A rotating disk contactor with perforated disks was used as a continuously operating equipment in liquid membrane separation. The hydrodynamic characteristics and rate of mass transfer in the contactor were studied. Experimental results indicated that perforated disks are effective in improving liquid-liquid dispersion and rate of mass transfer and can therefore be recommended for liquid membrane systems.

Investigations of the transition from bubbling to turbulent fluidizationwere conducted in a gas-solid fluidized bed using on-line computer analysis of pressure fluctuations. In order to study the effects of density and size of the particles as well as bed geometry on the transition, a series of experiments were performed, resulting in the following conclusions. 1. The technique of pressure fluctuation analysis was found effective and reliable in fluidization studies. 2. A correlation was obtained for the effect of powder properties and fluidized bed geometry on the transition velocity, Uc from bubbling to turbulent fluidization: 3. Addition of a new Yc-axis to Geldarts classification diagram is suggested, ia order to predict such flow pattern transition.

A liquid distribution model based on packing structure was developed for metallic corrugated plate packings. The model was verified by experimental results obtained in a 200mm-diameter and a 500mm-diameter column packed with type-4.5 or type-6.3 stainless steel corrugated plate packings. Further investigation showed that metallic corrugated plate packings are superior in liquid distribution behavior to Raschig rings, Pall rings and Cascade rings. From experimental measurements, the density of available wetting points is proposed for liquid distributor design.

A dripping plate column fitted with perforated trays was used in studying the mass transfer for non-Newtonian liquids. The rate of vaporization of aqueous polymer solutions rheological behavior with power-law was measured in the column. It was found that temperature, pressure, mass flow rate and the non-Newtonian behavior of the solution are significant factors influencing mass transfer rate, which increases with increasing temperature and mass flow rate, and with decreasing pressure and fluid consistency. Based on these experimental data, a correlation for the Colburn factor, jm, was developed: jm= 2.2708 × 10-5Re-0.7013 It can be used to design flash columns for monomer discharge from polymer solutions.

In applying computation technique to improve the operation of existing chemical processes, a computer-aided process optimization method is proposed-the improved two-tier method based on Successive Quadratic Pro-gramming(SQP). By inserting extra variables in SQP, a new quadratic programming problem is constructed, and a decomposition strategy for the quadratic programming is also presented, which has been found to be always feasible when the matrix is singular. This new two-tier method based on SQP has better performance both in numerical stability and in convergence behavior. Satisfactory result is obtained in applying this algorithm to optimize a methanol synthesis process, thus leading to implementation in actual practice.

An electrolytic cell with a dual cathode consisting of a Ni screen cathode and a poroun gas-diffusion oxygen cathode separated from each other by a porous diaphragm, has been constructed for measuring the loss of current efficiency. The reduction of oxygen and the reduction of HClO or C1O- could take place only on the oxygen cathode and Ni cathode, respectively. While the Ni cathode potential is higher than that for the evolution of H2, the Ni cathode current is responsible for the cathodic secondary reactions. The dependence of cathodic secondary current on the concentration of NaClO3, pH of the solution and the presence of Na2Cr2O7 has also been investigated. This method is also applicable to investigating the loss of cathode current efficiency for the production of chloro-alkali, BrO3-, IO3- and C1O4-.

The crystal phase adjustment and the after treatment of Copper Phthal-ocyanine pigments were studied experimentally by means of DTA, BET, X-ray diffraction, electro-microscopy, infra-red spectroscopy, trichromatic observation, wetting angle determination, particle size evaluation, etc. It was found that the color difference between the α-and β-Copper Phthalocyanine was mainly due to the different manners of the piling-up of the primary pigment crystals. The influence of dilution temperature on the properties of α-Copper Phthalocyanine by the sulphuric acid pasting process was studied. A series of organic solvents which did not convert α-to the β-Copper Phthalocyanine was listed. The pigment properties of β-Copper Phthalocyanine after re-fluxing with these solvents were found to be materially improved. A crystal seeding method for the preparation of β-Copper Phthalocyanine from α-Cop-per Phthalocyanine was demonstrated to be superior to the commercial kneading method, on account of less energy consumption.

The influence of temperature and duration of treatment of α-Copper Phthalocyanine in toluene was investigated experimentally, and the activation energy E in toluene was calculated to be 60.2 kJ/mol. A comparative study of different types of solvents on the rate of crystal phase transformation of α-Copper Phthalocyanine was carried out, and it was concluded that the solubility of Copper Phthalocyanine played a decisive role in the phasetransformation. The solubilities of α-Copper Phthalocyanine in different solvents could be estimated through the characteristic absorption peaks by means of the photo-electric spectrometer. It was also pointed out that the phase conversion process of α- to β-Copper Phthalocyanine pigment in those solvents where α-phase dissolves only with difficulty, is not related to Ostwald crystal growth.

