Assessing the influence of agrochemicals on the nature of copper corrosion in the vadose zone of arable land Part 3: Geochemical modelling

Abstract

This is the third in a series of papers from a pilot project that has attempted to answer the question ‘does the application of agrochemicals accelerate the corrosion of archaeological metals in the top 50cm of the soil?’. We have approached it by a combination of field-based experiments, by establishing laboratory microcosms and by using geochemical modeling techniques to understand the processes involved. This paper reports on the geochemical modelling simulations that trace the potential corrosion pathways followed in two sets of laboratory microcosm experiments (‘Lab Beakers’ and ‘Lab Bins’) and one field experiment (at Palace Leas). This approach uses soil solution as the fluid mediating corrosion in the soil vadose zone. Soil solution was displaced following controlled exposure to fertilizers. Modelling using The Geochemists Workbench was carried out to mimic the experimental conditions, and predictions were compared with image analysis results, limited XRD analysis and published corrosion observations. We focus here on a sub-set of the data relating to the behaviour of the thinnest samples of copper in each case. As with the field and laboratory data previously reported, the results are sometimes contradictory, but on balance this project has demonstrated that applied agricultural chemicals are likely to accelerate the rate of corrosion of metal objects within 50cm of the surface. In particular, it is likely that any fertilizers containing KCI (‘potash’) will be particularly aggressive. Geochemical modeling generates plausible corrosion predictions based on post-depositional interaction between archaeological copper and soil solution, and appears to be useful in helping to simplify and understand corrosion pathways in naturally complex systems.