1. Park Frost Temperature, K A-R scheme A-D scheme A-R scheme A-D scheme 10000 25.7 8.4 116.1 9.0 20000 16.8 8.2 74.5 9.3 30000 11.3 8.2 38.0 9.2 40000 12.5 8.6 23.5 9.3 50000 12.5 8.4 17.5 9.3 the two schemes for the electronic transitions. For this test case, n=1020m-3, mole fractions are 10 % of N and 90 % of e-, the number of simulated particles is approximately 1.6 million, Δt = 10-9s, and the number oftimestepsis30,000. TheprocessorusedforthistestisIntelXeonE5530 andtheCPUtimeforthecollision subroutine were compared. It is shown in the table that at low temperatures, the efficiency of two schemes are approximately the same. However, at high temperatures, the A-D scheme is more efficient than the A-R scheme. At maximum, the computational time with the A-D scheme was decreased by approximately 8 % for the Park model case.
2. Park Frost Temperature, K A-R scheme A-D scheme A-R scheme A-D scheme 10000 151 151 149 149 20000 215 206 209 211 30000 267 248 315 309 40000 299 290 406 394 50000 329 304 455 453
3. According to the experimental results, an external energy source is necessary to predict the O2-Ar ICP flow field. Therefore, a heat source for electrons (or charged species) was artificially appended to keep the electron temperature uniformly high, and DSMC computations were carried out. The flow field obtained by the DSMC is presented in Figs. 17 and 18. Comparisons of number density and temperature distributions alongthecenterlineareshowninFig. 17. ItcanbeseenthattheTe,Tel,andTvibinDSMCareapproximately the same, and comparable to the experimental electronic excitation temperature results. Meanwhile, the translational temperature in DSMC is higher than the experimental result. In DSMC, it is increased up to 3,500 K while it is lower than 600 K in the experiment. As mentioned earlier, this result indicates that the model of the e-T energy transfer should be improved. The DOI is fairly uniform along the centerline and is approximately 0.5-0.6 %, which is not surprisingly different from the case without any external energy source. However, the O+becomes the dominant charged species instead of O+2 because the EI ionization reaction rates are increased.
4. Using the newly developed DSMC code, the O2-Ar ICP flows have been investigated. In DSMC, 5 temperatures (translational, rotational, vibrational, electronic excitation and electron temperatures) were computed and compared with the experimental results. First, it was found that the electronically excited levels of Ar and O have significant impact on the chemistry in the ICP flow field. The production rate of O+2, O+and Ar+is greatly increased. The DOI is approximately 0.5-0.6 % with the assumption of the neutral freestream condition. Second, the e-Eelenergy transfer effect was evaluated using the Q-K model, and the Teldecrease was observed due to the low electron temperature. Since electronically excited Ar and O particles are consumed by ionization reactions, the electronic excitation populations are deviated from the equilibrium condition. Third, the DSMC computations were carried out using an external energy source in order to improve the comparison between the DSMC and the experiment. Our results suggest that the inclusion of the external energy source be important for the prediction of the ICP flow field.