We have an iron wire in a gaseous environment of HCl in a carrier gas of H2 and N2 at about 800 Deg C. We investigated the Fe_xCl_y concentration in the resulting gas at different HCl concentrations in the carrier gas. At first the dependence line in a double logarithmic plot had a slope of about one. After some experiments this slope decreased. The iron wire diameter stayed about the same but we can now see bowl like shapes on the surface. Our guess is that we first etched away the impurities and the pure iron remained. How would you expect the slope to change when changing the reaction from Fe_xOy + HCl to Fe + HCl?