&EPA EPA/600/R-12/644 | December 2012 | www.epa.gov/gateway/science United States Environmental Protection Agency Evaluation of Radial Flow Fluidized Filter (R3F) Followed by Microfiltration and Ultrafiltration Systems in Calimesa, California <. '# •' •» Office of Research and Development ------- EVALUATION OF RADIAL FLOW FLUIDIZED FILTER (R3F) FOLLOWED BY MICROFILTRATION AND ULTRAFILTRATON SYSTEMS IN CALIMESA, CALIFORNIA Craig L. Patterson, P.E. Water Supply and Water Resources Division National Risk Management Research Laboratory Cincinnati, Ohio 45268 Rajib Sinha, P.E. Shaw Environmental & Infrastructure (E&I), Inc. Cincinnati, Ohio 45212 This report was compiled in cooperation with Shaw Environmental & Infrastructure (E&I), Inc. Under EPA Contract EP-C-09-041, Work Assignment No. 0-03, Task 2 Submitted to National Risk Management Research Laboratory Office of Research and Development U.S. Environmental Protection Agency Cincinnati, OH 45268 and South Mesa Water Company 391 W. Avenue L Calimesa, CA 92320 September 18, 2012 ------- Notice The U.S. Environmental Protection Agency, through its Office of Research and Development, funded and managed, or partially funded and collaborated in, the research described herein. It has been subjected to the Agency's peer and administrative review and has been approved for publication. Any opinions expressed in this report are those of the author (s) and do not necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred. Any mention of trade names or commercial products does not constitute endorsement or recommendation for use. ------- Acknowledgements EPA acknowledges the contributions from George Jorritsma of the South Mesa Water Company. His efforts included coordinating and managing the research studies in Calimesa, California. EPA acknowledges the in-kind support and contributions of John Martin with R3f Filtration (Email: martinr3f@yahoo.com) and Kenneth W. Clark with NOK Corporation (Contact: Mr. Y. Inaba, Email: yosinaba@nok.co.jp, Tel: +81-3-3432-8415). Mr. Martin provided the R3f pilot- unit and Mr. Clark provided the MF and UF housings and membranes and participated in the research studies in Calimesa, CA. EPA Contributors Dr. John Ireland served as Project Officer on EPA Contract No. EP-C-04-034 and Mr. Craig L. Patterson P.E., served as the Work Assignment Manager for this research project. Mr. Jeffrey Q. Adams and Dr. Joel Allen performed technical reviews of the document. Mr. John Olszewski was the EPA Quality Assurance Managers, and was responsible for the quality assurance review of the document. Dr. Bruce Macler with U.S. EPA Region 9 in San Francisco, California reviewed the experimental plan for this research study. ------- Abstract U.S. EPA coordinated a field study with South Mesa Water Utility to look for treatment alternatives for California State Project Water in the small community of Calimesa, California. EPA evaluated the performance of a system comprised of Radial Flow Fluidized Filtration (R3f) followed by microfiltration (MF) and ultrafiltration (UF) through a series of turbidity and microbial challenges. The R3f and MF-UF system was challenged to remove turbidity, particles, and bacteria and viral surrogates (B. Subtilis, MS-2) from State Project Water. Turbidity, particle counts and Total Organic Carbon (TOC) were also monitored to obtain additional information on the performance and maintenance requirements of the systems. The report documents the results of these tests and also summarizes the critical observations on the maintenance of the R3f-MF-UF system and the use of turbidity and particle count as a surrogate for measuring the performance of the treatment plant. ------- Contents Notice ii Acknowledgements iii Abstract iv Contents v List of Figures vi List of Tables vii Acronyms and Abbreviation viii 1.0 Introduction 1 2.0 System Description, Operation and Testing Procedures 3 2.1 Field Test Site 3 2.2 Radial Flow Fluidized Filter (R3f) with MF and UF Membrane Polishing Filters 3 2.3 System Operation and Test Conditions 4 2.3.1 Microbial and Turbidity Challenges 4 2.3.2 Sampling and Analytical Procedures 5 3.0 Test Results 6 3.1 MS2 bacteriophage Test Results 6 3.2 B. subtilis Test Results 8 3.3 Turbidity, Particle Count and TOC Results 10 3.3.1 Turbidity Results 10 3.3.2 Particle Count Results 11 3.3.3 Pressure Buildup during the Challenge Tests 16 3.3.4 TOC Results 18 4.0 Conclusions 20 5.0 References 21 ------- List of Figures Figure 1. Conceptual Flow Diagram of the R3f System 3 Figure 2. Typical R3f Filter Setup 3 Figure 3. MF Filters Placed in a Shelco Multi-Cartridge Housing Unit 4 Figure 4. Correlation between Turbidity and 1.2 jim Size Particle 15 Figure 5. Correlation between Turbidity and 2 - 5 jim Size Particles 16 ------- List of Tables Table 1. Summary of the Sampling Strategy 5 Table 2. Analytical Methods For Physical and Microbial Parameters in this Study 5 Table 3. Summary of Challenge Tests Conducted on R3f-MF-UF System 