Web http://www.jcse.org/

Volume 1 Paper 17


A Novel Crevice Corrosion Experiment Using A Wire Beam Electrode

Yong-Jun Tan

Western Australian Corrosion Research Group, School of Applied Chemistry, Curtin University of Technology, GPO Box U1987, Perth 6001, Western Australia

Abstract

This paper describes a new experiment for studying crevice corrosion. A wire beam electrode was employed in this experiment to measure electrochemical parameters directly from crevice area and these parameters were used to calculate instantaneous crevice corrosion kinetics. A clear correlation between calculated corrosion depth map and real corrosion appearance was obtained.

Keywords: Corrosion measurement, crevice corrosion, the wire beam electrode

Introduction

Crevice corrosion often occurs in fissure or occluded region where the local solution is stagnant and is isolated from the bulk solution. Although crevice corrosion has been studied extensively and been modelled mathematically over the past decades, some aspects of crevice corrosion remain unclear and the completeness and accuracy of the proposed crevice corrosion mechanisms and models are often not experimentally verified. This situation may largely be due to technical difficulties associated with conventional crevice corrosion experiments.

Crevice corrosion is a heterogeneous electrochemical process in nature. When crevice corrosion occurs, there is a distinct separation of the anodic and cathodic regions on the metal surface and different electrochemical reactions occur on the anodic and cathodic areas. Electrons continuously travel from the anode areas to the cathode areas through the electrode body and at the same time ions travel between anode and cathode areas through the electrolyte, resulting in rapid corrosion penetration on crevice areas. To characterise a crevice corrosion process, electrochemical parameters at local areas of a working electrode surface such as local anodic reaction current, have to be determined. However, conventional electrochemical techniques have major difficulties in doing this. This is mainly because conventional electrochemical techniques use a one-piece electrode, an electrode which is constructed by a single piece of metal or other electrically conductive material of a chosen size. When such a one-piece electrode is used, normally only mixed and averaged electrochemical parameters, for instance a mixed potential, over the whole electrode surface are measurable. These measured electrochemical parameters are neither related specifically to the anodic zone nor the cathodic zone of the heterogeneous electrode surface. Again, with a one-piece electrode, it is impossible to measure the galvanic currents that flow in the electrode body between localised anodic and cathodic sites since an ammeter is not able to be inserted between anodic and cathodic sites which are located on a single piece of metal surface. Furthermore, conventional electrochemical techniques have limitations in determining the kinetics of heterogeneous electrode processes. The fundamental formulation describing the electrochemical kinetics over a metal surface, the Butler-Volmer equation, is based on a uniform electrochemical mechanism. Traditional electrochemical techniques which are based on the Butler-Volmer equation such as the Tafel polarisation method, the linear polarisation method and the AC impedance spectroscopy, in principle, are applicable only to measure the electrochemical kinetics of a uniform electrode surface.

To overcome some of the difficulties associated with conventional electrochemical methods. A multi-piece electrode, namely the wire beam electrode (WBE), has been developed [1-8]. The WBE is fabricated from a metallic wire bundle embedded in insulating materials. The metal wire bundle is constructed from many electrically insulated metal wire sensors which are made from metal wires of identical, dissimilar, galvanised, heat-treated or stressed metal materials with the terminals of the wire bundle connected together. The WBE can be variously designed to meet different industrial and research requirements. Its surface shape normally simulates the surface shape of a practical work-piece. A schematic diagram showing the conceptual design of the WBE and its application in crevice corrosion testing is in Figure 1.

Figure 1. A schematic diagram showing the WBE and its application in crevice corrosion testing.

