An ever-present problem associated with nanofiltration membranes for water treatment is fouling. Fouling of membrane elements often causes a significant increase in hydraulic resistance and applied pressure drop, which increases operating cost and decreases life of the membrane. This project focused on the (1) analysis of fouled sulfonated polysulfone membranes and (2) determining which foulants could not be removed by cleaning, and thus form irreversible fouling layers.

In Task 1 distilled water containing 10 mM Na⁺ was filtered through sulfonated polysulfone membranes at 70 psi for 2 hours for initial pore precompaction and for flux stabilization. The membranes were fouled with a solution of 2 mg/L of bovine serum albumin (protein), and 1 mM calcium in distilled water. Experiments were run for 1 minute (instantaneous fouling), 5 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 6 hours, 8 hours (permeate flux is equal to 50% of the initial permeate flux (i.e. J/Jo = 0.5). Time intervals were tested to model the evolution of hydraulic and cake resistances as a function of time. After each experiment, membrane autopsies characterized the fouling layer (i.e. the involuntary/adverse layer) with respect to chemical structure and morphology by the use of attenuated total reflectance fourier transform infrared spectroscopy (ATR-FTIR), atomic force microscopy (AFM), scanning electron microscopy (SEM), total organic carbon analysis (TOC), and conductivity equipment.

The protein solution (2 mg/L of bovine serum albumin, and 1 mM calcium in distilled water) showed a lower flux than the pore precompaction solution (distilled water containing 10 mM Na⁺) initially (Figure 1). Furthermore, the protein solution's flux slowly declined for 6 hours when it started to stabilize. The AFM analysis showed an increase in surface roughness across the membrane as fouling time increased (Figures 2 and 3 show the instantaneous fouling and the fouling on the membrane after 8 hours). The ATR-FTIR analysis showed changes in the bond structure of the membranes, furthermore, peaks around 1200, 1400 and 1700cm^-1 were observed to have changed (Figure 4). The conductivity analysis shows that over time (e.g. as fouling time increases) the solution conductivity decreases and that the rejection percentage increases (Figure 5).

The information in Task 1 only provides half of the picture. The other half is related to the irreversibility of the fouling layer. Task 2 aims at relating foulant characteristics and cake resistance to the reversibility of the cake layer. In Task 2, the experimental runs in Task 1 will be repeated and following each run, the membranes will be hydraulically cleaned. Membrane resilience will be determined by backflushing the membranes with distilled water for 30 minutes at 70 psi. After backflushing the membranes an instantaneous flux will be measured (recovered clean water flux). Relative to Task 1, experiments will be performed as a function of different time intervals in order to relate fouling layer chemical structure and morphology to the effectiveness of the physical (e.g. backwashing) method. By performing identical experimental runs as in Task 1 the amount of foulant detachment will be determined. Task 2 will be completed by August 2005.