Co-current Gas Flow Research Articles

Trickle‐bed reactor flow simulation

AbstractA steady‐state, two‐dimensional model for cocurrent gas and liquid flow in a trickle‐bed reactor is described. The model includes gas‐liquid interaction and isothermal, incompressible flow without phase change or chemical reaction. The momentum equation proposed by Saez and Carbonell is used, along with their expressions for relative permeabilities of gas and liquid phases. Capillary pressure equations determine the difference in the pressures of flowing phases. The results of the model simulation agree with experimental data on flow distribution, liquid spreading, and phase segregation in a two‐dimensional trickle bed with air‐water flow in both low‐ and high‐interaction flow regimes.

Periodic and solitary waves on thin, horizontal, gas-sheared liquid films

Measurements of wavelengths and speeds of periodic waves which occur at the interface of thin, horizontal, liquid films sheared by a cocurrent gas flow are compared with predictions of linear theory. In general, linear predictions accurately match the data near the point of neutral stability but begin to deviate as the gas velocity increases. The behavior away from neutral stability is explained by the examination of the dynamical system associated with steady solutions to a nonlinear wave equation based on boundary-layer approximations. Periodic waves are seen to occur in the neighborhood of the Hopf curve which serves as a boundary between periodic waves and a flat film. Solitary waves (which correspond under some circumstances to disturbance or roll waves) are observed to lie in another region of parameter space at gas velocities well above neutral stability. These results suggest that a complete nonlinear analysis of the boundary-layer (or associated wave) equations would be useful for predicting amplitudes and speeds for periodic waves and the region of occurrence for solitary waves.

The effect of cocurrent and countercurrent gas shear on entrance region mass transfer in turbulent falling liquid films

Predictions of average film thickness and mass transfer coefficients in the entrance region of gas absorption with high cocurrent and countercurrent gas flow in a turbulent falling film are presented. The model used is a modified van Driest eddy diffusivity in the inner wall region and an interface damping eddy diffusivity in the outher region of the film modified to include the effect of interfacial shear. The calculations show that the entrance length for mass transfer decreases with increasing film Reynolds number and interfacial shear for cocurrent flow. Also the model predicts a decrease in mass transfer coefficient with an increase in Schmidt number in accordance with experimental data. On the other hand, for counterflow and at a fixedRe, an increase in upward gas shear increases the film thickness and eventually leads to rising film flow. The entrance length for mass transfer decreases with increasing Reynolds number but slightly increases with countercurrent interfacial shear. The calculations show that, in practice, the entrance region for mass transfer for gas absorption with shear-thinned film can be neglected for cocurrent shear but often cannot be neglected for countercurrent shear with a shear-thickened film.

EXPERIMENTAL DETERMINATION AND MODELLING OF LIQUID-SOLID MASS TRANSFER COEFFICIENTS IN A THREE-PHASE REACTOR

Apparent mass transfer coefficients for solid dissolution in a liquid with and without a chemical reaction were experimentally determined in a fixed bed three phases reactor with downward cocurrent gas and liquid flows. The chemical system selected was benzoic acid, sodium hydroxide in aqueous solution, and atmospheric air. Continuous gas, pulse and dispersed bubble regimes were studied and the results were correlated obtaining apparent mass transfer coefficient as a function of liquid and fluid volumetric flow. It was found that gas flow effect on mass transfer coefficient was small over continuous gas and dispersed bubble regimes, but appreciable over pulse regime. Additionally, it was found that the mathematical model that best described the mass transfer process under pulse regime, by using the increment factor due to the instantaneous chemical reaction, is the film theory

Wave characteristics of falling film flows in a co-current gas flow

Data is presented from measurements of the frequencies, lengths, amplitudes, and phase velocities of waves in a falling annular flow of water and air in a pipe.

Longitudinal Flow Characteristics of Vertically Falling Liquid Films without Co-Current Gas Flow (Effect of Liquid Viscosity)

Longitudinal Flow Characteristics of Vertically Falling Liquid Films without Co-Current Gas Flow (Effect of Liquid Viscosity)

The Kinetics of Limestone/Dolomite with H2S Under Rich Combustion Conditions

Abstract The kinetics of the reaction of limestone/dolomite with H2S have been examined under fuel rich combustion conditions. The experiments were done in a tubular laminar flow reactor with co-current flow of gas and solids. In the temperature region, 1065-1310°C, the results were consistent with the absorption reaction of H2S first order with respect to calcium and half order with respect to H2S. Reactivity of the stones fell off substantially between 1065-1310°C because of severe desurfacing occurring at these temperatures to the limestone and dolomite. A diffusion controlled reaction is believed to be the dominant mechanism.