6 Table 4. Results of MS2 bacteriophage Challenges on the R3f-MF-UF System 2 Table 5. Summary of R3f-MF-UF System Performance in Removing MS2 bacteriophage 8 Table 6. Results of B. subtilis Challenges on the R3f-MF-UF System 9 Table 7. Summary of the R3f-MF-UF System Performance in Removing B. subtilis 10 TableS. Turbidity Results for the R3f-MF-UF System 11 Table 9. Summary of the R3f-MF-UF System Performance in Removing Turbidity 11 Table 10. Results of 1.2 |im Size Particle Count Monitoring for the R3f-MF-UF System 13 Table 11. Results of 2 - 5 |im Size Particle Count Monitoring for the R3f-MF-UF System 14 Table 12. Summary of the R3f-MF-UF System Performance in Removing Particles 14 Table 13. Pressure Buildup in R3f System during the Challenge Tests 17 Table 14. Pressure Buildup in the MF and UF Units during the Challenge Tests 18 Table 15. Results of TOC Monitoring for the R3f-MF-UF System 19 ------- Acronyms and Abbreviations B. subtilis Bacillus subtilis E. coli Escherichia coli gpm gallons per minute L liter LRV log removal value LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule mg milligram mL milliliter MF microfiltration (microfilter) MS2 MS2 bacteriophage NTU Nephelometric Turbidity Unit psi pounds per square inch PSL polystyrene latex QA quality assurance QAPP Quality Assurance Proj ect Plan QC quality control R3f Radial Flow Fluidized Filter Shaw Shaw Environmental & Infrastructure (E&I), Inc. SMWC South Mesa Water Company Std. Dev. standard deviation T&E Test and Evaluation TOC total organic carbon UF ultrafiltration (ultrafilter) ug microgram jam microns ------- 1.0 Introduction The U.S. Environmental Protection Agency (EPA) evaluated the performance of a system comprised of Radial Flow Fluidized Filtration (R3f) followed by Microfiltration (MF) and Ultrafiltration (UF) through a series of turbidity and microbial challenges. The tests were performed at the EPA Test and Evaluation (T & E) Facility in Cincinnati, Ohio and at the South Mesa Water Company (SMWC) site in Calimesa, CA. These tests were performed by EPA and Shaw Environmental & Infrastructure (E&I), Inc. (Shaw) under Contract No. EP-C-09-041, Work Assignment No. 0-03 with the EPA and under a separate contract with SMWC. The R3f and MF-UF system was challenged with the following contaminants: • Escherichia coli (E. coli) as a representative organism for bacterium • Bacillus subtilis (B. subtilis) as a representative organism for aerobic spores • MS2 bactedophage, a surrogate for enteric virus • Polystyrene Latex (PSL) beads as a surrogate of Cryptosporidium • Cryptosporidium. One test was conducted on the R3f system only using Cryptosporidium to confirm the surrogacy equivalence of PSL beads. SMWC had previously evaluated the use of the R3f system as a treatment device for a surface water source. The R3f system had successfully produced water that met the required turbidity standard of 0.30 Nephelometric Turbidity Unit (NTU). However, California required SMWC to achieve at least a 1-log removal of viruses from the filtration step. Since the media size in the R3f system (33 jim) is too large for removal of bacteria and viruses, a MF and UF system downstream of the R3f system was considered for inclusion for microbial removal. EPA had evaluated the R3f system at the T&E facility and then had evaluated an R3f system followed by a MF and UF system at a field site in Ely, Minnesota and the results showed nearly complete removal of bacteria and viruses. The same treatment system was shipped to SMWC and tested in this study to confirm the removal of bacteria and viruses in this water matrix as well as to evaluate the likely cartridge backwash and changeout frequency. Turbidity, particle counts and Total Organic Carbon (TOC) were also monitored to obtain additional information on the performance and maintenance requirements of the systems. This report documents the results of these tests and also summarizes the critical observations on the maintenance of the R3f-MF-UF system and the use of turbidity and particle count as surrogate for measuring the performance of the treatment plant. ------- 2.0 System Description, Operation and Testing Procedures This section provides a summary description of the systems evaluated and the associated operation and testing procedures employed at the field site and at the T&E Facility in Cincinnati, OH. The test procedures employed are presented in the following EPA-endorsed Quality Assurance Project Plans (QAPPs): • EPA QA ID No. 627-Q-10-0 (EPA and Shaw, 2007a) • Amendment 1 to EPA QA ID No. 627-Q-10-0 (EPA and Shaw, 2008) 2.1 Field Test Site The field test site was located at SMWC in Calimesa, California. The tested units were installed in the test trailer owned and operated by SMWC. 2.2 Radial Flow Fluidized Filter (R3f) with MF and UF Membrane Polishing Filters The radial flow fluidized filtration (R3f) system utilizes radial flow (similar to the flow pattern of a cartridge filter) through use of non-bonded