基于双侧阵列电极的电阻抗成像原位膜污染监测

基于电阻抗成像(electrical impedance tomography,EIT)技术设计了一种双侧阵列电极的膜污染原位监测装置,并应用于动态膜污染过程监测。通过模拟和实验比较了单侧和双侧电极阵列的成像效果。结果表明:电极双侧阵列方式依靠离子流动和外界电流产生额外电势差,监测更灵敏,重建图像与真实图像之间的相关系数0.89,能够更准确获得污染物信息。以高岭土、海藻酸钠(SA)和湖水为污染物,探究了不同污染物在超滤膜面的污染过程,得到了通量和膜污染成像图。结果表明,高岭土污染先集中于进水口附近,之后向中间区域和出水口迁移;湖水絮凝污染物主要分布在出水口附近;SA污染主要集中在进水口附近区域。建立了阻抗变化和污染物厚度之间关系,结果显示随厚度增大,阻抗变化增长速率也增大。     

动态膜处理污水时阻力分布及污染机理

对高岭土动态膜处理污水过程中膜污染阻力分布及膜污染机理进行了研究。通过测定和计算得知不同操作条件下各部分阻力的比例及其变化情况。动态膜过滤污水的阻力主要由动态膜膜孔堵塞阻力和膜面污染层阻力所控制。随着错流速度的增大,总阻力减小,各部分阻力的比例有所变化,但动态膜膜孔堵塞阻力和膜面污染层阻力依然占主导地位。利用传统膜过滤时有关膜污染的堵塞模型和滤饼过滤模型对动态膜处理污水时的实验数据进行分析,结果表明,动态膜过滤污水过程中,过滤初期约10 min左右,污染以膜的微孔堵塞为主,此后,膜的污染情况因操作条件不同而有所差异,影响最显著因素为错流速度,当错流速度较小时,膜的污染以膜面沉积污染物为主,符合滤饼过滤模型。适当提高错流速度有利于减小过滤阻力。     

铝系混凝剂减缓膜污染的红外光谱-多变量曲线分辨分析

超滤是一种高效的水处理技术，近年来被广泛应用于工业废水处理、生活污水回用、海水淡化预处理等领域。然而，超滤长期运行会造成膜污染。本文采用了在线混凝结合超滤工艺，使用不同形态的铝系混凝剂（硫酸铝、氯化铝或聚合氯化铝），处理含有不同溶解性有机质组分（腐殖酸、牛血清白蛋白和高岭土）的模拟原水，研究不同铝形态、不同组分及其相互作用对超滤膜污染过程的影响。本研究建立了流量衰减模型模拟膜污染过程，结合衰减全反射红外光谱（IR-ATR）和多变量曲线分辨-交替最小二乘法（MCR-ALS）的数据处理方法对膜上的多种污染物进行定性和定量分析。结果表明硫酸铝和氯化铝混凝剂均可明显提高膜比通量，减缓膜污染。该工艺混凝剂投加量低于常规处理工艺即可明显减缓膜污染。混凝剂投加量为0.4mg/L时，氯化铝混凝效果较好，混凝剂投加量为2.4mg/L时，硫酸铝混凝效果较好。低投加量（0.2mg/L、0.4mg/L）下，PAC对缓解膜污染程度不明显，反而加重膜污染。牛血清白蛋白对超滤膜的污染比腐殖酸严重。因为牛血清白蛋白的存在大大降低了混凝的效果，阻碍疏松滤饼层的形成。向原水中投加硫酸铝混凝剂，膜污染主要发生在过滤前期，即膜孔窄化、堵塞。过滤后期，膜表面形成疏松滤饼层，对膜通量影响不大，膜污染减缓。     

