Simulation analysis of radiometer effect and outgassing based on molecular particle source perturbation

The Laser Interferometer Space Antenna (LISA) mission is designed to detect space gravitational wave sources in the millihertz band. A critical factor in the success of this mission is the residual acceleration noise metric of the internal test mass (TM) within the ultra-precise inertial sensors. Existing studies indicate that the coupling effects of residual gas and temperature gradient fluctuations significantly influence this metric, primarily manifesting as the radiometer effect and the outgassing effect. However, current theoretical research methods are inadequate for accurately decoupling and predicting the contributions of these two effects. To this end, this paper conducts an in-depth decoupling analysis of the impacts of the radiometer effect and outgassing effect using a simulation method based on molecular particle source perturbation. By constructing a finite element simulation model that couples residual gas and temperature gradient fluctuations, we simulate molecular thermal motion based on fundamental theories such as Maxwell's distribution function, the free path distribution law, and Knudsen's adsorption layer hypothesis, in order to study the two manifestations of the radiometer effect and the outgassing. By analyzing and comparing the simulation results, theoretical results, and ground torsion results, we find that the simulation results align effectively with the measured ground torsion results. This alignment enhances our understanding of how the radiometer effect and outgassing impact the internal pressure changes in the sensitive probe, which cannot be sufficiently decoupled by theoretical calculations alone. This conclusion demonstrates that the simulation method can effectively analyze the coupling effects of temperature gradient fluctuation and residual gas in inertial sensors, providing an important theoretical basis and practical significance for developing physical models of inertial sensors aimed at predicting and optimizing overall performance.