garnet media that can be fluidized and backwashed. The technology uses fine garnet media (33 microns) for depth filtration. The filter system housing consists of an upper chamber and a lower chamber. The upper chamber provides the necessary head for flow through the media and for backwashing. Water to be treated enters the system through the annular section at the top of the inner core. The lower chamber contains the media and water travels radially from the annular section through the media to the bottom inner core as a plug flow. The treated water then flows out of the system through the pipe connected at the bottom of the inner core. The plug in the inner core prevents any reverse flow. A typical R3f system is 150 mm (6 inches) in diameter and 1.8 m (6 feet) high as shown in Figure 1. The media can be fluidized and backwashed very quickly resulting in backwash volumes significantly lower than other technologies. The R3f system is suitable for small community potable water systems. It can also serve as pre-filter to membranes or disposable final filtration system. The R3f system is comprised of one lead and one lag unit in series. A typical R3f system setup is shown in Figure 2. ------- Filtration Cycle v- t • ' i \ & ^ J^^-™ • * * * I S~TL~ "***^__ _ ii : 1 • ! i i 1 1 ^^ C- ' -3 > * # «, * « j; H. ^^zl. : * t * 1 ""jL^ __ __*-!**' m i i i y 1 1 « ^^ ^ -.> • -:> ; > 1 ^ ^- Figure 1. Conceptual Flow Diagram of the R3f System Lag Filter Inlet Lead Filter Control Panel Effluent Figure 2. Typical R3f Filter Setup The UF and MF polishing filters were tested by placing them in a 7-cartridge Shelco Multifilter Cartridge Housing (Figure 3). Two Shelco housings were placed in series - the first housing contained MF cartridges while the second housing contained UF cartridges. The system was configured so that both Shelco housing units could be backwashed at the same frequency as the R3f system. ------- Figure 3. MF Filters Placed in a Shelco Multi-Cartridge Housing Unit 2.3 System Operation and Test Conditions 2.3.1 Microbial and Turbidity Challenges Tests were carried out to evaluate the performance of the R3f-MF-UF systems in removing bacteria and virus using the following test matrix, organisms or surrogates: • For bacteria removal, the systems were challenged with B. subtilis endospores, • For evaluation of virus removal potential, the systems were challenged with MS2 bacteriophage, a surrogate for enteric viruses. • For evaluation of turbidity removal performance, the systems were operated using surface water from the lake as the feed source with as-received turbidity levels. Samples were collected for turbidity and particle count during the microbial challenges. Additional samples were collected for TOC analysis for evaluating removal of organic matter. For B. subtilis and MS2 bacteriophage, a 1-mL stock suspension with an approximate concentration of 109 cells/surrogates per mL was mixed with 500 mL of 0.01% Tween 20 in a 1- L glass beaker. A sub-sample was collected to determine the actual concentration of the injection suspension. The 500 mL suspension and the rinseate were added into the influent stream of the system using a peristaltic pump. The total injection time was approximately 60 ------- minutes. The B. subtilis and MS2 bacteriophage stock solutions were obtained from the Raven (Omaha, NE) and BioVir (Benicia, CA) laboratories, respectively. 2.3.2 Sampling and Analytical Procedures Tables 1 and 2 summarize the sampling strategies and analytical procedures used in this study. The detailed sampling and analytical procedures are described in aforementioned QAPPs. Table 1. Summary of the Sampling Strategy Parameter MS2 bacteriophage B. subtilis Turbidity and Particle Counts TOC Sampling Frequency in Each Test At TO, T5, T15, T15 dup, T30 and T60 minutes after the start of the injection from the influent and effluent streams of different treatment units. At TO, T5, T15, T15 dup, T30 and T60 minutes after the start of the injection from the influent and effluent streams of different treatment units. At TO, T30 and T60 minutes after the start of the injection from the influent and effluent streams of different treatment units. At T30 minutes after the start of the injection from the influent and effluent streams of different treatment units. Table 2. Analytical Methods For Physical and Microbial Parameters in this Study Parameter Turbidity B. subtilis MS2 bacteriophage Particle Count TOC Analytical Method/Instrument T&E SOP 507 Using Hach Turbidity Meter, Model 21 OOP (EPA and Shaw, 2006a; Hach Co., 1991) T&E SOP 301 Using Heat Shock and Membrane Filtration (EPA and Shaw, 2006b; Rice et al. 1994) T&E SOP 302 Based on the EPA method 1602 (EPA and Shaw, 2007b; EPA, 2001) HIAC Royco particle count analyzer (Pharmspec Version 1 .4) Phoenix TOC Analyzer, Model 8000 (EPA, 1999) ------- 3.0 Test Results This section summarizes the results of tests conducted on the R3f-MF-UF systems to evaluate the performance in