XDLVO理论分析混合有机污染物的膜污染

通过XDLVO理论分析了混合有机污染物的膜污染状况,对腐殖酸、海藻酸钠、牛血清蛋白这3种典型的模型污染物及其不同比例的混合物及PVDF膜的界面自由能进行了定量分析,从而判断膜污染性能,并通过过滤实验对预测结果进行了验证。结果表明,混合污染物和单一污染物界面性质完全不同,但根据界面自由能判断膜污染程度与过滤实验相符;溶液pH对混合污染物的溶液性质影响较大,最终影响膜过滤实验的污染程度,当pH较大时,膜污染程度相对较轻;添加不同的金属盐会影响混合污染物的界面性质,进而影响膜污染程度,添加钙离子相对镁离子和钠离子更容易导致膜污染。     

紫外/氯氧化预处理对超滤膜污染的控制研究

以紫外/氯氧化为预处理工艺，研究了该工艺预处理腐殖酸与高藻水对后续膜污染的影响。结果表明，紫外/氯氧化预处理对两种不同类型污染物引起的膜污染有显著不同的影响。预氧化后，腐殖酸引起的超滤膜污染得到有效控制，在次氯酸钠投加量为0.5 mmol/L时，可逆膜污染阻力降低了81.3%。高藻水经过紫外/氯预氧化后，膜污染反而加重，在次氯酸钠投加量为1 mmol/L时，不可逆污染阻力增大7.65倍。紫外/氯预氧化能够使进水溶液中腐殖酸发生矿化，使进水DOC显著降低；高藻水经预氧化后，溶液中大分子有机物转化为小分子有机物，使滤液DOC显著升高。扫描电镜结果表明，预氧化后腐殖酸在膜表面形成的滤饼层显著减少；尽管预氧化后藻细胞发生破裂，但破裂的藻细胞仍然会在膜表面形成滤饼层。模型拟合结果表明，紫外/氯氧化预处理能够缓解腐殖酸引起的膜污染主要是由于滤饼层过滤与膜孔堵塞的控制；而高藻水经紫外/氯氧化预处理后，滤饼层过滤与膜孔堵塞仍然占主导地位。     

不锈钢着黑色工艺

不锈钢去除表面自然形成的保护膜后,采用铬酸－硫酸溶液进行化学处理,随着工艺条件（温度、时间）变化,膜层厚度发生相应变化,膜层会出现不同的颜色,黑色就是其中的一种。般采用着色电位控制颜色,但此法受条件、设备等限制。很难得到推广应用。使溶液保持特定温度,控制着色时间,并不断观察膜层色泽,是不锈钢着黑色一种简捷有效的途径。     

PAC-UF处理含蛋白质类溶液过程中的膜污染研究

使用粉末活性炭(PAC)-超滤(UF)组合工艺处理牛血清白蛋白(BSA)溶液,研究了不同PAC投量下组合工艺的膜污染情况。结果表明,PAC本身对膜污染无明显贡献,PAC吸附BSA后使得膜通量急剧下降;PAC对改善BSA溶液通过超滤的膜通量和膜污染阻力均有一个最佳投加值;PAC在膜表面形成的滤饼层对膜通量改善作用明显;PAC滤饼层主要形成可逆污染,而水中BSA则是不可逆膜污染。     

预涂PAC动态膜生物反应器长期运行特性研究

利用普通工业滤布和粉末活性炭（PAC）形成预涂动态膜（PDM），然后置入活性污泥池中形成预涂动态膜生物反应器（PDMBR）。通过两个周期共计153d的试验表明，PDMBR长期运行稳定性良好。试验过程中膜通量为13.7 L·m^2·h^1，HRT为10h，平均COD去除率为90.70％，平均NI4^＋ -N去除率为96.15％。SEM分析知，造成膜污染的最主要的因素是PDM膜面的生物凝胶层，其次是压实的PDM膜层，而膜基质（滤布）由于受到外层PDM和生物凝胶层的双重保护而几乎没有受到污染，阻力变化不大，因而膜清洗容易。     

