Recovery rate of flux membrane operation

In this post I want to know about the flux recovery rate. From the book Basic Principles of Membrane Technology I found out the explanation about recovery rate. It define the ratio between qp and qr. Qp is permeate flow rate while Qr is feed stream flow rate. Before I move on, please see this doc, Recovery Diagram. Simple diagram process. Q is flowrate, f is feed, r is retentate and p is permeate.

So, recovery or r is defined by Qr/Qp. My research compute the value of Qr in m³/sec, while Qp in term of gram/sec. So, assume that the density of feed water equals to 1,000 kg/m³, then I can equalize the unit. Recovery is usually used to describe the membrane performance. In practical, membrane installation is usually set to produce recovery as high as possible. On the opposite on the lab in university, where generally recovery is very low to allow complete separation.

What I would like to find out is whether this has something to do with permeability or not. Permeability is simply a unit measurement indicating quantity of pure water passing through the membrane. It is known that the permeability of TiO₂ coating membrane decreased, meaning that denser pore size is formed after the coating. Qualitatively as has being said by Bideau et al (1995), the coating forms irregular shape, dimension and thickness, also develop vacant spot between TiO₂ particles. So higher possibility adsorption may occur.

The average recovery rate for humic acid membrane operation are as follows:

Naked membrane
50 kD : 1.14%
15 kD : 1.05%
1 kD : 0.64%

TiO₂-UV membrane
50 kD : 1.10%
15 kD : 0.85%
1 kD : 0.56%

Average recovery for coated membrane are lower than naked membrane. More fouling I guess, because of TiO₂ particles on membrane surface. The ratio of permeate and retentate for coated-UV membrane are lower than naked membrane, so should be lower permeate flux. And indeed the fact tells that flux for coated-UV membrane are lower than flux of naked membrane. So recovery is proportional with permeability and TiO₂-UV will only decrease the flux as the result of higher fouling.

Another question may arise, which one contribute fouling more: adsorption, inner pore blockage, TiO₂ or UV?

I can probably answer between TiO₂ or UV contributes more fouling but not between adsorption or internal pore blocking. This is because I did not do experimentation to investigate adsorption or internal pore blocking.