SOME GENERAL considerations on the use of computerized systems in the corrosion field for performing electrochemical measurements, and collecting and processing data, using suitable software are given, and the usefulness of determining instantaneous corrosion rates for metallic materials using the polarization-resistance method is emphasized. It has been stressed that the use of the Stern and Geary method is empirically justified by its ability to provide reliable information. A useful hint for computing the actual value of E-m when experimental data are stored on a magnetic support is given, and their selection criterion is illustrated. Some results relating to the behaviour of iron in 0.5M H2SO4 solutions containing aliphatic amines at various concentration and at 25 degrees C demonstrate clearly that the best-fit with the monomial i = R(p)(-1)Delta E over the [-10, 10]mV interval associated to the Stern and Geary scheme fairly faithfully reproduces the theoretical values of I-c obtained using the POLCURFIT program, as illustrated by a numerical case, and the relative error never exceeds the value of 4%. Furthermore, as documented by a graphic example, it has been proved that in the vicinity of Delta E = 0, the trend of experimental points is very close to the best-fit straight line, in agreement with Skold and Larson's observations.The scheme for determining R-p through the best-fit with the polynomial i = a Delta E + b Delta E-2 + c Delta E-3 was adopted in the SOFTCOR-DC-PSI program. In this case, the reproduction of the trend of polarization curves over the Delta E [-20, 20]mV interval is very satisfactory, as documented by the case of iron in 5% by weight HCl solutions containing a commercial product and at temperature values ranging from 35 to 90 degrees C.Finally, it should be stressed that the use of a third-degree polynomial allows application of the Bonhoffer and Jena definition with a high level of confidence.

The behaviour of iron in Im HCl and 0.5m H2SO4 solutions at various temperatures with values ranging from 25 to 65degreesC was investigated in.order to determine the, actual dependence of B-a, B-c, and 1(c), and the activation heat of the dissolution process by assuming the validity of Arrhenius' law. The values of these parameters were obtained by processing each polarization curve with a non-linear method and accounting for the influence of the solution resistance. The direct corrosion current density 1(g) was obtained by a spectrophotometric analysis of the solution samples. Experimental results demonstrated that polarization curves, of this transient type are able, quite faithfully, to represent the actual. iron dissolution, even if, the percentage value of (1(c)- 1(g))(-1) increases as temperture rises, the highest discrepancy being observed in the case of H2SO4 solutions.The increase of B-a with rising temperature seems to indicate that the accumulation of Fe2+ ions close to the specimen,s surface could influence.the electrode kinetics. On the other hand, B-c does, not appear to be sensitive to the temperature influence. The calibration chart method seems to be a valid tool to determine the actual value of the corrosion rate when R-p values are-computed with reference to the potential interval, and to the contribution of the ohmic drop to the electrode potential is removed. Finally, Arrhenius's law is followed with accuracy by the' two electrochemical systems, and the extrapolation of the empirical relation.ships is thus a valid procedure for determining 1(c) at temperatures higher than 65degreesC. Furthermore' the activation heat allows a ranking of the' aggressiveness of the two acid media.

THE BASIC ASPECTS of the theories proposed by North and Pryor [12] and Gaparini et al. [14] to explain the growth of a protective lepidocrocite layer on the surface of copper and its alloys exposed to seawater conditioned with ferrous salts, and the approach dealing with a heterogeneous reaction occurring through an electrochemical path, are discussed. Some general considerations concerning the physical meaning of aluminium brass corrosion monitoring are developed, and the usefulness of performing in-field measurements to detect anomalous situations during the operation of steam condensers cooled with seawater is carefully examined in order to check the actual validity of information produced using suitable electrochemical techniques. The monitoring system, constituted by three- and five-electrode probes and a computerized corrosion rate meter, installed on the steam condensers of two units of an Italian power station, is described in detail. Concise considerations concerning the computer program developed to control the operation of the corrosion meter and collect experimental data are also given.Some data obtained from the electrode probes installed on the tube sheets of the second pass of the Unit 3 condenser are examined in detail to establish empirical guidelines for an optimal injection of Fe2+ ions into seawater. Finally, experimental data collected during a three-year period were rather useful for proper operation of the ferrous chloride injection system, for defining the optimal length of the continuous seawater treatment, detecting failures of the protective film, and carrying out discontinuous conditioning operations.

