wires remained as anodes. The maximum galvanic current density increased steadily from 0.068 mA/cm2 to 0.481 mA/cm2 during 20 hours of exposure, suggesting that corrosion became more and more localised and concentrated. This result is surprising since extended exposure to corrosion environment is expected to cause continued degradation of the oil film and thus more anodic sites. The mechanism of this phenomenon requires further detailed investigation.
During the whole experiment period, the WBE and SRET maps correlated to each other. The WBE maps, in particular the potential distribution typically as shown in Figure 12, appear to give more fine details on the behaviour of corrosion anodes and cathodes. This experiment confirms that the combined WBE-SRET method was able to provide useful information on macro-cell electrochemical corrosion processes that involve macro-scale separation of anodes and cathodes. The WBE-SRET method is useful for understanding the initiation, propagation and electrochemical behaviour of localised corrosion anodes and cathodes, and also their dependence on externally controllable variables such as the existence of surface coatings .
Studying corrosion protection by coatings and cathodic protection
Application of the WBE has also been extended to evaluation of corrosion prevention techniques. In an experiment, the WBE was applied as a tool for monitoring the anodic electrodeposition of polyaniline (PANI) coatings and also for understanding the anti-corrosion performance and mechanism of the PANI coatings . Anodic polarisation currents were measured from various locations over the WBE surface to produce anodic polarisation current density maps. Experimental results revealed that if an AA1100 WBE was not pre-treated, the map would show a localised anodic current density distribution, resulting in a nonuniform PANI deposit. If the AA1100 electrode was pre-treated using a cathodic polarisation process, the map would show a random anodic current density distribution, and the PANI coating would cover the whole WBE surface. These results indicated that the WBE is a practical tool for monitoring, characterising, optimising and evaluating electrodeposited surface coatings such as PANI coatings .
One another application is the evaluation of cathodic protection current density distribution over an electrode surface with and without the presence of an organic coating. In industry, cathodic protection is often used in conjunction with organic coatings to prevent localised corrosion at weak areas in the coating films. In the case of cathodic protection of a coated metal structure using a sacrificial anode, protection current (galvanic current) is not uniformly distributed over the metal structure surface. Locations that are far away from the sacrificial anode site or under a high resistance coating could have low protection current density and thus may not be effectively protected. This is a major problem that has to be addressed when a cathodic protection system is designed. Similar problems arise when impressed cathodic protection current is applied to a metal structure such as a long pipeline with protective current density decaying as the distance to impressed current source increases. Thus locations far away from the current source and sites covered by high resistance media may not be effectively protected. In a sample experiment, the WBE was tested to measure the nonuniform distributions of protective current over a metal surface that was covered with a porous organic coating film .
At the end of the water-drop exposure experiment described in Figure 9, a zinc wire with a surface area of approximately 0.015 cm2 was introduced into the water drop, replacing the position of wire '1' in the WBE. This zinc wire behaved as a sacrificial anode to prevent localised corrosion at weak areas of the rustproof oil film. Protection current distribution was measured using an experimental design shown in Figure 13. Cathodic protection was thus applied with the zinc wire behaving as a sacrificial anode. As shown in Figure 14, the zinc wire became the only anode in the WBE system and produced a large protection current density (0.50 mA/cm2) to protect other mild steel wires, especially those located at the weak areas of the oil film, in the water drop from further corrosion.
Figure 13. A schematic diagram showing measurements of cathodic protection current distribution .
Figure 14. Cathodic protection current distributions over a WBE surface, filmed with a thin layer of engine oil, exposed to a drop of 0.05 N NaCl solution (approximately 12 mm in diameter).
These experimental studies showed that the WBE is a practical method for studying localised electrode processes under an organic coating or a rustproofing film. The WBE is able to determine the exact distribution of cathodic protection currents over a protected surface. This technique could provide important parameters for designing effective cathodic protection systems in order to avoid over-protection or under-protection of some sections of a coated metal structure.
Zhang et al investigated the corrosion of steel under defective coatings in 3.5% NaCl solution by the WBE and EIS techniques . Le Thu et al.  also used a modified wire beam electrode consisting of 210 minielectrodes to study the processes of local coating delamination in seawater under cathodic protection conditions and to evaluate compatibility between organic coatings and cathodic protection. They measured galvanic corrosion current flowing between microelectrodes with applied cathodic protection current. Nonuniformity of the coating was easily shown and the delamination rate near the artificial defect was estimated. When the coating is intact, EIS reveals a highly resistive behaviour for 10 months which is usually the case with commercial thick coatings devoted to be associated to cathodic protection. However, current measurements with WBE show preferential delamination zones after 10 months of immersion under strong cathodic protection. They suggested the possibility of applying the WBE as an efficient method of evaluating the compatibility between organic coatings and cathodic protection.
5. Concluding remarks
Recent advances in research methods such as scanning probes and the wire beam electrode have enabled better understanding of inhomogeneities in surface coatings and its effects as a critical factor on underfilm corrosion. Scanning probes and the wire beam electrode have both been shown to be able to provide spatial and temporal information on underfilm localized corrosion. However inhomogeneity still remains one of the less understood coating properties, more detailed research is needed in this area. The WBE could be applied in conjunction with scanning probe techniques in order to gain more detailed understanding on the initiation, propagation and localized corrosion under organic coatings and organic surface films. For instance it may be possible to correlate WBE galvanic current density, coating resistance or potential distribution maps with Volta potential profile measurable using the Scanning Kelvin Probe.