A Trap Model for Clogging and Unclogging in Granular Hopper Flows

Granular flows through narrow outlets may be interrupted by the formation of arches or vaults that clog the exit. These clogs may be destroyed by vibrations. A feature which remains elusive is the broad distribution $p(tau)$ of clog lifetimes $tau$ measured under constant vibrations. Here, we propose a simple model for arch-breaking, in which the vibrations are formally equivalent to thermal fluctuations in a Langevin equation; the rupture of an arch corresponds to the escape from an energy trap. We infer the distribution of trap depths from experiments and, using this distribution, we show that the model captures the empirically observed heavy tails in $p(tau)$. These heavy tails flatten at large $tau$, consistently with experimental observations under weak vibrations, but this flattening is found to be systematic, thus questioning the ability of gentle vibrations to restore a finite outflow forever. The trap model also replicates recent results on the effect of increasing gravity on the statistics of clog formation in a static silo. Therefore, the proposed framework points to a common physical underpinning to the processes of clogging and unclogging, despite their different statistics.