Speaker
Description
REVISED SHOCK LAYER RADIATION MODELING FOR AIR
Brett A. Cruden(1), Aaron M. Brandis
(1)AMA Inc, NASA Ames Research Center, Moffett Field, CA 94035, USA, E-mail:Brett.A.Cruden@nasa.gov
ABSTRACT
At the previous RHTG workshop, we reported a series of shock layer radiation measurements collected between 7-9 km/s and the corresponding disagreement to predicted radiance through the NEQAIR and DPLR codes. This presentation summarizes revisions to modeling practices in DPLR and NEQAIR as a result of this data and a revised comparison to both the shock tube data and radiometer data collected from the Orion Exploration Flight Test 1 (EFT-1). Significant improvements are demonstrated, with some remaining open questions.
1. INTRODUCTION
At RHTG-7, a presentation was given by the authors on a series of shock layer radiation measurements performed in air at velocities between 7-9 km/s, with focus on the non-equilibrium region of the shock. Comparisons against CFD and radiation predictions were performed using three different chemical kinetic model sets, two electron impact excitation databases and two different approximations of the 2-temperature model (i.e. Te=Tt or Te=Tv). The findings included the following: 1) N2 and NO radiation was underpredicted by all models; 2) the quality of N2+ radiation predictions were dependent upon the pressure of the test and the model employed; and 3) predictions of atomic radiation were mixed, with some lines predicted well under certain modeling assumptions and conditions, while other lines were predicted better at different conditions/assumptions [1].
This paper reviews these specific disagreements and the adjustments made to radiation models to improve the agreement. The adjustments include a re-examination of the quasi-steady state (QSS) models in NEQAIR and the reaction kinetics employed for air chemistry. The impact of these revisions upon the comparisons to shock-tube data will be discussed, and the new model will then be used to analyze radiometer data from the Orion Exploration Flight Test 1 (EFT1).
2. MODELING
The quasi-steady state calculations within NEQAIR were re-examined in the light of the new data. For the most part, NEQAIR’s QSS model has not been altered since the original work of Park [2].
The heavy-particle impact rates in NEQAIR have been revised to be consistent with literature values for the quenching of molecular states. The state-specific heavy-particle dissociation rates were also updated to be consistent with dissociation rates used in the CFD models. This involved scaling the state-specific dissociation rates such that the total dissociation rate is matched for a Boltzmann population of electronic states. Changes were also made to electron impact processes in the molecular QSS formulation. Besides updating excitation and dissociation rates, corrections were made to the integrations performed over both the electron energy distribution and the distribution of rotational states. Similar corrections were incorporated into the electron impact dissociation rates. These electron impact dissociation rates were then used to revise or introduce electron impact dissociation rates into the CFD models. Other processes that were added into the molecular QSS calculation include predissociation, and interactions of the excited states with other molecules and atoms through exchange and associative ionization reactions. In all cases, the inverse of every process is included in order to satisfy microscopic reversibility.
The atomic radiation models were also updated. Chiefly the update consisted of an examination of electron impact processes from different sources. Ultimately, a merged database containing the excitation rates of Huo [3] for allowed transitions, and Park [2] for spin forbidden transitions, was employed. Heavy-particle excitation is also introduced based on the work of Lemal, et al. [4] A final consideration is the interaction of excited atomic species with molecular ions, such as the dissociative recombination of N2+ into N. For much of the non-equilibrium regime, this is the dominant reaction determining the N atom concentration, thus should be included in the QSS balance. Assumptions regarding which states are involved in this process have significant impact on atomic radiation.
The reaction kinetics for the flowfield model is also re-examined. The major issue regards the rates of dissociation and exchange involving NO and their impact upon NO radiation. The NO exchange and dissociation processes are updated to be consistent with rates reviewed in the Journal of Physical and Chemical Reference Data [5, 6]. A single adjustable parameter was employed to match the measured data: the ratio of atomic to molecular dissociation rate for NO. A ratio of 5 was found to reproduce the data well over a range of pressures spanning almost two orders of magnitude. This is in contrast to the factor of 22 employed in Park [2] and Johnston [7] models.
3. DATA COMPARISONS
As alluded to in the previous section, the ability to predict molecular non-equilibrium radiation is now greatly improved. Figure 1 shows a comparison of radiance at 7.3 km/s and 0.7 Torr freestream pressure as a function of wavelength and position behind the shock. The non-equilibrium radiative overshoot is captured more accurately, as is the spectral profile. Further comparisons will be given in the final paper.
Fig. 2. Error in predictions of EFT1 flight data using the original and revised radiation models
This model has also been applied to the conditions of the EFT1 flight which returned to Earth from highly elliptical orbit at about 8.5 km/s. The data in its raw form is restricted, therefore the relative disagreement in prediction is presented in Fig. 2. The initial baseline model underpredicted the radiometer signal by about 80-90% (meaning the measurement was 1.8-1.9x the prediction). The revised radiation and kinetics model now are within 30-40% of the data at two time points that are within the radiative heating pulse. If CO2 and Ar are included in the flowfield model, this serves to increase the radiance further, so that the prediction is now within 10% of the measurement, which is considered an excellent agreement. The first data point shown is early in the flight where the absolute magnitude of radiation is not large. All models overpredict at this time point. This overprediction is primarily attributed to radiation from N2+.
4. REFERENCES
- Cruden, B. A., and Brandis, A. M., Measurement of Radiative Non-equilibrium for Air Shocks Between 7-9 km/s, AIAA AVIATION Forum, Vol. 2017-4535, 2017.
- Park, C. Nonequilibrium Hypersonic Aerothermodynamics. New York: John Wiley & Sons, 1990.
- Huo, W. M., Liu, Y., et al., Electron-Impact Excitation Cross Sections for Modeling Non-Equilibrium Gas, Vol. AIAA Paper 2015-1896, 2015.
- Lemal, A., Jacobs, C. M., et al., Air Collisional-Radiative Modeling with Heavy-Particle Impact Excitation Processes, Journal of Thermophysics and Heat Transfer, 1-14, 2015.
- Tsang, W., and Herron, J. T., Chemical Kinetic Data Base for Propellant Combustion I. Reactions Involving NO, NO2, HNO, HNO2, HCN and N2O, Journal of Physical and Chemical Reference Data, Vol. 20, 609, 1991.
- Baulch, D., Cobos, C., et al., Evaluated kinetic data for combustion modeling. Supplement I, Journal of Physical and Chemical Reference Data, Vol. 23, 847-848, 1994.
- Johnston, C. O. Study of Aerothermodynamic Modeling Issues Relevant to High-Speed Sample Return Vehicles. Vol. STO-AVT-218 2014.
Summary
Shock Tube data collected in air at the last workshop has been analyzed and used to revise the radiation model for air. The new model has been tested against EFT1 flight data.
