19990707: ``standardized'' configuration file, but different from 19990702

GOAL

Comparison of oxygen MSD's from two qualitatively (in the sense of the dlsub code) identical, but differently randomized configuration files, to see if ``standardization'' helps in obtaining any reproducibility in the simulation runs

Results

The configuration file was qualitatively identical to that in 19990702, in the sense that there were neither $Ce^{3+}$ nor oxygen vacancies in nearest neighbor position (see the .dlsub file).

Alas! Apparently, running simulations with qualitatively identical configuration files does not give reproducible results, as shown by the following plot of oxygen core MSD's for 19990702 and present experiment:

The oxygen diffusion coefficient from the present simulation comes out $2.34941885\times10^{-7}\;cm^2/s$ (MSD fitted in the range $[40-80]\;ps$, slope= $0.0140965131\;\mbox{\textit{\AA}}^2/ps$, intercept= $0.616267039\;\mbox{\textit{\AA}}^2$). This is about half the value obtained in 19990702 in the interval $[33-76]\;ps$.

So????