Search

The aim of our research was to develop a numerical model of microflows occurring in the culture chambers (CC) of a microfluidic device of our construction for high-throughput drug screening. The incompressible fluid flow model is based on the lattice-Boltzmann equation, with an external body force term approximated by the He-Shan-Doolen scheme and the Bhatnagar-Gross-Krook approximation of the collision operator. The model accuracy was validated by the algebraic solution of the Navier–Stokes equation (NSE) for a fully developed duct flow, as well as experimentally. The mean velocity prediction error for the middle-length cross-section of CC was 1.0%, comparing to the NSE algebraic solution. The mean error of volumetric flow rate prediction was 6.1%, comparing to the experimental results. The analysis of flow hydrodynamics showed that the discrepancies from the plug-flow-like velocity profile are observed close to the inlets only, and do not influence cell cultures in the working area of CC. Within its workspace area, the biochip provides stable and homogeneous fully developed laminar flow conditions, which make the procedures of gradient generation, cell seeding, and cell-staining repeatable and uniform across CC, and weakly dependent on perturbations.

We describe a high-performance implementation of the lattice-Boltzmann method (LBM) for sparse geometries on graphic processors. In our implementation we cover the whole geometry with a uniform mesh of small tiles and carry out calculations for each tile independently with a proper data synchronization at tile edges. For this method we provide both the theoretical analysis of complexity and the results for real implementations for 2D and 3D geometries. Based on the theoretical model, we show that tiles offer significantly smaller bandwidth overhead than solutions based on indirect addressing. For 2-dimensional lattice arrangements a reduction of memory usage is also possible, though at the cost of diminished performance. We reached the performance of 682 MLUPS on GTX Titan (72\% of peak theoretical memory bandwidth) for D3Q19 lattice arrangement and double precision data., Comment: Accepted in IEEE Transactions on Parallel and Distributed Systems, 14 pages, 9 figures, 5 tables

We describe a high-performance implementation of the lattice Boltzmann method (LBM) for sparse 3D geometries on graphic processors (GPU). The main contribution of this work is a data layout that allows to minimise the number of redundant memory transactions during the propagation step of LBM. We show that by using a uniform mesh of small three-dimensional tiles and a careful data placement it is possible to utilise more than 70% of maximum theoretical GPU memory bandwidth for D3Q19 lattice and double precision numbers. The performance of our implementation is thoroughly examined and compared with other GPU implementations of LBM. The proposed method performs the best for sparse geometries with good spatial locality., 20 pages, 20 figures, sent to Computer Physics Comunications, after including Reviewers comments

The construction and performance of a new apparatus for coating and encapsulation of particle cores with diameters lower than 1 mm are discussed in detail. The process of electrostatic dry powder coating was carried out in a diluted, fast circulating spout-fluid bed. The spatial separation of the draft tube entrainment zone from the coating zone leads to the reduction of effects of particle-to-apparatus wall adhesion and pneumatic nozzles plugging. As a result, the fine product of high quality without aggregates, agglomerates and granules, in a short time (approx. 10 min) and at a low temperature (approx. 45 °C) was obtained. Coating powder was charged electrostatically and continuously fed to the apparatus column. In order to verify the performance of the device, a number of tests were carried out with the application of materials used for the production of enteric drugs in the pharmaceutical industry. The quality of obtained coats was investigated by microscopic methods and the kinetics of dissolution of color marker in aqueous solutions was determined. As a result of the research, we obtain microcapsules with a constant release rate of the marker substance in alkaline environment, over 25 times faster than in the acid medium, with only 0.31% release of marker in acid solution in 2 h, which is a sufficient level for gastric resistance.

Batch sorption experiments were carried out for the removal of cationic dyes (methylene blue and malachite green) from their aqueous solutions using sorbent made from fly ash-a waste material. Effects of various experimental parameters: initial dye concentration, contact time, pH, adsorbent dosage, solution temperature, surfactant addition and ionic strength on the fly ash sorption of dyes were evaluated. The isothermal data for sorption followed the Langmuir model. The maximum sorption capacity obtained for methylene blue and malachite green was 36.05 mg/g and 40.65 mg/g, respectively. Kinetic studies indicate that sorption on fly ash follows the pseudo-second order kinetics. Present research suggests that fly ash could be an appropriate adsorbent for the removal of basic dyes from aqueous solutions.

The aim of this work was determining optimal parameters of the investigated jet-loop reactor by means of the built by authors low cost PIV system. The criterion of optimization was value of flow rate in a draft tube. It was found that flow rate in the draft tube increases with the increase of flow rate in the supplying nozzle, what is caused by increase of outlet velocity of supplying liquid stream. With increase of distance between the nozzle and the upper edge of the draft tube the diameter of the stream with high velocity increases, what causes growth of flow rate of liquid at the outlet. Long draft tube causes disturbance of flow, backward circulation and decrease of flow rate, because of decrease of distance from its lower edge to the bottom of the reactor. As a result, the liquid circulation may be stopped. Liquid circulation is more intensive in wider draft tubes, wider than diameter of flowing out stream. In the case of a narrow draft tube the liquid stream flowing out of the nozzle may flow outside the draft tube wall disturbing circulation.

