Stochastic dynamics of granular hopper flows: a slow hidden mode controls the stability of clogs

Granular flows in small-outlet hoppers exhibit intermittent clogging behavior: while a temporary clog (pause) can last for an extended period before flow spontaneously restarts, there are also permanent clogs that last beyond experimental timescales. Here, we introduce a phenomenological model with multiplicative noise that provides a dynamical explanation for these extreme events: they arise due to coupling between the flow rate and a slow hidden mode that controls the stability of clogs. The theory fully resolves the statistics of pause and clog events, including the non-exponential clogging times and non-Gaussian flow rate distribution and explains the stretched-exponential growth of the average clogging time with outlet size. Our work provides a framework for extracting features of granular flow dynamics from experimental trajectories.