Computational analysis of compressible flow using finite element method for two-dimensional multi-component diffusion of isotopes

• A 2D finite element code to solve the multi-component diffusion equation without radial averaging has been developed. • A turbulent flow model for a more physically realistic simulation of the gas centrifuge is implemented. • The machine cut (θ) as the most influential parameter on separation performance has been identified. • The overall separation factor (γ 0 ) of 1.15 for a five-component UF 6 mixture was obtained. Gas centrifuge technology represents a critical and efficient method for isotope separation. This study presents a comprehensive numerical investigation of flow and separation processes within a gas centrifuge, addressing significant simplifications in previous works. Unlike conventional approaches that assume laminar flow and employ radial averaging approximations for the diffusion equation, this research implements a more physically realistic turbulent flow model in ANSYS Fluent. It develops a custom two-dimensional finite element code to solve the full multi-component diffusion equation without simplifying assumptions. Validation of the finite element code against established methods for binary mixtures shows excellent agreement. For a five-component UF 6 mixture, the model calculates an overall separation factor (γ 0 ) of 1.15. A rigorous variance-based global sensitivity analysis using Latin Hypercube Sampling and Sobol indices identifies and quantifies the influence of key parameters, revealing that the machine cut (θ) is the most influential parameter, accounting for approximately 37% of the output variance, followed by rotor speed (ω). In contrast, the scoop drive force has a negligible impact. These findings provide a robust framework for optimizing centrifuge design and operation by highlighting the most critical factors affecting separation performance.