The Chemical Engineering Journal最新文献

The volumetric mass transfer coefficient k_La for a mechanically agitated gas—liquid contactor was measured by employing the dynamic method developed recently by Panja and Phaneswara Rao (Trans. Inst. Chem. Eng., 69 (1991) 302), which consists of bubbling pure CO₂ into an aqueous solution and following the concentration of the bicarbonate ion continuously with the help of an electrical conductivity probe. A mathematical model was used which enabled the value of k_La to be determined by a non-linear least-squares method. In the present work the above method has been extended to study the effect of process variables such as temperature, solids concentration, viscosity and interfacial tension on k_La in a tank of diameter 16.4 cm with the standard tank configuration. The value of k_La was found to increase (1) with increasing temperature of the liquid, (2) with increasing amount of dissolved isopropanol and amyl alcohol, which cause a reduction in surface tension, (3) with increasing solids concentration in the range 0–6 wt.% and (4) with increasing ion concentration. However, the value of k_La was found to decrease (1) with increasing viscosity and (2) with increasing solids concentration above 6 wt.%. A new correlation for k_L was obtained from the values of K_L calculated from the present k_La data by knowing the value of interfacial area from the equation of Calderbank (Chem. Eng., 45 (1967) CE-209). The effect of the presence of an alcoholic compound (amyl alcohol) under miscible and immiscible conditions on gas holdup was also investigated experimentally. The gas holdup was found to first decrease and then increase in the miscible range of amyl alcohol, again decrease and finally increase at a higher volume fraction of amyl alcohol, when an immiscible phase of amyl alcohol also formed.

The optimization problem for fed-batch fermentation is formulated in a framework which allows comparison of optimal operating policies with four different control variables. These variables include the original control variable, namely the substrate feed rate, and three transformed control variables, i.e. the substrate concentration in the fermenter, the substrate concentration in the feed and the mass flow rate of substrate. The optimality conditions for the three transformed control variables are identical but are generally different from the optimality condition for the original control variable. The optimality condition for the transformed control variables does not take into account (i) the dependence of the specific rates on the concentrations of cell mass and product and (ii) the substrate consumption dynamics.

The control variable transformation leads to suboptimal operating policies for fed-batch fermentation as compared with the operating policies resulting from the original control variable. Examples are provided to demonstrate the effect of the choice of the control variable on the optimization of fed-batch fermentation.

It has been shown that over the range of concentration used, the adsorption of Cephalosporin C can be characterized by a single linear adsorption constant. Additionally, the kinetics has been modelled successfully using both the linear adsorption isotherm and a single effective internal diffusion coefficient. The modelling work showed that the observed insensitivity of the goodness of fit to the effective Sherwood number was due to the fact that the method of agitation employed in the experiments was effective. The results hold in spite of the twofold variation in adsorption concentration in the liquid phase and the threefold variation in the square of the particle radius.

We report experimental observations on the use of a batch pulsatile flow reactor. The irreversible reaction between sodium hydroxide and ethyl acetate, NaOH+CH₃COOC₂H₅→CH₃COONa+C₂H₅OH, is followed as a model example in both a pulsatile flow and a stirred tank reactor. Reaction species were premixed with carboxymethylcellulose sodium salt in order to increase the viscosity of the solution and to make the fluid mixing within the vessel a critical parameter. The results show that the reaction kinetics are similar for the two devices and air entrainment problems are limited to the stirred tank reactor. Power density considerations show that the pulsatile flow reactor is more energy efficient than the stirred tank reactor.

Experiments were conducted to investigate the solid—liquid mass transfer behaviour in a non-newtonian liquid fluidized bed. The column diameter and height of the fluidized bed are 8 cm and 100 cm respectively. Carboxymethyl cellulose (CMC) solution was used as the non-newtonian fluid in this study. Dissolution of benzoic acid pellets into CMC solution was employed to obtain the solid—liquid mass transfer coefficient in the bed. The benzoic acid concentration in CMC solution was measured by titration. The in-bed benzoic acid concentration was examined to justify the use of an axial dispersion model in the evaluation of the mass transfer coefficient in this study. It was found that the mass transfer coefficient was essentially independent of the liquid velocity and particle size. However, it decreased as the CMC solution concentration was increased.

Mass transfer between an electrolyte and the inner core of an annular cell fitted with a tangential inlet is experimentally investigated using an electrochemical method. Attention is focused on the influence of the electrode length and position with respect to the tangential inlet to take into account both the development of the hydrodynamic and concentration boundary layers. According to the value of the ratio of the gap width e, to the inlet diameter Φ_e, two kinds of swirling motion are studied: pure swirl flow for e = Φ_e and contraction swirl flow when Φ_e >e. Depending on the geometric factors and the Reynolds number, enhancement in mass transfer of up to 550% is achieved in comparison with that obtained for fully developed axial flow. Two general correlations of the experimental data, taking into account the aforementioned parameters, are proposed for laminar and turbulent swirling flows.

In view of the importance of catalytic combustion in the field of air pollution control, a study on the total oxidation of lean mixtures of hexane (approximately 0.5 mol%) in air was undertaken. Silica-supported cobalt oxide catalysts prepared by the impregnation of silica hydrogel by ammoniacal cobalt oxalate precursor effectively catalysed the total oxidation at 553 K. The dependence of the catalyst activity on the temperature and environment during decomposition was investigated. A low decomposition temperature and an inert environment yielded better activity of the catalysts. This has been explained with the help of the decomposition mechanism of cobalt oxalate and has been supported by estimating the particle size distribution by X-ray diffraction analysis.