324 



A. W. AGAR AND R. S. M. REVELL 



600''C. By means of microdiffraction, the two re- 

 stricted directions can be determined to correspond 

 to the direction of [110] of a-brass. 



(2) Secondary scattering effect. The secondary 

 scattering effect for CuoO is shown in fig. 11. For 

 ZnO, the irrational diffraction spots due to this 

 effect are distributed more densely than those of 

 CuaO. Fig. 12 shows schematically this effect for 

 only the first quarter of the electron diffraction image 

 of a-brass. 



(3) Selective oxidation. For low pressure of oxygen 

 and the temperature under the melting point of 

 zinc, the surface of a-brass becomes rich in copper 

 atoms, because the vapour pressure of zinc is greater 

 than that of copper and the mobility of the copper 

 atom is greater than that of zinc. Accordingly, it 

 results in the formation of CuoO at an initial stage. 



At the high temperature over the melting point of 

 zinc, 419 C, the zinc atom will easily combine 

 with the oxygen atom, because the zinc atom be- 

 comes easily displaceable in the specimen. Further- 

 more, Cu.>0 has some tendency of reduction by the 

 above high diffusion of zinc or the dissociation itself 

 in vacuo. 



References 



1. BRiJCK, L., Ann. Physik 26, 233 (1936). 



2. GoTTSCHE, H.,Z. Physil< 134, 517 (1953). 



3. HONJO, G.. Proc. Pins. Soc. Jap. 8, 113 (1953). 



4. VON Laue, M.. Ann. Pliysik 26, 55 (1936). 



5. Menzer, G.,Z. Krist. A 99, 410 (1938). 



6. MiYAKE, S., Sci. Papers Inst. Phys. Cheni. Research 29, 



167(1936). 



7. Takahashi, N., Rikagal<iikenl<yi4Jo-iho 27, 536 (1939). 



(In Japanese.) 



8. WiLMAN, H., Proc. Roy. Soc. A 64, 329 (1951). 



The Electropolishing of Aluminium 



A. W. Agar and R. S. M. Revell 



Research Laboratory, Associated Electrical Industries Limited, Alclermaston, Berks., 

 and Metropolitan-Vickers Electrical Co. Ltd., Trafford Park, Manchester, 17 



When aluminium or its alloys are polished electro- 

 lytically, they are frequently found to exhibit a 

 fine structure when examined by replicas in the elec- 

 tron microscope. Various forms of these structures 

 have been described (1, 2, 3, 6). Typical forms are 

 shown in fig. 1 which may be described as furrows, 

 dots and whorls respectively. A whole range of struc- 



Fig. 1. Electropolished aluminium showing "whorl'" struc- 

 ture, and a boundary between this and one of the interme- 

 diate forms with "furrow" and "dotty"" structures. Shadow- 

 cast formvar replica. Magnification 10,000. 



ture intermediate between these has also been ob- 

 served. 



The above authors all employed aluminium oxide 

 replicas obtained by anodisation of the specimen 

 after polishing. It seemed possible that the structure 

 could have been produced as well by the anodising 

 as by the polishing process; consequently, the pol- 

 ished surfaces were examined by plastic replicas in the 

 present investigation to avoid any possible confusion. 

 Since the structures still appeared, it was clear that 

 they arose during the electropolishing process. 



The structure types may all appear on one speci- 

 men, and it is not possible to predict which type may 

 appear after the electropolishing. It seems likely (3, 

 6) that the type of structure is determined by the 

 underlying crystal orientation. Sharp boundaries 

 observed between one type of structure and another 

 lend support to this supposition (fig. 1). 



When the size of the structure, as opposed to its 

 type, is considered, however, it is found to be related 

 in some way to the polishing conditions. Since, in a 

 few cases, structureless surfaces were obtained, it 

 was decided to attempt to define the conditions for 

 obtaining structureless surfaces at will. 



The electrolyte used in this investigation was the per- 

 chloric acid/alcohol mixture (4). This was preferred as 

 being safer than the electrolyte containing acetic anhy- 

 dride and used by Jacquet. The specimens were super 

 purity aluminium (99.992 °o) so as to avoid any metallo- 

 graphic structures which might have confused the results. 



It was at first found very difficult to obtain reproducible 

 results from the electropolishing process, and it was found 



