Cross-flow microfiltration is a pressure-driven membrane process for
separating dispersed matters in the range of sizes from 0.05 to 10
mm from liquids. The principle of solid/liquid separation can be described as a pressing of pure liquid
through the membrane pores and through the deposit of solid particles on the
membrane wall. In the absence of membrane fouling, the increasing thickness of the
deposit of particles (cake formation) causes the initial rapid decrease of flux and limits
the wider application of microfiltration. In spite of this, the separation process
finds increasing use in water purification, wastewater treatment, brewing
and biotechnology industries, amongst others.
The main factor limiting the application of cross-flow microfiltration and
other pressure-driven membrane processes is flux decline due to membrane fouling
and concentration polarization (Schafer et
al., 2000). The steady-state value of permeate flux depends, besides hydrodynamic conditions (cross-flow velocity, the
Reynolds number, shear stress and shear rate at the membrane surface) in the
membrane system (Broussous et al., 2000) on physicochemical particle-particle and
particle-membrane interactions (Huisman et
al., 1999). One way of characterizing these
particle-particle and particle-membrane interactions is the knowledge of
the z-potential of both particle and membrane. This electrostatic characterization
of membranes is a useful way to predict and interpret the performance
of microfiltration process. The magnitude of the z-potential gives the information
of the stability of the system. Near the isoelectric pointIEP (the value of pH
where the charge and therefore z-potential of particles is equal to zero) the system
is unstable and the particles tend to flocculate. Therefore, the stability of the
particles and the particle-membrane system could affect the separation process. Many
studies showed that permeate flux, J, can be easily changed by pH (kind of added
salt), and salt concentration of the microfiltration dispersion (Nazzal and Wiesner,
1994; Mullet et al., 1997; Moritz et
al., 2001; and Martín et al., 2003). |