VOID FORMATION IN NEUTRON IRRADIATED METALS.

The characteristics of void formation in high purity metals are reviewed and summarized in terms of irradiation and material variables.Fluence, irradiation temperature, flux and spectral effects, and the role of impurities are discussed.Results are compared and contrasted with those for stainless steel and evaluated in terms of proposed mechanisms for the nucleation and growth of voids.Although general trends of behavior are established for most of the experimental variables, exceptions are invariably found.The observations indicate that details of the damage structure are purity and structure sensitive.the range of conditions under which voids are known to form.Although the problem is a complex one involving vacancies, interstitials and transmutation products and their migration through the host material, comparatively little "work has gone into studies of relatively pure materials.In 1968 Mastel and Brimhall 2 first found voids in an irradiated pure metal, nickel, and observed that their formation was greatly accelerated over that in stainless steel.Voids were found in nickel at fluences two to three orders of magnitude lower than in stainless steel.Since that time several studies have established that voids form in a wide range of materials crystallizing in the facecentered cubic, body-centered cubic, and hexagonal close-packed crystal systems.A summary of all observations of voids in relatively pure metals is tabulated in the Appendix to this paper.In this paper we shall examine only the effects of irradiation and material variables on the formation of voids.The characteristics of void formation in pure metals will be compared and contrasted with those in stainless steel, and the observations will also be used to evaluate proposed models for.' the formation of voids.The effects of the voids on mechanical i an** physical properties will not be considered nor will be the response of the voids to annealing treatments.The latter subject is the topic of another paper at this symposium. 3At this point the reader should be aware that the picture is by no means complete.It is somewhat like a jigsaw puzzle at a stage where only a few isolated areas have been fitted together.There is no tendency or general trend that is not violated at one time or another.Part of the problem may arise because we do not yet recognize the full range of variables