CFD simulation and PIV measurement of the flow field generated by modified pitched blade turbine impellers

The hydrodynamics generated by modified pitched blade turbine (m-PBT) impellers with down-pumping mode were systematically investigated through particle image velocimetry (PIV) measurements and computational fluid dynamics simulations. The simulated mean axial velocity, mean radial velocity, and turbulent kinetic energy by the standard k–? turbulent model were validated against the measured PIV data. This shows that the standard k–? turbulent model predicts mean velocity well, but underestimates turbulent kinetic energy near the blade. The flow field and power consumption as well as pumping number for the m-PBT and the standard PBT impeller were predicted. The simulation results demonstrate that a few simple changes of the blade shape influence the velocity distribution, i.e., increasing the magnitude of mean velocity in the vicinity of impeller, and that the m-PBT impeller has a higher pumping efficiency than the standard one.     

Preparation and power consumption of surfactant-fuel oil-water emulsions using axial, radial, and mixed flow impellers

Surfactant-oil-water emulsions could have applications in enhanced oil recovery and the bio-desulfurization process applied to crude oil and some fractions. A simple way to prepare oil in water (O/W) emulsions is using a tank and an agitation device. The aim of this work is to propose a technology to prepare surfactant-fuel oil-water emulsions by means of a system involving a tank equipped with baffles, and an agitation device. The employed fuel oil was a high-viscosity fraction, which makes it difficult to handle. Axial, radial, and mixed flow impellers were assessed in the preparation of O/W emulsions, with and without the presence of baffles. Sixteen commercial surfactants were evaluated on the O/W emulsion formation. The effect of the storage temperature on the emulsions stability was assessed. The presence of salt on the surfactant-fuel oil-water emulsion was also investigated. Power vs. Reynolds numbers, extremely important data for the scaling up of the process, were calculated in basis of the power drawn when preparing the emulsions. Total consumption energy applied to the system, as well as pumping capacity were measured and related to the quality of the O/W emulsions obtained.     

水解酸化-膜生物反应器处理化工综合废水

采用水解酸化-浸没式膜生物反应器工艺处理化工综合废水。小试结果表明:在进水CODCr的质量浓度为1500～2400mg/L,BOD5与CODCr的质量比为0.28～0.35,pH值为6～9,水解酸化、膜生物反应器HRT分别为12、18h,膜通量约16L/(m2.d),污泥负荷约为0.38[CODCr]/(kg[MLVSS].d)时,经该工艺处理,CODCr、挥发酚去除率分别达到92%、97%以上,且膜生物反应器出水浊度小于1NTU,未检测到SS。     

自产气气提与纤维床反应器耦合发酵生产丁醇

采用发酵产物中的二氧化碳（CO₂）和氢气（H₂）作为循环气提气源,对丙酮丁醇梭菌（Clostridium acetobutylicum CGMCC 5234）发酵产物进行原位气提,实现丙酮、丁醇和乙醇混合物（ABE）的连续纤维床固定化发酵生产。连续发酵实验进行了12批次共309 h,总溶剂ABE当量浓度为133.3 g·L^-1（其中丁醇 83.5 g·L^-1,丙酮38.4 g·L^-1,乙醇11.4 g·L^-1）,葡萄糖消耗率为1.29 g·（L·h） ^-1,总溶剂ABE产率为0.431 g·（L·h） ^-1,转化率为0.333 g·g^-1,其中丁醇产率为0.270 g·（L·h） ^-1,转化率为 0.209 g·g^-1,发酵液中丁醇浓度控制在8～12 g·L^-1,显著优于游离发酵的结果。气提提取之后冷凝的ABE溶液出现分层现象,其中丁醇相丁醇浓度高达603.7 g·L^-1,极大地减缓后续分离提纯的负担。结果表明,自产气循环气提与纤维床固定化耦合连续发酵生产ABE（特别是丁醇）的工艺具有可行性和竞争力。     

Shear and skin friction on particles in power-law fluids agitated by flat-blade and fluid foil impellers

