1、Designation: G 96 90 (Reapproved 2001)e1Standard Guide forOn-Line Monitoring of Corrosion in Plant Equipment(Electrical and Electrochemical Methods)1This standard is issued under the fixed designation G 96; the number immediately following the designation indicates the year of originaladoption or, i
2、n the case of revision, the year of last revision.Anumber in parentheses indicates the year of last reapproval.Asuperscriptepsilon (e) indicates an editorial change since the last revision or reapproval.e1NOTEThe Appendix section was editorially corrected in October 2001.1. Scope1.1 This guide outli
3、nes the procedure for conducting on-linecorrosion monitoring of metals in plant equipment underoperating conditions by the use of electrical or electrochemicalmethods. Within the limitations described, these test methodscan be used to determine cumulative metal loss or instanta-neous corrosion rate,
4、 intermittently or on a continuous basis,without removal of the monitoring probes from the plant.1.2 The following test methods are included: Test MethodAfor electrical resistance, and Test Method B for polarizationresistance.1.2.1 Test Method A provides information on cumulativemetal loss, and corr
5、osion rate is inferred. This test methodresponds to the remaining metal thickness except as describedin Section 5.1.2.2 Method B is based on electrochemical measurementsfor determination of instantaneous corrosion rate but mayrequire calibration with other techniques to obtain true corro-sion rates.
6、 Its primary value is the rapid detection of changes inthe corrosion rate that may be indicative of undesirablechanges in the process environment.1.3 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user of this stan
7、dard to establish appro-priate safety and health practices and determine the applica-bility of regulatory limitations prior to use. Specific precau-tionary statements are given in 5.6.2. Referenced Documents2.1 ASTM Standards:D 1125 Test Methods for Electrical Conductivity and Re-sistivity of Water2
8、G1 Practice for Preparing, Cleaning, and Evaluating Cor-rosion Test Specimens3G3 Practice for ConventionsApplicable to ElectrochemicalMeasurements in Corrosion Testing3G4 Guide for Conducting Corrosion Tests in Field Appli-cations3G15 Terminology Relating to Corrosion and CorrosionTesting3G59 Test M
9、ethod for Conducting Potentiodynamic Polar-ization Resistance Measurements3G 102 Practice for Calculation of Corrosion Rates andRelated Information from Electrochemical Measurements33. Terminology3.1 DefinitionsSee Terminology G15 for definitions ofterms used in this guide.4. Summary of Guide4.1 Tes
10、t Method AElectrical ResistanceThe electricalresistance test method operates on the principle that theelectrical resistance of a measuring element (wire, strip, or tubeof metal) increases as its cross-sectional area decreases:R 5slA(1)where:R = resistance,s = resistivity of metal (temperature depend
11、ent),l = length, andA = cross-section area.In practice, the resistance ratio between the measuringelement exposed to corrosion and the resistance of a similarreference element protected from corrosion is measured, tocompensate for resistivity changes due to temperature. Basedon the initial cross-sec
12、tional area of the measurement element,the cumulative metal loss at the time of reading is determined.Metal loss measurements are taken periodically and manuallyor automatically recorded against a time base. The slope of the1This guide is under the jurisdiction of ASTM Committee G01 on Corrosion ofM
13、etals and is the direct responsibility of ASTM Subcommittee G01.11 onElectrochemical Measurements in Corrosion Testing.Current edition approved March 30, 1990. Published May 1990.2Annual Book of ASTM Standards, Vol 11.01.3Annual Book of ASTM Standards, Vol 03.02.1Copyright ASTM International, 100 Ba
14、rr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.curve of metal loss against time at any point is the correctionrate at that point. The more frequently measurements are taken,the better is the resolution of the curve from which thecorrosion rate is derived.4.1.1 The elec
15、trical resistance of the metal elements beingmeasured is very low (typically 2 to 10 mV). Consequently,special measurement techniques and cables are required tominimize the effect of cable resistance and electrical noise.4.1.2 Various probe element cross-sectional areas are nec-essary so that a wide
16、 range of corrosion rates can be monitoredwith acceptable resolution.4.2 Test Method BPolarization Resistance:4.2.1 The polarization resistance test method involves inter-action with the electrochemical corrosion mechanism of metalsin electrolytes in order to measure the instantaneous corrosionrate.
