The Chemical Engineering Journal最新文献

Electrodialysis of aqueous solutions of acetic and nitric acids encountered in the industrial practice of deacidification of raw glyoxal solutions has been carried out to investigate the permselectivity of the system. The dependence of the permselectivity coefficient τ on experimental parameters such as flow velocity (V), current density (I), concentration of anions (C_i), etc. has been evaluated. A correlation between τ and the critical parameters has been obtained. Well-known theoretical equations have been used to calculate τ at limiting current density I_lim and as I approaches zero. Calculated results are compared with observed values. Limitations of theoretical procedures when applied to a system containing a weak acid are explained.

One of the most essential problems in dealing with solid-liquid suspensions in stirred vessels is the determination of a reliable scale-up rule from small stirred tanks to large vessels on an industrial production scale. According to a new approach based on physical modelling of the complex fluid dynamics, the necessary power input in stirred suspensions can be calculated as a sum of the circulation power and the sinking power of a particle swarm. The following results, which are compared with a great variety of experimental data in the literature, reveal that there is no simple and constant scale-up rule applicable to describe the power input for a large range of suspension properties, tank size, geometrical conditions or comparable suspension criteria.

A theoretical model is derived to decide whether biphasic systems or single-phase aqueous systems are chosen as the media for microbial transformations. The yield ratio and product ratio at chemical equilibrium, which depend on the partition coefficients, equilibrium constants, activity coefficients and initial concentrations, are proposed as appropriate indices by the analysis of four common types of biochemical reaction under two extreme initial conditions.

A model for pressure drop is proposed for gas—liquid flow through packed beds on the basis of the observed absence of radial pressure gradients and taking into consideration the structure of the bed and the physical properties of the fluids. The model divides the total voitage of the bed into internal and external voidage and appropriately distributes the total liquid holdup into internal and external holdup.

Over 2500 experimental data, from the present study as well as those reported in literature, are correlated by the model with an r.m.s. deviation of less than ±9%. The significant parameters affecting the two-phase pressure drop are found to be the bed porosity, the Reynolds number, and the product of the Eötvos and the Morton numbers.

Souders' method of predicting the viscosities of liquid compounds from structural information has been extended to include the contribution from the phosphorus atom. The phosphorus contribution has been predicted from the critical pressure to be 67.1. The prediction method has been tested using data available on 15 phosphorus-containing compounds. The results are comparable with those reported in the literature for group contribution methods in general.

A numerical model of a non-isothermal reactor was investigated subject to a general wall heat transfer boundary condition. Both laminar and plug flow were considered. As part of the investigation, a series of simulations were performed under different conditions with respect to the dimensionless groups appearing in the balance equations. It was observed that the influence of these groups was dependent on the degree of heat transfer within the reactor and with the environment that was occurring. Specifically, three distinct heat transfer regimes were identified. These regimes were themselves functions of the internal and external heat transfer driving forces.

In this work the problem of determining the feed rate policy of fresh enzyme into stirred bioreactors to offset the loss of activity due to enzyme deactivation is addressed. On the basis of a series-type deactivation mechanism with arbitrary deactivation kinetics, it is shown that the feed rate policy is a feedback function of the deactivation rate. From both the theoretical and the practical viewpoints, these results provide a useful tool in understanding enzyme deactivation, and in designing bioreactor operation.

The chemical role of the catalyst in the catalytically stabilized thermal combustion of methane is investigated; platinum and manganese oxide catalysts were tested. Comparison of the performances of an externally heated catalytic wall tubular reactor and a non-catalytic alumina wall tubular reactor shows that complete conversion of methane is obtained at a slightly higher temperature in the catalytic flow reactor as a consequence of reactant depletion close to the wall due to heterogeneous reactions. However, a highly active catalyst such as platinum changes the selectivity dramatically, almost preventing the production of intermediate compounds and giving high combustion efficiencies.

This paper contributes to the present knowledge on heat transfer during the boiling of liquids in thin layer evaporators (TLEs). The application of a vibrating element in TLEs of the static type substantially changes both the hydrodynamics and the rate of heat transfer. A total of 689 individual heat transfer coefficients (IHTCs) were determined experimentally in a TLE with an exchange surface area of 0.16 m². They comprise 86 measurements in a TLE of the static type (without a vibrating element) and 603 measurements in the TLE equipped with two different spiral elements. Distilled water and pure ethylene glycol were investigated in the process of boiling at normal pressure and under vacuum. In the investigated range of frequency and amplitude of vibrations an increase in IHTC values of between 23.7% and 104.8% has been observed.