نشریه جداسازی و پدیده های انتقال
سال بیست و دوم شماره 1 (1390)

The secondary settling tanks are used in activated sludge wastewater treatment processes to separate treated water from the microbial mass leaving the aeration tank. In this study, a single-phase, 2dimentional, and unsteady state model was presented and solved to predict the concentration and velocity distribution in secondary settling tanks. The results showed a good agreement with the experimental data. Due to the non-Newtonian nature of activated sludge and its influence on the obtained flow field, the viscous models such as Bingham, Casson, modified Herschel-Bulkley, Bokil, and constant water viscosity model were used in modeling and their results were compared. The results obtained from different models show that the non-Newtonian models have a considerable effect on the determination of concentration and velocity distribution. The Bingham and Casson models overestimate the sludge blanket depth due to higher shear stress than other models and dont present acceptable results.

Direct hydrogen reduction of cuprous sulfide in the presence of lime is a method which has many advantages such as fewer operating stages and no SO2 pollution in comparison with the conventional methods to produce Cooper's metal. A mathematical model for simulation of the reaction is presented in this paper, which contains effects of structural changes and also non-isothermal reactions. In this model, mass and heat balance equations have been solved as functions of time and location to obtain concentration and temperature distribution in the pellet, and dusty gas model have been used to describe diffusion of the gas in the pellet. Mathematical equations in this model are the generalized grain model which has been applied to more than one reaction. Finally, the data predicted from the model have been compared with those determined experimentally. Temperature, quantity of lime, initial porosity of the pellet, and pellet diameter are parameters that are used to validate the model. In this research, the grain of lime was assumed to be porous and made up of non porous sub-grains. Results of the model had better agreement with experimental data by this assumption.

In this study, the effect of operating parameters (temperature and pressure) and feed properties (feed concentration and pH) on performance of nanofiltration process (permeate flux, NaCl rejection and total hydraulic resistance) was investigated and genetic algorithm–artificial neural network (GA-ANN) method was used to model these parameters during the nanofiltration of wastewater obtained from ion-exchange resins regeneration in decolorizing columns of sugar industry. The polyamide tubular AFC80 membrane provided by PCI Company was used for experiments. In order to predict the permeate flux, NaCl rejection and total hydraulic resistance, multi-layer perceptron neural network with 4 inputs and 3 outputs was used. Genetic algorithm method was used to optimize the number of neurons in ANN hidden layer. The results showed that the total hydraulic resistance increases with increase in transmembrane pressure and feed concentration. The permeate flux increases with increasing the temperature and pH, whereas the total hydraulic resistance declines. Average permeate flux was found to be 7.7±3.7 kg/m2h. The rejection of sodium chloride was between 16% and 42.7%. The results of GA-ANN modeling method showed that the combination of ANN and GA gives better results and by combining these methods, analysis rate and precision of modeling increases. By using a network with 24 neurons in one hidden layer, the sigmoid transfer function and 30%/20%/50% of data for training/testing/validating process, the permeate flux (0.98), NaCl rejection (0.94) and total hydraulic resistance (0.96) can be well predicted in the nanofiltration of decolorization column wastewater.

One of the most effective geometrical parameters on performance of the wavy fin plate compact heat exchanger (CHE) is the wave amplitude. New correlations are required to consider the wave amplitude for designing this type of exchangers. In this paper, the effect of wave amplitude on Colburn factor (j) and Fanning friction factor (f) was investigated. The J factor is a criterion of heat transfer rate and the f factor is proportional to pressure drop. In this study, the three dimensional computational fluid dynamics simulation of an exchanger’s wave channel was performed. The channel’s height was 8 mm, its pitch was 2 mm, and its lengths were 43.2 and 64.8 mm. The published experimental data of this channel with wave amplitude of 1.5 mm is available. Therefore, the simulation was firstly carried out using this wave amplitude in Reynolds number range of 600 to 6500. The simulation results were in good agreement with the experimental data and the mean error values obtained for j and f were 7% and 4% respectively. In the second step, the channel simulation was implemented using wave amplitudes of 0.5, 1, 2, 2.5, and 3 mm. The results indicated that the wave amplitude is effective on j and f factors. Both factors increase with increasing the wave amplitude in laminar and turbulent flows; however, the wave amplitude is more effective on the f factor. Using the multiple curve fitting method, two correlations were obtained for j and f versus Reynolds number and different geometric dimensions of the wavy channel, especially the wave amplitude. In order to determine these correlations, previous experimental and calculated data and new calculated results were applied and the effect of wave amplitude was considered. These equations are useful for wavy fin plate compact heat exchanger using air as the cooling fluid.

In this research, separation of an aromatic compound (benzene) from aliphatic hydrocarbons (hexane and heptane) was investigated using the ionic liquid 3-methyl-1-octyle imidazolium sciaenid, [Omim][SCN]. The liquid-liquid equilibrium data of the two ternary systems, i.e. {benzene + hexane + [Omim][SCN]} and {benzene+heptanes+[Omim][SCN]} were determined at 298.15 K and atmospheric pressure. Selectivity and distribution coefficient of the systems were derived from the tie line data in order to determine the capability of the ionic liquid in separation of the aromatic compound from the aliphatic compounds. The distribution coefficient depended on the type of the aromatic compound present in the system. However, the selectivity of systems which contain aliphatic compounds with longer chains was higher. The consistency of tie line data was studied using the Othmer-Tobias and Hand equations. Finally, the experimental data of the ternary systems were well correlated with the NRTL model. The equilibrium data reported in the present work were not presented by other researchers in the previous literature.

In this article, microgel rheological properties has been studied at different Carbopol 940 concentrations in water using the rotational and capillary rheometer. The results obtained at shear rates less than 1000 1/s showed that the gels follow the Herschel-Bulkley rheological model. Investigation of the rheological properties’ temperature variation indicated the stability of the gels up to 50 °C. The effect of time on the rheological properties of the gels showed their appropriate stability in a period of 40 days at 25 °C. With respect to the effectiveness of pH of the medium on the gels’ behavior, the effect of pH was studied by addition of caustic soda and it was indicated that in a neutral medium, the gels have the maximum viscosity. The capillary tube rheometer was used to investigate the fluid behavior at high shear rates (shear rate range of 190000-600000 1/s). The results of the application of oscillatory rheometer showed a nonlinear and weak viscoelastic behavior of this kind of gels up to a Carbopol concentration of 0.5% compared to other viscoelastic fluids such as water-methyl cellulose solutions.

In this research, operational parameters of uranium recovery from the waste product of a uranium conversion facility’s fluorination reactor were investigated experimentally using solvent extraction method. In this work, the effect of parameters such as temperature, contact time, type of diluent, acidity of the solution, and initial uranium concentration were investigated. The extraction experiments were carried out at uranium initial concentration range of 450-2200 mg/l and organic to liquid phase ratio of 2. The experiments were conducted at different temperatures and the results showed that at lower temperatures, the extraction of uranium has its maximum value. Different diluents were used in the experiments and the results revealed that when kerosene is applied, the amount of extraction has its highest value. The stripping experiments were carried out using different stripers and the experimental results indicated that using a 1.5 M phosphoric acid solution results in the highest uranium recovery in the striping process.