State -resolved Aerothermodynamic Models for Air Using Quasi Classical Trajectory data
Abstract
Hypersonic flight through the atmosphere generates extreme aerothermodynamic conditions. Specifically, a thin, high-temperature, shock layer is formed around the leading edges of a vehicle that can exceed 10,000 degrees Kelvin. Within the shock layer, directed kinetic energy is converted into random “thermal” energy of the air molecules as well as internal energy (rotation and vibration). Once the rotational and vibrational energy of gas molecules is sufficiently excited, the molecules may dissociate into reactive atomic species. These atomic species, such as atomic oxygen (O) and nitrogen (N), then impact the vehicle surface resulting in material ablation and possibly failure. Furthermore, dissociation reactions in the shock layer are precursors to electronic excitation and ionization. These processes lead to radiation emitted from the high-temperature gas surrounding the vehicle, which may have a particular signature depending on the vehicle parameters and flight conditions. The big challenge is that all of these processes occur at finite rates and, given the high flow velocities, these processes do not reach equilibrium, rather the flow remains in a state of strong nonequilibrium. Understanding and prediction of hypersonic nonequilibrium flows involves molecular level physics with strong coupling between various processes. A clear example is the fact that molecules in high vibrational energy states are favored to dissociate. It is difficult (often impossible) to infer such molecular level detail from macroscopic hypersonic wind tunnel experiments. Such experiments can measure surface pressure, surface heat flux, material ablation rates, and in some cases limited radiation spectra. These results represent the combined effect of a large number of coupled molecular processes and, therefore, such experiments cannot provide all of the data required to construct molecular-level models. Recently, due to continued advancements in computational chemistry and large-scale parallel computing, molecular level physics can be investigated using ab-initio computations. The most common analysis technique is called Quasi-Classical-Trajectory (QCT) analysis, wherein a large number of individual molecular collisions are simulated using a potential energy surface (PES) that dictates the forces on individual atoms comprising the molecules. Since the molecular systems of interest for air are small (typically 2-4 atoms), the PES can be fit to highly accurate electronic structure calculations. In the past 5 years a number of such results have been reported. The challenge now is to reduce this huge amount of data into accurate, yet tractable, models for predictive hypersonic flow simulations. The goal of the proposed research is to develop accurate, yet tractable, models for both the direct simulation Monte Carlo (DSMC) particle method as well as for continuum computational fluid dynamics (CFD) methods. In particular, the proposed research will determine precisely what model input is required for the new DSMC models, how this molecular level data can be efficiently generated (via QCT for example), and how the molecular DSMC models bridge to continuum CFD models.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 20, 2017
- Source ID
- FA94531710081
Entities
People
- Thomas Schwartzentruber
Organizations
- Air Force Research Laboratory
- United States Air Force
- University of Minnesota