removing different microbiological contaminants, turbidity and particle counts. Table 3 summarizes the challenge tests conducted on the R3f-MF-UF systems at the field site Table 3. Summary of Challenge Tests Conducted on R3f-MF-UF System Date 2/24/10 2/24/10 2/24/10 2/24/10 2/25/10 2/25/10 Test ID MS2 bacteriophage Test 1 MS2 bacteriophage Test 2 MS2 bacteriophage Test 3 B. subtilis Test 1 B. subtilis Test 2 B. subtilis Test 3 System Configuration R3f Lead + R3f Lag + MF + UF R3f Lead + R3f Lag + MF + UF R3f Lead + R3f Lag + MF + UF R3f Lead + R3f Lag + MF + UF R3f Lead + R3f Lag + MF + UF R3f Lead + R3f Lag + MF + UF 3.1 MS2 bacteriophage Test Results Table 4 presents the results of the MS2 bacteriophage challenge tests conducted on the R3f-MF- UF system. The log removal value (LRV) for each system component is summarized in Table 5. The average influent concentration in the three challenge tests ranged from 2.4 x 103/100 mL (3.38 log) to 1.2 x 103/100 mL (4.08 log). The complete system (R3f-MF-UF) achieved complete removal of MS2 bacteriophage with the UF system responsible for the majority of the removal. The overall LRV of the system in removing MS2 bacteriophage is at least >4.10 based on the highest influent concentrations. The R3f system achieved an average LRV of 0.65 in the three tests; however, the performance of the lag unit deteriorated after the first test indicating continuous desorption of MS2 bacteriophage from the R3f system. The MF system achieved a LRV of 0.88 of MS2 bacteriophage in the first test but in subsequent tests the LRV deteriorated indicating breakthrough of MS2 bacteriophage. No breakthrough of MS2 bacteriophage was observed in the UF system. These test results are similar to the performance demonstrated by the R3f-MF-UF in tests conducted at Ely, Minnesota where the system showed an overall LRV of 3.70 (EPA and Shaw, 2009a). Negligible removal of MS2 bacteriophage was achieved by the R3f system alone in tests conducted at the T&E Facility (EPA and Shaw, 2009b). ------- Table 4. Results of MS2 bacteriophage Challenges on the R3f-MF-UF System Sample ID MS2 bacteriophage Conc./lOO mL Influent Effluent RSfLead Effluent RSfLag Effluent MF Effluent UF Test 1: 02/24/10 TO T5 T15 T15 Dup T30 T60d Mean [Std. Dev.] ± CIb N/Aa 3.7 x 103 1.2 x 104 8.0 x 103 1.6 x 104 28 9.9 x 103 [5.3 x 103] ± 5.2 x 103 N/Aa 1.1 x 103 6.0 x 103 9.0 x 103 1.2 x 104 181 7.0 x 103 [4.6 x 103] ± 4.6 x 103 N/Aa N/Aa 7.7 x 103 4.0 x 102 5.0 x 102 N/Aa 2.9 x 103 [7.9 x 104] ± 3.7 x 103 N/Aa N/Aa 370 400 370 N/Aa 3.8 x 102 [14] ± 4.3 x 102 N/Aa N/Aa N/De 1 N/De N/Aa N/De Test 2: 02/24/10 T0d T5 T15 T15 Dup T30 T60d Mean [Std. Dev.] ± CIb 8 1.5 x IQ4 1.5 x IQ4 9.4 x 103 9.1 x IQ3 14 1.2 x 104 [3.3 x 103] ± 3.3 x 103 0 130C 1.4 x 103 1.5 x IQ3 1.6 x 103 530 1.5 x 103 [100] ± 97 0 oc 1.7 x 103 1.5 x IQ3 1.0 x 103 48 1.4 x 103 [3.6 x 102] ± 3.2 x 102 0 60C 8.2 x 103 1.5 x IQ3 1.7 x 103 15 3.8 x 103 [3.1 x 103] ± 4.3 x 103 N/De N/De N/De N/De N/De N/De N/De Test 3: 02/24/10 T0d T5d T15 T15 Dup T30 T60d Mean [Std. Dev.] ± CIb 0 9.5 x IQ2 1.1 x IQ3 1.2 x 103 4.8 x IQ3 70 2.4 x 103 [2.1 x 103] ± 2.1 x 103 0 17 8.0 x 102 6.0 x 102 9.5 x IQ2 10 7.8 x 102 [1.8 x 102] ± 1.7 x 102 0 7 6.0 x 102 7.0 x 102 1.3 x IQ3 43 8.7 x 102 [2.4 x 102] ± 3.3 x 102 0 2 7.3 x IQ2 9.2 x 102 2.3 x IQ3 41 9.9 x 102 [2.4 x 102] ± 2.7 x 102 N/De N/De N/De N/De N/De N/De N/De 1 Not available, plaques did not grow probably due to either slightly overgrown host E. coli or problems in media caused by malfunction of autoclave. b Confidence Interval at 95% significance 0 Not considered for performance evaluation due to inconsistent value d Not considered for performance evaluation due lower influent concentration e Not detected, no plaques were observed on any of the 10 plates for the sample ------- Table 5. Summary of the R3f-MF-UF System Performance in Removing MS2 bacteriophage Test ID 1 2 3 Contribution to Log Removal Value (LRV) RSfLead 0.15 0.90 0.49 RSfLag 0.38 0.03 0.00 MF 0.88 0.00 0.00 UF Complete Removal (>2.58a) Complete Removal (>3.58a) Complete Removal (>3.00a) Overall Complete Removal (>4.00b) Complete Removal (>4.10b) Complete Removal (>3.38b) 1 LRV based on the influent concentration of the LTF unit. 3 LRV based on the influent concentration of the overall system 3.2 B. subtilis Test Results Table 6 presents the results of the B. subtilis challenges conducted on the R3f-MF-UF system. The log removal value (LRV) for each system component is summarized in Table 7. The average influent concentration in the three challenge tests ranged from 7.4 x 105/100 mL (5.86 log) to 8.6 x 105/100 mL (5.93 log) and the whole system achieved complete removal of B. subtilis with the MF system responsible for majority of the removal. The overall LRV of the system in removing MS2 bacteriophage is at least >5.93 based on the highest influent concentrations. The R3f system achieved an average log removal value of 1.0 in the three tests; however, the performance of the