Corrosion of archaeological bronze artefacts in acidic soil

Corrosion of bronze in soil is a well-known phenomenon. In particular, archaeological artefacts which may remain in the soil for thousands of years are subject to severe corrosion. However, bronze objects excavated 50-100 years ago seem to be less corroded than those found today. Therefore, recent pollution of the soil is suspected to accelerate the corrosion. An interdisciplinary project has been started in Sweden to search for correlations between the degree of bronze corrosion, corrosion products, general archaeological and environmental conditions, and parameters characterizing the soil chemically. From three archaeological sites near Stockholm (Birka, Fresta, and Valsta), 33 bronze artefacts and related samples of soil have been investigated. All corrosion products and the metal core (if any) were analysed by SEM/EDS and XRD. Metal oxides, carbonates, sulphates, chlorides and phosphates have been identified. Each soil sample has been geologically classified, and a number of chemical analyses have been undertaken: pH in water and KCI, resistivity, loss on ignition, exchangeable acidity, chloride, phosphate, sulphur contents, acid-soluble cations extracted in two different ways, etc. About 8000 data have been compiled in an EXCEL data base. A statistical evaluation including multivariate modelling and analysis utilizing the SIMCA-S system, has been undertaken. The results so far obtained are only tentative but suggest that high concentrations of soot, sulphur or phosphate in the soil may have accelerated the corrosion of the investigated bronze objects. The influence of low pH values, though, is less clear.