The WBE was initially developed and used to detect local defects in organic coating films [1-4]. Its application has been extended to the study of localised corrosion phenomena [3-9] and the study of various heterogeneous electrochemical processes such as cathodic protection processes [3, 7]. The WBE has several characteristics which make it uniquely suitable for studying localised corrosion and other heterogeneous electrochemical processes. Firstly, each wire in a WBE is an individual addressable electrochemical sensor. This enables a WBE to measure electrochemical parameters from local areas of an integrated electrode surface by means of wires located at those areas. Secondly, the working surface of a WBE, integrated by connecting all wire terminals, can effectively simulate the electrochemical corrosion processes occurring on a conventional one piece electrode surface [6, 7, 9]. Similar corrosion patterns were experimentally observed over a WBE and a large area one piece electrode surface when both electrodes were exposed to an identical environment designed to cause localised corrosion [6, 7] and this experimental result has been explained [6]. Thirdly, the surface area of each wire in an WBE is much smaller than the total working electrode area, thus corrosion and electrochemical processes on each wire surface can be assumed to be uniform even if the whole electrode surface is electrochemically nonuniform [6]. This approach is an analogy to calculus principles. This assumption allows electrochemical techniques and theories of describing uniform electrochemical processes to be applied to each wire in an integrated WBE, i.e. traditional electrochemical theories and techniques are extended to study localised corrosion and other heterogeneous electrochemical processes. Based on this assumption and the Butler-Volmer equation, new equations describing the electrochemical kinetics of each individual wire have been derived [6] and applied to study the kinetics of localised corrosion processes [6, 8, 9].

This paper describes a simple crevice corrosion experiment. A wire beam electrode was employed in this experiment as a key tool to measure electrochemical parameters directly from crevice area and these parameters were used to calculate instantaneous crevice corrosion kinetics. Some preliminary experimental results are discussed in this paper.

Experimental

Experimental setup

A simple electrochemical cell (Figure 1) was employed in this work to study crevice corrosion. A WBE was put on a ceramic semi-ball (with a diameter of 8.5 cm) under steady state conditions at room temperature to allow crevice corrosion to occur. The electrochemical cell contained 350 mL of 3% NaCl brine.

The WBE used in this work was made from one hundred identical mild steel (UNS No. G10350) wires. These wires were embedded in an epoxy resin, insulated from each other with a very thin epoxy layer. Each wire had a diameter of 0.18 cm and acted both as a minielectrode (sensor) and as a corrosion substrate. The working area (the area occupied by the wire beam) was approximately 3.24 cm2 (1.8 cm 1.8 cm). The total metallic area was approximately 2.55 cm2. The distribution of these minielectrodes in the working surface of the WBE is shown in Figure 1. The working surface of the WBE was polished with 400, 800 and 1200 grit silicon carbide paper and cleaned with ethanol and isopropanol.

Data acquisition

Corrosion potential distribution was obtained by measuring the open circuit potential of each wire of a WBE (Ek for the wire k) against a 3M Ag/AgCl reference electrode. The terminals of wires in the electrode were temporarily disconnected from the WBE system and connected in sequence to an automatic zero resistance ammeter (autoZRA, ACM Instruments, England) using a computer controlled automatic switch device (custom made). The autoZRA, which is able to measure both current and voltage, was used to record the corrosion potential. Its data logging software runs in the Microsoft Windows environment complete with a real time Excel link. Thus the potential and current data analysis and plotting can be performed with Microsoft Excel and Mathcad software. Corrosion potentials from local areas of the electrode were measured repeatedly during the experimental period, and were plotted to produce a corrosion potential distribution map.

The autoZRA was also connected in sequence between a chosen individual wire terminal and all other terminals shorted together using the computer controlled automatic switch to measure galvanic currents (Igk for wire k) flowing between each individual wire and the wire beam system. The autoZRA enables current (325 mA to 10 pA) and voltage to be measured accurately and to be recorded automatically. Galvanic corrosion currents from local areas of the electrode are measured and plotted to produce a galvanic corrosion current distribution map. The measurements were repeated during the experimental period.

Data analysis for corrosion rate mapping

Previously derived equations [6] can be used to calculate localised corrosion rates at any selected location of the WBE surface. In this work, corrosion reaction current of a selected wire k, Ika, was calculated by,

Ika = Igk/{1 - exp[- ()(Esys - Ek)]} (1)

Where Igk is galvanic current flowing between wire k and the rest, Esys is the overall corrosion potential of the whole WBE system, Ek is the potential of wire k, bak and bck are Tafel slopes (estimated).

Applying equation 1 to each wire in a WBE (wires 1 to 100), corrosion reaction current at all locations of the WBE can be calculated and be transferred to corrosion rate distribution over the WBE surface. In this way the corrosion rate and its distribution over the WBE can be mapped. A corrosion rate distribution map gives information on the processes, rates and pattern of localised or general corrosion.