Electro-oxidation of benzene in a fixed bed reactor

A fixed bed electrochemical reactor was used in the laboratory to oxidize benzene to quinone. The reactor consisted of a 3 mm thick bed of 1 mm lead shot, 0.5 m long by 0.05 m wide, sandwiched between a lead feeder plate and an asbestos diaphragm which was compressed against a stainless steel cathode plate. A dispersion of benzene in aqueous sulphuric acid was passed through the reactor and the rates of production of quinone, hydroquinone, carbon dioxide, oxygen and hydrogen, together with the cell voltage and pressure drop, were obtained for a range of operating conditions. Concentrations of quinone in the benzene product varied from 0.04 to 0.31 M and current efficiencies for quinone were in the range 22 to 55%. In a single pass of 1 M acid and benzene through the reactor at 25 ° C the quinone efficiency fell from 53% to 39% as the average superficial current density increased from 0.4 to 2.0 kA m -2. At an average superficial current density of 2.0 kA m -2 the quinone efficiency decreased with an increase in process temperature (25 to 50 ° C), but increased with increases in acid concentration (1 to 4 M), acid flow (0.5 to 1.0 cm 3 s -1 ), benzene flow (0.05 to 10 cm 3 s -1 ) and co-current nitrogen gas flow (0 to 32 cm 3 s -1 at STP). Recycling the 4 M sulphuric acid at 25 ° C raised the concentration of quinone in the product benzene but decreased the net current efficiency for quinone. Corresponding changes were observed in the cell voltage and in the current efficiencies for hydroquinone, carbon dioxide and oxygen. The results are discussed in terms of the process stoichiometry, electrode kinetics and mass transfer for three-phase flow in a fixed bed reactor.

The effect of axial dispersion on mass transfer between gases and liquids in trickle bed reactors

The cocurrent flow of gas and liquid through a packed bed is an extensively used operation in chemical industries where mass transfer and fluid dynamics affect the design equations. Reported overall mass transfer coefficients between gases and liquids are considered usually with the assumptions of plug flow for both phase [1]. That may be valid for gas phase, however liquid backmixing, where axially dispersed plug flow model is an adequate representation, are expected especially for short trickle bed reactors [2]. The effect of axial dispersion on mass transfer coefficients should be minimized and the true overall mass transfer coefficients should be used for design purposes. According to the model transient mass conservation equations for gas and liquid phases are ▪ where C is concentration, u is superficial velocity, K La is overall gas liquid mass transfer coefficient, D L is axial dispersion coefficient, ϵ is hold up volume fraction, H is reciprocal of Henry's law constant. The subscripts G and L stands for gas and liquid respectively. Initial and boundary conditions can be stated as; at t = 0, C G = C L = O for all z; at z = 0, C G = Mδ(t), and −D L∂C L/∂z + u LC L = 0; at z = Z, ∂C L/∂z = 0 at any time t. Simultaneous solution of equations 1 and 2 with boundary conditions resulted the following expression for m * OL that is the fraction of species transferred to liquid phase in infinite time at column height z. ▪ where ▪ When axial dispersion is neglected, A = 1 and B is a function of K La only [3]. The model may consider adsorption b including a similar mass conservation equation written for the species in the pores of catalyst particles and mass transfer term from liquid to solid eqn. ( 2). Experimental studies are done with nitrogen flowing concurrently with water at 20 °C and 1 atmosphere in a laboratory size trickle bed reactor packed with active carbon pellets. Impulse of sulfur dioxide is given to gas phase.

The residence time distribution of solid and liquid in multistage bubble columns in the cocurrent flow of gas, liquid and suspended solid

Multiphase flow in a multistage bubble column reaction is investigated experimentally. Within the range of parameters studied, it is possible to neglect the influence of plate geometry and liquid throughput on the liquid hold-up. Additionally, the distance between plates and solid concentration had negligible effect on the liquid hold-up.

Pressure drop, gas hold‐up, and oxygen transfer in tower systems

AbstractPressure drop, gas hold‐up, and oxygen transfer were investigated in a sieve tray column, a column with Koch motionless mixers, and a bubble column. The oxygen transfer experiments were conducted using cocurrent flow of gas and liquid under steady‐state conditions with oxygen transfer from the gas to the liquid phase. The oxygen transfer rates and efficiencies of the sieve tray column and the column with Koch mixers were found to be superior to those of the conventional bubble column. Gas hold‐up was also greater when sieve trays or Koch mixers were inserted in the tower. The pressure drop was found to be primarily due to the liquid head in all three columns.

