Reduction layer-cake

These systems are designed for little or no cake formation at lower levels of concentration. It can be shown that filtration rates increase with a reduction in cake thickness. However, some materials (especially gels such as aluminum hydroxides) are so compressible that 90% of the available pressure drop is absorbed by a "skin layer" formed on top of the filter media, wliile the remainder of the cake remains soupy and unconsolidated. Consequently, reducing die cake thickness (in examines such as unwashed "gels") in eijuiianent that uses tediniques of "thin cakes" would not result in any significant improvements. It is, therefore, advantageous to minimize the formation of a "skin layer". 

Mlcrofiltra.tlon, Various membrane filters have been used to remove viral agents from fluids. In some cases, membranes which have pores larger than the viral particle can be used if the filtration is conducted under conditions which allow for the adsorption of the viral particle to the membrane matrix. These are typically single-pass systems having pore sizes of 0.10—0.22 lm. Under situations which allow optimum adsorption, between 10—10 particles of poHovims (28—30 nm) were removed (34—36). The formation of a cake layer enhanced removal (35). The titer reduction when using 0.10—0.22 p.m membrane filters declined under conditions which minimized adsorption. By removal standards, these filters remove vimses at a rate on the low end of the desired titer reduction and the removal efficiency varies with differences in fluid chemistry and surface chemistry of viral agents (26). 

The evolution of the dimensionless density profile across the soot layer is shown in Fig. 23. The initial gradual replenishment of the soot in the catalytic layer (at t = 140 s) is followed by sudden penetration events (t — 262 and 326 s) before the establishment of a steady state profile (at =531 and 778 s). Regarding the non-catalytic (thermal) layer only a gradual reduction of its thickness, accompanied by a very small reduction of its uniform density is observed. This simple microstructural model exhibits a rich dynamic behavior, however we have also established an experimental program to study the soot cake microstructure under reactive conditions. 

External fouling is caused by the formation of a cake layer of cells or other materials on the membrane surface, leading to a reduction in permeate flux (defined as the volume of permeate produced per time and membrane area). Internal fouling is caused mainly by proteins and particles smaller than membrane pores. Proteins and protein aggregates can adsorb or deposit at the pore entrance or inside the pores and cause pore blockage or narrowing, leading to increased hydraulic resistance (2). 

The procedure as given is generally applicable for the reduction of esters to alcohols in excellent yields. When preparing the solid normal saturated alcohols, the procedure may be modified, if desired, to permit the recovery of the acid from the unreduced ester. After the alkali is removed the alcohol layer is washed with two successive portions of 20 per cent salt solution which are discarded. Neither the strong alkali nor the salt solutions remove an appreciable amount of organic acid. A solution of 50 g. of calcium chloride in 150 cc. of water is added to the butyl alcohol solution, the mixture is steam-distilled until the butyl alcohol is removed, and the flask and contents are allowed to cool. A hole is made in the cake of solid alcohol and the water layer removed. Two liters of toluene is added and the flask warmed and shaken until the alcohol dissolves and only fine crystals of the calcium salt of the unreduced acid remain. The solution is cooled to 350 and filtered with suction. The calcium soap is removed from the filter, warmed with about 500 cc. of toluene, cooled, filtered, and washed with a little more toluene. The combined toluene solutions may be concentrated and the alcohol crystallized, or the toluene may be completely distilled and the residue vacuum distilled. The insoluble calcium... 

Aimar et al. [19] noted that in the UF of complex liquids, such as cheese whey, which contains proteins, salts and casein fragments, concentration polarization, and adsorption and cake formation play a role in flux behavior during crossflow filtration. They may induce osmotic pressure in the retentate side since the chemical potential of the solute-rich polarized layer is lower than that of the permeate, and therefore at equilibrium, a positive osmotic pressure develops in the retentate to equal that of the permeate. The smaller the solute, the greater is its contribution to the osmotic pressure of the liquid, so that in milk, lactose and the minerals have the biggest contribution to osmotic pressure. In skim milk or whey, the osmotic pressure is around 7 bar (700 kPa) and this must be exceeded in RO to commence permeation in UF, only the proteins contribute to the osmotic pressure, which increases exponentially with protein concentration [56]. In any case, a TMP greater than the osmotic pressure is required for solvent to flow from the retentate side to the permeate side. This leads to the reduction in the effectiveness of applied TMP as driving force to permeation. 

While size reduction or blockage of pores may be considered to increase the resistance of the membrane (Rm) to permeate flux, accumulation of materials in the cake and concentration-polarization layers (so-called polarised solids) presents additional resistances to permeation (denoted here as R c and Rep respectively). These resistances var) as a function of the composition and thickness of each layer, which in turn are determined by the feed water quality and the characteristics of mass transfer in the membrane module. In most instances encountered in water and wastewater treatment, it appears that the concentration-polarisation layer, if it is formed, contributes negligible resistance to permeate flux i.e. Rep << Rc and, therefore. Rep may be neglected (Mallevialle et al (1996)). While this is in reality not always the case (as shown in later sections of this chapter and in Chapter 7) for the filtration of... 

A number of key issues are highlighted through direct comparison of the three processes. First, a cutoff for natural organics rejection at a pore diameter of clean membranes of about 6 to 8 nm is observed. Below this pore diameter rejection is <20%, while above it is mostly >80%. All twelve membranes lie on this cut-off curve. When the membranes are fouled this cut-off is shifted to >20 nm and the cut-off itself is not as distinct. This can be attributed to a pore size reduction due to internal pore adsorption, the formation of self-rejecting cake layers, and solute-solute interactions in the boundary lay er. If ferric chloride is added this cut-off ceases to exist. With chemical addition, rejection... 

The MFI is based upon cake filtration theory that particles are retained on the membrane surface during filtration. According to the resistance in series model, the reduction in flux due to the presence of cake layer and the additional resistance from the membrane under constant operating filtration can be described as ... 

Because of the high flux reduction immediately after the start of the experiment with 1% feed solution, it was speculated that there was instant pore blocking and cake formation on the membrane surface for all the membranes used, including SPPO coated membranes. Once a cake layer is... 

Pure water permeation of ultrafiltration membranes made of polyethersulfone was substantially reduced when a thin layer of SPPO material was coated on their surface. A very sharp flux reduction was observed immediately after the start of the experiment for most of the membranes including SPPO coated membranes. However, the flux reduction was lower for the SPPO coated membranes. The cake layer resistance was also low with SPPO coated membrane. The cake layer resistance was higher when the feed solution contained both clay and SBR compared to when only clay was present in the feed solution. 

---