Phase distribution including a bubblelike region in supercritical fluid

Pseudoboiling in supercritical fluid (SF) has been paid great attention in recent years. Available works mainly focus on thermodynamics analysis. Fewer studies were reported on the spatial time phase distribution. Here, SF is investigated in a multiphase fluid framework using molecular dynamics (MD) simulations. A simulation box contains 10 976 argon atoms, with periodic boundary conditions applied on all the box surfaces. Pressure and temperature are well controlled. Based on MD simulation results, an onset pseudoboiling temperature ${T}^{\ensuremath{-}}$ and a termination pseudoboiling temperature ${T}^{+}$ are defined using the neighboring molecules method, the radial distribution function method, and the two-body excess entropy method. The two transition temperatures divide the whole phase diagram into three regimes of liquidlike, two-phase-like (TPL), and gaslike, and the MD determined ${T}^{\ensuremath{-}}$ and ${T}^{+}$ well matched the thermodynamics-determined values. In the TPL regime, nanovoids are observed to have two distinct characteristics: (1) Particles are sparsely distributed to have gas density inside the void, but are densely populated to have liquid density outside the void. (2) Voids have a curved interface. These characteristics are very similar to bubble characteristics in subcritical pressure. Hence, voids in the supercritical state are called ``bubblelike'' in this paper. Nonlinear dynamics demonstrates chaotic behavior in the TPL regime, similar to the two-phase regime in the subcritical domain. The above findings give strong evidence that SF in the TPL regime consists of a mixture of bubblelike voids and surrounding liquids. Our work highlights the multiphase feature of a SF, hence, the well-established multiphase theory in subcritical pressures can be introduced to handle the complex SF.