1、Item No. 24187NACE Publication 5A195This Technical Committee Report has been preparedby NACE International Task Group T-5A-31*onFluid Flow Enhanced CorrosionState-of-the-Art Report on Controlled-FlowLaboratory Corrosion Tests December 1995, NACE InternationalThis NACE International technical committ
2、ee report represents a consensus of those individual members whohave reviewed this document, its scope, and provisions. Its acceptance does not in any respect preclude anyonefrom manufacturing, marketing, purchasing, or using products, processes, or procedures not included in this report.Nothing con
3、tained in this NACE International report is to be construed as granting any right, by implication orotherwise, to manufacture, sell, or use in connection with any method, apparatus, or product covered by LettersPatent, or as indemnifying or protecting anyone against liability for infringement of Let
4、ters Patent. This reportshould in no way be interpreted as a restriction on the use of better procedures or materials not discussed herein. Neither is this report intended to apply in all cases relating to the subject. Unpredictable circumstances maynegate the usefulness of this report in specific i
5、nstances. NACE International assumes no responsibility for theinterpretation or use of this report by other parties.Users of this NACE International report are responsible for reviewing appropriate health, safety,environmental, and regulatory documents and for determining their applicability in rela
6、tion to this report prior to itsuse. This NACE International report may not necessarily address all potential health and safety problems orenvironmental hazards associated with the use of materials, equipment, and/or operations detailed or referred towithin this report. Users of this NACE Internatio
7、nal report are also responsible for establishing appropriate health,safety, and environmental protection practices, in consultation with appropriate regulatory authorities if necessary,to achieve compliance with any existing applicable regulatory requirements prior to the use of this report.CAUTIONA
8、RY NOTICE: The user is cautioned to obtain the latest edition of this report. NACE Internationalreports are subject to periodic review, and may be revised or withdrawn at any time without prior notice. NACEreports are automatically withdrawn if more than 10 years old. Purchasers of NACE Internationa
9、l reports mayreceive current information on all NACE International publications by contacting the NACE InternationalMembership Services Department, P.O. Box 218340, Houston, Texas 77218-8340 (telephone +1 281228-6200).ForewordThis compilation of experimental techniques is intendedto provide the most
10、 up-to-date information available onevaluating the effect of velocity on corrosion. This reportis limited to the effects of single-phase flow andencompasses the effects of mass transfer, hydrodynamicshear, and differential mass transfer. The critical issuesfor experimental design are the appropriate
11、 choice of thetest protocol, the identification of the appropriate way tocorrelate the bench-top results to the full-scale system,and the identification of experimental systems for whichsuch correlation is feasible. Test design considerationsare discussed for a number of experimental systems.This te
12、chnical committee report is intended as an aid toengineers and technicians involved in laboratorycorrosion testing. The report updates and replaces NACEStandard TM0270, Method of Conducting ControlledVelocity Laboratory Corrosion Tests, which has beenwithdrawn and is available from NACE as a histori
13、caldocument only.This report was prepared by NACE International TaskGroup T-5A-31 on Fluid Flow Enhanced Corrosion, acomponent of Unit Committee T-5A on Corrosion inChemical Processes, and is issued by NACEInternational under the auspices of Group Committee T-5on Corrosion Problems in the Process In
14、dustries.* Chairman David C. Silverman, Monsanto, St. Louis, Missouri.NACE International2ContentsIntroduction2Interpretation of Laboratory Measurements.2Effect of Mass Transfer2Effect of Shear3Effect of Differential Mass Transfer3Effect of Multiphase Flow.3Philosophy Behind Experimental Design.3Hypo
15、thesis #1: Mass-Transfer Control3Hypothesis #2: Shear-Induced Removal of a Film.4Hypothesis #3: Differential Mass Transfer.5Experimental Systems5Rotating Disk.6Rotating Cylinder.7Impinging Jet.11Flow-Loop Systems.12References.16Appendix A - Summary of Features forExperimental Systems.17Appendix BGov
16、erning Equations.17Rotating Disk17Rotating Cylinder18Impinging Jet18Flow-Loop System19Appendix CSample Design Procedure.20Modeling of Corrosion Accelerationby ConvectiveMass Transfer.20Modeling of Corrosion Acceleration byShear Mechanism21Appendix D - The Chilton-Colburn Analogiesand Corrosion Rate.
