▎ 摘　　要

In this paper, a hybrid approach that combines a light trapping formalism and a metaheuristic optimization approach is suggested to enhance speed and sensing capabilities of a thin film/4H-SiC photodetector (PD). To overcome the shadowing problem, interdigitated electrodes made of graphene are employed. Besides, Ag nanoparticles are inserted to enhance photon absorption in the 4H-SiC active area. To avoid the complexity of considering all randomly deposited nanoparticles, the Maxwell-Garnet theory-based dielectric method is employed. The impact of plasmonic engineering factors on the photodetector characteristics is studied. By using embedded Ag nanoparticles and highly transparent graphene electrodes, the proposed device exhibits a maximum responsivity of 269.6 mA/W, a high photocurrent to dark current ratio (PDCR) of 1.7 x 106, a linear -dynamic-range (LDR) of 272 dB, a response time tau = 13.45 mu s, and a detectivity of 5.4 x 1013 Jones. In addition, the suggested model is regarded as a fitness function to be used for a Multi Objective Genetic Algorithm (MOGA) optimization method which optimizes the sensing ability of the device and its speed. The optimized design discloses optimum performances in terms of responsivity (564.5 mA/W), PDCR (3.75 x 106), LDR (301 dB), and detectivity (1.25 x 1014 Jones). Also, it balances the compromise between sensing performance and response time (tau = 4.7 mu s). Our results show that using an efficient light trapping formalism and a metaheuristic opti-mization technique, it is possible to notably improve the photodetector sensitivity.