Comparison of cross-sectional image analysis with weight change measurements for assessing non-uniform attack during corrosion testing of aluminium

Abstract

Samples of aluminium have been exposed in a cyclic wet—dry exposure test using a mixed salt spray. Conventional assessment of corrosion of the aluminium, by the weight changes after corrosion product stripping, implied a loss in metal cross-section of 28% after 22 weeks’ testing. This result was compared with observation of cross-sectional losses using photographs of randomly selected cross-sections combined with quantitative image analysis. Results indicated that, in a significant number of locations, the loss of cross-sectional area was over 20% after 8 weeks and, after 22 weeks, the maximum loss in area was over 50% compared with an average loss of 30%. Such behaviour is not surprising because of the localised nature of aluminium corrosion. Thus, weight loss measurements, which are often used for corrosion rate assessment after accelerated salt spray testing, are clearly not appropriate in the case of non-uniform corrosion. However, image analysis using modern low cost equipment is a useful tool for the objective assessment of corrosion loss and may, depending upon the lifetime criterion in use, result in a useful reduction in overall testing time.