The counterintuitive observation that placing an obstacle in front of a bottleneck can improve the flow of discrete materials has been noted across various systems. This phenomenon, initially observed in pedestrian evacuation studies, has been reproduced by several models. However, robust experimental evidence remains contradictory, especially in pedestrian dynamics. While some studies report improved flow rates with strategically placed obstacles, others do not. In contrast, there is strong experimental evidence supporting the effectiveness of obstacles in other systems, such as animal movement through narrow passages and granular material discharge from silos. Granular systems, with their simpler interaction rules and ease of experimental repetition, provide an ideal platform to study this phenomenon in detail. Previous research on granular silos has shown that a well-placed circular obstacle can dramatically reduce clogging probability without affecting the flow rate. This reduction is attributed to the creation of an empty region below the obstacle, hindering the stabilization of arches that typically cause clogging. Further, increased granular temperature in the horizontal direction has also been linked to this effect. Existing explanations focus on system dynamics; this study investigates whether the impact of the obstacle diminishes as the grain motion is minimized, approaching a quasi-static limit. The researchers hypothesize that even in a quasi-static regime, the obstacle still plays a significant role in clogging prevention.

The literature review extensively covers existing research on the effect of obstacles on flow in various systems, highlighting the contradictory findings in pedestrian dynamics and the more consistent observations in granular systems. Studies on pedestrian evacuation, animal movement, and granular material flow are reviewed, emphasizing the need for a deeper understanding of the underlying mechanisms. Prior work on granular silo clogging and the impact of obstacles is summarized, including the observation of an empty region below the obstacle, increased horizontal granular temperature, and the relationship between upward particle movement and clogging reduction. The authors also mention previous research on granular flow dynamics, jamming transitions, and the relationship between avalanche size distributions and clogging probability.

The experimental setup consisted of a two-dimensional silo constructed from glass sheets and aluminum gauges, containing monodisperse stainless steel spheres. A conveyor belt below the silo controlled the discharge rate, allowing for manipulation of system dynamics from the free-discharge regime to near quasi-static conditions. A circular methacrylate obstacle was strategically placed above the silo's orifice. An automated measurement protocol, using a high-speed camera and a balance, was used to determine the clogging probability (Pc) from the distribution of avalanche sizes (the number of particles flowing out before clogging). The protocol involved the use of a vibrating system to break clogging arches. Video recordings were analyzed to extract kinematic information, including particle velocities in both horizontal (vx) and vertical (vz) directions, within a specified region above the orifice. The parameter vo, the mean particle velocity at the outlet, was used to characterize the system dynamics. To analyze the static properties, the contact fabric tensor was calculated for various configurations, providing information on the anisotropy of particle-particle contacts. The solid fraction (φ) was also computed to quantify the packing density in the region of interest.

The experiments revealed a dual effect of the obstacle on clog prevention. At high extraction rates (high vo), the obstacle altered the kinematic properties near the orifice, creating an empty region below it and increasing horizontal velocities. This hampered arch formation and stabilization. Conversely, in the quasi-static limit (low vo), the primary mechanism was a change in the contact fabric tensor, introducing significant anisotropy below the obstacle. Horizontal contacts became more prevalent than vertical ones, hindering arch stability. The clogging probability (Pc) decreased with increasing belt velocity (and thus vo), consistent with previous findings for silos without obstacles. However, the obstacle consistently reduced Pc by over 50%, even in the quasi-static limit. The data were fitted to an equation that considers both geometrical and dynamic parameters. Different fitting parameters for the scenarios with and without obstacles suggest a dual role for the obstacle in preventing clogging. Analysis of particle velocity distributions showed a widening of the horizontal velocity distribution and a reduction of high positive vertical velocities with the obstacle. The proportion of upward particle displacements also increased with the presence of the obstacle, especially at higher extraction rates. The solid fraction (φ) was lower with the obstacle, indicating a less dense packing. The contact fabric tensor analysis showed a clear anisotropy in the presence of the obstacle: horizontal components of the tensor were significantly higher than the vertical components in the region above the orifice, suggesting a horizontal squeezing effect that destabilizes arches.

The findings confirm the dual role of the obstacle in preventing silo clogging. The alteration of kinematic properties and the introduction of anisotropy in the contact fabric tensor are key mechanisms. This dual effect is captured by a single equation, previously developed for silos without obstacles, which now incorporates different fitting parameters reflecting the impact of the obstacle. The results highlight the importance of both dynamic and static factors in determining clogging probability. The observation that the obstacle significantly reduces clogging even in quasi-static conditions contradicts the notion that the obstacle's effectiveness is purely linked to dynamic factors. The altered contact fabric tensor and the resulting anisotropy explain the effectiveness of the obstacle in reducing clogging in low-velocity regimes. The study's findings provide valuable insights into the complex interplay of geometrical and dynamic factors that govern clogging in granular materials. The insights could contribute to improved design and operation of silo systems and could also help to resolve contradictory findings in related systems.

The study experimentally confirms that an obstacle above a silo's exit reduces clogging through two mechanisms: kinematic alterations at high extraction rates and fabric tensor anisotropy in the quasi-static limit. A single equation successfully encapsulates the observed behavior. This work provides valuable insight into clog prevention in granular materials and offers a more comprehensive understanding of the obstacle's role, extending beyond solely dynamic effects. Future research could explore how particle shape, deformability, and obstacle properties affect the two mechanisms.

The study is limited to two-dimensional systems with monodisperse, hard, spherical particles. The generalizability to three-dimensional systems and particles with more complex shapes and deformability requires further investigation. The specific size and placement of the obstacle used in this study might not be universally optimal for all silo configurations. Further research could explore the sensitivity of the findings to changes in obstacle parameters.