Applying Statistics to Corrosion Data

The main insight into the kinetics and mechanism of iron dissolution and corrosion has come from electrochemical studies using complex E-t or I-t perturbation programs. Some of the conclusions that have emerged from this large body of work are summarized here. 

The electrochemical quantities relevant for the kinetics and mechanism, obtained from experimental measurements, are the interdq endence between potential, current, and species in the bulk of the solution (especially hydronium ions and anions) and time. Theoretically, one needs all the characteristic slopes SE/S log ij, Stj/Slogii, SE/Slogig, Sf] /S log is, S log i/S log aj, S log is/S log aj, Sa/St, S log ii/S pH, S log is/S pH, and S E/dt, as well as the characteristic times, that is, the waiting, peak, and polarization times tpeak and Tp, respectively. Practically, one cannot maintain all the conditions under control, but some of them are important, and others irrelevant. 

In spite of numerous investigations, a high variability in the data characterizes the electrochemical measurements, leading to the conclusion that important variables have been totally ignored or believed to be constant. Only a few examples are given below. 

Lorenz and co-workers reported for Fe in HCl + NaCl solutions a steady-state Tafel slope, b, close to 60 mV dec for pH 0-1.5 and 30 mV dec for pH 1.5-3, whereas Oftedal, apparently working under the same conditions, obtained fe = 30 mV dec in solutions of pH 1.5 and 60-70 mV dec for pH 1.5. Atkinson and Marshall found for Fe in IM HCl a steady-state Tafel slope close to 44 mV dec at lower current density and 85 mV dec at higher current density, at variance with Drazic and Drazic, who reported 120 mV dec for chlorine ion concentrations in the range between 10 and 2M at low current densities. Moreover, if 

During the past 10 years, there has been much effort to find out the causes of the irreproducibility of the data. The possible explanations are as follows. 

---