Segregation of an intruder in a heated granular dense gas

TL;DR

This paper revisits the segregation criterion for intruders in heated granular gases, improving theoretical accuracy with second Sonine approximation and validating results through DSMC simulations, especially for lighter intruders at high inelasticity.

Contribution

The study introduces a second Sonine approximation and accounts for temperature ratio dependence on volume fraction, enhancing the theoretical prediction of segregation behavior in granular gases.

Findings

Second Sonine approximation improves accuracy for lighter intruders.

Theoretical phase diagrams are sensitive to Sonine order without gravity.

Gravity reduces differences between Sonine approximations in segregation predictions.

Abstract

A recent segregation criterion [V. Garz\'o, Phys. Rev. E \textbf{78}, 020301(R) (2008)] based on the thermal diffusion factor $\Lambda$ of an intruder in a heated granular gas described by the inelastic Enskog equation is revisited. The sign of $\Lambda$ provides a criterion for the transition between the Brazil-nut effect (BNE) and the reverse Brazil-nut effect (RBNE). The present theory incorporates two extra ingredients not accounted for by the previous theoretical attempt. First, the theory is based upon the second Sonine approximation to the transport coefficients of the mass flux of intruder. Second, the dependence of the temperature ratio (intruder temperature over that of the host granular gas) on the solid volume fraction is taken into account in the first and second Sonine approximations. In order to check the accuracy of the Sonine approximation considered, the Enskog equation is also numerically solved by means of the direct simulation Monte Carlo (DSMC) method to get the kinetic diffusion coefficient $D_0$. The comparison between theory and simulation shows that the second Sonine approximation to $D_0$ yields an improvement over the first Sonine approximation when the intruder is lighter than the gas particles in the range of large inelasticity. With respect to the form of the phase diagrams for the BNE/RBNE transition, the kinetic theory results for the factor $\Lambda$ indicate that while the form of these diagrams depends sensitively on the order of the Sonine approximation considered when gravity is absent, no significant differences between both Sonine solutions appear in the opposite limit (gravity dominates the thermal gradient). In the former case (no gravity), the first Sonine approximation overestimates both the RBNE region and the influence of dissipation on thermal diffusion segregation.