REJECTION EXPERIMENTS

Graphs for Equations 15, 16, and 20 are shown in Figure 2, and for Equation 17 in Figure 3. Horizontal lines in Figure 2 represent the asymptotic limit for C /C for an experiment with a given P6cl6t number. For example, for a rejection experiment of myoglobin from saline solution D = 0,172 x 10 cm/sec, r =... 

To summarize, the proposed boundary layer theory for rejection experiments across imperfect membranes involves analytic expressions for two regions of boundary layer development, corrected for R<1 by Equation 23, and corrected by the condition for switching. Equation 22. How integrations are performed to obtain an implicit relation between R, and R is shown later. Values calculated for R at various values for J are then used to calculate a and P for the fiber membrane, according to Equation 7. 

Working Equations for Rejection Experiments in Saline. Raw da a from the rejection experiments are (total filtrate rate, cm /sec), Qg (total inlet feed stream low rate, cm /sec), C (feed stream inlet concentration, g/cm ), and C (average fiitrate concentration, g/cm ). R, is calculated acco ding to Equation 8,... 

Kilduff and Weber (1992) determined a dependence on ionic strength for the rejecdon of random-coil polymers or natural humic molecules. Concentration polarisation also changed rejection. This influences the results obtained in rejection experiments and size determination methods such as fractionation. 

IC was used for chloride determination for NF rejection experiments. Anions could not be analysed using IC, as humic substances interfere with the analysis (Hoffmann et al. (1986)). A MilHpore Waters Model 590 instrument was used with a Model 430 Conductivity detector. The eluent used was 0.68 gL boric acid (HjBOj), 0.235 gL gluconic acid anhydride (C6HioO< ) and 0.3 gL lithium hydroxide (LiOH 6 H2O). 

A standard protocol for rejection experiments is shown in Table 7.2. A standard rejection experiment took 4 to 7 hours, depending on the membrane used. 120 mL of the permeate were collected from a feed volume of 185 mL, which, at 100% rejection, leads to a threefold concentration in the cell. 

Recycle experiments were designed to determine membrane fouling. Steps 1, 2, and 4 for the recycle experiments arc identical to rejection experiments, as shown in Table 7.2. In recycle experiments, step 3 involves the same amount of feed being added. No permeate samples are taken. After this cycle is completed, the permeate is returned to the stirred cell. This is repeated two times, so that the permeate has been filtered three times. At the end, one permeate sample is taken to calculate overall rejection and the concentrate is weighed and also sampled. 

Table 7.2 Standard protocolfor rejection experiments in the stirred cell.

The aim of the pore size measurement was not to produce an absolute value for pore size or molecular weight cut off, but rather to determine which of the four membranes is more open, A dextran 1000 standard was chosen and rejection experiments were carried out at a dextran concentration of about 50 mgL and pH 8. The feed DOC was 19.4 mgL k Dextran was chosen as it is not expected to interact strongly with membrane material (Combe et al. (1999)). 

Fulvic acid (FA) was used for rejection experiments, as FA is smaller than HA. Therefore, the rejection due to size of this compound should be lowest. While the NOM is still smaller rhan FA, the impurities in the NOM and the fact that its contents are largely uncharacterised make the use of FA more attractive. 

The results confirm that charge interactions are important for organics rejection. While the humics and hydrolysates are retained by size exclusion effects, the smaller compounds demonstrate pH effects. Rejection experiments in this section failed to show this effect due to the high proportion of organics larger than pore size in the samples. 

In conclusion for the rejection experiments, the TFC-SR membrane promises best performance in water treatment with a stable flux, and low sodium and high calcium rejection. The organic rejection of this membrane is the highest overall. 

The rejection experiments have shown that flux depends on the salt solution composition. Figure 7.17 shows results for the different components of the chosen background solution with FA for the TFC-S membrane which, due to its high salt rejection, should be most sensitive to the salt used. 

---