Pure Water Flux

For the 100 kDa membrane, 450 mL of feed solution were introduced into the reserv oit. Pressure was adjusted to 100 kPa and the filtradon cell was fiUed up. The permeate was sampled once (sample volume 50 mL) and then recycled into the reservoir together with the retentate and filtration was repeated. This was repeated a third rime. This recycling experiment enabled the separation of concentration polarisation effects from fouling effects. The 110 mL of retentate was then also sampled. Pure water flux was measured again after each experiment with approximately 300 mL and 1000 mL of MilliQ filtered for the lOkDa and lOOkDa membranes, respectively. Membrane samples were kept in a Petri dish for deposit analysis. 

The amount M of solute or colloid deposited (in m on the membranes was calculated by using mass balance (see equation (6.1)), where Vf, Vr, and Vpi, are the volumes of feed, retentate and permeate (sample i), respectively, and cf, cr, and cpi the concentrations of feed, retentate and permeate. 

This can also be described as percent of mass in feed solution deposited (M ) 

The results for DOC should be treated with care, as the error is expected to be large, especially at low concentrations. This is due to the analytical method and risk of contamination of the samples. 

The regenerated cellulose UF membranes used were described in Chapter 4. These membranes were selected due to tbeir hydrophilicity, which should show reduced adsorption of organic compounds (Jucker and Clark (1994)). 

---

Water flux is sometimes normalized according to the initial or pure water flux,/ as or as flux drop, defined by... 

For very small AP, flux is linear with pressure. Figure 7 shows a graph of flux versus pressure. Curve A is the pure water flux from equation 1, curve B is the theoretical permeate flux (TPE) for a typical process. As the gel layer forms, the flux deviates from the TPF following equation 7 and curve D results. Changing the hydrodynamic conditions changes K and results in a different operating curve, curve C. 

Equation (20-80) requires a mass transfer coefficient k to calculate Cu, and a relation between protein concentration and osmotic pressure. Pure water flux obtained from a plot of flux versus pressure is used to calculate membrane resistance (t ically small). The LMH/psi slope is referred to as the NWP (normal water permeability). The membrane plus fouling resistances are determined after removing the reversible polarization layer through a buffer flush. To illustrate the components of the osmotic flux model. Fig. 20-63 shows flux versus TMP curves corresponding to just the membrane in buffer (Rfouimg = 0, = 0),... 

A typical plot illustrating the slow decrease in flux that can result from consolidation of the secondary layer is shown in Figure 6.5 [14], The pure water flux of these membranes is approximately 50 gal/min but, on contact with an electrocoat paint solution containing 10-20% latex, the flux immediately falls to about 10-12 gal/min. This first drop in flux is due to the formation of the gel layer of latex particles on the membrane surface, as shown in Figure 6.4. Thereafter, the flux declines steadily over a 2-week period. This second drop in flux is caused by slow densification of the gel layer under the pressure of the... 

The effect of the gel layer on the flux through an ultrafiltration membrane at different feed pressures is illustrated in Figure 6.7. At a very low pressure p, the flux Jv is low, so the effect of concentration polarization is small, and a gel layer does not form on the membrane surface. The flux is close to the pure water flux of the membrane at the same pressure. As the applied pressure is increased to pressure p2, the higher flux causes increased concentration polarization, and the concentration of retained material at the membrane surface increases. If the pressure is increased further to p3, concentration polarization becomes enough for the retained solutes at the membrane surface to reach the gel concentration cgel and form the secondary barrier layer. This is the limiting flux for the membrane. Further increases in pressure only increase the thickness of the gel layer, not the flux. 

Flush the system thoroughly with water to remove all traces of detergent measure the pure water flux through the membrane modules under standard test conditions. Even after cleaning, some degree of permanent flux loss over time is expected. If the restoration of flux is less than expected, repeat steps 1-3. 

Generally, the pure-water flux through a membrane layer, uw is directly proportional to the applied hydrostatic pressure difference (transmembrane pressure, AP) according to Darcy s law as follows ... 

Fig. 6. Pure water flux (y) as a function of the transmembrane pressure (x) (AP) for NF membranes (NF270, NF90) and low-pressure RO membrane (BW30) at T — 25 °C, R being the linear regression coefficient.

During reverse osmosis and ultrafiltration membrane concentration, polarization and fouling are the phenomena responsible for limiting the permeate flux during a cyclic operation (i.e., permeation followed by cleaning). That is, membrane lifetimes and permeate (i.e., pure water) fluxes are primarily affected by the phenomena of concentration polarization (i.e., solute build up) and fouling (e.g., microbial adhesion, gel layer formation, and solute adhesion) at the membrane surface [11]. 

In the pure water flux (PWF) of Membranes A to D that these membranes have different ranges of pore size. The PWF of Membranes, A, B, C, and D were 0.7, 2.4, 2.8, and 0.03 [m day atm ], respectively. These data indicate that the PWF of the membranes developed in this work were much better than those of carbon whisker membranes repeated in the previous works. The resultant PWF over 0.5 [m day atm ] at the pressure of 50 [kPa] is large enough for practical applications, since commercial membranes have PWF typically of 2 [m day atm" ]. 