The shear rates that exert angular deformation on spherical particles have been measured. The particles are mimiced by a spherical probe. The probe has been immersed in various impeller-agitated power law fluids. The fluids are aqueous dispersions of polymers, e.g. CMC, xanthan gum and starch. The probe has been positioned in various points of a stirred vessel and at various angles. Angle-averaged shear rate distributions were produced. The distributions obtained are characteristic for the specific impeller flow patterns. The flow patterns have been identified by computational fluid dynamics (CFD). Two types of impellers representative for the flat and the fluid-foil blade design, i.e., a Rushton flat-blade turbine (RT) and a Narcissus impeller (NS) are studied. The effects of rheological properties and blade design on the ‘shear-rate-on-particles’ distribution are examined. The local shear field non-uniformity has been uncovered and compared in terms of the CFD-generated time-averaged velocity and deformation rate profiles. The ‘shear-rate-on-particles’ distribution apart from the impeller is found to follow qualitatively the time-averaged inner flow shear rate distribution. Referring to impeller speed 5-12.5 Hz, the dimensionless wall shear rate varied between 200 and 1000. In power law fluids, the shear rate on particles decreased up to 50%. The fluid-foil NS-generated shear field was found comparable to the shear field induced by conventional flat-blade turbines and appeared in cases less sensitive to polymer presence. The shear rate produced by the fluid-foil impeller in the highly shear-thinning model solution (n∼0.4) exceeded the flat-blade RT-imposed shear rate. The analysis has been extended to skin friction drag on particles. It is shown that, while exerting an undoubtedly greater angular deformation in water-like fluids, in polymer presence the conventional flat-blade turbine introduces a flow geometry that imposes particle drag that is close or in some cases even less than the one generated by the fluid-foil impeller. The fact implies a weak shape effect of radial turbines on shear-sensitive particles or particle dispersions in power law liquids.     

Shear flow of assemblies of cohesive granular materials under constant applied normal stress

We have studied plane shear flow of nearly homogeneous assemblies of uniformly sized, spherical, cohesive particles in periodic domains under constant applied normal stress. Our focus has been on (a) exploration of the effect of inter-particle attractive forces on the flow behavior manifested by dense assemblies under constant applied normal stress, and (b) comparison of the rheological characteristics observed under constant-applied normal stress and constant-volume conditions. As a model problem, the cohesion resulting from van der Waals force acting between particles is considered. Simulations were performed for different strengths of cohesion, shear rates, and applied stresses. From each simulation, the volume fraction, shear stress and the average coordination number have been extracted. We find that cohesive assemblies sheared under constant applied normal stress shear differently from those sheared at constant volume only in the dynamic sense, while the time-averaged rheological characteristics are essentially indistinguishable. At constant volume, the fluctuations in shear stress are larger than, but have the same dependence on cohesion as under constant applied normal stress. This study has also exposed a pronounced dependence of the apparent coefficient of friction on particle volume fraction in the quasi-static flow regime.     

Shear stress at polymer/metal interface during melting in extrusion

Prediction of the screw horsepower requirement involves, among many others, the calculation of the shear stress (τ_s) between the solid polymer and the barrel surface during melting. Prediction of the solid bed down-channel velocity also requires the calculation of τ_s. However, the pseudoplastic nature and strong temperature dependence of melt viscosity make the mathematics of calculating τ_s extremely difficult. As a first step of developing a reasonable mathematical model for calculating τ_s, experimental measurements of τ_s were made over a wide range of metal temperature and sliding speed for five commercial polymers using molded, block samples. Although dependences of τ_s on metal temperature and sliding speed were found to have similar functionality to those of the dependences of melt viscosity on melt temperature and shear rate, this study showed that τ_s could not be expressed as a sole function of the melt rheological properties. Our subsequent study, to be reported in a follow up paper, will show that τ_s must be expressed as a function of the thermodynamic properties and melt density of the polymer as well as the melt rheological properties and the melting conditions.     