The results of dry coating of pharmaceutical materials (micro-particles), carried out in spout-fluid bed apparatus, designed by the authors are presented in the paper. The design has been developed starting from the Wurster conception device. The examination consisted of the determination of the optimal composition of coating mixture powder to plasticizer proportion, as well as the determination of basic process parameters (flow rates of individual streams). Resistance of coating on the conditions in gastrointestinal tract was chosen as optimization criterion. The purpose of the process is to defend an active substance placed in a core from the surroundings in the stomach. This active substance should be excellent dissolved only just in the small intestine (in case of an intestine medicine application).

Biosorption of Cu(II) and Cr(III) ions from aqueous solutions by peanut shell biomass was investigated as a function of initial pH, initial biomass concentration and temperature. The optimum sorption conditions were studied for each metal separately. The kinetics and equilibrium of biosorption were examined in detail. Four kinetic models (pseudo-first order, pseudo-second order, power function equation, and Elovich model) were used to correlate the experimental data and to determine the kinetic parameters. Four well-known adsorption isotherms were chosen to describe the biosorption equilibrium. The experimental data were analyzed using two two-parameter models (Langmuir and Freundlich) and two three-parameter models (Redlich–Peterson and Sips). The equilibrium biosorption isotherms showed that peanut shells possess high affinity and sorption capacity for Cu(II) and Cr(III) ions, with monolayer sorption capacities of 25.39 mg Cu2+ and 27.86 mg Cr3+ per 1 g biomass, respectively. All results showed that peanut shells biomass is an attractive, alternative low-cost biosorbent for removal of heavy metal ions from aqueous media.

The aim of the study was to investigate the simultaneous removal of p-cresol and Cr(III) ions using a membrane contactor enhanced by a micellar solubilization. The investigations of solutes removal were performed taking advantage a polysulfone capillary membrane contactor in a counter-current mode and a micellar sodium dodecyl sulfate ions solution. The coefficients for a resistance in series mass transfer model were determined.

Measurements of spouting characteristics were carried out in an apparatus consisting of two conical sections connected to a cylindrical section. Inside the lower conical section a draft tube was fixed. Results of minimum circulation velocity were obtained for an apparatus equipped with a draft tube of different inside diameters and lengths for different kinds of particles as well as for different static bed heights. Experimental data have been further analyzed to develop new correlations.

Further results of simulations of the drying kinetics in a spouted bed dryer with a draft tube (ICFB) will be presented in this paper. As noted earlier [Ind. Eng. Chem. Res. 2004, 43, 1113-1124], the Eulerian-Eulerian multifluid modeling approach was applied to predict gas-solid flow behavior. The heat- and mass-transfer model was coupled with computational fluid dynamics (CFD) code FLUENT 6.1, through application of user-defined functions (UDF). The falling rate period of drying was described by the linear and nonlinear lumped-parameter models. The new, robust point-by-point solution procedure was proposed to predict the kinetics of long-time batch processes and to overcome a lack of sufficient computational performance. The results of the simulations were compared with experimental data and with values obtained from various correlations. The drying kinetics during the constant and falling rate periods of drying of inorganic particles was predicted with sufficient accuracy and efficiency for engineering calculations. The mean relative errors were 3.24% and 19.8% for drying periods I and II, respectively. The biological texture of rapeseed caused higher discrepancies; however, for both types of grain, the results from the CFD simulations were more similar to the experimental data than to the values obtained from correlations. A CFD modeling technique, coupled with classical drying kinetics models, provided useful results for engineering purposes and allows use of the model throughout all phases of research and development.

A hydrodynamic model based on continuity and conservation equations was used to predict flow behavior in a spouted-bed dryer with a draft tube. The Eulerian-Eulerian multi-fluid modeling approach was applied to predict a gas–solid flow behavior. The model was qualitatively validated and the results of simulations were compared with the experimental data published by various researchers. The model predicts flow behavior well with fluctuations, circulations of phases, and cluster formations in the loading region, as was observed in experiments. The predicted solids circulation rate and the fraction of gas bypassing were close to experimental values published for various systems. The coarse 2D axisymmetric grid, used in simulations, can handle the main process dependencies with a sufficient accuracy for engineering calculations.

