The results of non-equilibrium dynamic Monte Carlo (DyMC) simulations on the permeation of methane and ethylene in MFI-type silicalite membranes are reported. The required parameter for the rate of jumping is estimated from our previous MD studies, and the validity of two different lattice models is examined in terms of the reproducibility of the self-diffusion coefficients using these estimated parameters. Our simulation results indicate that different lattice models are required for methane and ethylene. The permeability through silicalite membrane is estimated by non-equilibrium DyMC and compared with the expected values that are predicted by a previously reported method, the combined method of molecular simulation techniques with a permeation model (CMP). The simulated permeability of methane shows agreement with CMP values, and that of ethylene is reasonable when compared with the CMP values.
We have constructed a new pressure-velocity coupling method, named SIMPLERS (SIMPLER-Strict), based on SIMPLER (SIMPLE-Revised). In SIMPLER, unlike in SIMPLE, the pressure is calculated directly, and velocity is estimated from the calculated pressure field. In addition, to estimate the pressure field, velocity is simultaneously estimated considering the effect of neighboring points, and in SIMPLERS the pressure correction equation is solved to enable the velocity correction to be used. This treatment is expected to improve the convergence rate of the calculation. To validate this method, we apply SIMPLERS to back step flow (a flow case that is often used as a benchmark test) and the combustion field, and compare the convergence rate with those of SIMPLE and SIMPLER.
The mass transfer rate between solid particles and liquid and the critical frequency for complete suspension in a reciprocal shaking vessel were studied. The solid-liquid mass transfer coefficient of the reciprocal shaking can be correlated with the same correlation as that of a rotational shaking at higher shaking frequencies than the critical shaking frequency for complete suspension, NJS.
kLdp/D = 0.112[dp4PV/(ρν3)]0.29Sc1/3 (NJS < N < Ntr3)
The critical shaking frequency NJS was 20% higher than that of the rotational shaking, which was correlated with the following equation.
A mathematical model has been developed based on the nonisothermal Sharp Interface Model (SIM) for Fluid-Solid Non-Catalytic (FSNC) reactions for nonporous solid reactants. Fully transient analysis has been carried out for particles with both constant and varying sizes incorporating powerlaw kinetics. The Finite Volume Method (FVM) has been employed for numerical solution of the moving boundary problem. Its suitability in such applications has been established. The effects of various model parameters on the reaction characteristics have been studied. The phenomena of instabilities have also been analyzed.
Fully transient analysis of Fluid-Solid Non-Catalytic Reactions (FSNC) based on the Sharp Interface Model (SIM) has been carried out using the Finite Volume Method (FVM). The effects of nonlinear kinetics and relative resistances have been evaluated at the isothermal condition for particles with constant as well as varying size. The numerical solution agrees well with the analytical solution for the first order reactions and the results for non-linear kinetics compared favourably with the results reported in literature. Various types of multiple reactions like independent, parallel and consecutive reactions have also been studied. The Finite Volume Method has thus been established to be an effective numerical tool for modeling such reactions.
A theoretical and experimental investigation is carried out on the dynamic behavior of a dehumidification system used to dry out the wet floor in a beverage factory. The dehumidification system consists of an absorption-type rotary dehumidifier with lithium chloride salt coated on the honeycomb monolith. Experimental runs were carried out in the factory. By coupling a previous dynamic model of the dehumidifier with an evaporation model of the wet concrete floor, good agreement has been obtained between the simulated and experimental results, thus confirming the applicability of the integrated model. To obtain better performance of the integrated dehumidification system, the model is next used to investigate the effects of various system inputs and parameters on its dynamic performance. The results show that the paired velocities of dehumidified and regeneration outlet airs, and the pre-heating temperature of the regeneration air have more prominent effect on dehumidification efficiency than the other factors. It is possible to both reduce the room relative humidity below 50% and dry out the water on the floor within 12 hours of dehumidification. The most suitable dehumidification conditions among those investigated are: velocity of room air 0.06 m/s, inlet temperature of regenerated hot air 343 K, velocity of regenerated hot air 0.5 m/s, air velocity in dehumidifying section 0.68 m/s and rotational speed of honeycomb 10 rph.