A comparison of cavern development in mixing a yield stress fluid by rushton and intermig impellers

Intermig impellers have been postulated as very efficient for mixing highly viscous non-Newtonian fluids (such as xanthan and mycelial broths). However, no formal characterisation has been published and no fair comparisons have been made, based on accurate power drawn measurements and using equal number of impeller stages and equal diameter, if compared (for example) with the performance of Rushton turbines. Characterisation of the shape, size, and evolution of the well-mixed zones or “caverns” were correlated with power drawn, for single and dual Rushton turbines and for one- as well as two-stage Intermig unslotted impellers. Cavern evolution studies were carried out in a mixing tank (diameter=0.205 m, H/T=1.6) equipped with an accurate air bearing dynamometer. Carbopol 940 (0.25 wt.-%) was used as a model, transparent fluid. Impeller to tank diameter ratio was 0.53 for both impellers. Caverns were visualised by injecting methylene blue in the well-mixed zones. A single Rushton turbine developed larger caverns if compared with one-stage Intermig of the same diameter under power drawn below 1.5 kW m^?3. At higher power drawn, both impellers behaved very similarly, reaching a limit in cavern volume of about 40% of the total liquid volume, even at very high (20 kW m⁰³) power drawn. A similar trend characterised dual combinations: below 3 kW m^?3, dual Rushtons gave larger cavern volume if compared with the performance of two-stage Intermigs. In either case, power drawn higher than 3 kW m⁰³ was sufficient to mix more than 90% of the liquid volume. The presence or absence of the slot in the Intermig did not influence cavern development. Experiments with a smaller if compared with those obtained with the larger Intermig (D/T = 0.53).     

Steady-state mass transport at stationary discs under divergent laminar radial flow conditions

This paper deals with global and local wall-to-liquid mass transfer under divergent laminar radial flow conditions between two parallel stationary discs. Approximate theoretical expressions for local and overall transport rates in piston flow and Poiseuille flow, and empirical correlations from the literature, are compared to experimental observations utilizing the electrochemical method of rate measurement. The experimental results support the theoretical approach for this laminar flow regime, the latter expanding the scope of the applicability of electrochemical methods presented in the literature.     

Nitrification in a packed bed bioreactor integrated into a marine recirculating maturation system under different substrate concentrations and flow rates

BACKGROUND: A packed bed bioreactor (PBBR) activated with an indigenous nitrifying bacterial consortia was developed and commercialized for rapid establishment of nitrification in brackish water and marine hatchery systems in the tropics. The present study evaluated nitrification in PBBR integrated into a Penaeus monodon recirculating maturation system under different substrate concentrations and flow rates. RESULTS: Instant nitrification was observed after integration of PBBR into the maturation system. TAN and NO₂‐N concentrations were always maintained below 0.5 mg L^?1 during operation. The TAN and NO₂‐N removal was significant (P < 0.001) in all the six reactor compartments of the PBBR having the substrates at initial concentrations of 2, 5 and 10 mg L^?1. The average volumetric TAN removal rates increased with flow rates from 43.51 (250 L h^?1) to 130.44 (2500 L h^?1) gTAN m^?3 day^?1 (P < 0.05). FISH analysis of the biofilms after 70 days of operation gave positive results with probes NSO 190 ((β ammonia oxidizers), NsV 443 (Nitrosospira spp.) NEU (halophilic Nitrosomonas), Ntspa 712 (Phylum Nitrospira) indicating stability of the consortia. CONCLUSION: The PBBR integrated into the P. monodon maturation system exhibited significant nitrification upon operation for 70 days as well as at different substrate concentrations and flow rates. This system can easily be integrated into marine and brackish water aquaculture systems, to establish instantaneous nitrification. Copyright © 2011 Society of Chemical Industry     

