AWWA ACE56280

As utilities consider nanofiltration (NF) technology, it is important to realize that membrane selection is a critical factor to ensure proper system operation and cost effectiveness. However, due to the large variety of membrane manufacturers, types, and characteristics, pilot-scale testing for selecting the most appropriate membrane can prove time consuming and costly. Presented in this paper is a short-term bench-scale membrane screening test (MST) that can be performed within four hours. MST rejection results are compared with results from more detailed bench- and pilot-scale tests to address scale-up issues. NF results obtained through MST experiments were compared with results gathered through longer, more detailed rapid bench-scale membrane tests (RBSMTs) (5 days of operation). The MST protocol provided comparable rejection results to the RBSMTs. Feed rejection results for the tested parameters differed by an average of 4%. Bulk rejection data from the MSTs predicted RBSMT membrane performance with an average difference of about 2%. In addition to these bench-scale evaluations, this paper compares MSTs with pilot-scale and RBMST results for three different waters. The data shows comparable bulk rejections with pilot and RBSMT results differing by less than 10%. Overall, the data show how results from short-term MSTs can be used to accurately predict membrane performance at the pilot-scale. By using MSTs for membrane selection, a utility can cost-effectively and accurately select membranes at an early stage based on constituent rejection. Hydraulic performance and fouling can then be evaluated the pilot-scale with a limited number of membranes. Includes 13 references, tables, figures.