R3f system deteriorated after the first test indicating continuous desorption of B. subtilis from the R3f system. No breakthrough of B. subtilis was observed in the MF system. These test results are similar to the performance demonstrated by the R3f-MF-UF system in tests conducted at Ely, Minnesota where the whole system achieved complete removal of B. subtilis. The R3f system alone achieved a 1.0 log removal (EPA and Shaw, 2009a). The R3f system achieved an average log removal value of 0.44 in removing B. subtilis in tests conducted at the T&E Facility (EPA & Shaw, 2009b). ------- Table 6. Results of B. subtilis Challenges on the R3f-MF-UF System Sample ID B. subtilis Conc./lOO mL Influent Effluent RSfLead Effluent RSfLag Effluent MF Effluent UF Test 1: 02/24/10 T0d T5 T15 T15 Dup T30 T60 Mean [Std. Dev.] ± CIb 0 8.1 x 105 7.8 x 105 7.4 x 105 6.3 x 105 N/Aa 7.4 x 10s [7.9 x 104] ± 7.7 x 104 0 1.2 x 105 2.6 x 105 3.2 x 105 3.0 x 105 N/Aa 2.5 x 10s [9.0 x 104]± 8.8 x 104 0 9.0 x 103 1.7 x 104 1.7 x 104 5.4 x 104 1.4 x 104 2.4 x 104 [2.0 x 104] ± 1.8 x 104 N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De Test 2: 02/25/10 T0d T5 T15 T15 Dup T30 T60 Mean [Std. Dev.] ± CIb N/Aa 8.6 x 105 8.8 x 105 9.0 x 105 8.1 x 105 5.0 x 103c 8.6 x 10s [3.9 x 104] ± 3.8 x 104 8.6 x 103 1.1 x 105 4.8 x 105 4.9 x 105 5.6 x 105 1.6 x 104 4.1 x 10s [2.0 x 10s] ± 2.0 x 10s 8.0 x 103 .0 x 103 .0 x 105 .1 x 105 .1 x 105 .0 x 104 1.1 x 10s [5.8 x 103] ± 5.1 x 103 N/De 1.6 x 103c N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De Test 3: 02/25/10 T0d T5 T15 T15 Dup T30 T60 Mean [Std. Dev.] ± CIb 600 6.9 x 105 8.6 x 105 7.2 x 105 9.8 x 105 1.2 x 105 7.9 x 10s [1.6 x 10s] ± 1.6 x 10s 500 2.1 x 105 5.2 x 105 4.8 x 105 5.2 x 105 9.1 x IQ4 4.4 x 10s [1.5 x 10s] ± 1.5 x 10s 28 N/Aa 1.5 x 105 1.8 x 105 1.6 x 105 3.9 x 104 1.6 x 10s [1.5 x 104] ± 1.4 x 104 N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De N/De 1 Not available due to under-dilution of samples b Confidence Interval at 95% significance 0 Not considered for performance evaluation due to inconsistent value d Not considered for performance evaluation due to lower influent concentration e Not detected, no plaques were observed on any of the 10 plates for the sample ------- Table 7. Summary of the R3f-MF-UF System Performance in Removing B. subtilis Test ID 1 2 3 Log Removal Value (LRV) RSfLead 0.47 0.32 0.25 RSfLag 1.0 0.57 0.43 MF Complete Removal (>4.38a) Complete Removal (>5.04a) Complete Removal (>5.20a) UF Not Applicable0 Not Applicable0 Not Applicable0 Overall Complete Removal (>5.86b) Complete Removal (>5.93b) Complete Removal (>5.90b) a LRV based on the influent concentration of the LTF unit. b LRV based on the influent concentration of the overall system 0 Not applicable as complete removal was achieved by the MF unit 3.3 Turbidity, Particle Count and TOC Results 3.3.1 Turbidity Results Table 8 presents the results of turbidity monitoring during the microbial challenges. The performance of each system component in removing turbidity is summarized in Table 9. The average influent turbidity was 0.60 NTU and the R3f-MF-UF system achieved an average effluent turbidity of 0.13 NTU. This effluent turbidity satisfies the Long Term 2 Enhanced Surface Water Treatment Regulation (LT2ESWTR) requirement of effluent turbidity < 0.30 NTU (EPA, 2006). The R3f system alone produced water with an effluent turbidity of 0.27 NTU which also satisfies the LT2ESWTR requirements. The R3f-MF-UF system removed an average of 78.3% of turbidity of which 55% removal was contributed by the R3f system. The MF and UF systems contributed 23.3 % turbidity removal. These test results are similar to the performance demonstrated by the R3f-MF-UF system in tests conducted at Ely, Minnesota where the final effluent turbidity for a feed water of 1.88 NTU was 0.10 NTU (EPA and Shaw, 2009a). The R3f system achieved effluent turbidity of 0.61 NTU from an influent turbidity of 4.30 NTU in tests conducted at the T&E Facility without the use of any coagulant (EPA and Shaw 2009b). 10 ------- Table 8. Turbidity Results for the R3f-MF-UF System Date & Time 2/24/10; 10:00 2/24/10; 10:30 2/24/10; 11:00* 2/24/10; 13:15 2/24/10; 13:45 2/24/10; 14:15* 2/24/10; 15:00 2/24/10; 15:30 2/24/10; 16:00 * 2/24/10; 17:55 2/24/10; 18:25 2/24/10; 18:55 * 2/25/10; 08:45 2/25/10; 09: 15 2/25/10; 09:45 ** 2/25/10; 11:40 2/25/10; 12:20 2/25/10; 12:40*** Mean [Std. Dev.] ±CIb Turbidity (NTU) Influent 0.52 0.60 0.70 0.93 0.90 0.70 0.52 0.84 0.70 0.42 0.42 0.58 0.52 0.61 0.54 0.39 0.45 0.54 0.60 [0.16] ± 0.07 Effluent RSfLead 0.70a 0.40 0.33 0.31 0.36 0.30 0.36 0.36 0.40 0.24 0.27 0.29 0.31 0.28 0.31 0.29 0.29 0.34 0.32 [0.04] ± 0.02 Effluent RSfLag 0.42 0.28 0.64a 0.28 0.30 0.24 0.32 0.25 0.30 0.24 0.19 0.22 0.25 0.22 0.24 0.22 0.40 0.26 0.27 [0.06] ± 0.03 Effluent MF 0.20 0.14 0.14 0.20 0.16 0.14 0.15 0.13 0.13 0.13 0.15 0.14 0.14 0.18 0.15 0.19 0.18 0.18 0.16 [0.02] ± 0.01 Effluent UF 0.15 0.70a 0.14 0.16 0.14 0.13 0.12 0.12 0.12 