17、 Its particular advantage is its speed of response tocorrosion rate upsets. On a corroding electrode subject tocertain qualifications (see 12.1), it has been shown that thecurrent density associated with a small polarization of theelectrode is directly proportional to the corrosion rate of theelectr
18、ode.4.2.2 The polarization resistance equation is derived in TestMethod G59. See Practice G3for applicable conventions. Forsmall polarization of the electrode (typically DE up to 20 mV),the corrosion current density is defined as:icorr5BRp(2)where:B = a combination of the anodic and cathodic Tafel s
19、lopes( ba,bc), andRp= the polarization resistance with dimensions ohmcm2.B 5babc2.303 ba1 bc!(3)4.2.3 The corrosion current density, icorr, can be convertedto corrosion rate of the electrode by Faradays law if theequivalent weight (EW) and density, r, of the corroding metalare known (see Practice G
20、102):corrosion rate 5 K1icorrrEW (4)where:K1= a constant.4.2.4 Equivalent weight of an element is the molecularweight divided by the valency of the reaction (that is, thenumber of electrons involved in the electrochemical reaction).4.2.5 In order to obtain an alloy equivalent weight that is inpropor
21、tion with the mass fraction of the elements present andtheir valence, it must be assumed that the oxidation process isuniform and does not occur selectively; that is, each element ofthe alloy corrodes as it would if it were the only elementpresent. In some situations these assumptions are not valid.
22、4.2.6 Effective equivalent weight of an alloy is as follows:1(lmnifiWi(5)where:fi= mass fraction of ithelement in the alloy,Wi= atomic weight of the ithelement in the alloy,ni= exhibited valence of the ithelement under the condi-tions of the corrosion process, andm = number of component elements in
23、the alloy (normallyonly elements above 1 mass % in the alloy areconsidered).Alloy equivalent weights have been calculated for manyengineering metals and alloys and are tabulated in PracticeG 102.4.2.7 Fig. 1 represents an equivalent circuit of polarizationresistance probe electrodes in a corroding e
24、nvironment. Thevalue of the double layer capacitance, Cdl, determines thecharging time before the current density reaches a constantvalue, i, when a small potential is applied between the test andauxiliary electrode. In practice, this can vary from a fewseconds up to hours. When determining the pola
25、rizationresistance, Rp, correction or compensation for solution resis-tance, Rs, is important when Rsbecomes significant comparedto Rp. Test Methods D 1125 describes test methods for electri-cal conductivity and resistivity of water.4.2.8 Two-electrode probes, and three-electrode probes withthe refe
26、rence electrode equidistant from the test and auxiliaryelectrode, do not correct for effects of solution resistance,NOTE 1Rs= Solution Resistance (ohmcm2) between test and auxiliary electrodes (increases with electrode spacing and solution resistivity).Ru= Uncompensated component of solution resista
27、nce (between test and reference electrodes) (ohmcm2).Rp= Polarization Resistance Rp(ohmcm2).Cdl = Double layer capacitance of liquid/metal interface.i = Corrosion current density.FIG. 1 Equivalent Circuit of Polarization Resistance ProbeG 96 90 (2001)e12without special electronic solution resistance
28、 compensation.With high to moderate conductivity environments, this effect ofsolution resistance is not normally significant (see Fig. 2).4.2.9 Three-electrode probes compensate for the solutionresistance, Rs, by varying degrees depending on the positionand proximity of the reference electrode to th
29、e test electrode.With a close-spaced reference electrode, the effects of Rscan bereduced up to approximately ten fold. This extends the oper-ating range over which adequate determination of the polar-ization resistance can be made (see Fig. 2).4.2.10 A two-electrode probe with electrochemical imped-
30、ance measurement technique at high frequency short circuitsthe double layer capacitance, Cdl, so that a measurement ofsolution resistance, Rs, can be made for application as acorrection. This also extends the operating range over whichadequate determination of polarization resistance can be made(see
31、 Fig. 2).4.2.11 Even with solution resistance compensation, there isa practical limit to the correction (see Fig. 2). At highersolution resistivities the polarization resistance technique can-not be used, but the electrical resistance technique may beused.4.2.12 Other methods of compensating for the
32、 effects ofsolution resistance, such as current interruption, electrochemi-cal impedance and positive feedback have so far generally beenconfined to controlled laboratory tests.5. Significance and Use5.1 General corrosion is characterized by areas of greater orlesser attack, throughout the plant, at
33、 a particular location, oreven on a particular probe. Therefore, the estimation ofcorrosion rate as with mass loss coupons involves an averagingacross the surface of the probe. Allowance must be made forthe fact that areas of greater or lesser penetration usually existNOTE 1See Appendix X1 for deriv
