Membrane processes may offer small facilities a less expensive alternative for the removal of particles and organic materials. Costs of several ultrafiltration and nanofiltration processes are compared with the cost of conventional liquid‐solid separation with and without GAC adsorption for small facilities. Data on raw‐water quality, permeate flux, recovery, frequency of backflushing, and chemical dosage obtained from a pilot study were used with a previously developed model for membrane costs to calculate anticipated capital and operating costs for each instance. Data from the US Environmental Protection Agency were used to estimate conventional treatment costs. All of the membrane process calculations showed comparable or lower total costs per unit volume treated compared with conventional treatment for small facilities (<200,000 m3/d or about 5 mgd).

No abstract

No abstract

No abstract

The main objective of this study was to evaluate microbial passage through membrane elements with various levels of compromised integrity while installed within full-scale UF/MF systems. Challenge tests were performed for this purpose with the microbial contaminant surrogates Bacillus subtilis spores at two drinking water treatment plants. The experimental units used were one of several parallel racks part of each plant in which one of the elements was installed with various levels of compromised integrity achieved by precutting several hollow fibers and inserting removable pin-plugs atthe corresponding ends of the broken fiber segments. The UF rack was operated with 38 elements online (the rack has 50 elements) and was designed for an inside-out operation, and the MF rack included 50 elements and was operated in an outside-in mode (with a permeate outlet at one fiber end only). Spore removals observed for both the UF and MF racks with all precut fibers plugged were equal to or greater than 99.9992%, and as expected the removal efficiency deteriorated with an increasing number of unplugged fibers. Predictions made with a model based on the use of the Hagen-Poisseuille equation for laminar flow and the Darcy-Weisbach expression for turbulent flow inside broken fibers were found to provide an adequate conservative representation of experimental results. Additional simulations performed with the verified model revealed the occurrence of a greater microbial passage for an inside-out configuration compared to an outside-in mode. A lower microbial passage was predicted for the outside-in element configuration with one permeate outlet as compared to an element with permeate outlets at both fiber ends. The model offers a useful tool that together with other considerations such as membrane fouling, cleaning, and durability would assist in the selection of low-pressure membrane element configuration.

No abstract

No abstract

No abstract