Iron has been electropolished over a range of voltage and current i r~ thc acetic acidperchloric acid bath. Under optirnum conditions an "atomically smooth" surface has been obtained as evidenced by marked refraction clfects in the electron dilfraction patterns. There is also evidence for a thin film on the electropolished surface which thickens on further exposure to air to form a prcferelltially oriented cubic iron oxide. Light etching in nitric acid removed the film and formed a surface which, on diffraction, gave the more normal spot pattern of iron.The behavior of metals in corrosio~l and oxidation studies is very dependent on the original method of surface preparation. Mechanical polishing leads to plastic deformation of the structure which lnay penetrate quite deeply into the metal. Some impurities and original oxide film may be introduced into the metal by the polishing agent. Subsequent etching to relnove the deformed layer can lead to surfaces with high and variable surface roughness factors. Hydrogen reduction may introduce hydrogen into the metal and roughen the surface. For these reasons there has been a great deal of interest in the use of electropolishing as a neth hod of surface preparation.Satisfactory electropolishing gives rise to an undulating surface kvhich lnay best be described as "atornically smooth" (I). During reflection electron diffraction examination, the incident electron beam enters such a surface at varying angles of grazing incidence. Each resulting diffraction image is thus built up of a series of diffracted beams suffering slightly different refraction displacements towards the shadow edge. The over-all effect is the elongated spot pattern commonly found with electropolished surfaces (2).Pashley ( I ) indicates that on a flat substrate uniform films of the order of 2 A thiclr (monolayers) may be detected by electron diffraction under favorable conditions. Available information suggests that a layer of nlaterial 25-50 A thiclr co~npletely obscures the low angle diffraction maxima of a flat substrate (3, 4, 5). I t follows that inarlced refraction streaking of the substrate pattern and low background scatter strongly suggest freedom from such extensive contamination.There have been a number of attempts to produce "clean" and sillooth iron surfaces by electropolishing. Recently Ladage (6) has reported the satisfactory electropolisl~ing of iron in an ortho-phosphoric-chromic acid bath. Boswell (7) has electropolished iron in a chromic-acetic acid bath. Both these baths may contalninate the surface with chromium or phosphorus compounds.Electropolishing polycrystalline iron in the perchloric-acetic anhydride bath (8), Cohen (9) obtained surface films which obscured the substrate. Iron surfaces polished in a bath of different composition (10) were also contaminated by a film estimated to be about 100 A thiclr by electron microscopy. Wilma11 et al. (11,12,13) have polished iron single crystals in the perchloric-acetic anhydride bath (8) and claim to have produced clean "atomica...

No abstract

No abstract

oxides on evaporated iron films that the activation energy for the steady growth process (most probably cation diffusion) is much lower than that for the diffusion of Fe + + through bulk magnetite. The significance of this observation in relation to the diffusion mechanism in iron oxide films has been discussed by Adams and Roberts (23).While it is difficult to obtain measurements of the activation energy for diffusion of Fe + + in microcrystalline Fe304 and its variation with particle size, it appears that a more detailed investigation of this aspect of the oxidation process may help clarify some of the problems of oxidation anisotropy.

The oxidation of electropolished iron in the temperature region of 200~ is highly dependent on the surface orientation in respect to both the total thickness and phase composition of the oxide. In general, surface orientations close to the (001), (011), and (111) planes oxidize to give films with a large proportion of Fe304. The thick oxide formed on the (001) surface has a thin outer layer containing a high percentage of ~-Fe203. Specimens with orientations remote from the above low index planes oxidize to form films containing a smaller percentage of Fe304. On the (112) surface, ~-Fe203 develops in the outer layer of the oxide at an earlier stage than on the (001). No discrete phase boundaries exist between the oxide phases. These trends are also shown at higher temperatures. Thin films of Fe304 (up to about 100A) on the (001) and (112) surfaces behave differently with regard to cathodic reduction, with mainly formation of Fe + + from magnetite on the (001) surface, and a mixed reaction with a large amount of reduction to metallic iron on the (112) surface. A change in surface preparation can change the distribution of phases in the oxide.