Transition properties in dynamical and statistical features of drying crack patterns

In this study, we experimentally investigated the time dependence of the statistical properties of two-dimensional drying crack patterns to determine the functional form of fragment size distribution. Experiments using a thin layer of a magnesium carbonate hydroxide paste revealed a "dynamical scaling" property in the time series of the fragment size distribution, which has been predicted by theoretical and numerical studies. Further analysis results based on Bayesian inference show the transition of the functional form of the fragment size distribution from a log-normal distribution to a generalized gamma distribution. The combination of a statistical model of the fragmentation process and the dynamics of stress concentration of a drying thin layer of viscoelastic material explains the origin of the transition.