17、21Appendix E - Notation22IntroductionThe sensitivity of corrosion to fluid flow has been testedby a variety of methods. The method adopted in NACEStandard TM0270 involved stirring the test solution in acontainer relative to a stationary inner drum on which thetest samples were mounted. While this me
18、thod offered arelatively simple and safe construction and the ability toconduct simultaneous tests on multiple coupons, theunknown hydrodynamics of the system made thelaboratory test results difficult to use to predict corrosionrates in the field. Furthermore, the test method could notbe used to obt
19、ain mechanistic information regarding rate-controlling steps in the corrosion process. The purposes of this report are (1) to describe state-of-the-art test methods that give quantitative results, (2) toprovide information on selection of the most appropriatetest method, and (3) to provide informati
20、on on theinterpretation of test results. There are two key issues in selecting an appropriatelaboratory experimental system for characterizing flow-enhanced corrosion. The first issue is identification of theappropriate way to correlate the bench-top results to thefull-scale system. A number of meth
21、ods have beensuggested, including equating mass-transfer coefficientsand hydrodynamic shear stress at the wall. The relativemerits of the methods for relating bench-topmeasurements to industrial systems are discussed. Thesecond issue is identification of experimental systems forwhich these quantitie
22、s are well defined. No singleexperimental system is ideal for all cases; therefore,several systems are described in this report. Thediscussion pertains to single-phase fluid flow. The references cited in this report are intended to providebackground of tutorial nature rather than acomprehensive revi
23、ew of the literature. For this reason,emphasis is placed on citation of textbooks and reviewarticles.Interpretation of Laboratory MeasurementsThe methods suggested for correlating small-scale teststo industrially significant systems have included equatingmass-transfer coefficients and hydrodynamic s
24、hearstress. The choice of correlation parameter depends onthe suspected mechanism for flow enhancement. Anumber of papers provide excellent reviews of theconcepts employed in this report.1,2Effect of Mass TransferCathodically limited corrosion may be expected formetals in contact with solutions that
25、 contain reduciblespecies such as dissolved oxygen or ferric ions. In thesecases, fluid flow will influence corrosion rates byinfluencing the rate at which these species aretransported to the metal surface. Cathodically limitedcorrosion rates in large-scale systems may beapproximated in the laborato
26、ry by choosing flow systemswith comparable mass-transfer coefficients. Anodicallylimited corrosion may also be subject to mass-transfereffects if the flow-induced removal of corrosion by-products from the metal surface reduces the effectiveNACE International3thickness of corrosion-product films. In
27、the case ofaqueous solutions where molecular oxygen is thecathodic reactant, the mass-transfer-limited currentdensity is often independent of temperature; the solubilityof oxygen at constant partial pressure decreases as thetemperature increases, counterbalancing the increase ofthe diffusivity with
28、increasing temperature. Consequently,measurements of the diffusion-limited current density at asingle temperature can be applied over a broadtemperature range. This phenomenon is not observedwith other oxidizing species; therefore, comparisonsbased on mass-transfer coefficients are generally used. D
29、esign of experiments to address this form of velocity-enhanced corrosion is comparatively easy sincecorrelations for mass-transfer coefficients are readilyavailable for a wide variety of flow configurations.Effect of ShearSurface shear stresses have been used to provide acorrelation between flow geo
30、metries. The concept isbased on the observation that some erosion-corrosionphenomena seem to involve shear-induced damage toprotective films. Similarity of the local hydrodynamicenvironment has been achieved if the shear stresses aremaintained the same between the actual environment andthe laborator
31、y simulation. Hydrodynamic shear mayinfluence the persistence of filming inhibitors, which areexpected to be more loosely bound to metals than tooxide films. For flows in which shear effects remove orprevent formation of protective films, corrosion rates inlarge-scale systems may be approximated in
32、thelaboratory by choosing flow systems with comparablesurface shear stress.Effect of Differential Mass TransferIn many cases, the effect of flow is to create galvaniccoupling between regions of the metal that experiencediffering flow fields. Corrosion in the anodic region maybe uniform or may be in
33、the form of pitting. This form oferosion-corrosion is difficult to predict because predictionrequires detailed knowledge of the flow field and of theanodic and cathodic characteristics of the metal systemin use.Effect of Multiphase FlowThe flow regime in a multiphase flow piping system canhave a sub
