Abstract
The characteristics of the deformation and fracture of metals under repeated cycles of stress, generally described as “fatigue phenomena,” have received very considerable attention, both experimental and theoretical. Usually, consideration has been devoted exclusively to conditions in which the metal subjected to fatigue has its free surface exposed to the ordinary atmosphere. In many cases in actual practice, however, metals are subjected to fatigue action while surrounded by a fluid—either gaseous or liquid—which is of a corrosive nature and the endurance or “life” of the metal is controlled by the simultaneous conjoint action of the applied stresses and the corrosive agent. To such conditions the term “corrosion-fatigue” has been applied. Attention was first directed to this aspect of fatigue phenomena in 1917 by Haigh, who demonstrated experimentally that, in general, fatigue stresses and corrosive influences may be mutually accelerative, producing more destructive effects than either influence when acting separately, or when the stressing is applied subsequent to the corrosion stage. The subject then appeared to escape further attention for a period of about nine years, after which the results of the first of a series of important researches were published; in this connection reference should be made to the work of Lehmann, McAdam, Speller, McCorkle and Mumma, Binnie, Fuller, Haigh and Jones, etc. From the researches of these investigators, and particularly from those of McAdam, a very large amount of data is now available regarding the corrosion-fatigue resistance of a wide range of metals and alloys, and the separate effects of such variables as frequency of stress cycle, number of cycles, corrosion time, applied range of stress, etc., also of corrosion inhibitors and accelerators. No attention has apparently been given hitherto to the changes in microstructure occurring during a corrosion-fatigue test and, as a result, no information existed on such fundamental points as (i) the general course of a corrosion-fatigue crack; whether intercrystalline or transcrystalline; (ii) the actual point of initiation of the crack; whether it is situated at a crystal boundary, or on the site of previous slip bands, or at local corrosion pits bearing no distinct relation to these special positions. In planning a research with these general objects in view it was decided to make experiments, under corrosion-fatigue conditions, on (
a
) a single crystal, (
b
) a specimen consisting of two large crystals with the separating boundary, and (
c
) a specimen consisting of the usual finely-divided aggregate of crystals. By using single crystals and large crystal specimens and employing X-ray, microscopical and mechanical methods, it was hoped to correlate corrosion-fatigue phenomena with the fine structure. The present report describes the observations made on what is believed to be the first corrosion-fatigue test on a single crystal.