Results and Discussion

As shown in Figure 2, the electrode surface was electrochemically heterogeneous during the 90 hours exposure. The potential maps show a nonuniform potential distribution and there is a potential difference of over 200 mV between anodic areas (inside region of the crevice zone) and cathodic areas (outside region of the crevice zone). The most negative corrosion potential was recorded around a ring-shaped area where the galvanic corrosion current was the most positive (indicating electrons leaving the area).

The potential and galvanic current distribution data were used to calculate corrosion rate distribution over the corroding electrode surface, using equation 1. It is clear that corrosion concentrated on the ring shaped area where corrosion potential was the most negative. Corrosion rates are the highest on this ring shapes area (Figure 2).

With the extension of the electrode exposure, as shown in Figure 2, the anodic corrosion sites generally remained at similar locations on the electrode surface although the specific corrosion rates did vary. As an accumulative result, relatively heavy localised corrosion occurred around the ring shaped area. This is shown in an accumulative corrosion depth map (Figure 3a) that was obtained by summing the corrosion depths calculated over various periods of exposure. For wire k, its total accumulative corrosion depth TotalDepthk is,

TotalDepthk = CorrDepth(1)k + CorrDepth(2)k + CorrDepth(3)k +…CorrDepth(i)k +… (2)

Where CorrDepth(i)k is the incremental corrosion depth calculated using corrosion rate data measured during a specific exposure period i, which is normally 24 hours.

The accumulative corrosion depth map (Figure 3a) correlates quite well with real corrosion appearance (Figure 3b). It is interesting to note that heavy localised corrosion did not occur at the center of the crevice area. This phenomenon may be related to the dependence of crevice corrosion on solution chemistry, geometries of the crevice and other factors. Details of crevice corrosion is under further investigation.

Fig. 2a. Current at 0.5 hr Fig. 2b. Potential at 0.5 hr  Fig. 2c. Corrosion rate at 0.5 hr
Fig. 2d. Current at 2.5 hr Fig. 2e. Potential at 2.5 hr Fig. 2f. Corrosion rate at 2.5 hr
Fig. 2g. Current at 65 hr Fig. 2h. Potential at 65 hr Fig. 2i. Corrosion rate at 65 hr
Fig. 2j. Current at 90 hr Fig. 2k. Potential at 90 hr Fig. 2l. Corrosion rate at 90 hr

Figure 2. Galvanic current (in mA/cm2), potential (in V) and corrosion rate (in mm/y) distribution maps measured after a WBE was exposed to a crevice corrosion environment for various periods.

Figure 3a Figure 3b (click on the image for an enlarged view)

Figure 3. The accumulative corrosion depth map (Figure 3a) and real corrosion appearance (Figure 3b) after a WBE was exposed to a crevice corrosion environment for 90 hours.

4. Conclusions

A WBE has been employed in a simple experiment for studying the processes of crevice corrosion and its kinetics. Corrosion potential and galvanic current distribution maps were experimentally measured and were used to calculate and plot corrosion rate distribution maps. Comparison between calculated corrosion depth map and corrosion appearance shows a good correlation. This work shows that the WBE is a practical tool for studying crevice corrosion.

References

1. Y. J. Tan, Progress in Organic Coatings, 19, 89 (1991).

2. Y. J. Tan and S. T. Yu, Progress in Organic Coatings, 19, 257 (1991).

3. Y. J. Tan, US patent application No. 09/093,576.

4. C. L. Wu, X. J. Zhou, and Y. J. Tan, Progress in Organic Coatings, 25, 379 (1995).

5. Y. J. Tan, Corrosion (NACE), 50, 266 (1994).

6. Y. J. Tan, Corrosion (NACE), 54, 403 (1998).

7. Y. J. Tan, Corrosion Science, 41, 229 (1999).

8. Y. J. Tan, S. Bailey, B. Kinsella and A. Lowe, Mapping corrosion kinetics using the wire beam electrode in conjunction with electrochemical noise resistance measurements, J. Electrochem. Soc., in press (1999).

9. Y. Tan, S. Bailey and B. Kinsella, Mapping non-uniform corrosion in practical corrosive environments using the wire beam electrode method (I), (II) and (III), Corrosion Science, a series of three papers, to be submitted (1999).