Radial mixing in horizontal aerated tubular reactor

Modeling of modified airlift loop reactor with a concentric double-draft tube

Liquid circulation and mixing time studies were carried out in a laboratory bubble column and modified bubble columns. Liquid circulation and mixing time are both strongly dependent on the geometrical configuration of the reactor. Internal recycle loop and external recycle loop shorten the mixing time compared to the bubble column.

Pore-level modeling of effective longitudinal thermal dispersion in non-isothermal flows through granular porous media

US: Venator – iron oxide pigments

The influence of static mixers on mixing time and circulation time in an external‐loop airlift bioreactor with gas‐induced and forced‐liquid circulation was investigated.The study was carried out with water and three viscous, non‐Newtonian starch solutions. The mixing time was determined for the overrall flow loop using a classical tracer response technique.It was found that the mixing time is highly dependent on the superficial gas velocity and the presence of static mixers. The shortest mixing time is achieved in a forced‐loop airlift reactor without static mixers, where the average mixing time value is only 1/3 of the time necessary for mixing in the airlift reactor with gas‐induced liquid circulation and static mixers.The pseudoplasticity of the liquid phase insignificantly influences the mixing time and the circulation time.

The effect of mode of sparging gas on the mixing parameters of an internal loop airlift bioreactor was investigated. Two bioreactors of identical volume of 14×103 cm3 and the optimum riser to downcomer cross sectional area ratio of 0.6 were studied. In one bioreactor a gas sparger was located in the draft tube and in the annulus in another. Liquid mixing characteristics, i.e., mixing time and circulation time, were employed to describe the performance of the bioreactors. The tracer injection method was used to determine the mixing parameters. A mathematical modeling based on the tanks-in-series model was employed to characterize the hydrodynamics behavior of the bioreactors. Matlab 7.1 software was used to solve the model equations in the Laplace domain and determine the model parameter, the number of stages. A comparison between the simulation results and experimental data showed that the applied model can accurately describe the behavior of the bioreactors. The results showed that when the gas sparger was located in the draft tube, the liquid mixing time, circulation time, and the number of stage were less than while the gas sparger was located in annulus. This is due to more wall effects, more energy losses and pressure drop in the case of gas injection in the annulus.

Gas Holdup, Liquid Circulation and Mixing Behaviour of Viscous Newtonian Media in a Split-Cylinder Airlift Bioreactor

Agro-food industrial processes produce a large amount of residues, most of which are organic. One of the possible solutions for the treatment of these residues is anaerobic digestion in bioreactors. A novel 18-L bioreactor for treating waste water was designed based on pneumatic agitation and semispherical baffles. Flow patterns were visualized using the particle tracer technique. Circulation times were measured with the particle tracer and the thermal technique, while mixing times were measured using the thermal technique. Newtonian fluid and two non-Newtonian fluids were used to simulate the operational conditions. The results showed that the change from Newtonian to non-Newtonian properties reduces mixed zones and increases circulation and mixing times. Circulation time was similar when evaluated with the thermal and the tracer particle methods. It was possible to predict dimensionless mixing time (θm) using an equivalent Froude number (Fr eq).

Agro-food industrial processes produce a large amount of residues, most of which are organic. One of the possible solutions for the treatment of these residues is anaerobic digestion in bioreactors. A novel 18-L bioreactor for treating waste water was designed based on pneumatic agitation and semispherical baffles. Flow patterns were visualized using the particle tracer technique. Circulation times were measured with the particle tracer and the thermal technique, while mixing times were measured using the thermal technique. Newtonian fluid and two non-Newtonian fluids were used to simulate the operational conditions. The results showed that the change from Newtonian to non-Newtonian properties reduces mixed zones and increases circulation and mixing times. Circulation time was similar when evaluated with the thermal and the tracer particle methods. It was possible to predict dimensionless mixing time (theta (m)) using an equivalent Froude number (Fr (eq)).

An efficient numerical method is described for studying the combined conductive and convective transport from a surface film on a planar boundary to a fluid in simple shear flow. Such problems arise most commonly in the use of hot film anemometers, electrochemical shear probes, or in simple models of chemical reactions. The method is illustrated by calculating the total flux (Nusselt number) and variation of the flux along the surface for isothermal circular disks at arbitrary values of the Peclet number (dimensionless shear rate) and is compared with asymptotic results valid at high and low Peclet numbers. The theoretical low Peclet number results of Phillips [Q. J. Mech. Appl. Math. (in press)] lie within 2% of the numerical results for Peclet numbers as high as P=2. At high Peclet numbers, a theoretical estimate for the neglected flux from the edge regions is used, together with the numerical simulations, to propose a correction to the standard one-term asymptotic expression. This approximate relationship remains within 7% of the numerical calculations for Peclet numbers as small as P=5. In addition, results for the total heat transfer as a function of Peclet number are presented for isolated elliptical disks at two orthogonal orientations with respect to the flow and an asymptotic expression for high Peclet numbers is presented for arbitrary disk orientations.