34、ation of curves and Table X1.1 for description of points A, B, C and D.NOTE 2Operating limits are based on 20 % error in measurement of polarization resistance equivalent circuit (see Fig. 1).NOTE 3In the Stern-Geary equations, an empirical value of B = 27.5 mV has been used on the ordinate axis of
35、the graph for “typical corrosion rateof carbon steel”.NOTE 4Conductivitymhos!cm51 000 000Resistivity ohmcm!NOTE 5Effects of solution resistance are based on a probe geometry with cylindrical test and auxiliary electrodes of 4.75 mm (0.187 in.) diameter,31.7 mm (1.25 ft) long with their axes spaced 9
36、.53 mm (0.375 in.) apart. Empirical data shows that solution resistance (ohmscm2) for thisgeometry = 0.55 3 resistivity (ohmscm2).NOTE 6A two-electrode probe, or three-electrode probe with the reference electrode equidistant from the test and auxiliary electrode, includes % ofsolution resistance bet
37、ween working and auxiliary electrodes in its measurement of Rp.NOTE 7A close-space reference electrode on a three electrode probe is assumed to be one that measures 5 % of solution resistance.NOTE 8In the method for Curve 1, basic polarization resistance measurement determines 2Rp+ Rs(see Fig. 1). H
38、igh frequency measurement shortcircuits Cdlto measure Rs. By subtraction polarization resistance, Rpis determined. The curve is based on high frequency measurement at 834 Hz withCdlof 40 F/cm2on above electrodes and 6 1.5 % accuracy of each of the two measurements.NOTE 9Curve 1 is limited at high co
39、nductivity to approximately 700 mpy by error due to impedance of Cdlat frequency 834 Hz.At low conductivityit is limited by the error in subtraction of two measurements where difference is small and the measurements large.NOTE 10Errors increase rapidly beyond the 20 % error line (see Appendix X1, Ta
40、ble X1.1).FIG. 2 Guidelines on Operating Range for Polarization ResistanceG 96 90 (2001)e13on the surface. Visual inspection of the probe element, coupon,or electrode is required to determine the degree of interferencein the measurement caused by such variability. This variabilityis less critical wh
41、ere relative changes in corrosion rate are to bedetected.5.2 Both electrical test methods described in this guideprovide a technique for determining corrosion rates without theneed to physically enter the system to withdraw coupons asrequired by the methods described in Guide G4.5.3 Test Method B ha
42、s the additional advantage of provid-ing corrosion rate measurement within minutes.5.4 These techniques are useful in systems where processupsets or other problems can create corrosive conditions. Anearly warning of corrosive attack can permit remedial actionbefore significant damage occurs to proce
43、ss equipment.5.5 These techniques are also useful where inhibitor addi-tions are used to control the corrosion of equipment. Theindication of an increasing corrosion rate can be used to signalthe need for additional inhibitor.5.6 Control of corrosion in process equipment requires aknowledge of the r
44、ate of attack on an ongoing basis. These testmethods can be used to provide such information in digitalformat easily transferred to computers for analysis.TEST METHOD AELECTRICAL RESISTANCE(1-6)46. Limitations and Interferences6.1 Results are representative for average metal loss on theprobe element
45、. On wire-form measuring elements, pitting maybe indicated by rapid increases in metal loss reading after 50 %of probe life is passed. The larger cylindrical measuringelements are much less sensitive to the effect of pitting attack.Where pitting is the only form of attack, probes may yieldunreliable
46、 results.6.2 It should be recognized that the thermal noise andstress-induced noise on probe elements, and electrical noise onthese systems, occur in varying degrees due to the process andlocal environment. Care should be exercised in the choice ofthe system to minimize these effects. Electrical noi
47、se can beminimized by use of correct cabling, and careful location ofequipment and cable runs (where applicable) to avoid electri-cally noisy sources such as power cables, heavy duty motors,switchgear, and radio transmitters.6.2.1 The electrical resistivity of metals increases withincreased temperat
48、ure. Although basic temperature compensa-tion is obtained by measuring the resistance ratio of an exposedtest element and protected reference element, the exposedelement will respond more rapidly to a change in temperaturethan does the protected reference element. This is a form ofthermal noise. Var
49、ious probes have different sensitivities tosuch thermal noise. Where temperature fluctuations may besignificant, preference should be given to probes with thelowest thermal noise sensitivity.6.2.2 If probe elements are flexed due to excessive flowconditions, a strain gage effect can be produced introducingstress noise onto the probe measurement. Suitable probeelement shielding can remove such effects.6.3 Process fluids, except liquid metals and certain moltensalts, do not normally have sufficient electrical conductivity toproduce a significant shorting
