When measuring OCP over an inert thin film coating (e.g. silica on steel) most people measure a positive shift in OCP. These coatings are seldom defect-free and from our understanding the OCP over the coating should be equivalent to the OCP of the substrate. Why does the OCP change over a coating system?