SOME GENERAL considerations about eht acid cleaning of steam generators in thermal power plants are given, and the importance of specific corrosion inhibitors in restraining corrosion rates in carbon steels is stressed. The use of interphase capacitance as a suitable parameter for evaluating the inhibitor effectiveness and chemical stability of organic compounds was examined in the case of iron and two carbon steels interacting with 5% wt HCl solution at temperatures ranging from 65 to 90 degreesC and containing a proper corrosion inhibitor at various concentrations.Alternating current measurements were performed over the 0.08-2 x 10(-4)Hz frequency interval under current control by imposing for each experimental point the constraint 4 < (0) < 6mV RMS, and adopting a suitable integration time. The value of the interphase capacitance was computed over the 4 x 10(3) - 2 x 10(-4)Hz frequency range using various mathematical schemes in order to verify the validity of the equivalent electrical network. The main result is that small values of the double-layer capacitance are associated with high performance in corrosion inhibitors. On the other hand, the increase in time of the capacitance value indicates a decrease of the chemical stability of the inhibitor.

The BEHAVIOUR of iron in H2SO4 solutions at different pH values and at 25 degreesC was examined by producing polarization curves of various types. Potentiodynamic polarization measurements were carried out using the three potential sweep rates of 1, 10, and 60 mV/min. of current-transient polarization curves were produced using current pulses with a length of 0.1 sec. and collecting electrode potential and current values 0.04 sec. after system excitation. Experimental data, corrected to account for the influence of the ohmic drop of the electrode potential, were processed using the NOLI method to compute the values of B-a, B-c, and I-c, and in several cases the successive approximation technique was used.The experimental results indicate that the values of the electrochemical parameters depend on experimental procedures adopted to carry out the polarization curves. The current-transient polarization curves relating to iron in solutions containing H2SO4 and Na2SO4 show that there is a marked dependence of B-a and B-c, on the pH. A comparison of the direct and electrochemical determinations of the corrosion rates for the 1NH(2)SO(4) solution stresses the usefulness of optimizing the electrochemical technique when it is mandatory to determine the correct magnitude of I-c, for instance by using a monotonic-increasing function of the potential sweep rate.

Some basic considerations concerning the use of the polarization resistance for corrosion rate monitoring over potential intervals as wide as [E-c - 20, E-c + 20] mV are developed and a useful hint for determining the correct magnitude-order of i '(0) through the best-fitting of experimental data with a polynomial of the third degree is given. Some general concepts relating to the corrosion behaviour of copper alloy in sea-water are also recalled. The experimental application concerns the behaviour of the copper alloys 687, 706 and 715 exposed to artificial sea-water at pH 8.2 and 40 degrees C, and having flow rates of 1.0, 1.5 and 2.0 ms(-1). Polarization measurements of the dynamic type were performed under current control, the starting point being always E-c, and usually refer to the Delta E interval [- 30, 0] mV. The values of the current sweep-rate were 5.7, 17.5 and 28.9 mu A/min. Experimental results showed that polarization measurements of the galvanodynamic type and the scheme based of the first derivative of i(Delta E) are effective to monitor corrosion rates of copper alloys and permit to determine values of R-p as high as 2.8 x 10(5) Ohm cm(-2) with good accuracy and reproducibility. At last, for the alloy 715 this procedure is suitable to detect without uncertainty the random occurrence of localised corrosion attacks.

Some basic considerations about organic coatings are given and a simple electric network simulating the response of the metal-coating system, exposed to an aggressive environment, to a sinusoidal signal of high frequency is examined. It is also proved that this approach provides reliable and correct information on the coating performance when the film defects have a negligible area. The computer program, developed to perform measurements over the [6 x 10(3), 6 x 10(4)] Hz frequency interval and process experimental data with a prefixed model, is briefly discussed as concerns its main features which are the proper choice of the voltage amplitude to be applied to the system under investigation and the optimisation of the current range as a function of the operating frequency. Two interesting experimental applications are illustrated: the first one refers to carbon steel specimens painted with different commercial products and exposed in a salt-spray cabinet. The second application is dealing with an epoxy coating applied to aluminium brass specimens that were exposed to artificial sea-water at 40 degrees C and having different flow rates. In both cases results were very promising and demonstrated the usefulness and validity of this methodology which is an effective way to face the problem to monitor the resistance to corrosion of composite systems.