膜生物反应器中膜污染滤饼层渗透模型的研究

在膜生物反应器中,活性污泥在膜表面沉积形成滤饼层是造成膜污染的主要因素.采用扫描电镜和自动图像分析技术研究了膜过滤滤饼层的微观结构,并以分形理论和Darcy定律为基础,推导出用于预测膜污染滤饼层渗透性能的数学模型,验证了该渗透模型的可靠性,并利用该渗透模型进行了膜生物反应器中污泥浓度对滤饼层渗透性能影响的研究.结果表明,活性污泥颗粒沉积形成的柔性滤饼层具有明显的多孔结构,并且具有很好的分形特征;膜污染滤饼层的分形维数能够真实反映滤饼层的孔隙率大小;用该渗透模型得到的渗透系数K'可以作为有效预测滤饼层渗透性能的参数;滤饼层渗透系数K'随污泥浓度的增大而递减,污泥浓度低于10000 mg·L-1时,K'的变化趋势较小,污泥浓度达到10000 mg·L-1以上时,K'急剧下降.     

低膜面流速下曝气对管式膜水力特性及膜污染影响

采用曝气强化管式膜超滤高岭土混合液,考察了低膜面流速下曝气对强化膜分离过程影响,探讨了曝气对膜面水力特征及膜污染过程影响,并对过滤介质影响及膜污染阻力构成进行了研究。结果表明,在低膜面流速下,通过向管式膜引入曝气使膜表面形成气液两相流,可实现膜通量稳定保持在15L/(m²·h)以上。不仅如此,曝气的引入使膜表面雷诺数由1800~2500增至3300~4500,显著增强了膜表面湍流程度,并且实现了低膜面流速下使膜污染指数控制在较低水平,节省了运行能耗。此外,曝气的引入主要减轻了膜表面滤饼污染,使膜过滤总阻力减小且对高岭土截留效率影响不大,但强烈的膜面传质使高岭土粒径有减小趋势,并且膜表面形成污染阻力以不可逆污染层为主,不利于膜污染长周期控制。     

The effects of performance and cleaning cycles of new tubular ceramic microfiltration membrane fouled with a model yeast suspension

The efficiency of crossflow microfiltration processes is limited by membrane fouling and concentration polarization leading to permeate flux decline during operation. The experiments that were carried out in the laboratory were conducted to determine and investigate the performance, behaviour and the fouling susceptibility of new ceramic tubular microfiltration membranes during the crossflow filtration of yeast suspensions. The tubular membranes of nominal pore size 0.5 microns were fouled over a varied range of concentration, temperatures, pH, crossflow velocities and system pressures. The typical filtration conditions were at a temperature of 25°C, a system pressure of 1.5 bar and a concentration of 0.03 g/L yeast suspension. These parameters varied during subsequent investigations. After each experiment, the membrane and the rig were cleaned using a three stage cleaning process and was reused in order to replicate industrial filtration conditions. The effects of repeated fouling and cleaning cycles upon membrane flux over time and cleaning efficiency are investigated and their influence over time is also documented. For every experiment, the flux data was recorded over a 50 min period and the membrane was changed after the PWF declined considerably due to excessive fouling over time. Chemical cleaning consisted of a sequential application of a 1% caustic solution through the rig followed by a 2% hypochlorite solution and a 2% nitric solution, all at 50°C. The permeate flux was shown to decrease with filtration time during the development of the fouling layer. Once the fouling layer was developed and established, there appeared to be a leveling of permeate flux. The experimental results are presented in the report and the flux values at different conditions are presented.     

Synthesis and characterization of low-cost ceramic membranes from fly ash and kaolin for humic acid separation

Ceramic microfiltration membranes were prepared using five different compositions formulated with different amounts of fly ash and kaolin and sintered at 900 °C. The SEM analysis evidenced a large number of small pores on the surface of kaolin-rich membranes. The M4 membrane prepared using 25% fly ash and 50% kaolin was found to be optimum as it had a good combination of pore size (0.885 μm), porosity (42.7%), mechanical strength (43.6MPa), and chemical stability (<3% weight loss in acid and 0.02% in base), and this membrane was successfully applied in separation of humic acid from water. The permeate flux data fitted very closely with cake-filtration model, indicating the formation of a cake layer on membrane surface. Membrane fouling was found to be reversible and easily negated by cleaning and backflushing. The regenerated membrane showed better rejection of humic acid than fresh membrane with a flux recovery of above 80%.     

