Granular media filtration is used almost universally as the last particle removal process in conventional water treatment plants, but every particle cannot be removed. Laboratory-scale filtration experiments were performed at a filtration velocity of 5 m/h using spherical glass beads with mean diameter of 0.55 mm as collectors; the 10 cm depth ensured particles would be in the effluent. Particle suspensions (Min-U-Sil 5) with 1.7 μm mean particle size were filtered at pH values of 3.0, 4.0, and 5.0, all above the isoelectric point (near pH 2.0). Zeta potential distribution (ZPD) and particle size distribution (PSD) of influent and effluent particles were measured. More favorable particles, i.e. particles with smaller surface charge, were well attached to the collectors during the early stage of filtration. This selective attachment of the lower charged particles caused the ZPD of the effluent to move to a more negative range. However, the ZPD did not keep moving to a more negative range during later stages of filtration, and this result was thought to be caused by two reasons: ripening effects and detachment of flocs. In the early stages of filtration, particles were captured better at pH 3.0 than at the higher pH values (as expected), but this trend reversed at a cumulative hydraulic loading (CHL) of only 5–10 m3/m2. Particle breakoff apparently occurs as flocs of particles that are initially captured individually, as indicated by the trends for different particle sizes. At pH 3.0, particles in the 3–5 μm range had worse removal after a CHL of 5 m3/m2, whereas removal improved through the entire CHL of 20 m3/m2 for all size particles at pH 4.0. Chemical parameters such as zeta potential can be important during the initial stage of filtration, but their importance can decrease over time depending on the specific chemical conditions. The influent particle size distribution and the removal of certain size particles during the initial stage can significantly influence ripening which, in turn, can influence the overall particle removal efficiency. Better initial particle removal does not necessarily mean better overall particle removal.
A Packing Computational Method Relating Fractal Particle Size Distribution and Void Fraction in Granular Media
