Modeling Graphene Extraction Process Using Generative Diffusion Models

EasyChair Preprint 9782

Abstract

Graphene, a two-dimensional material composed of carbon atoms arranged in a hexagonal lattice, possesses a unique array of proper- ties that make it a highly sought-after material for a wide range of appli- cations. Its extraction process, a chemical reaction’s result is represented as an image which shows areas of synthesized material. Knowing initial conditions (oxidizer) the synthesis result could be modeled by generating possible visual outcome. A novel text2image pipeline to generate exper- imental images from chemical oxidizers are proposed. Key components of such pipeline are a textual input encoder and a conditional generative model. In this work the capabilities of certain text model and generative diffusion model are investigated and some conclusions are drawn provid- ing further suggestions for further full text2image pipeline development.

Keyphrases: CLIP, Graphene, diffusion, text2image

@booklet{EasyChair:9782,
  author    = {Modestas Grazys},
  title     = {Modeling Graphene Extraction Process Using Generative Diffusion Models},
  howpublished = {EasyChair Preprint 9782},
  year      = {EasyChair, 2023}}