0.12 0.12 0.14 0.11 0.11 0.10 0.14 0.10 0.14 0.13 [0.02] ± 0.01 Blank 0.09 0.09 0.09 0.10 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.10 0.10 0.10 0.11 0.11 0.11 0.10 [0.01] ± 0.003 a Inconsistent data b Confidence Interval at 95% Significance. * R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush. ** Both MF and UF units was replaced. *** Test completed; MF and UF systems continued to run. Table 9. Summary of the R3f-MF-UF System Performance in Removing Turbidity Test ID Whole Test Period Contribution to % Removal RSfLead 46.7 RSfLag 8.3 MF 18.3 UF 5.0 Overall 78.3 3.3.2 Particle Count Results The particle counts of the influent and effluent samples were monitored for obtaining secondary information on the performance of the system in removing contaminants. Particle count data was also used to evaluate the potential of R3f system in reducing particle load to the MF and UF systems. Table 10 shows the particle count results for 1.2 jim size particle which is the approximate equivalent size of B. subtilis. Table 11 describes the particle count results for the 2 |im - 5 |im size range particles which are the approximate equivalent size of Cryptosporidium. Table 12 summarizes the performance of the R3f-MF-UF system in removing particles of the 11 ------- aforementioned sizes. The average influent concentration of the 1.2 jim size particle was 2198/mL and the final effluent concentration of the 1.2 jim size particle achieved by the R3f- MF-UF system was 173/mL. The R3f system contributed noticeably to the reduction of 1.2 jim size particle. The overall performance in removing 1.2 jim size particles by the whole system was 92.2% of which 71.0% reduction was contributed by the R3f system. The average influent concentration of the 2 - 5 jim size particles is 247/mL and the final effluent concentration of the 2-5 |im size particle achieved by the R3f-MF-UF system was 32/mL. The R3f system contributed noticeably to the reduction of 2-5 jim size particles. The overall performance in removing 2-5 jim size particles by the whole system is 87.0% of which 60.0% reduction was contributed by the R3f system. For the R3f system, most of the reduction was contributed by the R3f lead filter. Most of the additional reduction was contributed by the MF system; no additional removal was contributed by the UF system. Particle counts were not monitored for the system during the tests conducted at Ely, Minnesota. 12 ------- Table 10. Results of 1.2 |im Size Particle Count Monitoring for the R3f-MF-UF System Date & Time 2/24/10; 10:00 2/24/10; 10:30 2/24/10; 11:00* 2/24/10; 13:15 2/24/10; 13:45 2/24/10; 14:15* 2/24/10; 15:00 2/24/10; 15:30 2/24/10; 16:00 * 2/24/10; 17:55 2/24/10; 18:25 2/24/10; 18:55 * 2/25/10; 08:45 2/25/10; 09: 15 2/25/10; 09:45 ** 2/25/10; 11:40 2/25/10; 12:20 2/25/10; 12:40*** Mean [Std. Dev.] ±CIb Turbidity (NTU) Influent 2032 1509 2748 2684 4155 4576 3334 5050 5673 630 813 1436 1298 753 563 418 550 1342 2198 [1701] ±786 Effluent RSfLead 8101a 586 408 702 922 966 1706 1395 1156 217 982 138 889 768 184 95 419 225 692 [449] ± 223 Effluent RSfLag 3149a 522 6746a 669 2993a 411 1123 606 1709 203 553 300 192 672 196 900 1057 368 632 [420] ± 213 Effluent MF 805a 299 112 198 54 67 58 84 92 68 226 40 50 96 65 215 250 239 130 [86] ± 40 Effluent UF 1090a 3974a 4079a 534a 172 92 102 78 190 52 49 114 70 133 124 217 750 283 173 [179] ± 88 Blank 4.2 4.8 3.2 1.8 2.0 1.0 3.3 4.1 2.1 0.11 0.11 2.1 3.8 1.7 2.3 3.0 2.7 2.7 2.5 ± 0.60 a Inconsistent data, probably due to backflush of the R3f system and inappropriate startup of the MF and UF systems. Not considered for performance evaluation. b Confidence Interval at 95% significance * R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush. ** Both MF and UF units was replaced. *** Test completed; MF and UF systems were continued to run. 13 ------- Table 11. Results of 2 - 5 |im Size Particle Count Monitoring for the R3f-MF-UF System Date & Time 2/24/10; 10:00 2/24/10; 10:30 2/24/10; 11:00* 2/24/10; 13:15 2/24/10; 13:45 2/24/10; 14:15* 2/24/10; 15:00 2/24/10; 15:30 2/24/10; 16:00 * 2/24/10; 17:55 2/24/10; 18:25 2/24/10; 18:55 * 2/25/10; 08:45 2/25/10; 09: 15 2/25/10; 09:45 ** 2/25/10; 11:40 2/25/10; 12:10 2/25/10; 12:40*** Mean [Std. Dev.] ±CIb Turbidity (NTU) Influent 372 168 353 247 494 416 317 447 334 113 113 375 197 94 63 88 61 209 247 [143] ± 66 Effluent RSfLead 1849a 166 50 63 35 40 360 190 112 46 104 15 405 119 21 16 69 36 108 [115] ± 54 Effluent RSfLag 957a 85 1832a 60 689a 38 96 54 271 19 81 48 30 94 21 155 272 41 91 [81] ±41 Effluent MF 247a 36 39 49 10 16 14 11 21 12 30 15 9 13 12 49 34 28 23 [14] ± 6 Effluent UF 214a 1774a 1841a 129a 41 20 23 20 47 13 11 23 12 16 21 36 101 60 32 [25] ± 12 Blank 4.2 4.8 3.2 1.8 2.0 1.0 3.3 4.1 2.1 0.11 0.11 2.1 3.8 1.7 2.3 3.0 2.7 2.7 2.5 ± 0.60 a Inconsistent data, probably due to backflush of the R3f system and inappropriate startup of the MF and UF systems. Not considered for performance evaluation. b Confidence Interval at 