Materials and CorrosionVolume 50, Issue 8 p. 476-476 Article Reply to the Discussion by F. Mansfeld and M. W. Kendig G. Rocchini, G. Rocchini Via Valera, I-20020 Arese (Italy)Search for more papers by this author G. Rocchini, G. Rocchini Via Valera, I-20020 Arese (Italy)Search for more papers by this author First published: 16 September 1999 https://doi.org/10.1002/(SICI)1521-4176(199908)50:8<476::AID-MACO476>3.0.CO;2-9AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat No abstract is available for this article. Volume50, Issue8August 1999Pages 476-476 RelatedInformation

The mathematical developments of the original and new versions of the Mansfeld method are given and the characteristic function G(u0) is introduced. The reliability and effectiveness of the use of G(u0) have been verified by processing theoretical data, computed over the ΔE interval [−150, 150]mV at 10mV steps and describing ideal systems with rather different values of Ic, where Ba does not exhibit large changes and Bc ranges from 70 to 120mV. The roots of G(u0) were computed using a false position method and single precision. Data processing after 30 iterations provided very accurate values of the electrochemical parameters. It was also verified that the accuracy of the improved method does not depend on the ΔE interval amplitude. The convergence criteria of the numerical sequence u0n − 1 + Δun have been discussed and some applications have been given. The results indicate that the convergence of u0n − 1 + Δun is very fast and very accurate evaluations of Ba, Bc and Ic are attained after a few iterations.

The physical meaning of monitoring corrosion rates and states of the metal-solution interphase using sinusoidal current at various frequencies is discussed. Three method concerning low-frequency measurements. area method associated to Nyquist's diagrams and the capacitance of the double layer are examined, The basic concepts underlying the use of these techniques are given and. in the case of low-frequency signals, some considerations on the mathematical theory based on the Kramers-Kronig relations between real and imaginary parts of Z(omega) are developed. The examples treated refer to laboratory and pilot-plant tt sts and illustrate the usefulness of monitoring electrochemical processes taking place in a very large class of systems, In the case of low-frequency measurements satisfactory-results were obtained, for instance, for iron in 1, 2 and 3 N H2SO4 solutions at 25 degrees C and for aluminium brass in a 5% wt HCl solution at 70 degrees C. The capacitance method was employed to examine the behaviour of iron in 5% wt HCl solutions containing a corrosion inhibitor at various concentrations at temperatures ranging from 35 to 90 degrees C. Experimental results show that the impedance technique is a valid tool to investigate the behaviour of electrochemical systems without any restraint on the shape of their voltage-current characteristic. Lastly, some considerations concern specific software developed to perform electrochemical measurements with commercial computerized systems and to process experimental data in real time.

On a etudie les effets de la temperature, du dioxyde de soufre (SO 2 ) et des echanges thermiques sur la corrosion de huit alliages conventionnels et avances, utilisables pour les echangeurs de chaleur et les composants non refroidis des systemes de combustion a lit fluidise. Les essais ont ete conduits a 650 °C et 850 °C sur des echantillons recouverts avec depots synthetiques de soufre et d'oxyde de calcium, avec des faibles additions de carbonate de potassium et de sodium, dans l'air et dans une atmosphere de combustion simulee avec 0.1% de SO 2 . Pour reproduire les conditions des echanges thermiques le flux de chaleur a ete simule a 650 °C avec une sonde refroidie a l'air, placee dans un four a 850 °C. Les resultats montrent une grande influence de tous les parametres consideres sur le comportement a la sulfuration des materiaux.

A direct method for monitoring the corrosion rate of metals or alloys, based on the use of computerized systems, is described. Polarization measurements were performed using a suitable system driven by a specific program which permits also to process experimental data with the NOLI method after the completion of each curve in order to compute Ic, Ba and Bc. The behaviour of iron in 1 N H2SO4 solutions at 25 °C was examined by carrying out polarization curves of the potentiodynamic type at potential sweep rates of 120, 300 and 600 mV min−1 and collecting simultaneously the values of the current and electrode potential. Each run lasted about 15 h; potentiodynamic polarization curves were performed over the potential interval [Ec − 90, Ec + 60]mV, starting from the cathodic region, at regular time intervals. NOLI method was successful in processing all the experimental data. The experimental results show that the corrosion current density is a monotonic increasing function of the potential sweep rate. The corrosion rate of iron was underestimated when the potential sweep rates of 120 and 300 mV min−1 were used. The best result was obtained using a potential sweep rate of 600 mV min−1. In this case electrochemical and direct determinations of the corrosion rate are in good agreement and the apparent dissolution valence takes the value of 1.93 which confirms the fact that iron dissolution occurs through the formation of Fe2+ ions.

