Among the few classes of computational approaches for examining rarefied gas dynamics, the most widely used technique for spatial scales relevant to suborbital spaceflight is the direct simulation Monte Carlo method. One area in which the direct simulation Monte Carlo method can be improved is the numerical modeling of the interactions between gas molecules and solid surfaces. Gas-surface interactions are not well understood for rarefied hypersonic conditions, although various models have been developed. The goal of this study is the assessment of gas-surface interaction models in use with the direct simulation Monte Carlo method. Assessment is made of the two most common gas-surface interaction models in use with direct simulation Monte Carlo: the Maxwell model and the Cercignani, Lampis, and Lord model. The assessment is performed by simulations of flat-plate wind-tunnel tests. Boundary-layer profiles are compared with existing wind-tunnel data. At about 90% accommodation, both models match the wind-tunnel profiles. Parametric studies demonstrate differences between the model predictions of scattering distributions, boundary-layer profiles, and surface-property distributions. The two models offer similar performance for computing flat-plate aerodynamics.= total enthalpy I 0 = modified Bessel function of the first kind and zeroth order K = scattering kernel Kn = global Knudsen number Kn DGL = density-gradient-length local Knudsen number m = mass of one molecule n = number density n = local normal unit vector of the solid surface P, P max = probability, maximum probability p = pressure q 1 = dynamic pressure, 0:5 1 V 2 1 R G = particular gas constant R u = universal gas constant St = Stanton number, (heat flux at the solid surface divided by q 1 V 1 ) T = temperature T = characteristic temperature of intermolecular potential T VHS = reference temperature for the variable-hard-sphere collision model t = local resultant tangent unit vector of the solid surface [t 1 t 2 =jt 1 t 2 j, t 1 and t 2 are orthonormal] V = magnitude of velocity jVj V = mass velocity, bulk velocity, or mean molecular velocity W p = reference particle weight, n=n simulation particles x, y = computational domain coordinates relative to the flat-plate leading edge Z rot;1 = maximum rotational collision number = energy accommodation coefficient VSS = deflection angle exponent of variable-soft-sphere collision model = Dirac delta function diameter = reference collision cross-sectional diameter = number of internal energy degrees of freedom vib = characteristic temperature of vibration = mean free path = absolute viscosity = absolute molecular velocity, V 0 0 = random molecular velocity = mass density = momentum accommodation coefficient = flux ! = viscosity index for the variable-hard-sphere collision model Subscripts E int = internal energy e = at the edge of the boundary layer i = incident M = Maxwell mp = most probable n = relative to surface normal vector Q = physical property r = reflected rot = rotational internal energy mode t = relative t...

No abstract

No abstract

The aerothermodynamics of spacecraft entering a planetary atmosphere are sensitive to the level of gas-surface accommodation governed by the gas-surface interaction. The modeling of this interaction plays an integral role in the solid surface boundary condition of the Direct Simulation Monte Carlo (DSMC) method. The Maxwell, and Cercignani, Lampis and Lord (CLL) gas-surface interaction models are examined. Existing windtunnel test results of rarefied hypersonic flow over flat surfaces enable the assessment of these gassurface interaction models for DSMC simulations for this kind of flow condition. These models gave the same boundary layer velocity profiles at 50 % to full gas-surface accommodation. Approximately, 90 % gas-surface accommodation yielded the overall best agreement between the simulations and windtunnel data, reported by Cecil and McDaniel [AIAA Paper 2005-4695]. Regarding molecular velocity distributions next to the surface, the gas-surface interaction models result in similar horizontal component distributions, but distinct vertical component distributions. Molecular velocity distributions also reveal translational nonequilibrium very near the surface due to surface reflected molecules, within 5 local mean-free-paths above the surface. Within a region of significant translational nonequilibrium, the distributions are better characterized by the most probable value, rather than the mean value. Regarding scattering distributions, the Maxwell model results in distributions with unrealistic peaks due to specular reflection; however, the CLL model results in petal-shaped distributions, similar to observations of molecular beam studies. Moreover, while the Maxwell scattering distributions experienced abrupt changes with increasing accommodation and position, the CLL distributions varied smoothly. Nevertheless, both yield good agreement with the PLIF windtunnel test boundary layer velocity profiles using a proper specification of gas-surface accommodation.

Future hypersonic vehicles are likely to use sharp leading edges on wings and engine inlets to reduce drag. The radius of curvature associated with such structures may be one centimeter or less. The very small size leads to high heating rates that can be accommodated using advanced materials. An additional aspect of the small size concerns the basic gas dynamics. At such small scales, the continuum approach invoked in formulating the Navier-Stokes equations may be invalid. In this study, the flow around various sharp leading edge shapes is computed for a high-altitude point on a typical re-entry trajectory. The particle-based direct simulation Monte Carlo method (DSMC) is employed. Effects are investigated of variation in free stream velocity and leading edge shape in terms of flow field and surface properties. The aerothermodynamic performance of the sharp leading edges is assessed using the shock standoff distance, the total drag, and the heat transfer rate. The data generated in the present study form the initial part of a more complete investigation involving the optimization of leading edges over a hypersonic re-entry trajectory.