• flooding;
  • wetted wall;
  • flow limitation;
  • high pressure;
  • two phase flow


Countercurrent wetted wall flow limitation was studied in a 20.6 mm i.d. cylindrical column, with water as the liquid. The gas was either air at up to 0.8 MPa, or argon at up to 1.3 MPa, thus covering a range of gas densities (1.6–21.3 kg/m3) of great practical interest for which very few data are available. The film Reynolds number varied in the range 515–3090. The velocity of gas for which signs of flooding instability first became apparent was typically 5 to 25% below the velocity at which liquid carryover started and this may be a result of the particular design adopted for fluid inlet to and outlet from the test section. Over the range of gas densities tested, the limiting gas velocity decreased with pressure from between 2.5 and 5.0 m/s to between 0.4 and 0.8 m/s, very approximately following the law: (gas density) × (limiting gas velocity)2 = constant. The data collected were used to test the correlation of Wallis, which gave remarkably good prediction of the limiting gas velocities, and a correlation of Alekseev and coworkers modified by McQuillan and Whalley in 1985, which gave poor predictions, particularly with regard to the effect of changes in liquid flow rate. The results of the present work are shown to be useful in interpreting recent data on the effect of gas density on the slug/churn flow transition in cocurrent vertical gas–liquid flow. © 2006 American Institute of Chemical Engineers AIChE J, 2006