95% significance * R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush. ** Both MF and UF units was replaced. *** Test completed; MF and UF systems were continued to run. Table 12. Summary of the R3f-MF-UF System Performance in Removing Particles Particle Size 1.2 |im 2-5 |im Contribution to % Removal RSfLead 68.5 56.2 RSfLag 2.7 7.0 MF 22.9 27.5 UF None None Overall 92.2 87.0 Although the R3f-MF-UF system achieved complete removal of B. subtilis and MS2 bacteriophage, a small number of 1.2 jam and 2-5 jim size particles were detected in the final effluent of the system. This is probably due to either the limitations of the particle size analyzer or characteristics of the non-biological particles. However, a good correlation was observed 14 ------- between the turbidity and the 1.20 jim particle count data (R : 9683) (Figure 4) and the 2-5 jim size particle count data (R2: 0.9915) (Figure 5). This implied that the particle count data may be used as secondary information on the performance of the system in removing turbidity and biological contaminants. y=4319.5x-489.98 R2 = 0.9683 •5T 250° N (7) g 2000 o 1 ^ 1500 3 O O (0 0- 1000 500 0.1 0.2 0.3 0.4 Turbidity (NTU) 0.5 0.6 0.7 Figure 4. Correlation between Turbidity and 1.2 jim Size Particle 15 ------- •5T 30° N V) y = 479.82x-42.026 R2 = 0.9915 o £ Cs 250 200 150 100 53 i o O 50 (0 0- 0.1 0.2 0.3 0.4 0.5 Turbidity (NTU) 0.6 0.7 Figure 5. Correlation between Turbidity and 2-5 jim Size Particles 3.3.3 Pressure Buildup during the Challenge Tests Table 13 shows the pressure buildup in the R3f system during the challenge tests. The initial differential pressure in the R3f lead filter was 9 psi and it did not increase during the challenge tests. The initial differential pressure in the R3f lag filter was 0 psi indicating that this unit was experiencing some back pressure due to the addition of MF and UF systems at the downstream location. However, the differential pressure in the R3f lag filter did not increase during the challenge tests. Although the differential pressure did not increase during the challenge tests, the R3f systems were flushed at the end of each test. Table 14 shows the pressure buildup in the MF and UF units during the challenge tests. The initial differential pressure in the MF unit of 2 psi did not increase during the challenge tests. The initial differential pressure in the UF unit of 12 psi increased to 20 psi at the end of the second B. subtilis test. Both the MF and the UF units were replaced before the start of the next test. The MF and the UF were stopped during the flush of the R3f units. The flow rate remained at 2.5 gpm during the challenge tests. 16 ------- Table 13. Pressure Buildup in R3f System during the Challenge Tests Date & Time 2/24/10; 10:00 2/24/10; 10:15 2/24/10; 10:30 2/24/10; 11:00* 2/24/10; 13:15 2/24/10; 13:45 2/24/10; 14:00 2/24/10; 14:15* 2/24/10; 15:00 2/24/10; 15:30 2/24/10; 16:00 2/24/10; 16:00 * 2/24/10; 17:55 2/24/10; 18:10 2/24/10; 18:25 2/24/10; 18:55 * 2/25/10; 08:45 2/25/10; 09:00 2/25/10; 09: 15 2/25/10; 09:45 ** 2/25/10; 11:40 2/25/10; 11:55 2/25/10; 12:10 2/25/10; 12:40*** Pressure (psi) In 59 59 58 58 61 61 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 Outla 50 50 50 51 52 52 54 53 53 53 53 53 52 52 52 52 53 53 53 53 53 53 53 53 Out 2b 52 51 52 52 52 54 55 54 54 54 54 55 54 54 54 54 54 54 55 54 54 55 54 55 Differential Pressure (psi) API3 9 9 8 7 9 9 6 7 7 7 7 7 8 8 8 8 7 7 7 7 7 7 7 7 AP2b 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 * R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush. ** Both MF and UF units was replaced. *** Test completed; MF and UF systems were continued to run aR3f Lead Filter bR3f Las Filter 17 ------- Table 14. Pressure Buildup in the MF and UF Units during the Challenge Tests Date & Time 2/24/10; 10:00 2/24/10; 10:15 2/24/10; 10:30 2/24/10; 11:00* 2/24/10; 13:15 2/24/10; 13:45 2/24/10; 14:00 2/24/10; 14:15* 2/24/10; 15:00 2/24/10; 15:30 2/24/10; 16:00 2/24/10; 16:00 * 2/24/10; 17:55 2/24/10; 18:10 2/24/10; 18:25 2/24/10; 18:55 * 2/25/10; 08:45 2/25/10; 09:00 2/25/10; 09: 15 2/25/10; 09:45 ** 2/25/10; 11:40 2/25/10; 11:55 2/25/10; 12:10 2/25/10; 12:40*** Pressure (psi) In 48 47 48 48 50 50 50 50 50 50 50 50 49 49 50 50 50 50 50 50 50 50 50 50 MFOut 46 45 47 46 48 48 48 48 48 48 48 48 46 46 46 47 48 48 48 48 47 48 48 48 UFOUT 34 33 34 33 32 34 34 33 34 33 33 32 31 30 29 29 29 29 28 28 38 36 35 35 Differential Pressure (psi) APMF 2 2 1 2 2 2 2 2 2 2 2 2 3 3 3 3 2 2 2 2 3 2 2 2 APUF 12 12 13 13 16 14 14 15 14 15 15 16 15 16 17 18 19 19 20 20 9 12 13 13 * R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush. ** Both MF and UF units was replaced. *** Test completed; MF and UF systems were continued to run 3.3.4 TOC Results Table 15 describes the results of TOC monitoring during the microbial challenges. The average influent concentration of TOC was 3.85 mg/L. No reduction of TOC was achieved the R3f-MF- UF system. 