Characterizations And Fabrications Of EUV Multilayers Based On Improved Evolutionary Algorithm

In the design and characterization of extreme ultraviolet(EUV)multilayers,there are certain problems in the genetic algorithms that are mainly adopted,mainly including the problems of slow solution speed and low mass population.In the research process of this paper,the main The use of cloud model quantum evolution algorithm(CQEA)and real-coded quantum evolution algorithm(RQEA)in the characterization of wide-angle aperiodic EUV multilayer films and periodic EUV multilayer films solved various problems mentioned above.First of all,in the study of this paper,the establishment of inversion fitting program for grazing incidence X-ray reflection spectroscopy(GIXR)of an isobaric Mo/Si multilayer film is realized.The established process is mainly based on real-coded quantum evolutionary algorithms such as RQEA and real-coded genetic algorithm(RGA).Secondly,on the basis of the above,this paper realizes the design and establishment of a design procedure for aperiodic wide-angle EUV multilayers,which is mainly based on cloud model quantum evolutionary algorithms and real-coded quantum evolutionary algorithms,and according to the corresponding The results were analyzed experimentally and theoretically.The analysis results show that:(1)The evolutionary algorithm is improved,and an improved quantum evolution algorithm based on the evaluation of the functional gradient information of the equal-period EUV multilayer film is proposed.The characterization accuracy of the multilayer film and the design efficiency of the aperiodic EUV multilayer film are improved;(2)The dual-objective evolutionary algorithm is used to establish a joint solution algorithm for the characterization of equal-period EUV multilayer films,which solves the multi-solution problem in multilayers characterization and obtains high-precision multilayer film micro-structure parameters;(3)Robust multilayer designs for aperiodic Mo/Si multilayer films with wide-angle EUV supermirrors,theoretically evaluating robust film design for different film thickness errors,and analyzing the maximum film thickness control acceptable for film design error.The verification analysis shows that the dual-objective and cloud model quantum evolution algorithm has a very broad application prospect in the field of EUV multilayer film characterization and design,which lays a foundation for further work.