As noted on the first page, the default e + D ionization rate in DEGAS 2 is smaller than that of EIRENE. Although we didn't go into detail about it, the normal DEGAS 2 charge exchange cross sections are smaller as well. To get an idea of how these data differences impact the simulation, we will now compare a run with the DEGAS 2 physics (the only differences relative to the first run are more flights, 80000 vs. 5000, and the improvement in the treatment of low energy reflections) against the DEGAS 2 run discussed at length on the previous page.
We have also taken the opportunity to compare runs with track length estimator scores of the charge exchange against collision estimator scores (which we have been forced to use heretofore because of limitations in the original Janev cross sections). These results differed, as expected, only in the ion energy and momentum sources. Those differences were less than the error bars, providing another check on the charge exchange coupling scores in DEGAS 2. The results shown below are from the track length estimator run.
With the higher ionization rates, we expect lower neutral densities with the "ehr2" rates. This is indeed what we observe. The ratios of peak densities in the DEGAS 2 physics run to those in the EIRENE physics run are: D - 0.83, D2 - 0.89, D2+ - 0.93. Plotting the D density with the same color palette as for the EIRENE physics run:
The sum of the particle source is constrained to be the same in the two runs (i.e., the incident current in each flux tube is the same). This leads to local particle sources which match much more closely than does the density:
As was also pointed out in the first page, the electron energy loss rates are very similar, so that the electron energy losses track closely the density differences. Hence, they are smaller in magnitude in the DEGAS 2 physics run than in the EIRENE physics case:
With the lower density and smaller charge exchange cross sections, we expect both the ion energy and momentum sources to decrease in magnitude:
Well, that's about it. From here we need to: