This research will replace conventional diffuser plates with porous, optimized lattice structures. These will be developed using physics-informed generative algorithms and multiscale simulation frameworks, then validated experimentally using Boise State’s MOCVD reactor system. The project leverages academic resources to overcome industrial lock-in, enabling retrofit-compatible designs that reduce cost while enhancing deposition uniformity.