Experimental study of gas absorption into turbulent falling films of water and ethylene glycol-water mixtures

Experiments are reported on the effects of liquid viscosity and interfacial shear on the liquid side mass-transfer coefficient for turbulent falling films. CO 2 gas was absorbed into a liquid film formed on the inside of a 2·05 cm ID, 1·98 m long vertical tube. The Reynolds number power dependence of the mass-transfer coefficient was found to depend markedly on viscosity. Transition from wavy laminar to turbulent flow occurred at Reynolds numbers from 1000 to 1200. With co-current gas flow a marked and linear increase in mass-transfer coefficient was found, while for counter-current gas flow the coefficient decreased.

Cocurrent gas and particle flow during pneumatic discharge from a bunker through an orifice

The flow of gas escaping from a bunker during pneumatic discharge of solid particles has been measured. A laboratory bunker (0.20 m diam.) was used from which sand (average size 205 μm) or glass ballotini (average size 100 and 200 μm) were discharged while simultaneously air was introduced at the bottom. It appeared that the flow of gas is primarily a function of the solids flow rate; orifice diameter and particle properties occurs as parameters in this relation. When combining this result with that of a previous investigation a unique relation is obtained between air flow rate, solids flow rate and pressure drop across the orifice.

Mass Transfer in Packed Beds with Two-Phase Flow

Mass transfer coefficients were measured at 25’C and 1 atm for cocurrent liquid (water) and gas flow in beds of small particles (0.054-0.29 cm) of napthalene and CuO-ZnO. The coefficients for transfer between particle and liquid are not greatly different for the three arrangements: upflow, downflow (trickle bed), or liquid full. At high gas rates and low liquid rates upflow gives somewhat higher transport rates, while trickle beds are favored over liquid-full operation at high liquid rates. Desorption of oxygen was measured to obtain liquid-to-gas coefficients kLa when mass transfer in the liquid phase controlled interphase transport. For downflow such coefficients were independent of gas flow rate, but in upflow kLa increased with gas rate. Upflow gave higher coefficients at high gas and liquid rates. Desorption of napthalene from water to air was employed to evaluate the gas-side coefficient in liquid-to-gas transfer for trickle-bed operation. The results showed a high sensitivity to gas flow rate and a modest sensitivity to liquid rate.

Effect of interfacial shear on gas absorption into a turbulent falling film with co-current gas flow

Effect of interfacial shear on gas absorption into a turbulent falling film with co-current gas flow

Effect of interfacial shear on gas absorption into a turbulent falling film with co-current gas flow

Experimental measurements of the liquid side mass transfer coefficient are reported for CO 2 absorption into a turbulent falling water film at 25°C with co-current gas flow. The liquid film Reynolds number was in the range 1830 to 8030. As the gas velocity was increased from 0 to 14 m/s there was a more than four fold increase in the mass transfer coefficient. The data is well correlated by ( K L K o L ) = 1+2.3( τ i τ o w ) . Data for the skin friction coefficient are also reported, showing a large increase above the dry wall value for high liquid and gas phase Reynolds numbers.

Axial heat transfer in packed beds. Gas flow through beds between 20 and 650°C

The experimental measurements and theoretical models described in part I are extended to the case of gas flow through the bed. Temperature profile measurements are given for cocurrent gas and axial heat flows through the bed over a temperature range 20–650°C and a particle Reynolds number up to 5. These results are analysed in terms of both a continuous (diffusion) model and a new discrete (mixing cell) model. The latter provides more quantitative information on the heat transfer mechanisms involved since, unlike the former, separate account is taken of solid conduction-radiation, solid contact resistance, as well as solid-gas convection and bulk gas flow convection. The suitability of the discrete model is verified by comparison with both the present experimental cocurrent results at low and high temperatures and with previously published countercurrent low temperature measurements using a range of packing materials.

Pressure gradient and liquid film thickness in co-current upwards flow of gas/liquid mixtures: Application to film-cooler design

Expressions have been developed for pressure gradient and liquid film thickness in upwards annular co-current flow of gas and liquid in the absence of disturbance waves and droplet entrainment. Experimental measurements of pressure gradient have been made with various phase flow rates, tube diameters and physical properties of the liquid. The results have been correlated in terms of the equivalent sand-roughness of the liquid film.