The definition and meaning of the polarization resistance, after Bonhöffer and Jena, are discussed, and its relationship with the corrosion rate is given. The basic mathematical ideas suitable to introduce the method of calibration charts, first proposed by Skold and Larson, are expounded by considering the current‐voltage characteristic i(ΔE) = Ic(eβE – e‐βE) and the correctness of their practical application based on a suitable relationship between Id and RP is proved. Some numerical methods which are useful to compute RP by processing experimental data in the vicinity of βE = 0 or inside intervals as wide as β−50, 50 mV are concisely given. Experimental applications refer to laboratory tests and show that usually Id as a function of RP must be best‐fitted with the function Id(Rp) = NRP−M, M being different from 1. In the case of iron in 0.5 m H2SO4 solutions, containing some primary aliphatic amines, the corrosion rate takes the form Id = NRP−1 and confirms the goodness of the polarization resistance method, even if the value of the proportionality constant N differs from that one predicted by the theory of Stern and Geary. The results were very satisfactory and proved that the calibration chart method is a valid tool to monitor the reaction kinetics when corrosion is of the uniform type.

The influence of the solution resistance between the reference and working electrodes on experimental polarization curves relating to a class of electrochemical systems described by the law i=Ic(eαΔEe-e−βΔEc) has been examined and a numerical method for computing Rs has been proposed. The best-fitting method, based on the function In i = P+ QΔE + Ri, P, Q and R being the unknown parameters, processes data belonging to the anodic and cathodic Tafel regions. Applications to artificial data, obtained from the analytical expression of i(ΔE), which accounts for the influence of Rs, and relating to the ΔE interval [− 300, 300] mV, show its usefulness and validity. Some cases simulated the shape of real polarization curves, characterised by Ic = 0.1 mA cm−2, Ba = 40 mV and Bc = 120 mV, as a function of Rs over the interval [0.1, 5] Ω. Their processing was successful and demonstrated that from a practical point of view the anodic and cathodic regions provide the same value of Rs. Furthermore the proposed method reproduces with a good accuracy the actual values of Ic, Ba and Bc. Rather similar results were observed when the distortion of polarization curves for R s = 1 Ω was studied as a function of Ic over the interval [0.01, 4.0] mA cm−2, even if the cathodic region seems to provide better information at high values of Ic. At last the proposed method can be improved using the successive approximation technique because in this case the distinction between the anodic and cathodic regions tends to disappear when ΔE intervals of suitable amplitude are considered.

Some concise ideas on the mathematical meaning of the linear polarization resistance are given and a method is proposed to determine the width of the ΔE interval around the point ΔE = 0 where the shape of an experimental polarization curve can be approximated by the monomial i = Rp−1 ΔE with a good accuracy consistent with the criterion adopted to search for its left and right bounds. The left and right bounds are determined by searching for the roots of a suitable function obtained from the best-fitting of experimental data with a polynomial of the fourth degree. The applications discussed refer to the behaviour of iron in H2SO4 solutions at various normality and 25 °C, and 1 N HC1 and H2SO4 solutions, containing suitable corrosion inhibitors, at 75 °C. Polarization measurements were performed using both potentiodynamic technique with different sweep rates and current transients. The dependence of Rp on ε, which determines the interval width, was examined using the LINPOL program. This program permits to best-fit experimental data, account for the influence of the ohmic drop on the electrode potential and to compute the left and right bounds of the ΔE interval as a function of ε. Examination of Rp values relating to the previous systems indicates that from a practical point of view this quantity can be considered independent of the width of the ΔE interval when ε ≤ 5 so that in many cases it is not mandatory to process only experimental data belonging to the [−0,10]mV interval.

The mathematical bases of the Barnartt and LeRoy three point methods are expounded by introducing an elementary algebraic identity which allows them to be unified, the difference in the known terms of the resulting equations depending only on the chosen criteria of the points (ΔEk, i(ΔEk)) (k = 1,2,3). These methods have been used to process experimental data on iron in H2SO4 solutions at various normalities and 25 °C, and 1 N H2SO4 and HCl solutions, containing commercial inhibitors at various concentrations, at 75 °C. All the experimental data were best-fitted over the ΔE interval [−35,35]mV with a polynomial of the fourth degree in order to meet the choice criteria for any value of ΔEi. Based on the polynomial best-fitting a criterion for checking the goodness of the three point method is suggested. The Barnartt method was applied to analyse artificial data obtained by setting Ic = 1 mA cm−2, Ba = 40 mV and Bc = 120 mV, and subjected to arbitrary changes, simulating random or systematic errors, in order to verify its value and reliability. Comparison of the two three-point methods and NOLI analysis shows that the determinations of Ic, Ba and Bc are rather satisfactory only for the systems at 25 °C. For the systems at 75 °C the Barnartt method works properly only as concerns the evaluation of Ic, whereas the use of the LeRoy method is more critical. At last, examination of artificial data points out that in some cases the equation of the second degree obtained using the Barnartt method has complex roots.

