Gravimetric corrosion monitoring and mathematical fitting of atmospheric corrosion data for carbon steel

The corrosion of carbon steel was studied in different atmospheres at sites in the Basque Country. A bilogarithmic law relating weight losses and exposure time is proposed. Coefficients a and b in the basic equation were formulated as functions of the measured atmospheric parameters. The system of equations was solved by a multiple linear least squares regression. The final equation In p = 2·689 + 2·784[S0₂{^0·4 + 3·102[NaClj^O·6 + 0·846 In t_w - 3·853[S0₂}0·5 [NaCl{0·6 - 0·212 In t + 0·024[S0₂} In t- 0·32[NaClj In t + 0·381 [S0₂{ [NaCl]} In t - 0·178t_w In t, where p is penetration (J1m) and t and tw are total exposure and time of wetness (years), fits the experimental results with a correlation coefficient of 0·985. Its predictive power, using results from the literature, was good when the environmental parameters were in the studied range.