Romanoff's soil corrosion data deconstructed

In 1957 Romanoff published the results of the monumental US soil corrosion field exposure program. It reported corrosion mass loss and penetration data mainly for a variety of ferrous metals exposed for up to 17 years in some 120 different, well-documented soils. For various reasons no coherent quantitative analysis of the data has yet been proposed. Herein the Romanoff data for steels is deconstructed, interpreted and to some extent quantified using the notion that the bi-modal model for immersion and atmospheric corrosion should apply. It also uses the notion that the conditions measured for the soils are not necessarily those at the metal-soil interface. Both maximum pit depth and corrosion mass loss are shown to be most severe under wet, low oxygen-availability conditions, and that any non-homogeneity in soils appears to be detrimental. In contrast, the presence of aggressive ions such as chlorides and sulphates appear to have only limited influence. Because of a lack of information about the availability of critical nutrients for bacterial activity the contribution of microbiologically influenced corrosion could not be assessed