Earlier today, I looked at the zTtrk dependence of the Run9 purity:

https://drupal.star.bnl.gov/STAR/blog/dmawxc/update-08202018-run-9-pp-few-checks-gamma-rich-purity

Here is the zTtrk dependence of the Run9 purity after averaging together adjacent bins of zTtrk (for 2 different pTtrk ranges):

Next, I calculated the purity vs. eTtrg using two different zTtrk criteria: with zTtrk = (0.2, 0.4) for all 3 bins of eTtrg, and with zTtrk = (0.2, 0.5) for eTtrg = (15, 20) GeV and zTtrk = (0.2, 0.3) for the other 2 bins:

And here are the two on top of each other:

The equivalent plots for pTtrk = (1.2, 20) GeV/c have been attached, but they're identical since the zTtrk cuts limit pTtrk to pTtrk > 1.8 GeV/c.