Electrochemical evaluation of under-deposit corrosion and its inhibition using the wire beam electrode method
Yongjun Tan ⇑,Young Fwu,Kriti Bhardwaj
Western Australian Corrosion Research Group,Department of Chemistry,Curtin University,GPO Box U1987,Perth,Australia
a r t i c l e i n f o Article history:
Received 11September 2010Accepted 15December 2010
Available online 23December 2010Keywords:A.Steel
B.Electrochemical calculation
C.Acid corrosion C.Acid inhibition C.Crevice corrosion
a b s t r a c t
A new experimental method has been applied to evaluate under-deposit corrosion and its inhibition by means of an electrochemically integrated multi-electrode array,namely the wire beam electrode (WBE).Maps showing galvanic current and corrosion potential distributions were measured from a WBE surface that was partially covered by sand.Under-deposit corrosion did not occur during the exposure of the WBE to carbon dioxide saturated brine under ambient temperature.The introduction of corrosion inhib-itor imidazoline and oxygen into the brine was found to significantly affect the patterns and rates of cor-rosion,leading to the initiation of under-deposit corrosion over the WBE.
Ó2010Elsevier Ltd.All rights reserved.
1.Introduction
The presence of solid deposits on metal surface can cause a localised form of corrosion,namely under-deposit corrosion (UDC).UDC is often pitting and mesa-type of attack that is fre-quently observed in cooling water systems where scales and fou-lants exist and in oil and gas pipelines where sand,debris,biofilm and carbonate deposit are present.In order to control UDC,aggressive pigging programs are used in industry to remove deposits on a regular basis.Unfortunately frequent pigging treat-ment is troublesome,expensive,and is not suitable for some pipe-line systems.Corrosion inhibitor treatment is often applied as an alternative or supplementary means of preventing UDC.Inhibitors are believed to mitigate UDC if applied in sufficient concentrations [1].However,the effects of inhibitors on UDC have not been effec-tively assessed in industry because UDC is considered nearly impossible to assess by normal corrosion testing techniques.Cur-rently there is only limited understanding on the efficiency of UDC inhibition.Most of the research and tests conducted in the past have been to explore the effects of inhibitors on carbon diox-ide (CO 2)corrosion of bare steel surfaces.Unfortunately corrosion data from bare steel testing cannot be used to determine the behaviour of under-deposit corrosion.It was reported that a crude oil pipeline that was found to have very low corrosion rates by con-ventional corrosion monitoring in the bulk fluid suffered severe under-deposit pitting corrosion on the bottom of the line where large volume of sediments existed [2].
Understanding the mechanism and factors affecting UDC is obvi-ously critical in determining optimal approaches to UDC problems and in developing effective corrosion inhibitors.Industry experience suggests that complex factors,such as the retention of aggressive species in the deposits,failure of inhibitors to penetrate the deposits,a large cathode to anode surface area ratio and the possible forma-tion of a localised differential concentration cell would affect UDC [1].Current explanation for UDC is often based on an assumption that the accumulation of deposits on metal surface prevents corro-sion inhibitors and biocides from accessing metal surface,leading to insufficient inhibitor concentration and accelerating anaerobic growth underneath deposits.These assumptions need to be verified through corrosion tests that are able to determine local electro-chemical parameters from under-deposit areas and that are capable of evaluating all major factors affecting UDC.
An effective UDC test not only needs to effectively simulate com-plex UDC environmental parameters and diffusion pathways,more importantly it should correctly simulate the mechanism that leads to the initiation and propagation of UDC.Ideally,it should be able to accurately measure local corrosion parameters on an instanta-neous basis.During recent years several techniques have been developed in various laboratories [3–9]in order to meet the industry requirements in UDC inhibitor testing and evaluation.For instance de Reus et al.[3]used an experiment device to measure inhibitor
performance under solid deposits.The device was made of two sets of three electrode arrays with one set covered with sand and another directly exposed to brine solution.Such device was designed to allow simultaneous electrochemical measurements at both uncov-ered and covered areas for direct electrochemical comparison.Using this device,de Reus et al.[3]found that a higher concentration of
0010-938X/$-see front matter Ó2010Elsevier Ltd.All rights reserved.doi:10.sci.2010.12.015
Corresponding author.Tel.:+61892663907;fax:+61892662300.
