In order to ensure that the initial conditions of the experiment are consistent and reach a steady-state as soon as possible, pre-wetting of the particles was carried out as follows according to the Levec pre-wetting method³⁴. The particles were completely immersed in water for 20 min, then placed in a stainless drum with a screen at the bottom and naturally filtered for 30 min. The pre-wetted particles were fed into the silo to ensure a continuous and stable flow of particles during the experiment. Since the first bed layer packed by free-falling of particles had lower bed porosity, it would take at least one bed volume time to ensure the densely packed bed had been entirely discharged¹⁴. And during the experiment, each time the experimental conditions were changed, it was necessary to wait for at least 10 min to ensure that the system had reached a steady state.

The pressure drop and exit liquid distribution were acquired simultaneously in the same bed. The pressure drop was measured by the differential pressure transducer (CYG1219 type, Baoji Research Center of Transducer, China). The measuring range is 0-30 kPa, with a relative accuracy of ±0.25% on the full scale. The sampling frequency was set as 400 Hz and the sampling time was set as 180 s in all cases. At the top and bottom of the column, two pressure pipes were inserted via drills (distance 750 mm). The front end of the pressure pipe was covered with a screen to prevent particles from entering the pressure measuring system. The gas-liquid separators were installed between the pressure pipes and the differential pressure transducer to maintain the pressure measurement completely in a single-phase gas condition according to Sai³⁵. The dynamic liquid holdup was measured by switching off the inlet gas and liquid flows and collecting the drained liquid within 30 min.

In order to quantify liquid distribution at the outlet, a maldistribution factor M_f was defined according to the following equation³⁶: