Figure 1.

The autoclave for the corrosion potential meaurements and Stress Corrosion Cracking experiments.

Figure 2

Calculated potentials for the Ag-AgCl-KCl electrode with reference to NHE

Figure 3

Corrosion potential of platinum, AISI 304 and preoxidized 17MnMoV6.4: the water was ammoniated to pH 8.5, 200°C, 50 bar. The linear flow in the autoclave was 0.001 m/s.

Corrosion Potential Measurements in Boiler Water: The Influenece of Oxygen Content

G.A.A. van Osch and W.M.M. Huijbregts

Corrosion, Vol. 42, No. 2, february 1986, pg 120 -123. (paper 32)

The corrosion potentials of AISI 304 stainless steel (SS), platinum, and 17MnMoV6.4 steel were measured in boiler water. The measurements were performed in a refreshing autoclave system at 200 C in ammoniated water of pH 8.5. The oxygen content of the water was varied. The corrosion potential of the SS increased trom -400 to + 200 mV normal hydrogen electrode (NHE) in the oxygen range of 1 to 10 p pb.

Oxidation reactions play an important role in various corrosion processes. The corrosion potential of a metal depends highly on the oxygen content in the water. The potential value determines which corrosion reactions can occur thermodynamically. Examples are stress corrosion cracking (S CC) of sensitized stainless steel (SS) under boiling water reactor (BWR) conditions (1-3) ,rippling in boiler tubes (4,5) and erosioncorrosion of unalloyed steels (6).

In water-steam circuits of conventional and nuclear boilers (pressurized water reactors, PWR's), the oxygen content is kept to a minimum. In connection with intergranular stress assisted corrosion of sensitized stainless steels in BWR's, corrosion potentials have been measured over a wide range of oxygen values (0 to 10,000 p pb). For example, the corrosion potential data of Indig and Lee (1,7) are compared with those of leibovitz. The results of the three authors are not fully identical; Lee observed a gradual increase in corrosion potential between 5 and 1000 p pb O2, whereas the results of Indig show a more rapid increase in the corrosion potential in the oxygen range between 5 and 100 p pb. Leibovitz did not observe such a rapid increase in the oxygen/potential curve, probably because the test ran only from 20 to 10,000 p pb O2.

It can be concluded, therefore, that the oxygen/corrosion potential relationship is not well characterized. However, progressively more researchers are using the results of Indig to interpret their test results (2,9)

We decided to do extensive research on the dissolved oxygen/corrosion potential relationship tor various materials in ammoniated water at 200 C.

The measurements were performed at 200 C in a continuously refreshed system. The linear flow speed of the water was 10-3 ms-1 (8.5 L/hr). Demineralized water was thermally degassed in a reflux steam generator. Ammonia was dosed into the feed water to give a pH value of 8.5 at the high pressure inlet. The dissolved oxygen concentration in the autoclave was regulated by dosing the degassed water with aerated CO2-free water either before the high-pressure pump or before the autoclave. The dissolved oxygen content, pH, and conductivity were measured at the inlet and outlet to the system. The inlet and outlet conductivities were both about 0.0007 S/m. A pressure of 50 bar was maintained in the buffer vessel by nitrogen.

The refreshing system contained an AISI 304 SS autoclave for electrochemical measurements and stress corrosion cracking experiments (Figure 1). Two idential Ag/AgCI reference electrodes were used and two other measuring electrodes, one of platinum and one of a 17MnMoV6.4 steel sample. The autoclave was electrically isolated trom the system and served as the third measuring electrode.

The Ag-AgCI-0.001 M KCI half-cell was chosen as a reference electrode. Indig and Leibovitz thoroughly tested this system, and it proved to work satisfactorily (1,8). The potential values with reference to the normal hydrogen electrode (NHE) were calculated (Figure 2). Leibovitz calculated the solubility of AgCI as a function of temperature. The reference electrode capsules were filled with pure water because some of the molten AgCI on the Ag wire would dissolve to saturation. The potentials of such electrodes with respect to NHE are also given in Figure 2.

According to Leibovitz, the solubility of AgCI at 200 C amounts to 0.001 M; the electrodes were thus filled with 0.001 M KCI before exposure. Minor leakage of this chloride into the water would be replenished by the dissolution of some AgCI. If the potential difference between the two reference electrodes was 0, it was assumed that both were functioning well.

The oxygen concentrations at the inlet and outlet of the system were measured with a Weston and Stack model 3400 analyzer. For reliable results, the electrolyte in the oxygen analysis cell was regularly refreshed, and the lead electrode was frequent I y cleaned.