E-mail address:yj.tan@curtin.edu.au (Y.Tan).
Fig.1.Schematic diagram showing a UDC experiment setup.
large cathode and a small anode.Their
three specimens with two specimens
directly exposed to brine solution.One of
was coupled to the uncovered specimen.
assessed by detecting galvanic currents
covered and uncovered specimen.It was
ered specimen was anodically polarised
corrosion attack.Indeed this device could
however,it may not work in high
phasefluid where a high electrolyte
vanic currentflowing between the sand
specimen.Another issue is that the
simulate localised chemical changes that
covered steel pipeline surfaces.
Another experiment device designed
artificial pit electrode method[5–7]
pled through a zero resistance ammeter
steel and immersed in a brine solution.
current is believed to relate to the rate of
change before and after inhibitor addition
cator for assessing inhibitor performance.
the experiment was introduced by Han et
and the cathode were kept closer in order to simulate real situation
that the pit electrode and the large piece of steel should be parts of a single metal surface.Although the artificial pit electrode method should be able to detect galvanic currentsflowing between local-ised anodes and cathodes over a partially covered metal surface, it has similar limitations to the method by Pedersen et al.[4]in detecting galvanic current in high resistive media.It also has diffi-culties in simulating ions diffusion and chemical changes over a partially covered metal surface.It would not provide spatial infor-mation on localised corrosion.Another critical issue is that this method does not take the fact that UDC can occur by both direct and indirect mechanisms into consideration.In direct UDC attack, for instance corrosion under a bacteria containing deposit,galvanic currents couldflow only between anodic and cathodic areas that are both under the deposit.It is well-known that the growth of bacteria,such as sulphate-reducing bacteria and the formation of galvanic current if the deposit has a large resistance or if the elec-trolyte resistance is high.Under this situation,anodic areas will lo-cate near cathodic areas and galvanic currentsflow only between these neighbouring areas without passing through the external ZRA.
Indeed techniques discussed above have some limitations in simulating UDC mechanism and in measuring local electrochemi-cal parameters affecting UDC.Tan et al.[10]reviewed these limita-tions and concluded that sufficient attention needs to be paid to localised corrosion mechanism and its eff
ect on corrosion testing because the severity and pattern of corrosion is often determined by the corrosion mechanism.If a corrosion test fails to simulate corrosion mechanism such as galvanic effects,it could result in unsuccessful and misleading results[11–15].Limitations in conventional techniques in measuring local electrochemical parameters are also responsible for difficulties in measuring UDC. Electrochemical techniques,such as linear polarisation resistance and electrochemical impedance spectroscopy can be used to mea-sure the corrosion rates on an instantaneous basis;however,they have limitation in measuring localised corrosion[16–18].The cyclic polarisation method is probably the only traditional electro-chemical method that is used to determine localised corrosion sus-ceptibility[19];however,it does not determine the rates of localised corrosion.Relatively new methods such as scanning probe techniques have been developed into useful research tools for localised corrosion studies,however,scanning probes are unable to scan corrosion occurring under solid deposits.Electro-chemical noise analysis and the wire beam electrode(WBE)meth-od are probably the only techniques that have potential in measuring localised corrosion rates.Although some controversial issues still exist in the interpretation of electrochemical noise data, electrochemical noise has been recognised to be a rich source of information on the corrosion process and localised corrosion [20].The WBE is a multi-electrode array that has been applied successfully in various localised corrosion testing and research [16–18].The WBE’s working
surface is electrochemically-inte-grated by coupling all the wire terminals in the solid phase and by closely packing all the wires in the solid/electrolyte interface. This electrochemical integration allows the working surface of a WBE to effectively simulate a conventional one-piece electrode surface in electrochemical behaviour.Indeed research has shown that similar corrosion patterns were produced over WBE and conventional one-piece electrode surfaces when both were
Fig.2.The molecular structure of imidazoline.Fig.3.A typical baseline galvanic current map measured using a bare WBE exposed to a brine–CO2corrosion environment without inhibitor present.