The definition of the polarization resistance, Rp, as the reciprocal of the first derivative of the function i = i(ΔE) at the point ΔE = 0 is introduced by considering the kinetic law of a system where more than two elementary electrochemical reactions occur. The importance of establishing a direct relationship between Ic and Rp is stressed. The theory of this new formulation, based on the Maclaurin series expansion of the ideal response f(ΔE), is expounded and some considerations on the numerical calculation of definite integrals are also given. Finally, some experimental applications of this theory are presented. They refer to different electrochemical systems and show that, even if their response close to the mixed potential is of the non-linear type, it is always possible to obtain a well-defined value of Rp. The systems examined are: iron and 1 N H2SO4 solutions at 25 °C containing KCl at various concentrations; five carbon steels in solutions having 100 g 1−1 of EDTA at pH 9 and 100 °C; aluminium brass in a solution containing 35 g 1−1 of NaCl at 35 °C. The definite integrals were computed using ΔE = 0 mV as lower bound and 15, 20 and 30 mV or −25, 15 and 25 mV as upper bounds. Usually experimental data were best-fitted with a polynomial of the fourth degree over the ΔE interval [−50, 50]mV. It was observed that the polynomial of the fourth degree reproduced very faithfully the values of i(ΔE) inside the interval [−30, 30]mV. Examination of the results shows that a good agreement between the proposed and other methods exists and stresses the self-consistence and accuracy of the experimental data (ΔEk, ik) (k = 1, 2,…, n) obtained with different electrochemical techniques.

The influence of the ohmic drop on the electrode kinetics controlled by the activation energy has been examined by assuming two prevailing processes and a simple representation of the metal-solution interphase by an electrical dipole. The mathematical formalism has permitted the introduction of the apparent polarization resistance, Rpa, and the form factor H which is essential for a correct use of the linear polarization method. Artificial data, relating to the interval [−20,20]mV, were obtained with reference to the curve described by Ba = 30 mV, Bc = 120 mV and Ic = 0.100 mA cm−2 by searching for the roots of a suitable function with the Newton method at various values of Rs. Other artificial data were generated by considering the same values of Ba and Bc, assuming Rs = 2 Ω and changing the value of Ic. The analysis of both sets of artificial data by the NOLI method was successful and permitted the dependence of the electrochemical parameters on Rs or Ic to be determined quantitatively. The analysis of data concerning Rs = 2 Ω and Ic ranging from 0.1 to 1.0 mA cm−2 stresses the importance of reducing the contribution of the ohmic drop to the electrode potential because Ica takes values greater than the true ones. Examination of some examples concerning artificial and experimental data showed that the form factor is essential for the calculation of the actual value of Ic when its determination is based on the use of Rpa.

The basic ideas of the inversion method for computing the corrosion rate of a given metal or alloy in an aggressive environment are expounded and the main mathematical equations are given. It is also shown that the correct application of this numerical technique needs, in principle, a preliminary knowledge of the values of the anodic and cathodic Tafel slopes, Ba and Bc. It has been proved that Ic does not seem to depend critically on the random choice of the values of Ba and Bc when they are rather close to the true ones. Examination of some artificial data showed that the approximate polynomial of the fourth degree gives very accurate values of ΔE as a function of w over the interval [− 0.5,0.5]. The inversion method has been used to compute Ic relating to Armco iron in H2SO4 solutions at various pH values and at 25 °C, and in 1 N H2SO4 and HCl solutions containing suitable corrosion inhibitors at 75 °C. Polarization measurements, performed using current transients and experimental data corrected to account for the contribution of the ohmic drop to the electrode potential, were generally best-fitted over the ΔE interval [− 50, 50] mV using the INTER1 program. The actual values of Ba and Bc were obtained by analysing the polarization curves examined by the NOLI method. The analysis of the behaviour of these electrochemical systems has given satisfactory results compared with those obtained by the NOLI and linear polarization methods.