Flux Enhancement in a Helical Microfiltration Module with Gas Injection

Abstract

A helical flow module with an inner rod mounted membrane was designed and built to reduce gel layer deposit and membrane fouling during microfiltration. Controlled centrifugal instabilities resulting from flow in a helically grooved channel, as well as the leakage flow between adjacent grooves, generated secondary vortex flows. The permeation fluxes for helical modules with Dean vortex flow were compared with flat crossflow modules at different operating pressures, concentrations, and feed flow rates. The permeation flux of the helical module for a feed solution containing 0.3 wt% kaolin solution at 1.2 bar was 57% higher than that of the flat module. Moreover, in addition to secondary vortex flow, compressed air was introduced to the membrane module. The increase in flux for the helical module with air injection was significant: the flux enhancements at 1.3 bar, 2 L‐solution/min and 1.3 L‐air/min for 0.1 wt% solutions of kaolin and bentonite were 47 and 73%, respectively.     

MODELING OF MEMBRANE FOULING AND FLUX DECLINE IN MICROFILTRATION OF OILY WASTEWATER USING CERAMIC MEMBRANES

One of the major problems in pressure-driven membrane processes is reduction of flux far below the theoretical capacity of the membrane. The results of an experimental study of fouling mechanisms of ceramic membranes in separation of oil from synthesized oily wastewaters are presented. Mullite microfiltration (MF) membranes were synthesized from kaolin clay as MF ceramic membranes. The rejection of total organic carbon (TOC) for the synthetic feeds was found to be more than 94% by these membranes. Hermia's models were used to investigate the fouling mechanisms of membranes. The effect of pressure, cross flow velocity (CFV), temperature, oil concentration, and salt concentration on flux decline were investigated. The results showed that the cake filtration model can well predict the flux decline of mullite ceramic membranes; average error of this model is less than 7%. The results show that by increasing pressure from 0.5 to 4 bar, porosity of the cake layer on the mullite membranes decreases from 25.68% to 14.98%. After the cake filtration model, the intermediate pore blocking model was found to well predict the experimental data with an average error less than 10.5%.     

基于小孔径PVDF内衬膜A/O-MBR膜污染分析

为研究膜生物反应器膜污染问题,在缺氧-好氧一体式膜生物反应器中,对模拟生活污水的处理效果进行考察,考察了4种不同孔径聚偏氟乙烯(PVDF)内衬膜对膜污染的影响规律。结果表明,基于小孔径PVDF膜的MBR工艺对模拟生活污水中COD、NH3-N的去除率分别达到96%、90%,出水浊度在0.35 NTU以下;在操作条件固定的情况下,随着膜孔径增加,临界通量逐渐降低;孔径越小,跨膜压差上升越缓慢,膜污染较轻。膜污染模型分析结果显示:无论何种孔径的膜,在MBR系统中,滤饼层和凝胶层污染所占的比例都最大(>80%),因此,在使用过程中减缓滤饼层及凝胶层的形成至关重要。     

About the mechanism of formation of iron hydroxide fouling layers on reverse osmosis membranes

The rate of fouling by deposition of iron hydroxide on tubular reverse osmosis membranes was studied under closely controlled experimental conditions resembling those of brackish water desalination. Feed brines of approximately 5,000 ppm NaCl were used to which small amounts of iron hydroxide were added. The effect of brine flow rate, product water flux, pH and oil contamination on the rate of fouling layer buildup was investigated. Results obtained suggested that the rate of membrane fouling due to iron hydroxide buildup is governed by two processes: nucleation and growth.