The use of linear regression methods and Pearson’s correlation matrix to identify mechanical–physical–chemical parameters controlling the micro-electrochemical behaviour of machined copper

Machining introduces residual stresses at the specimen surface and modifies the microstructure and the texture in a small volume close to the surface. Such changes are important in controlling the corrosion behaviour of machined workpieces in the presence of an aggressive environment. In the present paper, the relationships between cutting parameters and the surface and subsurface characteristics of machined copper were first quantified experimentally using a complete plan of L8 (23) and a linear regression method. The influence of surface characteristics on the local polarisation curves of machined surfaces was then investigated. The shear-type crystallographic orientation generated under certain machining conditions tends to stabilize the material by inhibiting anodic reactions. This beneficial effect was balanced by the deleterious impact of roughness and quadratic stress. In the presence of lubrication, the current densities determined in the passive range were generally very low. The evolution of the open-circuit potential OCP measured under free corrosion conditions was analysed considering the surface characteristics and the Pearson’s correlation matrix. OCP was shifted in the anodic direction with increasing quadratic stress, roughness and lubrication (classified according to their increasing influence on the OCP).