USE OF ARTIFICIAL-INTELLIGENCE TECHNIQUES FOR THE DESCRIPTION OF PROCESSES IN NI AL MULTILAYERS/

Knowledge discovery is a novel research area in the field of artificia l intelligence. Its aim is to discover empirical laws that govern the behavior of complex systems using measurements of system variables. In this paper a brief description of the GOLDHORN knowledge discovery sy stem is presented. GOLDHORN discovers differential equations and has f eatures for handling noisy data, including some digital filters. In th e present case, this method was used to describe analytically atomic m igration in thin layers. A multilayer structure of nickel and aluminum was deposited on a copper substrate using the triode sputtering syste m and hollow cathode CVD plasma deposition. The composition of the ele ments in the deposited layers was determined by Auger electron spectro scopy (AES). The structure was then annealed for different times. Afte r annealing, the samples were analyzed again. The AES data were then a nalyzed by the GOLDHORN software package in order to obtain an analyti cal description of atomic migration as a function of the relative conc entration of elements in a layer. The analysis shows that the rate of migration of Al in Ni depends on the relative concentrations of the el ements. Different phases appeared to be indicated via the changes in t he slope of the curve. Our results show that knowledge discovery is a very useful tool for analyzing complex processes such as atomic migrat ion in multilayer systems.