The Dual-Piston Pressure Swing Adsorption (DP-PSA) is a closed system where pistons moving in cylinders at each end of the adsorbent column induce the cycling of fluid flows and pressure variations in the fixed bed. This system is uniquely suited to test adsorbent materials under a wide range of conditions: vacuum to 25 bar; 20-200 °C; cycle times as fast as 1 s. The fact that the movement of the pistons can be controlled independently allows also to run the system under both linear and non-linear frequency response conditions.
The development of the DP-PSA includes experiment design, measurement of leak rates, dead volumes for packed and unpacked column, friction, pressure drop across the adsorbing column and temperature gradient along the column. The system is controlled through LabVIEW running on a real-time computer. This allows the automated running of a series of experiment without direct supervision. During these experiments the absolute pressure and the pressure drop along the column as well as the temperature in the gas and solid phase at two positions along the column is recorded. The experimental runs are analysed with the general adsorption simulator CySim.
The system can be run in two modes: single pellet experiments and full packed column runs. Single pellet experiments, where only the pellets connected to the thermocouples are inside the system, were carried out to study the heat transfer between the zeolite pellets and gases (He, N2, CO2, and mixtures of N2 and CO2). The other mode is that the adsorbing column is packed fully with zeolite 13X pellets. Experiments with pure gases and mixtures of N2 and CO2 were run with different configurations of the system: cycle time; phase angle; stroke length ratio and initial temperature. The measured signals were compared based on the amplitude and time shifts and numerical simulations were used to compare the predictions of a dynamic model of the system with the experimental results. The DP-PSA has been shown to generate a very large set of experimental results, with varying conditions which allow to determine physical parameters of dynamic models. This is achieved without consuming gases given that several experiments are automatically carried out in a closed system.
The authors would like to acknowledge the support from the Engineering and Physical Sciences Research Council under grants EP/G062129/1 and EP/F034520/1.