ABSTRACT
The nanofiltration (NF) drinking water production unit of the Mery-sur-Oise plant (Val d'Oise, France) consists of eight identical filtration trains composed of three stages positioned in steps for a production capacity of 140,000 m(3) day(-1). To gain a better understanding of the irreversible fouling of the NF membranes, spiral wound modules in operation for 8 years from each of the three stages of the plant were autopsied before and after chemical cleaning and analysis by Attenuated Total Reflection Fourier Transform Infrared spectroscopy, Inductive Coupled Plasma-Atomic Emission Spectrometry, contact angles, adenosine triphosphate (ATP) content measurements, and rheometry. The fouled membranes from the three stages had similar contact angles of approximately 60 degrees . Relative infrared signals typical of biofilms were classified in descending order from stage 1 to stage 3. The foulant matter of stages 1 and 2 contained similar but weaker ATP concentrations than stage 3. During rheometry experiments, rotation and oscillation analyses demonstrated that the biofilm of stage 3 was less viscous and less elastic than the biofilms of stages 1 and 2. After cleaning, all the parameters analyzed demonstrated a quantitative decrease in the fouling matter at the NF membrane surface, but a biofilm with intact viscoelastic properties (unchanged G' and G'' values) remained at the membrane surface for the three stages. The persistence of biofilm material with intact mechanical properties at the NF membrane surface after chemical cleaning may result in permanent permeability decreases.
Subject(s)
Biofilms/growth & development , Biofouling , Micropore Filters/microbiology , Nanotechnology/methods , Water Purification/methods , Adenosine Triphosphate/analysis , Biofilms/drug effects , Biofouling/prevention & control , Detergents/pharmacology , Nanotechnology/instrumentation , Rheology , Spectrophotometry, Atomic , Water Purification/instrumentationABSTRACT
Sources of organic pollution in the Oise river, France have been identified using a "mutual search" strategy. This consisted of searching the characterized discharges for compounds found in the river water and, conversely, searching river water extracts for compounds found in the discharges. The "matches" were facilitated by comparisons of retention indices on two capillary gas chromatographic columns of widely differing polarity, response factors of selective detectors, and mass spectra. These assignments were verified by the geographical correlation of the selected pollutant's point of appearance in the river with the location of the suspected source.