Ilias and Govind [1993] also used the CFD approach to solve coupled transport equations of momentum and species describing the dynamics of a tubular ultraflltration or reverse osmosis unit. An implicit finite-difference method was used as the solution scheme. Local variations of solute concentration, u ansmembranc flux and axial pressure drop can be obtained from the simulation which, when compared to published experimental data, shows that the common practice of using a constant membrane permeability (usually obtained from the data of pure water flux) can grossly overestimate... 

The trends in Figures 2 to 6 are consistent with the qualitative features of solute preferential sorption discussed earlier in this paper. The permeate flux is lower than the pure water flux due to pore blocking. This effect is enhanced by either decreasing the pore size or increasing the feed concentration. 

The solvent flux (J) is typically described by the equation J = AP/Rj. The total resistance to flow (Rj) is expressed as the sum of two resistances, R i + Rc, where Rm is the resistance due to the membrane and R is the cake-layer resistance. The resistance Rra can be determined by measuring the pure-water flux on an imfouled membrane, one limiting case corresponding to maximum solvent flux. This case is independent of feed flowrate. As Rc increases, the flux becomes independent of AP. This is illustrated in the Figure 9.8. 

Once the gel-layer is formed, it is often the limiting resistance to flow. Figure 3.32 shows two membranes with widely different membrane resistances (Rm). The pure water flux differs by a factor of 3.75 yet in the presence of protein (retained by both), the water flux differs by a factor of only 1.11, for pressures over the threshold pressure of 20 psi. It will be noted that the higher hydraulic permeability of PM 30 membrane results in a much lower threshold pressure (7 psi). 

The most critical parameter in the characterisation of membranes is their flux. For the characterisation of clean membranes flux is measured with MilliQ water as pure water flux . The definition of the instantaneous flux is given in equation (3.2), where V is the filtrate volume, t the filtration time, and A the membrane surface area. 

Pure water flux under laminar conditions through a tortuous porous harrier may be described, according to Carman (1938) and Bowen and Jenner (1995), by equation (3.4). 

Process Supplier Type Typical Operating Pressure [bar] Specifications Pore Size [pm] Molecular Weight Cut-Off [kDa] Pure Water Flux [Lm%- ] Surface Charge at pH 8 [mV]... 

This characterisation is rektivel) vague, as different methods are used by each manufacturer (Readman (1991), Thorsen et al. (1997)). As a more comparable parameter, the pure water fluxes as determined in the experiments are also given, as well as the membrane zeta potential at pH 8. A new membrane was used for each experiment (except for fractionation experiments). 

The results of surface charge measurements of the membranes as a function of pH, pure water fluxes and electronmicrographs are shown in the MF, UF, and NF chapters, respectively. 

Pure water flux was determined for each membrane using 3 L MilliQ water. The last 100 mL of filtrate were analysed for TOC as a control sample for organic contamination. The cell was then filled with the feed solution, the stirring switched to 270 rpm, and the pressure adjusted to 100 kPa (unless indicated otherwise). Two types of filtration protocols were used, a standard and a regck protocol. Pure water flux was determined after the recycle experiments only, using 1 L of MilliQ water. 

Flux decline is therefore the decline in pure water flux of the membranes before and after the experiments. The majority of experiments performed were recycle experiments. The apparent rejection was calculated using the following equation. Cpiis the concentration of permeate sample i taken directly from the membrane and not from the permeate container. It is thus the permeate concentration averaged over the filtration interval required to collect the sample. The concentration in the batch cell was measured at the end of each experiment. 

Table 5.1 Pure water flux of clean MF membranes at a transmembrane pressure of 100 kPa.

Figure 5.4 Flux ratio (flux after collection of 800 mL permeate over pure water flux) as a function of primary colloid sit e for stable colloids (pH3), aggregates in the absence of organics (pH8), stabilised colloids (OPS), and aggregates (SPO).

For fractionation experiments, the perspex stirred cells (see Chapter 4 for equipment description) were operated directly from the nitrogen bottle without a reservoir. Membranes were floated in a beaker of MilliQ water, skin side down, for at least one hour to remove the glycerin coating. Then at least 300mL of MilliQ water were filtered through the membrane. The filtrate was analysed with UV and DOC to confirm full removal of glycerin. The membranes were reused up to 5 times and stored in 0.1 % sodium azide at 4 C. Pure water flux was measured after the filtration of 500 mM of MiUiQ water prior to each experiment. The filtration protocols for serial and parallel fractionation were described in Chapter 4. In this Chapter parallel fractionation results will be shown. 

All experiments were stirred at 270 rpm unless otherwise indicated. A feed reservoir of 1.5 L was connected to the stirred cell to provide extended filtration volume. Pure water flux was measured after the filtration of 1 L of MilliQ water for both membranes. 

---