The distribution of total energy per particle for the ten initial configurations at in the soft disk mixture with their original momenta (dashed line) and for five initial configurations, each with 100 random assignments of momenta from the Maxwell-Boltzmann distribution (solid line).
The mean, range, standard deviation (stdev), and the ratio stdev/mean for the propensity distributions calculated for ten configurations each at , 0.46, 0.5, 0.6, 0.8, and 1. At each temperature, the configurations were separated by from each other, and the propensities were averaged over 100 runs of . Note the different -axis scales.
The propensity distributions over small and large particles for selected configurations at (a) and (b) . The propensities were averaged over 100 runs of .
(Color) The spatial distribution of propensities at for four configurations separated by . The propensities were averaged over 100 runs, and the scale is the same as in Fig. 7(b).
Cluster measures of spatial heterogeneity for particles with propensities in the top 10%. Data points are shown individually for ten configurations each at , 0.5, 0.6, 0.8, and 1. Statistics obtained using random values are shown for comparison. The dotted line represents the maximum variance possible for a given number of clusters (see text for more details).
Convergence of the relative uncertainty in the propensity [see Eq. (2)] as a function of the total number of runs for configurations at (a) and (b) . The error bars indicate the range of values at a given number of runs, and the curve joins the mean values of , where the average is taken over particles.
(Color) Convergence of the spatial distribution of propensity as a function of the number of runs for a configuration at . The propensities were calculated using (a) 50 runs and (b) 1000 runs. Note that there is little difference in the coarse grained spatial variation between the two plots.
The propensities and their 95% confidence intervals for particles along a line parallel to the axis in a configuration at . The propensities and their uncertanties were calculated using 1000 runs. Note that the error bars are just under twice the standard error.
The distribution of particle displacements over an isoconfigurational ensemble of 100 runs for a single particle at . Note the highly asymmetric and non-Gaussian shape of the distribution.
The distribution of single particle non-Gaussian parameters [see Eq. (3)] for configurations at and 1.0, calculated using ensembles of 1000 runs.
(Color) (a) The spatial distribution of the single particle non-Gaussian parameter for a configuration at . (b) The propensity map for the same configuration used in (a). Quantities were calculated using ensembles of 100 runs.
The particle directionality as a function of propensity for ten configurations each at (a) and (b) . Quantities were calculated using 100 runs.
Displacement vectors for selected particles at with high directionality and either (a) low or (b) high propensity. The vectors are from isoconfigurational ensembles of 100 runs. The axes indicate the and coordinates. Note the change in length scale between (a) and (b).
The correlation between the motion of a particle and all other particles as a function of the distance between and . The moving average has been indicated by a thick line, and the Pearson’s correlation coefficient between displacement magnitudes was calculated using data from an ensemble of 100 runs.
Article metrics loading...
Full text loading...