par Hou, Marc 
Référence Nuclear Instruments and Methods, 182-183, 1, page (153-162)
Publication Publié, 1981
Référence Nuclear Instruments and Methods, 182-183, 1, page (153-162)
Publication Publié, 1981
Article révisé par les pairs
| Résumé : | Component analysis is used to study the spatial distributions of point defects resulting from collision cascades in solids. The components are the three (orhogonal) eigenvectors of the covariance matrix of the spatial distribution. Those corresponding to the extreme eigenvalues determine the directions maximizing and minimizing the variance of the spatial distribution. The intermediate one is the direction maximizing the variance of the distribution projected on a plane perpendicular to the principal component. The standard deviations of the distribution projected on the three components give a measure of its size. This measure is only dependent on the cascade structure. Vacancy and interstitial distributions generated in metals by the computer code MARLOWE based on the binary collision approximation are analysed and compared in this picture. The simulation of hundreds of cascades generated by projectiles in the keV energy range incident on polycrystalline gold makes it possible to collect information on their average spatial anisotropy, energy density and on the cascade development. The dependence of those characteristics on the energy and the masses involved is discussed. © 1981. |