34、stantial effect on the mechanism of corrosionin that system.2As a result, the mechanism of corrosionunder multiphase flow conditions may not be the same asin single-phase flow experimentsdesigned to duplicatesuch parameters as mass transfer or hydrodynamicshear. For example, erosion-corrosion in oil
35、 and gasproduction piping has been proposed to be caused by theonset of an annular mist flow regime and the subsequentremoval of corrosion-product films by liquid-dropletimpingement. It is not possible to duplicate suchmultiphase corrosion mechanisms in a single-phaseexperiment. The design of experi
36、ments for velocity-enhanced corrosion in multiphase systems falls outsidethe scope of this report. If experiments are run in single-phase flow for experimental convenience, the results maynot represent actual field conditions and may besubstantially in error. Gaining a full understanding ofsystem co
37、nditions is important in order to avoidperforming experiments under inappropriate flow regimes.Philosophy Behind Experimental DesignCareful experimental design is essential when evaluatingthe influence of fluid flow on corrosion. Corrosion ratesare governed by the interaction of coupled physicalphen
38、omena that vary with each system.3The selection ofan appropriate experimental design can depend on theparticular hypothesis established for the cause of theobserved or expected velocity-enhanced corrosion.Hypothesis #1: Mass-Transfer ControlCorrosion rates may be influenced by mass transfer ofoxidiz
39、ing species to the surface if such species exist inthe fluid. The rotating disk or cylinder can be used to testwhether such mass-transfer effects are possible. Identification of a current plateau for polarization scans inthe anodic or cathodic directions is indicative of mass-transfer limitations to
40、 an anodic or a cathodic reaction,respectively. However, the appearance of a currentplateau may also be attributed to passivation. The knownhydrodynamic characteristics of the experimental systemcan be used to verify that the plateau is indeed caused byhydrodynamic mass-transfer effects. The logarit
41、hm ofthe limiting current is plotted against the logarithm of therotation rate and the slope of the line examined. If masstransfer controls the corrosion rate, then the slope will notdeviate significantly from 0.5 for a rotating disk and 0.7for a rotating cylinder.For example, a mass-transfer-limite
42、d current to a rotatingdisk is known to be proportional to the square root of therotation speed (see Figures 1 and 2). This relationshipmust be satisfied in order to state conclusively that thecurrent plateau is caused by limitations to hydrodynamicmass transfer. Note that salt films often form onco
43、rroding metals, and the diffusion resistance of thesefilms can cause deviations from the idealized behaviorshown in Figure 2. Note also that additional experimentsare performed to identify the species responsible for theNACE International4mass-transfer control apparent in the polarization scans. In
44、the event that clear plateaus are not obtained (caused,for example, by competing reactions) the square-rootdependence on rotation speed will still apply to mass-lossmeasurements and may apply to impedance polarizationmeasurements. A similar approach can be applied toidentify mass-transfer-limited re
45、actions with the rotatingcylinder or the impinging jet.Hypothesis #2: Shear-Induced Removal of a FilmA number of steps can be taken to explore the hypothesisthat a given occurrence of fluid-velocity-sensitivecorrosion can be attributed to shear-induced removal of afilm. The premise that the velocity
46、-sensitive corrosion iscaused by ordinary mass-transfer effects can beeliminated by the steps described in the previous section.Corrosion control by mass-transfer limitations to areacting species can be eliminated from consideration ifthe corrosion rate for a disk electrode, for example, is afunctio
47、n of rotation speed but is not a linear function ofthe square root of the rotation speed. Such results maysuggest that mass transfer through surface filmsinfluences corrosion. Shear-induced effects may beindicated if nonuniform corrosion is seen on a surface forwhich the mass-transfer rate is unifor
48、m but the shearstress is not. However, two strong notes of caution aremade: Corrosion is influenced by local potential drivingforces, concentration, and fluid velocity. Theconsequences of these influences are described inthe following section and in the section on rotatingdisk electrodes. Nonuniform
49、 corrosion caused byvariations in surface overpotential is sometimesattributed incorrectly to shear-induced phenomena. Shear-induced removal of a film is not proved bysimply correlating corrosion rates with shear stress. The mass-transfer coefficient depends on the sameflow parameters that influence the shear stress. Ifcorrosion rates correlate with shear stress, they willalso correlate with the mass-transfer coefficient. Additional experiments are therefore performed toprove the hypothesis. A convincing way ofconfirming shear damage is to locate remnants ofthe damaged corrosio