Microporous polypropylene (PP) membranes were prepared by the thermally induced phase separation (TIPS) technique and used as supports of liquid membranes. Several different quenching temperatures were used in the preparation of the microporous membranes. Membranes prepared by air-cooling and quenching in a water bath of 353 K showed higher porosity at the membrane surface compared with those made by quenching in a water bath of 303 K and in ice-water. A simplified one-dimensional heat transfer equation was solved to evaluate the temperature profile within the membrane to demonstrate the effect of quenching temperature on the final pore size distribution. The calculated result suggested that the asymmetric membrane structure formed by quenching in 303 K water and ice-water was not attributable to the temperature gradient. Supported liquid membranes (SLMs) were prepared by impregnating a carrier solution in the prepared porous polypropylene membranes, and uphill transport of Ce(III) was investigated using octyl(phenyl)-N,N-diisobutyl-carbamoyl-methylphosphine oxide (CMPO) as a carrier, tributyl phosphate (TBP) as a modifier, and dodecane as a solvent. The membranes prepared by air-cooling showed higher permeability of Ce(III) than the commercial membrane.
A methodology to determine the damping and elastic properties of particulate materials subjected to low intensity vibration was investigated. Acceleration transmissibility through a shallow bed was measured by sensing two acceleration signals at both top and bottom of the bed using a top-cap mass mounted on the bed surface. An analogue model technique was employed, which enables the damping and elasticity to be identified simultaneously and conveniently, compared to existing direct experimental techniques. The effects of the model forms upon the acceleration transmissibility data were theoretically examined. From the experimental resonant characteristic curve, the loss factor and elastic modulus were deduced using the analogue model technique. The data gave a reasonable agreement with data generated by existing techniques.
The effect of experimental variables upon the damping characteristics of particulate materials was investigated using the low intensity vibration method (Yanagida et al., 2003) of the present series of papers. The loss factor was measured from the acceleration transmissibility data. A significant stress dependence upon the elastic modulus was seen, whereas the loss factor showed the insensitivity to experimental variables including the compressive stress, the particle size and the bed height within the range of this study. A model based upon fundamental contact mechanics gave a qualitative interpretation for the insensitivity of the loss factor at small amplitudes.
Fluidized bed medium separation (FBMS) was carried out in order to separate ores by a dry method. The FBMS utilizes the floating and sinking of different-density objects in a gas-solid fluidized bed. Although the apparent density of the fluidized bed is the main factor for determining the floating and sinking of separating objects, it is also affected by the size and shape of separating objects, the fluidized bed dimensions, as well as the fluidization stability. In the present work, the distribution of apparent density was estimated by measuring the equilibrium height of spheres with various densities immersed in the fluidized bed. Silicastone and pyrophyllite were used as the ores for separation, the density of which differed by about 250 kg/m3. The apparent density of the fluidized bed was adjusted between the separating ores by mixing glass beads and steel shot which had about the same minimum fluidization velocity. For continuous separation, the FBMS was combined with devices that can continuously remove floating and sinking objects. As a result, a high separation efficiency of about 95% was attained for each ore although the ore recovery is dependent on the experimental conditions.
The hydrodynamic behavior of a circulating powder-particle fluidized bed (CPPFB) was examined in terms of total solid entrainment rate and fine particle hold-up. Experiments were carried out in a circulating fluidized bed of 2 m in height and 0.052 m in diameter, using FCC catalyst particles of 66 μm as the coarse particles and cohesive aluminum hydroxide powders of 0.5 μm, 1 μm and 3 μm as the fine particles. The effects of adding cohesive fine particles on the total solid entrainment rate were investigated under different operating gas velocities. The total solid entrainment rate at the same operating gas velocity was found to increase with the increase in the fine particle size and decrease with the increase in fine particle hold-up.
