Flooding instability of high-density gas slugs rising in vertical tubes filled with water

Abstract

When a long gas slug rises in a vertical tube filled with liquid, the liquid ahead of the slug is displaced and runs down along the tube wall, thus originating a counter-current annular flow. The present work was undertaken to experimentally demonstrate that when the density of gas slugs, rising in stagnant liquid, is raised above certain limits, the wetted-wall flow around the slugs is subject to flooding instability; curiously, the available correlations on flooding would not predict this occurrence at the low gas velocities involved. In our experiments, the movement of slugs of argon rising in water, inside tubes of internal diameters 21.0 and 32.8 mm, was followed by means of fast response differential pressure transducers. Experiments were performed at absolute pressures of up to 6.1 MPa and from the signals of the pressure transducers it was inferred that gas slugs, with a total length of around 2 m, became unstable above 2.1 MPa, in the smaller tube, and above around 1.0 MPa in the larger one. Simple physical arguments, about the origin of flooding instability in wetted-wall flows, are used to derive approximate scaling laws for the phenomenon. The predictions from these scaling laws are shown to be in very good agreement with the experimental findings in the present work and with those of Nakazatomi et al. (1992, JSME International Journal, Series II, 35, 388-394), in their study of the rise of long slugs of nitrogen in water, at pressures of up to 20 MPa. Finally, our work suggests that in wetted-wall tubes operated with high-density gas, the velocity at which the flooding instability sets in may be significantly lower than that above which there is liquid carry over, at the top of the tube, and this calls for increased caution in the definition of flooding. (C) 2000 Elsevier Science Ltd. All rights reserved. When a long gas slug rises in a vertical tube filled with liquid, the liquid ahead of the slug is displaced and runs down along the tube wall, thus originating a counter-current annular flow. The present work was undertaken to experimentally demonstrate that when the density of gas slugs, rising in stagnant liquid, is raised above certain limits, the wetted-wall flow around the slugs is subject to flooding instability; curiously, the available correlations on flooding would not predict this occurrence at the low gas velocities involved. In our experiments, the movement of slugs of argon rising in water, inside tubes of internal diameters 21.0 and 32.8 mm, was followed by means of fast response differential pressure transducers. Experiments were performed at absolute pressures of up to 6.1 M Pa and from the signals of the pressure transducers it was inferred that gas slugs, with a total length of around 2 m, became unstable above 2.1 M Pa, in the smaller tube, and above around 1.0 M Pa in the larger one. Simple physical arguments, about the origin of flooding instability in wetted-wall flows, are used to derive approximate scaling laws for the phenomenon. The predictions from these scaling laws are shown to be in very good agreement with the experimental findings in the present work and with those of Nakazatomi et al. (1992, JSME International Journal, Series II, 35, 388-394), in their study of the rise of long slugs of nitrogen in water, at pressures of up to 20 M Pa. Finally, our work suggests that in wetted-wall tubes operated with high-density gas, the velocity at which the flooding instability sets in may be significantly lower than that above which there is liquid carry over, at the top of the tube, and this calls for increased caution in the definition of flooding.