Blocking Law Analysis

The effect of stirring is large. This means that the blocking law analysis for these conditions is invalid. Initial fluxes were about 500 Lm -h and fluxes after the first cycle were 346 and 152 Lm % for sdrred and unstirred conditions, respectively (this corresponds to flux declines of 40 and 70%). 

Neither the cake filtration nor the complete blocking law show a linear relationship when considering that the analysis is invalid for stirred filtration. 

This means that the blocking law analysis was successful in the distinction between pore and surface fouling. Pore fouling is the dominant mechanism for the 100 kDa membrane as was su ested considering the pore size and the estimated size of organics after aggregation (see Chapters 4 and 7 for organic solubility). 

---

It may be possible to do a membrane autopsy to identify the foulant(s) and fouling mechanism. For microporous membranes the blocking law analysis [1], which uses permeate volume (V) vs. time (t) data, can supplement the observations. The generalized relationship at constant pressure and in dead-end filtration mode gives,... 

Results for the blocking law analysis are shown in the following sections for stable colloids in the absence of organics, and for both SPO and OPS cases. 

In this section, the blocking law analysis for SPO systems is shown as a function of pH (Figure 5.14 and Figure 5.15) and primary colloid size (Figure 5.16 and Figure 5.17). 

The blocking law analysis showed that particles do accumulate, block pores, and adsorb at the membrane surface, although cake formation is not always visible. Also, it is unknown how the model reacts to cake formation inside pore entrances. This is a possible scenario and may have a similar effect as the cake formed on the surface. A more quantitative analysis of the blocking laws of the systems used appears useful. 

Filtration of colloids at pH extremes served as a baseline of colloids, which are neither aggregated nor stabilised by an adsorbed organic layer. As expected from the literature review, the particle size closest to the membrane pore size (250 nm) caused largest flux decline. Rejection occurred in this case due to a combination of hydrophobic, specific, and van der Waals forces, and could not be explained by charge interaction only. Blocking law analysis showed evidence of pore blocking, cake formation, adsorption and pore closure at some stage of the filtration process. 

Blocking law analysis showed that cake formation is absent in the OPS case due to the very low rejection and deposition. 

First of all, pore blocking could cause more severe flux decline of the 100 kDa membrane. This is addressed in the blocking law analysis in the following section. 

For selected experiments blocking law analysis is carried out as for the calcium-organic systems in section 6.6.4. The mechanisms for floes formed at the two different dosages of ferric chloride were compared for the 100 kDa membrane. The results are shown in Figure 6.53 and Figure 6.54 for dosages of 25 and 100 mgL ferric chloride, respectively. 

Figure 6.53 Blocking law analysis for the 100 kDa membrane (25 mglj FeCl , 5 mgL as DOC FA, S=stirring at 270 rpm, NS-unstirred), (A) complete blocking and cake filtration and (B) intermediate blocking and standard blocking.

Blocking law analysis in UF indicated internal pore adsorption to be the principal mechanism by which natural organic aggregates were retained and resulted in a pore size reduction. When ferric chloride is added cake filtration becomes important. 

---