Two‐phase gas‐liquid flow has been investigated in a 1‐inch internal diameter vertical tube coil containing two risers and a downcomer all connected by “U” bends. Flow pattern data were obtained in the three vertical tubes, each 17.30 ft. long, for five different air‐liquid systems at about 25 psia over flow ranges of 0–700 lbm air/min‐ft2 and 140–25300 lbm liquid/min‐ft2. Liquid phase viscosities ranged from 1 to 12 cp. A flow pattern classification with six regimes including coring‐bubble, bubbly‐slug, falling film, falling bubbly‐film, froth and annular flow regimes was established for downflow. Flow patterns in the bends were also classified. Data from the present investigation were used to formulate an empirical flow pattern graphical correlation for both upflow and downflow which is based upon the coordinates (Rv)1/2 and FrTP/A, where Rv is the delivered gas‐to‐liquid volume ratio, FrTP is the mixture Froude number, and A = μs/(SLσs3)1/4 in which μs, SL, σs are specific viscosity, specific density and specific surface tension respectively of the liquid with reference to water. The correlation was satisfactorily tested with independent literature data for upflow systems, including air‐water, steam‐water at various pressures, nitrogen‐mercury and air‐heptane, and data from flowing gas‐oil wells. No independent literature data appear to be available for testing the correlation for downflow systems, but it is anticipated that the correlation will prove to be generally applicable. The coring phenomenon in downward bubble flow was examined by means of high speed motion photography and is explained by the development of a lift force on a bubble.

The feasibility of using fluidized bed technology for the production of an iron premix of ferrous fumarate encapsulated with soy stearine for iron fortification of table salt has been successfully demonstrated. Ferrous fumarate was selected as source of iron, in preference to ferrous sulphate heptahydrate, ferric sodium EDTA and reduced elemental iron, for its bioavailability, bland taste, stability and high iron content. Potassium iodate was used as iodine source. Fluid-bed processors of 7, 200 and 500 kg capacity were used to granulate ferrous fumarate with a solution containing, hydroxypropylmethylcellulose, sodium hexametaphosphate and titanium dioxide for binding, stabilization and colour masking, respectively. Encapsulation of granulated ferrous fumarate was achieved by top-spraying hot soy stearine containing more titanium dioxide at 98ºC on the fluidized bed. The premix contained 46.4 wt. % of ferrous fumarate. Double Fortified Salt (DFS), made by dry blending 1 part of iron premix and 150 parts of already iodized salt, contained 1000 ppm of iron and 50 ppm of iodine. Less than 10% of the iron in the premix, with a bulk density of 0.78 g/cc, dissolved in HCl at pH of 4.0 indicating good coating integrity. Particle size distribution of iron premix was consistently centred on 300 µm with more than 90% of the particles in the size range of 150 to 710 µm, similar to that of free flowing refined salt.

Two‐phase gas‐liquid flow has been investigated in a 1‐inch internal diameter vertical tube coil containing two risers and a downcomer all connected by “U” bends. Pressure drop, holdup and flow pattern data were successfully obtained simultaneously in the three vertical tubes, each 17.30 ft. long, for five different air‐liquid systems at about 25 psia and 50°F‐80°F over flow ranges of 0–700 lbm air/min‐ft2 and 140‐‐25300 lbm liquid/min‐ft2. Pressure drops and liquid holdups were plotted against gas volume flowrate with liquid flowrate as a parameter. From these plots it was found that for a combination of an increase in liquid viscosity and density, and a decrease in surface tension, the frictional pressure drop increased in down flow and decreased in upflow. Holdup, on the other hand, increased for both types of vertical flow with respect to the same combination of parameters. The Lockhart‐Martinelli scheme was satisfactory in correlating frictional pressure drop and holdup in all the flow regimes except the frothy‐slug regime in upflow. In downflow however, the Lockhart‐Martinelli scheme met with limited success because of a strong influence of liquid flowrate, physical properties and pipe orientation. Holdup in the falling film and falling bubbly film regimes in downflow were satisfactorily treated by the drift flux approach which emphasizes the relative motion of the two phases.

No abstract

Heat transfer in co‐current two‐phase upflow and downflow of air–water has been investigated in a 25.8 mm electrically heated vertical pipe at 172.3 kPa for water mass velocities of 54 to 172 kg/m2s and gas flow rates of 0 to 1.322 × 10−2 m3/s. It was found that although the injection of air in the liquid flow increased the two‐phase heat transfer coefficients significantly for both systems, upflow coefficients were generally higher than those for downflow for the same liquid flow rate. This could have important implications in the design of some chemical reactors and heat engineering processes. Changes in heat transfer rates were found to occur at the flow pattern transition boundaries. Two‐phase heat transfer coefficients were well correlated by an expression based on dimensional analysis for both upflow and downflow.