PIV study of the flow field generated by a sawtooth impeller

Stereoscopic and high-speed particle image velocimetry (PIV) techniques have been employed to study the flow field induced by a sawtooth (EkatoMizer) impeller, operated in the fully turbulent flow regime at an impeller speed of 1500 rpm. Ensemble-averaged mean flow fields and turbulence quantities were calculated for a region close to the impeller blades. The flow was found to be anisotropic near the impeller and exhibited return-to-isotropy behaviour further away from it. Macro-instabilities were found to have a high probability of occurrence in the discharge stream. All three velocity components from the stereo-PIV measurements were used to estimate the dissipation rate, by adopting a large eddy simulation (LES) analogy. Spurious vectors distorting the dissipation rate calculation were identified, and various standard deviation filters were applied for vector validation. By evaluating the filtered dissipation rate profiles against the multifractal intermittency model of Meneveau and Sreenivasan (1991), the global standard deviation filter was found to be the most suitable type. The ratio of the maximum to the mean dissipation rate for the EkatoMizer discharge stream was found to be similar to that reported for Rushton disk turbine and pitched-blade turbine impellers in the literature, raising questions about the reported high-shear advantage of sawtooth impellers. However it should be noted that these PIV experiments were conducted outside the sawtooth blades and it is possible that the maximum dissipation rate occurs within the impeller swept volume, where could be significantly higher.     

真空变压吸附制氧径向流吸附器的流动特性模拟

真空变压吸附制氧是一个复杂的动态过程,深入了解真空变压吸附制氧过程中吸附器内的流动特性是吸附器设计与完善的基础.基于Fluent中的多孔介质模型,通过用户自定义函数功能,建立了真空变压吸附制氧用径向流吸附器的二维轴对称模型,研究了真空变压吸附首次和第二次循环中径向流吸附器的流动特性,对比分析了吸附剂颗粒直径、流道截面积...     

Porous silicate cement membranes generated by the novel method combining freeze casting and heat-dry curing

Porous silicate cement membranes (PSCMs) fabricated by the freeze casting method show great potential to be utilized in seawater pretreatment, fermentation broth separation, and industrial wastewater treatment due to its merit of high-temperature resistance, low-cost, and hierarchically ordered porous structures, while the freeze casting method is complex and time-consuming. In this work, a combination of freeze casting and heat-dry curing was initially applied to generate PSCMs. The preparation periods of PSCMs could be shortened by simplifying preparation processes and reducing curing time. The resulting membranes presented double-layer structures, containing a nucleation zone (N-zone) with dense structures and a stability zone (S-zone) with lamellar pore structures. The X-ray diffraction pattern of membranes displayed the mixed hexagonal and rhombohedral structures. This novel method could save more than half of energy consumption compared with the traditional preparation technology of silicate cement samples. The membranes with a mesopore size of 3.794 nm showed high permeation performance with pure water flux reaching 207.23 L m⁻² h⁻¹ under 0.15 MPa and room temperature. The separation efficiency of oil-water was 78.05% under operating pressure of 0.05 MPa. Molecular dynamics simulation was applied to narrate the microscopic process of transformation during heat-dry curing, and obtained a good similarity of consequences between the computational method simulation and experimental operation.     

Optimum emulsion liquid membranes stabilized by non-Newtonian conversion in Taylor-Couette flow

The use of Taylor-Couette flow for extraction by emulsion liquid membranes has been characterized. In particular, the rate limiting step for the extraction of a weak acid (benzoic) in an aqueous continuous phase to a strong base (NaOH) in the encapsulated internal phase has been attributed to a surfactant resistance. Numerical and analytical solutions are developed to characterize the mass transfer at Biot numbers Bi<10. It is demonstrated for fixed droplet composition that a single curve exists for all N for the extent of extraction by stretching the time coordinate by a function of the rotation rate ∝N^1.2 that is proportional to the globule external area per unit total volume. Moreover, the leakage rate is shown to increase as t^0.6 and to be proportional to N^0.75, the Kolmogoroff eddy velocity. Comparison with stirred tank data demonstrates that greater droplet stability exists for Taylor-Couette flow.     

Synthesis at several length scales of zeolite A membranes using a continuous flow method

Zeolite A membranes were synthesized in the inner side of porous TiO₂ and α-Al₂O₃ tubular supports by a continuous method. The methodology was then used at several scales for preparing membranes on TiO₂ and α-Al₂O₃ supports with lengths of 6, 12 and 25 cm. Formation of an homogeneous zeolite film was confirmed by XRD and SEM in all supports. Single-gas permeation experiments (He, H₂, N₂ and n-C₃H₈) indicated that Knudsen diffusion was the predominant mechanism in both supports. All the synthesized membranes present high flux and moderate selectivity to water in water–ethanol mixtures.     

