Mesoscopic numerical investigation of dynamic mechanical properties of ice with entrapped air bubbles based on a stochastic sparse distribution mechanism

Hail ice inevitably contains entrapped air bubbles in nature. The dynamic mechanical properties of ice are highly affected by the dimension, distribution and amount of the entrapped air bubbles. The bubbles in ice neither could be manufactured by man-made or obtained from nature with controllable amount and distribution, nor simulated by classic FEM in mesoscale due to its small size. A stochastic algorithm has been proposed as the sparse distribution mechanism of air bubbles in ice to organize the air bubbles with controllable amount and agglomeration in numerical model. A mesoscopic numerical modeling method of ice with entrapped air bubbles has been developed by the meshfree ALE multi-material method. The effect of amount and agglomeration of the air bubbles in ice on dynamic mechanical properties has been investigated by simulated SHPB tests. The results show that the dynamic strengths of the ice with air bubbles are increase with the increasing of strain rates, nonlinear decrease with the increasing of volume ratios of the air bubbles. Both the volume ratios and distribution of air bubbles in ice affect the dynamic strengths at a low strain rates level; but the distribution rarely affects the dynamic strengths at a high strain rate level.