(532d) Experimental Characterisation of Wire Mesh Demisters

Four different wire mesh demisters have been studied in two different experimental apparatuses, one low pressure rig and one high pressure rig. A wire mesh demister is made by knitting wires to form a layer that can either be rolled into cylindrical elements or folded up in several layers. The materials used in the wire mesh demisters tested in this study were Polypropylene (PP) and stainless steel (SS). The porosity ranged from 95.4 – 99 % and the specific surface area ranged from 145 – 600 m2/m3. The design varied, including single layer, nested double layer, rolled and random configuration. The wire diameter varied from 0.27 – 0.30 mm. The height used was 100 mm. These data are summarized in table 1.

The low pressure apparatus include a 4 inch steel column, liquid circulation pump, compressed air supply, pneumatic nozzle for mist introduction and a wire mesh demister.The gas/liquid system used in this apparatus was air/water. Mesh A,B and C were tested. The high pressure apparatus include separate gas – and liquid loops, a 3 stage downstream separator and a 6 inch steel test scrubber with inlet device, wire mesh demister and cyclone pack. In this apparatus two gas/liquid systems were tested; Nitrogen/Exxsol D60 and a Synthetic natural gas mixture (NG). The pressures tested were 20, 50 and 92 bars.

The pressure drop over a wire mesh demister is low. For a mesh exposed to a dry gas, the pressure drop lies in the range 0-1000 Pa. Normally, the total pressure drop over a wire mesh demister is expressed as the sum of the dry mesh pressure drop and the wet mesh pressure drop (York and Poppele, 1963). Models presented in the litterature are in the same form as the Ergun equation used for calculating pressure drop in packed beds. Experimental data suggest that the pressure drop is not a function of the superficial gas velocity squared. The velocity dependency of the pressure drop varies with the mesh used. Based on the experimental data gathered, a preliminary model has been formulated. The model, although in need of refinement, is in good agreement with the experimental data.

Tests with wet gas were also performed. A plot of the pressure drop as a function of the gas load factor clearly shows the point where the wire mesh demister gets flooded. At this point the pressure drop will increase rapidly. Such plots have been presented in the literature (El-Dessouky, et.al., 2000), but not for high pressure experiments with real fluid systems and not for high gas load factors. The pressure drop curves for mesh D and B show that the pressure drop stabilises again when increasing the gas load further beyond the flooding point. Even a small reduction in pressure drop can be seen in some cases.

Bibliography

1. El-Dessouky, H. T., et.al., ”Performance of wire mesh mist eliminator”, Chem. Eng. Processes, 39, pp. 129-139 (2000) 2. York, O. H. and Poppele, E. W., “Wire mesh mist eliminators”, Chem. Eng. Progress, 59 (6), pp. 45-50 (June 1963)