Sensitivity of stress corrosion cracking of stainless steel to surface machining and grinding procedure

An investigation has been undertaken to establish the effect of surface preparation method on the susceptibility of a 304 stainless steel to stress corrosion cracking under simulated atmospheric corrosion conditions. MgCl₂ was deposited onto four-point bend specimens, which were then placed in a chamber with a relative humidity of 45% and temperature of 60 °C. These test conditions were designed to reflect external exposure of stainless steel components in industrial plant, including nuclear reactor components, situated in a coastal region, but with the severity of the exposure conditions enhanced to allow discrimination of the effect of surface preparation in a short timescale (up to 1500 h). Four surface preparation methods were evaluated: transverse grinding, longitudinal grinding, transverse dressing using an abrasive flap wheel, and transverse milling. For each case, surface topography, surface defect mapping, near-surface microhardness mapping, residual stress and electron back-scattered diffraction measurements were undertaken. Stress corrosion cracks were observed for the ground and milled specimens but not for the dressed specimens, with cracks apparently originating at corrosion pits. The density of cracks increased in the order: transverse ground, milled and longitudinal ground, with the cracks notably much smaller in length for the transverse ground condition. The propensity for cracking could be linked to the high residual stress and apparent nanocrystalline microstructure at the surface. There was a greater propensity for pitting to initiate at local defect sites on the surface (laps, deeper grooves). However, the tendency was not overwhelming, suggesting that other factors such as more general roughness or the distribution of MnS inclusions had an influence, perhaps reflecting the severity of the environment.