1256
exposed to identical corrosion environments and this has been ver-ified theoretically[16–18].
Two important characteristics of the WBE method that are par-ticularly valuable for UDC testing are,(i)the WBE is applicable to high resistance multi-phase environment,as demonstrated in a previous study[21],and is thus able to simulate UDC under high resistance deposits;(ii)the WBE can map corrosion on an instan-taneous and continuous basis,providing unprecedented spatial and temporal information on localised corrosion processes occur-ring under-deposits.Turnbull et al.[9]used a multi-electrode array to study UDC;however,their electrode array was made of far-spaced and uncoupled
multi-electrodes that are not electrochemi-cally integrated to simulate localised corrosion occurring over a continuous metal surface.
2.Experimental
Fig.1a illustrates a UDC experiment setup using the WBE.The WBE used in this work was made from one hundred identical mild steel(UNS No.G10350)wires embedded in epoxy resin,insulated from each other with a very thin epoxy layer.Each wire had a diameter of0.19cm and acted both as a mini-electrode(sensor) and as a corrosion substrate.The working area was abraded with 240,320and1200grit silicon carbide paper,rinsed with deionised water to remove water-soluble contaminants and by ethanol to re-move organic contaminants.As shown in Fig.1b,the WBE working surface was partially covered with a rubber‘O’shaped ringfilled with sand to simulate a localised under-deposit corrosion environ-ment.The partially covered WBE surface was then exposed to3l of synthetic brine(3%NaCl by weight,0.01%NaHCO3by weight) contained in a custom-made electrochemical testing cell at room temperature(approximately21°C).During CO2corrosion testing, CO2sparging was continued to maintain a virtually oxygen free environment.During corrosion exposure periods,all the wire terminals of the WBE were connected together and,therefore,elec-trons could move freely between wires,in a similar way as would be the case with larger one-piece electro
de.All chemicals used were of analytical grade supplied by Sigma and Aldrich and used without further purification.
UDC processes were monitored by mapping galvanic currents across the multi-electrode array to understand how localised cor-rosion initiated and propagated under sand and how it changed with the introduction of the inhibitor or oxygen(O2).Galvanic cur-rent was measured between a chosen individual electrode and all the other electrodes shorted together using a pre-programmed Autoswitch device and an ACM AutoZRA.Corrosion potentials were also mapped when necessary to help understand the mechanism of UDC.Galvanic current and corrosion potential data were analysed by procedures similar to that described in references[21]and[22] using a custom-designed analysis software.The measurements were taken regularly to determine changes in corrosion processes and
patterns.
current distribution map measured from a WBE surface that was partially covered by sand and was exposed
In this paper,the term‘galvanic current’is used to describe cur-rents caused by local potential differences that can be originated from variations in surface metallurgy or in surface chemistry.Mea-surement of potential is useful for determining the thermodynam-ics of a corrosion cell;while the measurement of galvanic currents is useful for evaluating corrosion kinetics.UDC under O2containing atmospheric environment where UDC is considered to be similar to conventional crevice corrosion due to the formation of an oxygen differential aeration cell inside and out-side the deposit-covered metal surfaces.As shown in Fig.4,anodic dissolution currents concentrated on areas covered by a rubber‘O’ring and sand,while cathodic currents that are mainly due to the
polarised(b)corrosion potential maps measured over a WBE exposed to a brine–CO2corrosion environment for 1258Y.Tan et al./Corrosion Science53(2011)1254–1261

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。