Five runs we re performed. Figure 6 shows the measured corrosion potentials on SS for the five runs. During these experiments, the oxygen content was changed at intervals by dosing the initially degassed water with air-containing water (carbondioxide gas free) before the high pressure pump or just at the autoclave inlet. When the increment in oxygen content was too large, particularly with low oxygen contents, it took several days for the corrosion potential on SS to become constant. When dosing oxygen just before the autoclave, an immediate response in corrosion potential was seen on the platinum electrode, even though the oxygen at the outlet remained low for several hours. The above effect may be explained by the oxidation of corrosion products in the high-pressure system. On the tubes and appendages, iron oxide (magnetite, Fe304) is formed that will oxidize to heamatite (Fe203) with increasing oxygen concentrations. Therefore, the oxygen content of the water was only changed after the potentials had become constant and an equilibrium had been established.

Figure 3 represents the results for SS, platinum, and 17MnMoV6.4 steel obtained in the last run. The latter steel was preoxidized for 7 days in water containing 200 ppb O2. The potentials of this sample are clearly lower than those of SS and platinum.

According to Murray, the pH values of pure and ammoniated water (0.07 mg/L NH3) at 200°C are 5.65 and 5.8, respectively (11). It is suggested that this small difference in pH will have little effect on the corrosion potential values.

Regarding the oxygen measurements, Indig and Lee used Beckman equipment, whereas the present investigatiors used the Weston and Stack type. For oxygen control, the desired amount of aerated water was injected, whereas Indig, Lee, and Leibovitz changed the oxygen content of the gas mixture above the water in the storage tank. Thus, it appears that a direct comparison of corrosion potential/oxygen content relationships may be ambiguous, either because the test conditions are reportedly different or because the data are incomplete.

In the present experiments, the corrosion potentials on SS at a given oxygen content varied by as much as 200 mV over the whole range of oxygen contents. On the other hand, for each individual test run, the potential values are on a characteristic curve, similar to those found in the experiments of Indig and Lee. The data of Lee and Indig show also a rather high scatter.

This scatter cannot be easily explained. However, we are of the opinion that the activity of the electrode surface is a primary factor for the reproducibility of the potential measurements. The passivity of the SS and the amount of the deposited oxide (magnetite and/or hematite) will influence the potential measurements.

An important difference between the measurements of Indig and Lee and ours is found in the oxygen value at which the corrosion potential increases rapidly for a given increment in dissolved oxygen content. It can be concluded that in the KEMA measurements at 200°C, the rapid increase in potential for an increment in dissolved oxygen content occurs at much lower oxygen concentrations (1 to 10 p pb) than those observed in Lee's measurements (5 to 1000 p pb) at 200°C.

1. The Ag-AgGI electrode is readily applicable to electrochemical measurements in high-temperature high-pressure water.The KEMA results show higher corrosion potentials with lower oxygen contents than the experiments in the US. The cause of this discrepancy is not yet clear.
2. A sudden increase in corrosion potential for a given increment in dissolved oxygen content was observed in 200°C water when the oxygen content was in the range of 1 to 10 p pb.
3. This critical potential range was also found to be higher in the US measurements (5 to 1000 p pb).

1. M. E. Indig, A. R. Mcllree, Gorrosion, Vol. 35, No.7, p. 288, 1979.
2. M. J. Povich, D. E. Broecker, Materials Performance, Vol. 18, No.10, p. 41, 1979.
3. A. K. Agrawal, G. A. Welch, J. A. Begley, R. W. Staehle, CORROSION/78, Paper No.187, National Association of Corrosion Engineers, Houston, Texas, 1978.
4. W. Schoch, H. Wiehn, E. Richter, H. Schuster, VGB-Mitteilungen, Vol. 52, No.3, p. 228, 1972.
5. W. M. M. Huijbregts, KEMA Scientific and Technological Report, Vol. 3, No.2, p. 33, 1985.
6. W. Kastner, K. Riedle, H. Tratz, VGB Kraftwerktechnik, Vol. 64, No.5, p. 452, 1984.
7. J. B. Lee, A. K. Agrawal, R. W. Staehle, ..Corrosion and Corrosion Cracking of Materials for Water-Cooled Reactors," EPRI Report NP-1741, Ohio State University, Electric Power Research Institute, Palo Alto, California, March 1981.
8. J. Leibovitz, W. R. Kassen, W. L. Pearl, S. G, Sawochka, "Improved Electrodes for BWR In-Plant ECP Monitoring," EPRI Report NP-2524, Electric Power Research Institute, Palo Alto, California, July 1982.
9. G. Herbsleb, VGB Kraftwerkstechnik, Vol. 64, No.2, p. 138, 1984.
10. K. J. Vetter, Electrochemical Kinetics, Theoretical and Experimental Aspects, Academic Press, New York, New York, p. 642, 1967.
11. W. M. M. Huijbregts, '.Oxygen and Corrosion Potential Effects on Chloride Stress Corrosion Cracking," KEMA Report SO 359/85, to be published.
12. R. C. Murray, J. W. Cobble, Proc. Int. Water Conf., Chemical Equilibria in Aqueous Systems at High Temperatures, Pittsburgh, Pennsylvania, p. 295, 1980.