DOE OSTI2016
Accurate simulations of fluid and plasma flows require accurate thermodynamic properties of the fluids or plasmas. This thermodynamic information is represented by the equations of state of the materials. For pure materials, the equations of state may be represented by analytical models for idealized circumstances, or by tabular means, such as the Sesame tables. However, when a computational cell has a mixture of two or more fluids, the equations of state are not well understood, particularly under the circumstances of high energy densities. This is a particularly difficult issue for Eulerian codes, wherein mixed cells arise simply due to the advection process. LANL Eulerian codes typically assume an “Amagat’s Law” (or Law of Partial Volumes) for the mixture in which the pressures and temperatures of fluids are at an equilibrium that is consistent with the fluids being segregated within the cell. However, for purposes of computing other EOS properties, e.g., bulk modulus, or sound speed, the fluids are considered to be fully “mixed”. LANL has also been investigating implementing instead “Dalton’s Law” in which the total pressure is considered to be the sum of the partial pressures within the cell. For ideal gases, these two laws give the same result. Other possibilities are nonpressure- temperature-equilibrated approaches in which the two fluids are not assumed to “mix” at all, and the EOS properties of the cell are computed from, say, volume-weighted averages of the individual fluid properties. The assumption of the EOS properties within a mixed cell can have a pronounced effect on the behavior of the cell, resulting in, for example, different shock speeds, pressures, temperatures and densities within the cell. There is no apparent consensus as to which approach is best under HED conditions, though we note that under typical atmospheric and near atmospheric conditions the differences may be slight.