18 ------- Table 15. Results of TOC Monitoring for the R3f-MF-UF System Test ID MS2 Test 1 MS2 Test 2 MS2 Test 3 BS Test 1 BSTest2 Mean [Std. Dev.] ± CIb TOC (mg/L) IN 3.63 3.59 3.65 4.38 3.98 3.85 [0.34] ± 0.30 Effluent RSfLead 3.75 3.67 3.71 4.45 N/Aa 3.90 [0.37] ± 0.33 Effluent RSfLag 3.76 3.75 3.74 4.47 3.92 3.93 [0.31] ± 0.27 Effluent MF 3.71 3.73 3.77 4.43 3.92 3.91 [0.30] ± 0.26 Effluent UF 3.64 3.64 3.75 4.29 4.02 3.87 [0.28] ± 0.25 1 Not Available, sample was not collected. 3 Confidence Interval at 95% Significance. 19 ------- 4.0 Conclusions The R3f-MF- UF system achieved complete removal of MS2 bacteriophage with most of the removal achieved by the UF system. The overall performance of the whole system was similar to that observed in previous experiments conducted on the same system at Ely, Minnesota. The R3f-MF- UF system achieved complete removal of B. subtilis with most of the removal achieved by the MF system. The overall performance of the whole system was similar to that observed in previous experiments conducted on the same system at Ely, Minnesota. The performance of the R3f system alone in removing B. subtilis was similar to that observed in tests conducted at the T&E Facility in Cincinnati, Ohio. For an influent turbidity of 0.60 NTU, the R3f-MF-UF system achieved an average effluent turbidity of 0.13 NTU which satisfied the LT2ESWTR requirement of effluent turbidity < 0.30 NTU (EPA, 2006). The R3f system contributed considerably in removing turbidity. The overall performance of the whole system was similar to that observed in experiments conducted on the same system at Ely, Minnesota. The performance of the R3f system alone in removing turbidity was better than that observed in tests conducted at Ely, Minnesota and at the T&E Facility in Cincinnati, Ohio. The R3f-MF-UF system demonstrated excellent removal of bacteria size (1.2 jim) and Cryptosporidium size (2-5 jim) particles. The overall performance in removing 1.2 |im size particles by the whole system was 92.2% of which 71.0% reduction was contributed by the R3f system. . The overall performance in removing 2-5 |im size particles by the whole system is 87.0% of which 60.0% reduction was contributed by the R3f system. The R3f system demonstrated good potential for use as a pretreatment unit to reduce particle load on the MF and UF systems. Although the R3f-MF-UF system achieved complete removal of B. subtilis and MS2 bacteriophage, a small number of 1.2 jim and 2 - 5 jim size particles were detected in the final effluent of the system. A good correlation was observed between the turbidity and the particle counts. Therefore, the particle count data may possibly be used as secondary information in assessing the performance of the system in removing turbidity and biological contaminants. Although no pressure buildup was observed in the R3f system during the challenge tests, the R3f units were backflushed at the end of each test. The MF did not experience any additional pressure buildup during the challenge tests. There was an 8 psi pressure buildup in the UF system during the challenge tests. The R3f-MF-UF system did not reduce TOC from the source water. The R3f system was not capable of reducing TOC in tests conducted at the T&E Facility. 20 ------- 5.0 References EPA (1999). "Method 415.2: Total Organic Carbon in Water (UV Promoted, Persulfate Oxidation)", Washington, D.C. EPA (2001). "Method 1602 for Detection and Enumeration of MS2 Bacteriophage in Drinking Water", Washington, D.C. EPA (2006). The Long-Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) Implementation Guidance, EPA-816-06-005, Washington, D.C. EPA and Shaw (2006a). T&E SOP 507: Turbidity Analysis, EPA T&E Facility, Cincinnati, OH. EPA and Shaw (2006b). T&E SOP 301: Enumeration of B. subtilis Using Membrane Filtration Method, EPA T&E Facility, Cincinnati, Ohio. EPA and Shaw (2007a). "Quality Assurance Project Plan for Glass Bead R3f and Multimedia Systems", EPA QA ID: 627-Q-10-0, EPA T&E Facility, Cincinnati, Ohio. EPA and Shaw (2007b). T&E SOP 302: Enumeration of MS2 Bacteriophage Based on EPA Method 1622, EPA T&E Facility, Cincinnati, Ohio. EPA and Shaw (2008). "Addendum to Quality Assurance Project Plan for Glass Bead R3f and Multimedia Systems", EPA QA ID: 627-Q-10-1, EPA T&E Facility, Cincinnati, Ohio. EPA and Shaw (2009a). "Final Report: Evaluation of Small System Filtration Technologies for the Treatment of Color, Disinfection Byproducts and Microbiological Contaminants in Surface Waters", EPA T&E Facility, Cincinnati, Ohio. EPA and Shaw (2009b). "Final Report: Comparative Evaluation of Glass Bead/Garnet R3f and Multimedia Filtration Systems", EPA T&E Facility, Cincinnati, Ohio. Hach Co. (1991). Hach 21 OOP Portable Turbidity Meter Procedure Manual, Denver, Colorado. Rice, E.W., Fox, K.R., Miltner, R.J., Lytle, D.A. and Johnson, J.H. (1994). "A Microbiological Surrogate for Evaluating Treatment Efficiency." In Proceedings of AWWA Water Quality Technology Conference (WQTC, San Francisco, California. 21 ------- |