The hydrodynamics of biotechnological processes is complex. So far, few studies were made with bioreactors of the airlift type with an enlarged degassing zone.In this work, the influence of solids loading, solids specific gravity and draught tube dimensions on mixing and circulation times and critical air flow rate for an internal loop airlift bioreactor with an enlarged sedimentation/degassing zone is studied.The results indicate that the critical air flow rate as well as the mixing time increase with an increase in solids loading in the bioreactor. Circulation time presents a maximum for a solids load between 5 and 10% (v/v). It is also shown that small variations in solids specific gravity, for values close to that of the liquid, have a significant influence on the critical air flow rate and on the mixing time.An optimal (minimal) value for the circulation time and for the critical air flow rate was obtained for a riser to down comer diameter ratio of 0.46. The minimum mixing time was obtained for a riser to down comer height ratio of 0.80.

ADVANCED MIXING MODELS

This paper describes a boundary element scheme for solving steady-state convection–diffusion problems at high Peclet numbers. A special treatment of the singular integrals is included which uses discontinuous elements and a regularization procedure. Transformations are performed to avoid directly evaluating Bessel functions for Cauchy principal value and hypersingular integrals. Test examples are solved with values of Peclet number up to 107 to assess the numerical scheme. © 1998 John Wiley & Sons, Ltd.

The effect of the addition of ethanol (10 g/l) to the liquid-phase on gas and solids holdup, circulation and mixing times and interstitial liquid velocity in a three- phase airlift reactor was investigated. The airlift reactor (60 l) is of the concentric draught-tube type with an en- larged degassing zone. Ca-alginate beads were used as solid-phase and airflow rate (from 1.9 to 90.2 l/min) and solids loading (0-30% (v/v)) were manipulated. Riser and downcomer gas holdup were found to increase with the addition of ethanol, leading to a decrease on the relative solids holdup. The presence of ethanol seems to have no influence on the circulation time. On the other hand, mixing time variation depends on the solids loading and airflow rate. Riser and downcomer interstitial liquid ve- locity are lower for ethanol solution than for water. List of symbols Ad downcomer cross-section area (m) Ar riser cross-section area (m) d vertical distance between two points of the riser and of the downcomer (m) H1)H2 pressure difference between two points of the riser and of the downcomer (cmH2O) tc circulation time (s) tm mixing time (s) uld downcomer interstitial liquid velocity (m/s) uli interstitial liquid velocity in section i (m/s) ulr riser interstitial liquid velocity (m/s) vgr riser superficial gas velocity (m/s) Vs solids volume measured in each sample (l) VT sample total volume (l) Dt time required by the tracer to travel between the two acquisition points in the riser and in the downcomer (s) egd downcomer gas holdup

In this work we characterize liquid phase mixing in a 150 L bubble column with a draft tube (internal airlift configuration) for a water-air-sand system at high solid concentrations. Liquid mixing is assessed by measuring the evolution of sodium chloride concentration after a pulse of concentrated NaCl solution is injected. Tracer concentrations were measured by means of electrical conductivity probes. The experimental set up consists of a 0.29 m internal diameter, 3 m length Plexiglas column with a conical bottom (cone apex angle of 60°) and a concentric draft tube with 0.14 m internal diameter and 2 m length. The gas superficial velocity based on the cross section of the column vaired from 0.057 to 0.22 m/s. Sand particles of 280 μm in average size were used, with slurry concentrations ranging from 120 to 500 kg/m3. From the tracer outputs, circulation time (time between peaks of the response curve) and mixing time (time required to achieve a 95% homogeneous solution) were determined after the pulse had been injected. The experimental data were analyzed by means of a tank in series model with recirculation. The circulation and mixing times were found to increase with solids concentration, and to decrease as the riser gas velocity was increased. The circulation limes were larger in continuous operation than in semibatch mode. The results show that the system has equivalent dispersion coefficients that are one order of magnitude lower than those found in a conventional bubble column.