In this paper, design of a complete reactive distillation (RD) system has been developed. The reactive distillation system in the study is the production of ethyl acetate (EtAc) via esterification of acetic acid (HAc) with ethanol (EtOH) using sulphuric acid as homogeneous catalyst. A suitable NRTL model parameter set for calculating of liquid activity coefficients has been established with excellent prediction of the compositions and temperatures for the four azeotropes in this system. In the VLE calculations, vapor association of acetic acid due to dimerization has also been considered. A reactive distillation column with an overhead decanter can be designed to achieve over 93 wt% of ethyl acetate composition at organic phase top product stream while the bottom product stream is designed to be rich in acetic acid so that it can be recycled and mixed with fresh acid make-up stream to serve as acid feed to the reactive distillation column. Since the purity of the optimum top organic product is still not good enough for the ethyl acetate product specification in industry, an additional column is designed to purify the ethyl acetate product of the reactive distillation column to over 99.5 wt%. The top draw of the second column will be recycled back to the decanter. In summary, the overall optimum design of this ethyl acetate reactive distillation system includes two columns (including the reactive distillation column and the second column), one decanter, and two recycle streams. The optimum operating condition of the overall system will also be studied to minimize the total operating cost of the overall system while meeting product specifications.
In many industries, the effective monitoring and control of batch processes is crucial to the production of high-quality materials. Several techniques using multivariate statistical analysis have been developed for monitoring and fault detection of batch processes. Multiway principal component analysis (MPCA) has shown a powerful monitoring performance in many industrial batch processes. However, it has shortcomings that all batch lengths should be equalized and future values of batches should be estimated for on-line monitoring. In order to overcome these drawbacks and obtain better monitoring performance, we propose a new statistical method for on-line batch process monitoring that uses different unfolding method and independent component analysis (ICA). If the measured data set contains non-Gaussian latent variables, the ICA solution can extract the original source signal to a much greater extent than the PCA solution since ICA involves higher-order statistics and is not based on the assumption that the latent variables follow a multivariate Gaussian distribution. The proposed monitoring method was applied to fault detection and identification in the simulation benchmark of the fed-batch penicillin production, which is characterized by some fault sources with non-Gaussian characteristics. The simulation results clearly show the power and advantages of the proposed method in comparison to MPCA.
In recent years new methods for active protein film fabrication have been rapidly developing. One of these methods is the electrospray deposition (ESD) method. This method allows the spontaneous deposition of many identical dots and has a remarkable spatial resolution as well as the ability to overcome limitations like low rates of deposition, low efficiency of substance transfer, and cross-contamination. In this study, an antibody-based protein array for high-throughput immunoassay was fabricated with the ESD method using a quartz mask with holes made by an abrasive jet technique. An antibody solution was electrosprayed onto an ITO glass and then deposited antibodies were cross-linked with a vapor of glutaraldehyde. The diameter of the spots was approximately 150 μm. The arrays were then incubated with corresponding target antigenic molecules and washed. The captured antigens were collectively detected by fluorescence and chemiluminescence. These signals were quantitatively visualized with a high-resolution charge-coupled device (CCD) detector. Sensitive (~1 ng/ml) and simultaneous detection of various antigens could be performed by enzyme linked immunosorbent assay or fluorescence immunoassay using this system.
Some metal compounds in coal vaporize and form fumes during the combustion. The fumes are usually exhausted through the flue gas. For coal-fired combined power generation systems such as pressurized fluidized bed combustion (PFBC), hot metallic vapors may contact with the surfaces of gas-turbine blades. Since this contact of the hot vapors with the surface has a corrosive effect, it is necessary to control the formation of those fumes, which contain alkali metal compounds. In this paper, the evolution behavior of alkali metal compounds, especially sodium compounds, has been studied, using an electrically heated drop tube furnace with a low-pressure impactor. The main objective in this study is to elucidate the conditions and the possible mechanisms to form alkali metal compounds in particulate matter during combustion. Two types of coal with different sodium contents were tested, where the coal conversion characteristics were established. Furthermore, the evolution and inclusion of sodium compounds into the sub-micron particles were studied in relation to the particle size distribution and sodium fraction distribution in the collected particulates. The study proved that the evolution and inclusion of sodium on sub-micron particles depended on the functions of the coal type. The reaction-controlled mechanism and heterogeneous condensation via chemical reactions during the combustion much more influenced the inclusion of sodium in sub-micron particles. At the coarse particles of above about 0.5 μm, the reaction that formed the particles was mainly via gas film diffusion surrounding the particle.