Issues on soot removal from exhaust gases by means of radial flow ceramic traps

Radial flow ceramic foam traps (uncatalytic and catalytic) for soot removal from the exhausts of combustion systems were prepared. Their performances were tested by treating the exhaust of a gas-oil burner. Two different trap volumes (0.5 and 1.5 l) were employed to evaluate the effect of gas velocity on the trap filtering performances. It was found that at high gas velocity through the trap critical conditions leading to soot blow off may be reached in dependence of gas velocity and temperature, soot load on the trap and burner operating conditions (quantity and quality of particulate matter). At lower velocity critical conditions are unlikely to be achieved but the behaviour is still strongly dependent of the burner operating conditions. The catalytic trap showed good performances being able to remove soot from the burner exhaust gases with an efficiency of 70% and allowing the simultaneous filtration and combustion of the captured soot as well of part of the soot accumulated when the trap operating conditions were not favourable for soot combustion.     

Investigation on a vertical radial flow adsorber designed by a novel parallel connection method

Due to the increasing global demand for industrial gas,the development of large-scale cryogenic air separation systems has attracted considerable attention in recent years.Increasing the height of the adsorption bed in a vertical radial flow adsorber used in cryogenic air separation systems may efficiently increase the treatment capacity of the air in the adsorber.However,uniformity of the flow distribution of the air inside the adsorber would be deteriorated using the height-increasing method.In order to reduce the non-uniformity of the flow distribution caused by the excessive height of adsorption bed in a vertical radial flow adsorber,a novel parallel connection method is proposed in the present work.The experimental apparatus is designed and constructed;the Computational Fluid Dynamics (CFD) technique is used to develop a CFD-based model,which is used to analyze the flow distribution,the static pressure drop and the radial velocity in the newly designed adsorber.In addition,the geometric parameters of annular flow channels and the adsorption bed thickness of the upper unit in the parallelconnected vertical radial flow adsorber are optimized,so that the upper and lower adsorption units could be penetrated by air simultaneously.Comparisons are made between the height-increasing method and the parallel connection method with the same adsorber height.It is shown that using the parallel connection method could reduce the difference between the maximum and minimum radial static pressure drop by 86.2％ and improve the uniformity by 80％ compared with those of using the height-increasing method.The optimal thickness ratio of the upper and lower adsorption units is obtained as 0.966,in which case the upper and lower adsorption units could be penetrated by air simultaneously,so that the adsorbents in adsorption space could be used more efficiently.     

混凝/厌氧/兼氧-好氧膜生物反应器组合新工艺处理制革废水

针对制革废水高悬浮物含量、高有机物浓度及高色度的特点,采用混凝/厌氧/兼氧-好氧膜生物反应器组合新工艺对其进行中试处理研究,重点考察混凝预处理的反应条件（pH、投药量等）、生物反应器的启动策略,以及水力停留时间（HRT）、溶解氧（DO）和水温等运行参数对制革废水处理效果的影响。结果表明：当混凝过程中pH为9.0～10.0,聚合氯化铝（PACl）投加量为350～450 mg·L^-1时,废水悬浮物浓度（SS）、色度、总铬和化学需氧量（COD）去除率的平均值分别为70.4%,73.9%,97.7%和37.9%;基于阶梯负荷启动策略,50 d左右完成联合厌氧折流板反应器的启动,厌氧环节在HRT为20 h、水温30℃左右的条件下能够去除68.2%左右的COD;通过对兼氧-好氧膜生物反应器中DO分布的研究和HRT的优化,该单元的COD和NH₄-N的平均去除率分别达到67.7%和81.3%（HRT 6 h,DO 2.0～3.0 mg·L^-1）。经过组合工艺的处理,系统出水各项主要指标（COD、NH4-N、SS、色度和总铬等）达到DB 44/26-2001一级排放标准,表明本文提出的新工艺在制革废水处理中具有良好的应用前景。