For a description of the heat and mass transfer occurring during the drying of grain in a spouted bed dryer with a draft tube, a CFD modeling technique was used. The Eulerian−Eulerian multifluid modeling approach was applied to predict gas−solid flow behavior. A heat- and mass-transfer model was added as compiled executable code by means of UDF programming to FLUENT 6.1, expanding its abilities. Moisture was treated as UDS, and its transport in each phase and exchange between phases were predicted. The kinetics of drying was described by classical and diffusional models for periods I and II of drying, respectively. Results of the simulations were compared with experimental data and with values obtained from various correlations. CFD simulations predict very well the mass-transfer rate but underpredict the heat-transfer rate. Predicted values of the Nusselt and Sherwood numbers are above 2, in agreement with theory. The heat- and mass-transfer rates obtained from the correlations and CFD simulations were u...

Phenols are often present in wastewaters from many industrial processes, such as oil refineries, petrochemical plants, ceramic plants, steel plants, coal conversion processes. Phenols are pollutants of high priority concerns because of their toxicity and possible accumulation in the environment. At present, several methods, such as microbial degradation, adsorption, chemical oxidation, incineration, solvent extraction, reverse osmosis and irradiation, are used for removing phenols from a wastewater. Conventional extraction processes are conducted presently in towers, columns or mixersettlers. Membrane-based solvent extraction (membrane contactor) can be used as an alternative technique to the classical solvent extraction. In such situation, the first liquid phase flows in the lumens of the fibers and the second phase through the shell side of the contactor. The liquid–liquid interface is immobilized in the pores of the membrane and can be regulated by maintenance of a suitable pressure difference between both phases. A membrane is only a physical support and does not play an active role in the separation. The advantages of membrane contactors over classical extractors are: high surface area per unit volume, possibility of changing the hydrodynamics of both phases independently, no need for different densities of the phases, no emulsification and no loading or flooding. The new concept of membrane contactor enhanced by the micellar solubilization will be presented. Application of micellar solubilization of organic component in the receiving phase reduces its concentration at a membrane wall and thus enhances the kinetic of the mass transfer. Phenols are solubilized within a micellar interior above the critical micelle concentration (CMC). Dimensions of micelles are much larger than the phenolic molecules. They contain about 50–100 surfactant molecules. The micellar solution flows counter-currently inside or outside the capillaries whereas the feed solution containing phenols flows on the other side of the capillary. The driving force of phenol diffusion through the membrane pore was enhanced thanks to solubilization of the phenol molecules within micelles. *Corresponding author. 424.fm 6 11

Classic methods of biosurfactant separation are difficult and require large amounts of organic solvents, thus generate high amounts of waste. This work presents and discusses in detail an original procedure to separate rhamnolipid from fermentation broth using high performance membrane techniques. Due to the unique properties of surface active agents, such as capability of forming aggregates above the critical micelle concentration, it is possible to easily purify the biosurfactant with high efficacy using inexpensive and commonly used membranes. In this article, two-stage ultrafiltration is proposed as a method for separating and purifying rhamnolipid from the culture medium. The obtained purified rhamnolipid solution was capable of reducing surface tension of water down to 28.6 mN/m at critical micelle concentration of 40 mg/l. Separation of rhamnolipid was confirmed by HPLC; three types of rhamnolipids were identified (RL1, RL2, RL4), with considerable predominance of RL2.

The results of experimental investigations of bed flow hydrodynamics in spouted beds are compared with CFD simulations (Eulerian-Eulerian approach) for two different column geometries. The experimental results of bed porosity and fluctuation frequency of mass flow rate of grain in the fountain region are compared with the corresponding results of simulations. The simulation results confirmed the observations of Muir et al. (1990, Chem. Eng. Comm. 88: 153–171) and Yang and Keairns (1978, AlchE Symp. Ser. No. 176 74: 218) that fluctuations of bed flow in DTSB are caused by particle cluster formation in the loading region at the bottom of column. The solids cross into the jet and cover the column inlet and are carried upward periodically through a draft tube. Subsequent figures obtained from simulations, which show stages of particle cluster formation at the entrance of column, exactly match visual observations. The frequency of fluctuations of grain mass flow rate predicted in simulations (∼5–6 Hz) is in the range of that experimentally determined. The fluctuating inflow of solids results in slug formation and explains the vertical variations of height and porosity of the fountain.

The hydrodynamics of a down flow jet-loop reactor with a gas-liquid-solid three-phase system in semi-industrial scale were investigated. The Eulerian-Eulerian modelling approach was applied to predict flow behaviour in the reactor. A commercially available, control-volume-based code FLUENT 6.1 was chosen to carry out the computer simulations. In order to reduce computational times and required system resources, the 2D axisymmetric segregated solver was chosen. The influence of different k-e turbulence models, as well as, different types of meshes on velocity profiles in each phase was analyzed. The unstructured mesh reduces discrepancies on the axis of symmetry caused by the axisymmetric solver and is more accurate. The prediction error of the water circulation rate ratio for a gas-liquid system was only 3.6 % and about 15 % for gas-liquid-solid system. Unfortunately the gas hold-up was not predicted properly.