Different methods of drying α-and β-Copper Phthalocyanine were tried, and the relation between the capillary aggregation of the pigment and particle size distribution was investigated. In drying α-Copper Phthalocyanine, organic solvent flushing and salt co-precipitation were found not to induce any phase transformation when a proper solvent was used. A series of surfactants, including anionic, cationic, amphoteric as well as a rosin soap, were tried for the surface treatment of α-Copper Phthalo-cyanine in order to control particle size, color shade and strength, and dis-persibility of the pigment. Critical surface tension of the solid was determined experimentally in order to verify the influence of surface, treatment on the properties of α-Copper Phthalocyanine. Two modes of adsorption of a surfactant on the pigment surface were proposed, that is, the hydrophilic single-point standing adsorption and the hydrophobic multi-point lying adsorption. The variation of the dispersion stability of the pigment after treatment with different surfactants was discussed in the light of the proposed modes of adsonption.

In investigating the performance of an extraction column in which axial concentration profiles were measured, a simplex method has been proposed for the simultaneous determination of axial dispersion and mass transfer coefficients from steady state data. The axial concentration at any point in continuous flow equipment is dependent on both the rate of material transfer and degree of dispersion. Consequently, parameters representing these effects should be obtainable by fitting the solution of a characteristic model equation to the experimentally determined cincentration profiles. In an extraction column of 5 cm inner diameter with a trapezoidal pulse, concentration profiles of Th in two phases (30% TBP (OK) and Th(NO3)4-HNO3-H2O) and two plate cartridges were measured. The diffusionmode] equation is used to describe the mass transfer process, and computed values obtained agree with experimental steady state profile data satisfactorily . Results show that axial dispersion coefficients of a trapfzoidal pulse column are smaller than those of a sine pulse, and that the column height for compensating axial dispersion is 32-44% of the apparent transfer unit height. When the simplex method is used for optimization, the approximation of the original differentia] equations with difference equations leads to linear systems of equations. The elimination method is used to solve directly each system (x or y) of equations, and the iterative method is used to solve a system (x and y) of equations. The program is very simple, the conver gence fast.Calculating a data set takes about 1 mm a minicomputer PDF 11/23.

A new type of internal separator for circulating FBB system is described in this paper. The main advantages of the separator are high separation efficiency and low pressure drop. It can be put to use with or without a special circulating passage. Without any sealing device the particles separated from the separator can return easily to the lean phase or to the dense phase through a circulating passage. Experimental results of eighteen sets of model installed in two bench scale facilities at cold condition are described. The cross sections of the bed and the freeboard of one facility are 360 × 360mm2 and 500× 500mm2 respectively. Another facility with a cylindrical combustion chamber of inside diameter φ300mm was also used in the experiments. When the bed material used is flyash from the freeboard of a bubbling bed boiler, the separation efficiency may reach up to 92-97%, while the pressure drop is usually lower than 4mm H2O. Recently, a full scale internal separator has been put into test in a Fluidized Bed Boiler. It has been shown that there is a good prospect for this separator to be used to improve combustion efficiency. Typical axial, tangential and radial velocity profiles measured by a probe and the fields of particle concentration determined by optical , fibre devices above this separator in the freeboard are described in this paper. Computers are used to calculate the critical particle diameter.

A method based on the orthogonal collocation technique was given for calculating the effectiveness factors of bidisperse catalysts. The set of coupled second-order differential equations for modeling the system was successfully solved. Legendre and Jacobi polynomials chosen as trial functions, and different combination patterns of collocation points in pellet and in particles, were tested for their influence on the rate of convergence and the final solution. Effectiveness factors were obtained for reversible and irreversible reactions. Values of the calculated effectiveness factor of first-order irreversible reactions agree very well with those of the exact solution, while those of second-order irreversible reactions are fairJy close to the trial-and-error solutions by Jayaraman. Effectiveness factors for reactions of the Langmuir-Hinshelwood kinetics were also obtained.

The generalized maximum likelihood principle and other statistical regression techniques were applied to 21 sets of ternary LLE, VLE and VLLE data for estimating the pair parameters of three local-composition models (NRTL, FHW and UNIQUAC), and the corresponding fitting accuracies were compared. Calculational results showed that the correlation and prediction of multicomponent fluid phase equilibria may be greatly improved by applying the maximum likelihood principle. Furthermore, the reliability and limitation of the cross-prediction among LLE. VLE and VLLE were investigated by utilizing the pair parameters obtained from binary and ternary data for the three models. The gap and interdependence between LLE and VLE were also discussed.