Probability density fittings of corrosion test-data: Implications on C6H15NO3 effectiveness on concrete steel-rebar corrosion

Abstract

In this study, corrosion test-data of steel-rebar in concrete were subjected to the fittings of the Normal, Gumbel and the Weibull probability distribution functions. This was done to investigate the suitability of the results of the fitted test-data, by these distributions, for modelling the effectiveness of C6H15NO3, triethanolamine (TEA), admixtures on the corrosion of steel-rebar in concrete in NaCl and in H2SO4 test-media. For this, six different concentrations of TEA were admixed in replicates of steel-reinforced concrete samples which were immersed in the saline/marine and the microbial/industrial simulating test-environments for seventy-five days. From these, distribution fittings of the non-destructive electrochemical measurements were subjected to the Kolmogorov–Smirnov goodness-of-fit statistics and to the analyses of variance modelling for studying test-data compatibility to the fittings and testing significance. Although all fittings of test-data followed similar trends of significant testing, the fittings of the corrosion rate test data followed the Weibull more than the Normal and the Gumbel distribution fittings, thus supporting use of the Weibull fittings for modelling effectiveness. The effectiveness models on rebar corrosion, based on these, identified 0.083% TEA with optimal inhibition efficiency, η = 72.17± 10.68%, in NaCl medium while 0.667% TEA was the only admixture with positive effectiveness, η = 56.45±15.85%, in H2SO4 medium. These results bear implications on the concentrations of TEA for effective corrosion protection of concrete steel-rebar in saline/marine and in industrial/microbial environments.